byp4a55 professor6T9

Linux server.ecdc.com.eg 5.14.0-611.54.3.el9_7.x86_64 #1 SMP PREEMPT_DYNAMIC Thu May 7 16:31:24 EDT 2026 x86_64
Apache
50.6.228.173

File Name : ChangeLog

2020-11-14  Akim Demaille  <akim.demaille@gmail.com>

version 3.7.4
	* NEWS: Record release date.

2020-11-13  Akim Demaille  <akim.demaille@gmail.com>

c++: shorten the assertions that check whether tokens are correct
	Before:

YY_ASSERT (tok == token::YYEOF || tok == token::YYerror || tok == token::YYUNDEF || tok == 120 || tok == 49 || tok == 50 || tok == 51 || tok == 52 || tok == 53 || tok == 54 || tok == 55 || tok == 56 || tok == 57 || tok == 97 || tok == 98);

After:

YY_ASSERT (tok == token::YYEOF
	               || (token::YYerror <= tok && tok <= token::YYUNDEF)
	               || tok == 120
	               || (49 <= tok && tok <= 57)
	               || (97 <= tok && tok <= 98));

Clauses are now also wrapped on several lines.  This is nicer to read
	and diff, but also avoids pushing Visual C++ to its arbitrary
	limits (640K and lines of 16380 bytes ought to be enough for anybody,
	otherwise make an C2026 error).

The useless parens are there for the dummy warnings about
	precedence (in the future, will we also have to put parens in
	`1+2*3`?).

* data/skeletons/variant.hh (_b4_filter_tokens, b4_tok_in, b4_tok_in):
	New.
	(_b4_token_constructor_define): Use them.

2020-11-13  Akim Demaille  <akim.demaille@gmail.com>

c++: don't glue functions together
	* data/skeletons/bison.m4 (b4_type_foreach): Accept a separator.
	* data/skeletons/c++.m4: Use it.
	And fix an incorrect comment.

2020-11-13  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: YY_ASSERT should use api.prefix
	Working on the previous commit I realized that YY_ASSERT was used in
	the generated headers, so must follow api.prefix to avoid clashes when
	multiple C++ parser with variants are used.

Actually many more macros should obey api.prefix (YY_CPLUSPLUS,
	YY_COPY, etc.).  There was no complaint so far, so it's not urgent
	enough for 3.7.4, but it should be addressed in 3.8.

* data/skeletons/variant.hh (b4_assert): New.
	Use it.
	* tests/local.at (AT_YYLEX_RETURN): Fix.
	* tests/headers.at: Make sure variant-based C++ parsers are checked
	too.
	This test did find that YY_ASSERT escaped renaming (before the fix in
	this commit).

2020-11-13  Akim Demaille  <akim.demaille@gmail.com>

c++: don't use YY_ASSERT at all if parse.assert is disabled
	In some extreme situations (about 800 tokens), we generate a
	single-line assertion long enough for Visual C++ to discard the end of
	the line, thus falling into parse ends for the missing `);`.  On a
	shorter example:

YY_ASSERT (tok == token::TOK_YYEOF || tok == token::TOK_YYerror || tok == token::TOK_YYUNDEF || tok == token::TOK_ASSIGN || tok == token::TOK_MINUS || tok == token::TOK_PLUS || tok == token::TOK_STAR || tok == token::TOK_SLASH || tok == token::TOK_LPAREN || tok == token::TOK_RPAREN);

Whether NDEBUG is used or not is irrelevant, the parser dies anyway.

Reported by Jot Dot <jotdot@shaw.ca>.
	https://lists.gnu.org/r/bug-bison/2020-11/msg00002.html

We should avoid emitting lines so long.

We probably should also use a range-based assertion (with extraneous
	parens to pacify fascist compilers):

YY_ASSERT ((token::TOK_YYEOF <= tok && tok <= token::TOK_YYUNDEF)
	               || (token::TOK_ASSIGN <= tok && ...)

But anyway, we should simply not emit this assertion at all when not
	asked for.

* data/skeletons/variant.hh: Do not define, nor use, YY_ASSERT when it
	is not enabled.

2020-11-13  Akim Demaille  <akim.demaille@gmail.com>

c++: style: follow the Bison m4 quoting pattern
	* data/skeletons/variant.hh: here.

2020-11-11  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: provide the Bison version as an integral macro
	Suggested by Balazs Scheidler.
	https://github.com/akimd/bison/issues/55

* src/muscle-tab.c (muscle_init): Move/rename `b4_version` to/as...
	* src/output.c (prepare): `b4_version_string`.
	Also define `b4_version`.
	* data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/d.m4,
	* data/skeletons/java.m4: Adjust.
	* doc/bison.texi: Document it.

2020-11-11  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-11-11  Akim Demaille  <akim.demaille@gmail.com>

style: make conversion of version string to int public
	* src/parse-gram.y (str_to_version): Rename as/move to...
	* src/strversion.h, src/strversion.c (strversion_to_int): these new
	files.

2020-11-11  Akim Demaille  <akim.demaille@gmail.com>

%require: accept version numbers with three parts ("3.7.4")
	* src/parse-gram.y (str_to_version): Support three parts.
	* data/skeletons/location.cc, data/skeletons/stack.hh:
	Adjust.

2020-11-11  Todd C. Miller  <Todd.Miller@sudo.ws>

yacc.c: fix #definition of YYEMPTY
	When generating a C parser, YYEMPTY is present in enum yytokentype but
	there is no corresponding #define like there is for the other values.
	There is a special case for YYEMPTY in b4_token_enums but no
	corresponding case in b4_token_defines.

* data/skeletons/c.m4 (b4_token_defines): Do define YYEMPTY.

2020-11-10  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-11-01  Akim Demaille  <akim.demaille@gmail.com>

doc: fix incorrect section title
	Reported by Gaurav Singh <gaurav.singh199709@yahoo.com>.
	https://lists.gnu.org/r/bug-bison/2020-11/msg00000.html

* doc/bison.texi (Rpcalc Expr): Rename as...
	(Rpcalc Exp): this, as the nterm is named 'exp'.

2020-10-28  Nick Gasson  <nick@nickg.me.uk>

doc: minor grammar fixes in counterexamples section
	* doc/bison.texi: Minor fixes in counterexamples section.

2020-10-14  Akim Demaille  <akim.demaille@gmail.com>

doc: fix typo
	* README: here.

2020-10-13  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-10-13  Akim Demaille  <akim.demaille@gmail.com>

version 3.7.3
	* NEWS: Record release date.

2020-10-13  Akim Demaille  <akim.demaille@gmail.com>

build: don't link bison against libreadline
	Reported by Paul Smith <psmith@gnu.org>.
	https://lists.gnu.org/r/bug-bison/2020-10/msg00001.html

* src/local.mk (src_bison_LDADD): here.

2020-10-13  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-09-27  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: fix: use symbol_name
	* data/skeletons/glr.cc: here.

2020-09-06  Akim Demaille  <akim.demaille@gmail.com>

build: fix a concurrent build issue in examples
	Reported by Thomas Deutschmann <whissi@gentoo.org>.
	https://lists.gnu.org/r/bug-bison/2020-09/msg00010.html

* examples/c/lexcalc/local.mk: scan.o depends on parse.[ch].

2020-09-05  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-09-05  Akim Demaille  <akim.demaille@gmail.com>

version 3.7.2
	* NEWS: Record release date.

2020-09-05  Akim Demaille  <akim.demaille@gmail.com>

build: disable syntax-check warning
	error_message_uppercase
	etc/bench.pl.in-419-static int yylex (@{[is_pure (@directive) ? "YYSTYPE *yylvalp" : "void"]});

* cfg.mk: here.

2020-09-05  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-09-05  Akim Demaille  <akim.demaille@gmail.com>

build: fix incorrect dependencies
	Commit af000bab111768a04021bf5ffa4bbe91d44e231c ("doc: work around
	Texinfo 6.7 bug"), published in 3.4.91, added a dependency on the
	"all" target.

This is a super bad idea, since "make all" will run this
	target *before* "all", which builds bison.  It turns out that this new
	dependency actually needed bison to be built.  So all the regular
	process (i) build $(BUILT_SOURCES) and then (ii) build bison, was
	wrecked since some of the $(BUILT_SOURCES) depended on bison...

It was "easy" to see in the logs of "make V=1" because we were
	building bison files (such as src/files.o) *before* displaying the
	banner for "all-recursive".  With this fix, we finally get again the
	proper sequence:

rm -f examples/c/reccalc/scan.stamp examples/c/reccalc/scan.stamp.tmp
	    /opt/local/libexec/gnubin/mkdir -p examples/c/reccalc
	    touch examples/c/reccalc/scan.stamp.tmp
	    flex   -oexamples/c/reccalc/scan.c --header=examples/c/reccalc/scan.h ./examples/c/reccalc/scan.l
	    mv examples/c/reccalc/scan.stamp.tmp examples/c/reccalc/scan.stamp
	    rm -f lib/fcntl.h-t lib/fcntl.h && \
	    { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \
	      ...
	    } > lib/fcntl.h-t && \
	    mv lib/fcntl.h-t lib/fcntl.h
	    ...
	    mv -f lib/alloca.h-t lib/alloca.h
	    make  all-recursive

Reported by Mingli Yu <mingli.yu@windriver.com>.
	https://github.com/akimd/bison/issues/31
	https://lists.gnu.org/r/bison-patches/2020-05/msg00055.html

Reported by Claudio Calvelli <bugb@w42.org>.
	https://lists.gnu.org/r/bug-bison/2020-09/msg00001.html
	https://bugs.gentoo.org/716516

* doc/local.mk (all): Rename as...
	(all-local): this.
	So that we don't compete with BUILT_SOURCES.

2020-09-02  Akim Demaille  <akim.demaille@gmail.com>

doc: updates
	* NEWS, TODO: here.

2020-08-30  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-08-30  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of sed portability issues
	Reported by David Laxer <davidl@softintel.com>.
	https://lists.gnu.org/r/bug-bison/2020-08/msg00027.html

* tests/output.at: Don't use + with sed.
	While at it, fix a quotation problem hidden by the use of '#'.

2020-08-30  Akim Demaille  <akim.demaille@gmail.com>

c: always use YYMALLOC/YYFREE
	Reported by Kovalex <kovalex.pro@gmail.com>.
	https://lists.gnu.org/r/bug-bison/2020-08/msg00015.html

* data/skeletons/yacc.c: Don't make direct calls to malloc/free.
	* tests/calc.at: Check it.

2020-08-30  Akim Demaille  <akim.demaille@gmail.com>

build: beware of POSIX mode
	Reported by Dennis Clarke.
	https://lists.gnu.org/r/bug-bison/2020-08/msg00013.html

* examples/d/local.mk, examples/java/calc/local.mk,
	* examples/java/simple/local.mk: Pass bison's options before its
	argument, in case we're in POSIX mode.

2020-08-30  Akim Demaille  <akim.demaille@gmail.com>

doc: history of api.prefix
	Reported by Matthew Fernandez <matthew.fernandez@gmail.com>.
	https://lists.gnu.org/r/help-bison/2020-08/msg00015.html

* doc/bison.texi (api.prefix): We move to {} in 3.0.

2020-08-11  Akim Demaille  <akim.demaille@gmail.com>

CI: intel moved the script for ICC
	* .travis.yml: Adjust.

2020-08-08  Akim Demaille  <akim.demaille@gmail.com>

fix: unterminated \-escape
	An assertion failed when the last character is a '\' and we're in a
	character or a string.
	Reported by Agency for Defense Development.
	https://lists.gnu.org/r/bug-bison/2020-08/msg00009.html

* src/scan-gram.l: Catch unterminated escapes.
	* tests/input.at (Unexpected end of file): New.

2020-08-07  Akim Demaille  <akim.demaille@gmail.com>

fix: crash when redefining the EOF token
	Reported by Agency for Defense Development.
	https://lists.gnu.org/r/bug-bison/2020-08/msg00008.html

On an empty such as

%token FOO
	           BAR
	           FOO 0
	    %%
	    input: %empty

we crash because when we find FOO 0, we decrement ntokens (since FOO
	was discovered to be EOF, which is already known to be a token, so we
	increment ntokens for it, and need to cancel this).  This "works well"
	when EOF is properly defined in one go, but here it is first defined
	and later only assign token code 0.  In the meanwhile BAR was given
	the token number that we just decremented.

To fix this, assign symbol numbers after parsing, not during parsing,
	so that we also saw all the explicit token codes.  To maintain the
	current numbers (I'd like to keep no difference in the output, not
	just equivalence), we need to make sure the symbols are numbered in
	the same order: that of appearance in the source file.  So we need the
	locations to be correct, which was almost the case, except for nterms
	that appeared several times as LHS (i.e., several times as "foo:
	...").  Fixing the use of location_of_lhs sufficed (it appears it was
	intended for this use, but its implementation was unfinished: it was
	always set to "false" only).

* src/symtab.c (symbol_location_as_lhs_set): Update location_of_lhs.
	(symbol_code_set): Remove broken hack that decremented ntokens.
	(symbol_class_set, dummy_symbol_get): Don't set number, ntokens and
	nnterms.
	(symbol_check_defined): Do it.
	(symbols): Don't count nsyms here.
	Actually, don't count nsyms at all: let it be done in...
	* src/reader.c (check_and_convert_grammar): here.  Define nsyms from
	ntokens and nnterms after parsing.
	* tests/input.at (EOF redeclared): New.

* examples/c/bistromathic/bistromathic.test: Adjust the traces: in
	"%nterm <double> exp %% input: ...", exp used to be numbered before
	input.

2020-08-07  Akim Demaille  <akim.demaille@gmail.com>

style: fix missing space before paren
	* cfg.mk (_space_before_paren_exempt): Be less laxist.
	* src/output.c, src/reader.c: Fix space before paren issues.
	Pacify the warnings where applicable.

2020-08-07  Akim Demaille  <akim.demaille@gmail.com>

style: fix comments and more debug trace
	* src/location.c, src/symtab.h, src/symtab.c: here.

2020-08-07  Akim Demaille  <akim.demaille@gmail.com>

style: more uses of const
	* src/symtab.c: here.

2020-08-07  Akim Demaille  <akim.demaille@gmail.com>

bench: fix support for pure parser
	* etc/bench.pl.in (is_pure): New.
	(generate_grammar_calc): Use code provides where needed.
	Use is_pure to call yylex properly.
	Coding style fixes.

2020-08-03  Akim Demaille  <akim.demaille@gmail.com>

portability: multiple typedefs
	Older versions of GCC (4.1.2 here) don't like repeated typedefs.

CC       src/bison-parse-simulation.o
	    src/parse-simulation.c:61: error: redefinition of typedef 'parse_state'
	    src/parse-simulation.h:74: error: previous declaration of 'parse_state' was here
	    make: *** [Makefile:7876: src/bison-parse-simulation.o] Error 1

Reported by Nelson H. F. Beebe.

* src/parse-simulation.c (parse_state): Don't typedef,
	parse-simulation.h did it already.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

style: revert "avoid warnings with GCC 4.6"
	This reverts commit d0bec3175ff5cf6582ffbf584b73ea6aaea838d0 (which
	should have read "We have a clash...", not "With have a clash...").
	Now that `max()` was renamed `max_int()`, we can use `max` again, as
	elsewhere in the code.

* src/counterexample.c (visited_hasher): Alpha reconversion.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

version 3.7.1
	* NEWS: Record release date.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

portability: we use termios.h and sys/ioctl.h
	Reported by Maarten De Braekeleer.
	https://lists.gnu.org/r/bison-patches/2020-07/msg00079.html

* bootstrap.conf (gnulib_modules): Add termios and sys_ioctl.

2020-08-02  Maarten De Braekeleer  <maarten.debraekeleer@gmail.com>

portability: rename accept to acceptsymbol because of MSVC
	MSVC already defines this symbol.

* src/symtab.h, src/symtab.c (accept): Rename as...
	(acceptsymbol): this.
	Adjust dependencies.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-08-02  Maarten De Braekeleer  <maarten.debraekeleer@gmail.com>

portability: use CHAR_LITERAL instead of CHAR because MSVC defines CHAR
	* src/parse-gram.y, src/scan-gram.l: here.

2020-08-02  Maarten De Braekeleer  <maarten.debraekeleer@gmail.com>

portability: use INT_LITERAL instead of INT because MSVC defines INT
	It is defined as a typedef, not a macro.
	https://lists.gnu.org/r/bison-patches/2020-08/msg00001.html

* src/parse-gram.y, src/scan-gram.l: here.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

portability: beware of max () with MSVC
	Reported by Maarten De Braekeleer.
	https://lists.gnu.org/r/bison-patches/2020-07/msg00080.html

We don't want to use gnulib's min and max macros, since we use
	function calls in min/max arguments.

* src/location.c (max_int, min_int): Move to...
	* src/system.h: here.
	* src/counterexample.c, src/derivation.c: Use max_int instead of max.

2020-08-02  Akim Demaille  <akim.demaille@gmail.com>

libtextstyle: be sure to have ostream_printf and hyperlink support
	Older versions of libtextstyle do not support them, rule them out.

Reported by Lars Wendler
	https://lists.gnu.org/r/bug-bison/2020-07/msg00030.html

and by Arnold Robbins
	https://lists.gnu.org/r/bug-bison/2020-07/msg00041.html and
	https://lists.gnu.org/mailman/private/gawk-devel/2020-July/003988.html

and by Nelson H. F. Beebe
	https://lists.gnu.org/mailman/private/gawk-devel/2020-July/003993.html

With support from Bruno Haible in gnulib
	https://lists.gnu.org/r/bug-gnulib/2020-08/msg00000.html
	thread starting at
	https://lists.gnu.org/r/bug-gnulib/2020-07/msg00148.html

* configure.ac: Require libtextstyle 0.20.5.
	* gnulib: Update.

2020-08-01  Akim Demaille  <akim.demaille@gmail.com>

CI: comment changes

2020-08-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-08-01  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: better location for type redeclarations
	From

foo.y:1.7-11: error: %type redeclaration for bar
	        1 | %type <foo> bar bar
	          |       ^~~~~
	    foo.y:1.7-11: note: previous declaration
	        1 | %type <foo> bar bar
	          |       ^~~~~

foo.y:1.17-19: error: %type redeclaration for bar
	        1 | %type <foo> bar bar
	          |                 ^~~
	    foo.y:1.13-15: note: previous declaration
	        1 | %type <foo> bar bar
	          |             ^~~

* src/symlist.h, src/symlist.c (symbol_list_type_set): There's no need
	for the tag's location, use that of the symbol.
	* src/parse-gram.y: Adjust.
	* tests/input.at: Adjust.

2020-07-30  Akim Demaille  <akim.demaille@gmail.com>

todo: updates for D

2020-07-29  Akim Demaille  <akim.demaille@gmail.com>

cex: style: comment changes
	* src/parse-simulation.c: here.

2020-07-29  Akim Demaille  <akim.demaille@gmail.com>

cex: style: prefer "res" for the returned value
	* src/derivation.c (derivation_new): here.

2020-07-29  Akim Demaille  <akim.demaille@gmail.com>

cex: style: prefer FOO_print to print_FOO
	* src/state-item.h, src/state-item.c (print_state_item): Rename as...
	(state_item_print): this.
	* src/counterexample.c (print_counterexample): Rename as...
	(counterexample_print): this.

2020-07-28  Akim Demaille  <akim.demaille@gmail.com>

scanner: don't crash on strings containing a NUL byte
	We crash if the input contains a string containing a NUL byte.
	Reported by Suhwan Song.
	https://lists.gnu.org/r/bug-bison/2020-07/msg00051.html

* src/flex-scanner.h (STRING_FREE): Avoid accidental use of
	last_string.
	* src/scan-gram.l: Don't call STRING_FREE without calling
	STRING_FINISH first.
	* tests/input.at (Invalid inputs): Check that case.

2020-07-28  Akim Demaille  <akim.demaille@gmail.com>

doc: refer to cex from sections dealing with conflicts
	The documentation about -Wcex should be put forward.

* doc/bison.texi: Refer to -Wcex from the sections about conflicts.

2020-07-28  Akim Demaille  <akim.demaille@gmail.com>

doc: factor ifnottex/iftex examples
	* doc/bison.texi: Factor the common bits out of ifnottex/iftex.

2020-07-28  Akim Demaille  <akim.demaille@gmail.com>

doc: fix colors
	The original Texinfo macros introducing colors were made for
	diagnostics, which are printed in bold.  So by copy-paste accident the
	styles we introduced for counterexamples were also in bold.  They
	should not.

* doc/bison.texi: Separate the styling of diagnostics from the styling
	for counterexamples.
	Don't use bold in the latter case.

2020-07-28  Akim Demaille  <akim.demaille@gmail.com>

doc: fixes
	* doc/bison.texi: Fix spello.
	Fix missing colors, and factor.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

version 3.7
	* NEWS: Record release date.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

style: avoid warnings with GCC 4.6
	With have a clash with the "max" function.

src/counterexample.c: In function 'visited_hasher':
	    src/counterexample.c:720:48: error: declaration of 'max' shadows a global declaration [-Werror=shadow]
	    src/counterexample.c:116:12: error: shadowed declaration is here [-Werror=shadow]

* src/counterexample.c (visited_hasher): Alpha conversion.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

doc: fix definition of -Wall
	* doc/bison.texi (Diagnostics): here.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	* bootstrap.conf: We need stpncpy.

2020-07-23  Akim Demaille  <akim.demaille@gmail.com>

tests: fixes
	Fix 6b78e50cef3c2cd8e6f4e7938be987e8769f8eef, "cex: make "rerun with
	'-Wcex'" a note instead of a warning"

* tests/conflicts.at (-W versus %expect and %expect-rr): Fix
	expectations.

2020-07-22  Akim Demaille  <akim.demaille@gmail.com>

cex: update NEWS for 3.7
	* NEWS: Update to the current style of cex display.

2020-07-22  Akim Demaille  <akim.demaille@gmail.com>

doc: catch up with the current display of cex
	Unfortunately I found no way to use the ↳ glyph in Texinfo, so I used
	@arrow{} instead, which has a different width, so we have to have all
	the examples doubled, once for TeX, another for the rest of the world.

* doc/bison.texi: Use the current display in the examples.
	* doc/calc.y, doc/ids.y, doc/if-then-else.y, doc/sequence.y: New.

2020-07-21  Akim Demaille  <akim.demaille@gmail.com>

cex: make "rerun with '-Wcex'" a note instead of a warning
	Currently the suggestion to rerun is a -Wother warning:

warning: 2 shift/reduce conflicts [-Wconflicts-sr]
	    warning: rerun with option '-Wcounterexamples' to generate conflict counterexamples [-Wother]

Instead, let's attach it as a subnote of the diagnostic (in the
	current case, -Wconflicts-sr):

warning: 2 shift/reduce conflicts [-Wconflicts-sr]
	    note: rerun with option '-Wcounterexamples' to generate conflict counterexamples

* src/conflicts.c (conflicts_print): Do that.
	Adjust the test suite.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.93
	* NEWS: Record release date.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

cex: label all the derivations by their initial action
	From

input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples]
	      Example: A b .
	      First derivation
	        a
	        `-> A b .
	      Second derivation
	        a
	        `-> A b
	              `-> b .

input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples]
	      Example: A b .
	      First reduce derivation
	        a
	        `-> A b .
	      Second reduce derivation
	        a
	        `-> A b
	              `-> b .

* src/counterexample.c (print_counterexample): here.
	Compute the width of the labels to properly align the values.
	* tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at,
	* tests/report.at: Adjust.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

cex: improve readability of the subsections
	Now that the derivation is no longer printed on one line, aligning the
	example and the derivation is no longer useful.  It can actually be
	harmful, as it makes the overall structure less clear.

* src/derivation.h, src/derivation.c (derivation_print_leaves): Remove
	the `prefix` argument.
	* src/counterexample.c (print_counterexample): Put the example next to
	its label.
	* tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at,
	* tests/report.at: Adjust.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

cex: don't issue an empty line between counterexamples
	Now that we use complain, the "sections" are clearer.

* src/counterexample.c (print_counterexample): Use the empty line only
	in reports.
	* tests/counterexample.at, tests/diagnostics.at, tests/report.at: Adjust.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

cex: use usual routines for diagnostics about S/R conflicts
	See previous commit.  We go from

input.y: warning: 3 reduce/reduce conflicts [-Wconflicts-rr]
	    Shift/reduce conflict on token "⊕":
	      Example              exp "+" exp • "⊕" exp
	      Shift derivation
	        exp
	        ↳ exp "+" exp
	                  ↳ exp • "⊕" exp

input.y: warning: 3 reduce/reduce conflicts [-Wconflicts-rr]
	    input.y: warning: shift/reduce conflict on token "⊕" [-Wcounterexamples]
	      Example              exp "+" exp • "⊕" exp
	      Shift derivation
	        exp
	        ↳ exp "+" exp
	                  ↳ exp • "⊕" exp

with an hyperlink on -Wcounterexamples.

* src/counterexample.c (counterexample_report_shift_reduce):
	Use complain.
	* tests/counterexample.at, tests/diagnostics.at, tests/report.at:
	Adjust.

2020-07-20  Akim Demaille  <akim.demaille@gmail.com>

cex: use usual routines for diagnostics about R/R conflicts
	This is more consistent, and brings benefits: users know that these
	diagnostics are attached to -Wcounterexamples, and they can also click
	on the hyperlink if permitted by their terminal.

We go from

warning: 1 reduce/reduce conflict [-Wconflicts-rr]
	    Reduce/reduce conflict on token $end:
	      Example              A b .
	      First derivation     a -> [ A b . ]
	      Second derivation    a -> [ A b -> [ b . ] ]

warning: 1 reduce/reduce conflict [-Wconflicts-rr]
	    input.y: warning: reduce/reduce conflict on token $end [-Wcounterexamples]
	      Example              A b .
	      First derivation     a -> [ A b . ]
	      Second derivation    a -> [ A b -> [ b . ] ]

with an hyperlink on -Wcounterexamples.

* src/counterexample.c (counterexample_report_reduce_reduce):
	Use complain.
	* tests/counterexample.at, tests/diagnostics.at, tests/report.at:
	Adjust.

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use hyperlinks to point to the only documentation
	* src/complain.c (begin_hyperlink, end_hyperlink): New.
	(warnings_print_categories): Use them.
	* tests/local.at (AT_SET_ENV): Disable hyperlinks in the tests, they
	contain random id's, and brackets (which is not so nice for M4).

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

doc: add anchors for warnings
	Unfortunately Texinfo somewhat mangles anchors such as `-Werror` into
	`g_t_002dWerror`, so let's not include the dash.

* doc/bison.texi (Diagnostics): here.

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

glyphs: fix types
	The code was written on top of buffers of `char[26]`, and then was
	changed to use `char *`, yet was still using `sizeof buf`, which
	became `sizeof (char *)` instead of `sizeof (char[26])`.

Reported by Dagobert Michelsen.
	https://lists.gnu.org/r/bug-bison/2020-07/msg00023.html

* src/glyphs.h, src/glyphs.c: Get rid of uses of `char *`, use only
	glyph_buffer_t.

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.92
	* NEWS: Record release date.

2020-07-19  Akim Demaille  <akim.demaille@gmail.com>

style: avoid strncpy
	syntax-check seems to dislike strncpy.  The GNU Coreutils replaced
	their uses of strncpy with stpncpy.

strlcpy is not an option.
	  http://sources.redhat.com/ml/libc-alpha/2002-01/msg00159.html
	  http://sources.redhat.com/ml/libc-alpha/2002-01/msg00011.html
	  http://lists.gnu.org/archive/html/bug-gnulib/2004-09/msg00181.html

* src/glyphs.c: Use stpncpy.

2020-07-18  Akim Demaille  <akim.demaille@gmail.com>

cex: display derivations as trees
	Sometimes, understanding the derivations is difficult, because they
	are serialized to fit in one line.  For instance, the example taken
	from the NEWS file:

%token ID
	    %%
	    s: a ID
	    a: expr
	    expr: expr ID ',' | "expr"

gave

First example        expr • ID ',' ID $end
	    Shift derivation     $accept → [ s → [ a → [ expr → [ expr • ID ',' ] ] ID ] $end ]
	    Second example       expr • ID $end
	    Reduce derivation    $accept → [ s → [ a → [ expr • ] ID ] $end ]

Printing as trees, it gives:

First example        expr • ID ',' ID $end
	    Shift derivation
	      $accept
	      ↳ s                      $end
	        ↳ a                 ID
	          ↳ expr
	            ↳ expr • ID ','
	    Second example       expr • ID $end
	    Reduce derivation
	      $accept
	      ↳ s             $end
	        ↳ a        ID
	          ↳ expr •

* src/glyphs.h, src/glyphs.c (down_arrow, empty, derivation_separator):
	New.
	* src/derivation.c (derivation_print, derivation_print_impl): Rename
	as...
	(derivation_print_flat, derivation_print_flat_impl): These.
	(fputs_if, derivation_depth, derivation_width, derivation_print_tree)
	(derivation_print_tree_impl, derivation_print): New.
	* src/counterexample.c (print_counterexample): Adjust.
	* tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at,
	* tests/report.at: Adjust.

2020-07-16  Akim Demaille  <akim.demaille@gmail.com>

cex: use the glyphs
	* src/derivation.c: here.
	* src/gram.h, src/gram.c (print_arrow, print_dot, print_fallback):
	Remove.

2020-07-16  Akim Demaille  <akim.demaille@gmail.com>

cex: factor the handling of graphical symbols
	* src/glyphs.h, src/glyphs.c: New.

2020-07-15  Akim Demaille  <akim.demaille@gmail.com>

cex: style changes
	* src/counterexample.c: here.

2020-07-15  Akim Demaille  <akim.demaille@gmail.com>

cex: simplify tests
	* tests/counterexample.at (AT_BISON_CHECK_CEX): Handle the keyword.
	Simplify the signature.

2020-07-15  Akim Demaille  <akim.demaille@gmail.com>

cex: more colors
	Provided by Daniela Becker.

* data/bison-default.css: More colors.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

style: comments changes
	* src/print.c: here.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

doc: update GLR sections
	Reported by Christian Schoenebeck.

* doc/bison.texi (GLR Parsers): Minor fixes.
	(Compiler Requirements for GLR): Remove, quite useless today.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: display shifts before reductions
	When reporting counterexamples for s/r conflicts, put the shift first.
	This is more natural, and displays the default resolution first, which
	is also what happens for r/r conflicts where the smallest rule number
	is displayed first, and "wins".

* src/counterexample.c (counterexample): Add a shift_reduce member.
	(new_counterexample): Adjust.
	Swap the derivations when this is a s/r conflict.
	(print_counterexample): For s/r conflicts, prefer "Shift derivation"
	and "Reduce derivation" rather than "First/Second derivation".

* tests/conflicts.at, tests/counterexample.at, tests/report.at: Adjust.
	* NEWS, doc/bison.texi: Ditto.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

style: s/lookahead_tokens/lookaheads/g
	Currently we use both names.  Let's stick to the short one.

* src/AnnotationList.c, src/conflicts.c, src/counterexample.c,
	* src/getargs.c, src/getargs.h, src/graphviz.c, src/ielr.c,
	* src/lalr.c, src/print-graph.c, src/print-xml.c, src/print.c,
	* src/state-item.c, src/state.c, src/state.h, src/tables.c:
	s/lookahead_token/lookahead/gi.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: factor memory allocation
	* src/counterexample.c (counterexample_report_state): Allocate once
	per conflicted state, instead of once per r/r conflict.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: use state_item_number consistently
	* src/counterexample.c, src/state-item.c: here.
	(counterexample_report_state): While at it, prefer c2 to j/k, to match
	c1.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: more consistent memory allocation/copy
	* src/counterexample.c, src/parse-simulation.c: It is more usual in
	Bison to use sizeof on expressions than on types, especially for
	allocation.
	Let the compiler do it's job instead of calling memcpy ourselves.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: minor renaming
	* src/counterexample.c (has_common_prefix): Rename as...
	(have_common_prefix): this.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: use better type names
	There are too many gl_list_t in there, it's hard to understand what is
	going on.  Introduce and use more precise types.  I sure can be wrong
	in some places, it's hard to tell without proper tool support.

* src/counterexample.c, src/lssi.c, src/lssi.h, src/parse-simulation.c,
	* src/parse-simulation.h, src/state-item.c, src/state-item.h
	(si_bfs_node_list, search_state_list, ssb_list, lssi_list)
	(state_item_list): New.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

cex: minor style changes
	* src/counterexample.h, src/derivation.h, src/derivation.c:
	More comments.
	Use `out` for FILE*, as elsewhere.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of version numbers from git describe
	* tests/report.at: Be robust to version numbers such as
	3.6.4.133-fbac-dirty.

2020-07-14  Akim Demaille  <akim.demaille@gmail.com>

tests: fix expectations
	Broken in ee86ea88399ed02243fbceb2704c9ea322a12bf9.

* tests/diagnostics.at: here.

2020-07-12  Akim Demaille  <akim.demaille@gmail.com>

doc: makeinfo wants @arrow{}, not @arrow
	* doc/bison.texi: here.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

cex: prefer → to ::=
	It does not make a lot of sense to use ::= in our counterexamples,
	that's not something that belongs to the Bison "vocabulary".  Using
	the colon makes sense, but it's too discreet.  Let's use the arrow,
	which we already use in some reports (HTML and Dot).

* src/gram.h (print_dot_fallback): Generalize into...
	(print_fallback): this.
	(print_arrow): New.
	* src/derivation.c: Use it.

* NEWS, tests/conflicts.at, tests/counterexample.at,
	* tests/diagnostics.at, tests/report.at: Adjust.
	* doc/bison.texi: Ditto.
	Unfortunately the literal `→` is output as `↦`.  So we need to use
	@arrow.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

style: cex: prefer the array notation
	Prefer `&foos[i]` to `foos + i` when `foos` is an array.  IMHO, it
	makes the semantics clearer.

* src/counterexample.c, src/lssi.c, src/parse-simulation.c,
	* src/state-item.c: With arrays, prefer the array notation rather than
	the pointer one.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

style: cex: remove variables that don't make it simpler to read
	* src/counterexample.c: With arrays, prefer the array notation rather
	than the pointer one.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: demonstrate caret-diagnostics
	* examples/c/bistromathic/parse.y (user_context): We need the current
	line.
	(yyreport_syntax_error): Quote the guilty line, with squiggles.
	* examples/c/bistromathic/bistromathic.test: Adjust.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: do not display parse errors on completion
	Currently autocompletion on a line with errors leaks the error
	messages.  It can be useful to let the user know, but GNU Readline
	does not provide us with an nice way to display the error.  So we
	actually break into the current line of the user.

So instead, do not show these errors.

* examples/c/bistromathic/parse.y (user_context): New.
	Use %param to pass it to the parser and scanner.
	Keep quiet when in computing autocompletion.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: don't stupidly reset the location for each token
	That quite defeats the whole point of locations...  But anyway, we
	should not see these messages at all.

* examples/c/bistromathic/parse.y (expected_tokens): Fix (useless)
	location tracking.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: promote yytoken_kind_t
	* examples/c/bistromathic/parse.y: Use yytoken_kind_t rather than int.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

html: capitalize titles
	* data/xslt/xml2xhtml.xsl: Use "State 0", not "state 0".
	As we do in text reports.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

html: don't define several times the same anchors
	Currently when we output useless rules, they appear before the
	grammar, but using the same invocation.  As a result, the anchor is
	defined twice, and the wrong one, being first, is honored.

* data/xslt/xml2xhtml.xsl (rule): Take a new 'anchor' parameter to
	decide whether being an anchor, or a target.
	Let it be true when output the grammar.
	* tests/report.at: Adjust.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

html: simplify
	* data/xslt/xml2xhtml.xsl: Merge two identical when-clauses.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

reports: let html reports catch up with --report and --graph
	* data/xslt/xml2xhtml.xsl: Show the symbol types.
	* tests/report.at: Adjust.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

reports: let xml reports catch up with --report and --graph
	The text and Dot reports are expected to be identical when generated
	directly (--report, --graph) or indirectly (via XML).  The xml
	testsuite had not be run for ages, let it catch up a bit.

* src/print-xml.c: Pass the type of the symbols.
	* data/xslt/xml2text.xsl
	Catch up with the new layout.
	Display the symbol types.
	Use '•', not '.'
	* tests/local.at: Smash '•' to '.' when matching against the direct
	text report.
	* tests/report.at: Adjust XML expectations.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

reports: update html ouput
	* data/xslt/xml2xhtml.xsl: Improve indentation.
	Use ul/li rather that pre.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

tests: check html
	* tests/report.at: here.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

style: factor complex expressions
	* src/print-xml.c, src/print.c: Introduce a variable pointing to the
	current symbol.

2020-07-11  Akim Demaille  <akim.demaille@gmail.com>

maint: make it easier to update expectations
	* tests/local.mk (update-tests): New.

2020-07-09  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-07-09  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.91
	* NEWS: Record release date.

2020-07-09  Akim Demaille  <akim.demaille@gmail.com>

news: update

2020-07-09  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-07-08  Akim Demaille  <akim.demaille@gmail.com>

examples: add license headers
	Prompted by Rici Lake.
	https://stackoverflow.com/questions/62658368/#comment110853985_62661621
	Discussed with Paul Eggert.

* doc/bison.texi, examples/c/bistromathic/parse.y,
	* examples/c/lexcalc/parse.y, examples/c/lexcalc/scan.l,
	* examples/c/pushcalc/calc.y, examples/c/reccalc/parse.y,
	* examples/c/reccalc/scan.l, examples/d/calc.y,
	* examples/java/calc/Calc.y, examples/java/simple/Calc.y:
	Install the GPL3+ header.

2020-07-05  Akim Demaille  <akim.demaille@gmail.com>

style: update comments
	* src/reader.c: action_obstack was removed in 2002...
	* src/parse-gram.y: Better names.
	* src/scan-code.h: More comments.

2020-07-05  Akim Demaille  <akim.demaille@gmail.com>

style: update comments in the skeletons
	* data/skeletons/c++.m4, data/skeletons/glr.c, data/skeletons/lalr1.d,
	* data/skeletons/lalr1.java, data/skeletons/yacc.c:
	Be more accurate about yychar and yytoken.
	Don't name local variables as if they were members.

2020-07-05  Akim Demaille  <akim.demaille@gmail.com>

doc: more details about symbols in m4
	* data/README.md: here.
	* README-hacking.md (Vocabulary): More.

2020-07-05  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-07-05  Akim Demaille  <akim.demaille@gmail.com>

examples: include the generated header
	* examples/c/bistromathic/parse.y, examples/c/lexcalc/parse.y,
	* examples/c/reccalc/parse.y: here.
	Add some comments.

* src/parse-gram.y (api_version): Pull out of handle_require.
	Bump to 3.7.

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.90
	* NEWS: Record release date.

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

news: update

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

package: fix syntax-check errors
	* examples/c/bistromathic/bistromathic.test, po/POTFILES.in: here.

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-07-04  Akim Demaille  <akim.demaille@gmail.com>

cex: give more details about -Wcex and -rcex
	* data/bison-default.css: Cobalt does not seem to be supported.
	* doc/bison.texi (Counterexamples): A new section.
	(Understanding): Show the counterexamples as it shows in the report:
	with its items.
	(Bison Options): Document -Wcex and -rcex.

2020-07-03  Akim Demaille  <akim.demaille@gmail.com>

news: update

2020-07-03  Akim Demaille  <akim.demaille@gmail.com>

dot: also use a dot in the output
	* src/print-graph.c (print_core): Use a dot instead of a point.
	* doc/figs/example-reduce.gv, doc/figs/example-reduce.txt,
	* doc/figs/example-shift.gv, doc/figs/example-shift.txt,
	* doc/figs/example.gv: Update.
	* tests/output.at, tests/report.at: Adjust.

2020-07-02  Akim Demaille  <akim@lrde.epita.fr>

doc: improve a sentence
	* doc/bison.texi: here.

2020-07-01  Akim Demaille  <akim.demaille@gmail.com>

news: formatting changes

2020-07-01  Akim Demaille  <akim.demaille@gmail.com>

news, todo: update

2020-06-30  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify that the pcontext interface is *.c only
	Reported by Rici Lake.
	https://lists.gnu.org/r/bug-bison/2020-06/msg00054.html

* doc/bison.texi (Syntax Error Reporting Function): Make it clear that
	this is not exported.
	Remove C++ details that landed in the C doc.

2020-06-30  Akim Demaille  <akim.demaille@gmail.com>

doc: use color in the cex examples
	* doc/bison.texi: here.
	And use smallexample when it no longer fits in PDF.

2020-06-30  Akim Demaille  <akim.demaille@gmail.com>

doc: repair the references to the Bibliography
	In commit c80cdf2db2b302db4137fabd4ae11e578fa51fca ("doc: simplify
	uses of @ref", Jan 27 2020, released in Bison 3.6), I broke the
	references to the Bibliography.  For instance:

For a more detailed exposition of the mysterious behavior in LALR parsers
	    -and the benefits of IELR, @pxref{Bibliography,,Denny 2008 March}, and
	    -@ref{Bibliography,,Denny 2010 November}.
	    +and the benefits of IELR, @pxref{Bibliography}, and
	    +@ref{Bibliography}.

which results in "see Bibliography" twice, instead of the more precise
	reference.

* doc/bison.texi (@pcite, @tcite): New.
	Use them instead of @ref to Bibliography.
	Cite only the first author (that's what we did for the other entries).

2020-06-30  Vincent Imbimbo  <vmi6@cornell.edu>

doc: cex documentation
	* NEWS, doc/bison.texi: Add documentation for conflict counterexample
	generation.

2020-06-30  Akim Demaille  <akim.demaille@gmail.com>

doc: update Doxygen template file
	* doc/Doxyfile.in: here.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: push: don't clear the parser state when accepting/rejecting
	Currently when a push parser finishes its parsing (i.e., it did not
	return YYPUSH_MORE), it also clears its state.  It is therefore
	impossible to see if it had parse errors.

In the context of autocompletion, because error recovery might have
	fired, the parser is actually already in a different state.  For
	instance on `(1 + + <TAB>` in the bistromathic, because there's a
	`exp: "(" error ")"` recovery rule, `1 + +` tokens have already been
	popped, replaced by `error`, and autocompletions think we are ready
	for the closing ")".  So here, we would like to see if there was a
	syntax error, yet `yynerrs` was cleared.

In the case of a successful parse, we still have a problem: if error
	recovery succeeded, we won't know it, since, again, `yynerrs` is
	clearer.

It seems much more natural to leave the parser state available for
	analysis when there is a failure.

To reuse the parser, we should either:

1. provide an explicit means to reinitialize a parser state for future
	   parses.

2. automatically reset the parser state when it is used in a new
	   parse.

Option 2 requires to check whether we need to reinitialize the parser
	each time we call `yypush_parse`, i.e., each time we give a new token.
	This seems expensive compared to Option 1, but benchmarks revealed no
	difference.  Option 1 is incompatible with the documentation
	("After `yypush_parse` returns a status other than `YYPUSH_MORE`, the
	parser instance `yyps` may be reused for a new parse.").

So Option 2 wins, reusing the private `yynew` member to record that a
	parse was finished, and therefore that the state must reset in the
	next call to `yypull_parse`.

While at it, this implementation now reuses the previously enlarged
	stacks from one parse to another.

* data/skeletons/yacc.c (yypstate_new): Set up the stacks in their
	initial configurations (setting their bottom to the stack array), and
	use yypstate_clear to reset them (moving their top to their bottom).
	(yypstate_delete): Adjust.
	(yypush_parse): At the beginning, clear yypstate if needed, and at the
	end, record when yypstate needs to be clearer.

* examples/c/bistromathic/parse.y (expected_tokens): Do not propose
	autocompletion when there are parse errors.
	* examples/c/bistromathic/bistromathic.test: Check that case.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: don't display undefined locations
	Currently, completion when there is a syntax error shows broken
	locations.

* examples/c/bistromathic/parse.y (expected_tokens): Initialize the
	location.
	* examples/c/bistromathic/bistromathic.test: Check that.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: simplify initialization of push parsers
	The previous commit ("yacc.c: declare and initialize and the same
	time") made b4_initialize_parser_state_variables useless.

* data/skeletons/yacc.c (b4_initialize_parser_state_variables): Inline
	into...
	(yypstate_clear): here.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: declare and initialize and the same time
	In order to factor the code of push and pull parsers, the declaration
	of the parser's state variable was common (being local variable in
	pull parsers, and struct members in push parsers).  This result in
	rather poor style in pull parser, with first variable declarations,
	and then their initializations.

The initialization is about to differ between push and pull parsers,
	so it is no longer worth keeping both cases together.

* data/skeletons/yacc.c (b4_declare_parser_state_variables): Accept an
	argument, and when it is set, initialize the variables.
	Adjust dependencies.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: style changes in push mode
	* data/skeletons/yacc.c: here.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: simplify yypull_parse
	Currently yypull_parse takes a yypstate* as argument, and accepts it
	to be NULL.  This does not seem to make a lot of sense: rather it is
	its callers that should do that.

I believe this is historical: yypull_parse was introduced
	first (c3d503425f8014b432601a33b3398446d63b5963), with yyparse being a
	macro.  So yyparse could hardly deal with memory allocation properly.
	In 7172e23e8ffb95b8cafee24c4f36c46ca709507f that yyparse was turned
	into a genuine function.  At that point, it should have allocated its
	own yypstate*, which would have left yypull_parse deal with only one
	single non-null ypstate* argument.

Fortunately, it is nowhere documented that it is valid to pass NULL to
	yypull_parse.  It is now forbidden.

* data/skeletons/yacc.c (yypull_parse): Don't allocate a yypstate.
	Needs a location to issue the error message.
	(yyparse): Allocate the yypstate.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

doc: tidy the text files
	* etc/README: Rename/reformat as...
	* etc/README.md: this.
	And ship it.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

bench: simplify the `rand` target
	* etc/bench.pl.in: There is no need to recompile the bench cases
	themselves.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

bench: make it easy to edit the generated files
	* etc/bench.pl.in (&compile): Generate rules that compile the
	generated files independently of the source files.

2020-06-29  Akim Demaille  <akim.demaille@gmail.com>

tests: don't use $VERBOSE
	It is used by the test suite itself, which results in this test
	failing.

* tests/c++.at: Use $DEBUG, not $VERBOSE.

2020-06-28  Akim Demaille  <akim.demaille@gmail.com>

doc: overhaul of the readmes
	* README-hacking.md (Working from the Repository): Make it first to
	make it easier to find the instructions to build from the repo.
	(Implementation Notes): New.
	* README: Provide more links.

2020-06-28  Akim Demaille  <akim.demaille@gmail.com>

java: rename package as api.package
	* data/skeletons/lalr1.java: here.
	* doc/bison.texi: Update.
	* src/muscle-tab.c: Ensure backward compat.
	* tests/java.at: Check it.

2020-06-28  Akim Demaille  <akim.demaille@gmail.com>

style: shift/reduce, not shift-reduce
	* src/reader.c: here.

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

style: rename endtoken as eoftoken
	* src/symtab.h, src/symtab.c (endtoken): Rename as...
	(eoftoken): this.
	Adjust dependencies.

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

news: fixes
	Reported by Jacob L. Mandelson.

* NEWS: here.

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

style: use 'nonterminal' consistently
	* doc/bison.texi: Formatting changes.
	* src/gram.h, src/gram.c (nvars): Rename as...
	(nnterms): this.
	Adjust dependencies.
	(section): New.  Use it.
	Replace "non terminal" and "non-terminal" by "nonterminal".

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

doc: parse.assert in C++ requires RTTI
	* doc/bison.texi (%define Summary): Say it.

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

c++: by default, use const std::string for file names
	Reported by Martin Blais and Yuriy Solodkyy.
	https://lists.gnu.org/r/help-bison/2020-05/msg00011.html
	https://lists.gnu.org/r/bug-bison/2020-06/msg00038.html

While at it, modernize filename_type as api.filename.type and document
	it properly.

* data/skeletons/c++.m4 (filename_type): Rename as...
	(api.filename.type): this.
	Default to const std::string.
	* data/skeletons/location.cc (position, location): Expose the
	filename_type type.
	Use api.filename.type.
	* doc/bison.texi (%define Summary): Document api.filename.type.
	(C++ Location Values): Document position::filename_type.
	* src/muscle-tab.c (muscle_percent_variable_update): Ensure backward
	compatibility.
	* tests/c++.at: Check that using const file names is ok.
	tests/input.at: Check backward compat.

2020-06-27  Akim Demaille  <akim.demaille@gmail.com>

ielr: fix crash on memory management
	Reported by Dwight Guth.
	https://lists.gnu.org/r/bug-bison/2020-06/msg00037.html

* src/AnnotationList.c (AnnotationList__computePredecessorAnnotations):
	Beware that SBITSET__FOR_EACH nests _two_ for-loops, so "break" does
	not actually break out of it.
	That was the only occurrence in the code.
	* src/Sbitset.h (SBITSET__FOR_EACH): Warn passersby.

2020-06-25  Akim Demaille  <akim.demaille@gmail.com>

style: factor the access to a rule from its items
	* src/counterexample.c (item_rule): Move to...
	* src/counterexample.h: here.
	* src/AnnotationList.c, src/counterexample.c, src/ielr.c: Use it.

2020-06-25  Akim Demaille  <akim.demaille@gmail.com>

style: clean up nullable
	* src/nullable.c: Reduce scopes.
	Prefer `r` to `rules_ruleno`, which is truly an ugly name.

2020-06-25  Akim Demaille  <akim.demaille@gmail.com>

style: clean up ielr
	* src/AnnotationList.c, src/ielr.c: Fix include order.
	Prefer `res` to `result`.
	Reduce scopes.
	Be free of the oldish 76 cols limitation when it clutters too much the
	code.
	Denest when possible (we're starving for horizontal width).

2020-06-23  Akim Demaille  <akim.demaille@gmail.com>

don't use strlen to compute visual width
	* src/output.c (prepare_symbol_names): Use mbswidth.

2020-06-23  Akim Demaille  <akim.demaille@gmail.com>

doc: use dot/'•' rather than point/'.'
	AFAICT, "dotted rule" is a more frequent synonym of "item" than
	"pointed rule".  So let's migrate to using "dot" only.

* doc/bison.texi: Use dot/'•' rather than point/'.'.

* src/print-xml.c (print_core): Use dot rather than point.  This is
	not backward compatible, but AFAICT, we don't have actual user of the
	XML output (but ourselves).  So...
	* data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl,
	* data/xslt/xml2xhtml.xsl, tests/report.at: ... adjust.

2020-06-23  Akim Demaille  <akim.demaille@gmail.com>

cex: display all the S/R conflicts, not just one per (state, rule)
	Before this commit, on

%%
	    exp
	    : "if" exp "then" exp
	    | "if" exp "then" exp "else" exp
	    | exp "+" exp
	    | "num"

we used to not display the third counterexample below:

Shift/reduce conflict on token "+":
	      Example              exp "+" exp . "+" exp
	      First derivation     exp ::=[ exp ::=[ exp "+" exp . ] "+" exp ]
	      Second derivation    exp ::=[ exp "+" exp ::=[ exp . "+" exp ] ]

Shift/reduce conflict on token "else":
	      Example              "if" exp "then" "if" exp "then" exp . "else" exp
	      First derivation     exp ::=[ "if" exp "then" exp ::=[ "if" exp "then" exp . ] "else" exp ]
	      Second derivation    exp ::=[ "if" exp "then" exp ::=[ "if" exp "then" exp . "else" exp ] ]

Shift/reduce conflict on token "+":
	      Example              "if" exp "then" exp . "+" exp
	      First derivation     exp ::=[ exp ::=[ "if" exp "then" exp . ] "+" exp ]
	      Second derivation    exp ::=[ "if" exp "then" exp ::=[ exp . "+" exp ] ]

Shift/reduce conflict on token "+":
	      Example              "if" exp "then" exp "else" exp . "+" exp
	      First derivation     exp ::=[ exp ::=[ "if" exp "then" exp "else" exp . ] "+" exp ]
	      Second derivation    exp ::=[ "if" exp "then" exp "else" exp ::=[ exp . "+" exp ] ]

* src/counterexample.c (counterexample_report_state): Don't stop of
	the first conflicts.
	* tests/conflicts.at, tests/counterexample.at, tests/diagnostics.at,
	* tests/report.at: Adjust.

2020-06-22  Akim Demaille  <akim.demaille@gmail.com>

cex: don't display twice unifying examples if there is no color
	It makes no sense, and is actually confusing, to display twice the
	same example with no visible difference.

* src/complain.h, src/complain.c (is_styled): New.
	* src/counterexample.c (print_counterexample): Display the unified
	example a second time only if it makes a difference.
	* tests/conflicts.at, tests/counterexample.at, tests/report.at: Adjust.
	* tests/diagnostics.at: Make sure we do display the unifying examples
	twice when colors are enabled.  And check those colors.

2020-06-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix reporting of null nonterminals
	I implemented this to print A ::= [ ], but A ::= [ %empty ] might be
	clearer.

* src/parse-simulation.c (nullable_closure): Don't generate null
	nonterminal derivations as leaves.
	* src/derivation.c (derivation_print_impl): Don't print seperator
	spaces for null nonterminal.
	* tests/counterexample.at: Update test results.

2020-06-22  Akim Demaille  <akim.demaille@gmail.com>

cex: use the bullet in HTML
	* data/xslt/xml2xhtml.xsl: here.

2020-06-19  Akim Demaille  <akim.demaille@gmail.com>

cex: style changes
	* src/counterexample.c: Simplify a bit.
	* src/parse-simulation.c, src/parse-simulation.h: Enforce coding style.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

c++: get rid of global_tokens_and_yystype
	This was a hack to make it easier for people to migrate from yacc.c to
	lalr1.cc and from glr.c to glr.cc: when set, YYSTYPE and YYLTYPE were
	`#defined`.  It was never documented (just mentioned in NEWS for Bison
	2.2, 2006-05-19), but was used to simplify the test suite.  Stop that:
	adjust the test suite to the skeletons, not the converse.

In C++ use yy::parser::semantic_type, yy::parser::location_type, and
	yy::parser::token::MY_TOKEN, instead of YYSTYPE, YYLTYPE and MY_TOKEN.

* data/skeletons/glr.cc, data/skeletons/lalr1.cc: Remove its support.
	* tests/actions.at, tests/c++.at, tests/calc.at: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: don't assume the terminal supports "•"
	Use of print_unicode_char suggested by Bruno Haible.
	https://lists.gnu.org/r/bug-gettext/2020-06/msg00012.html

* src/gram.h (print_dot_fallback, print_dot): New.
	* src/gram.c, src/derivation.c: Use it.
	* tests/counterexample.at, tests/report.at: Adjust the test suite.
	* .travis.yml, README-hacking.md: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: also include in the report on --report=counterexamples
	And let --report=all include the counterexamples.

* src/getargs.h, src/getargs.c (report_cex): New.
	* src/main.c: Compute counterexamples when -rcex is specified.
	* src/print.c: Include the counterexamples when -rcex is specified.

* tests/conflicts.at, tests/existing.at, tests/local.at: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: also include the counterexamples in the report
	The report is the best place to show the details about
	counterexamples, since we have the state right under the nose.

For instance:

State 7

1 exp: exp . "⊕" exp
	    2    | exp . "+" exp
	    2    | exp "+" exp .  [$end, "+", "⊕"]
	    3    | exp . "+" exp
	    3    | exp "+" exp .  [$end, "+", "⊕"]

"⊕"  shift, and go to state 6

$end      reduce using rule 2 (exp)
	    $end      [reduce using rule 3 (exp)]
	    "+"       reduce using rule 2 (exp)
	    "+"       [reduce using rule 3 (exp)]
	    "⊕"       [reduce using rule 2 (exp)]
	    "⊕"       [reduce using rule 3 (exp)]
	    $default  reduce using rule 2 (exp)

Conflict between rule 2 and token "+" resolved as reduce (%left "+").

Shift/reduce conflict on token "⊕":
	        2 exp: exp "+" exp .
	        1 exp: exp . "⊕" exp
	      Example                  exp "+" exp • "⊕" exp
	      First derivation         exp ::=[ exp ::=[ exp "+" exp • ] "⊕" exp ]
	      Example                  exp "+" exp • "⊕" exp
	      Second derivation        exp ::=[ exp "+" exp ::=[ exp • "⊕" exp ] ]

Reduce/reduce conflict on tokens $end, "+", "⊕":
	        2 exp: exp "+" exp .
	        3 exp: exp "+" exp .
	      Example                  exp "+" exp •
	      First derivation         exp ::=[ exp "+" exp • ]
	      Example                  exp "+" exp •
	      Second derivation        exp ::=[ exp "+" exp • ]

Shift/reduce conflict on token "⊕":
	        3 exp: exp "+" exp .
	        1 exp: exp . "⊕" exp
	      Example                  exp "+" exp • "⊕" exp
	      First derivation         exp ::=[ exp ::=[ exp "+" exp • ] "⊕" exp ]
	      Example                  exp "+" exp • "⊕" exp
	      Second derivation        exp ::=[ exp "+" exp ::=[ exp • "⊕" exp ] ]

* src/conflicts.h, src/conflicts.c (has_conflicts): New.
	* src/counterexample.h, src/counterexample.c (print_counterexample):
	Add a `prefix` argument.
	(counterexample_report_shift_reduce)
	(counterexample_report_reduce_reduce): Show the items when there's a
	prefix.
	* src/state-item.h, src/state-item.c (print_state_item):
	Add a `prefix` argument.
	* src/derivation.h, src/derivation.c (derivation_print)
	(derivation_print_leaves): Add a prefix argument.
	* src/print.c (print_state): When -Wcex is enabled, show the
	conflicts.
	* tests/report.at: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: indent the diagnostics to highlight the structure
	Instead of

Shift/reduce conflict on token D:
	    Example              A a • D
	    First derivation     s ::=[ A a a ::=[ b ::=[ c ::=[ • ] ] ] d ::=[ D ] ]
	    Example              A a • D
	    Second derivation    s ::=[ A a d ::=[ • D ] ]

display

Shift/reduce conflict on token D:
	      Example              A a • D
	      First derivation     s ::=[ A a a ::=[ b ::=[ c ::=[ • ] ] ] d ::=[ D ] ]
	      Example              A a • D
	      Second derivation    s ::=[ A a d ::=[ • D ] ]

* src/counterexample.c (print_counterexample): Indent.
	* tests/counterexample.at: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: don't report the items
	Showing the items (with the state numbers) is really something we
	should restrict to the report.

* src/counterexample.c (counterexample_report_shift_reduce)
	(counterexample_report_reduce_reduce): Don't show the pointed rules,
	we will do that in the report.
	* tests/counterexample.at: Adjust.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: make sure traces go to stderr
	* src/parse-simulation.h, src/parse-simulation.c (print_parse_state):
	here.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

cex: add an argument to the reporting functions to specify the stream
	* src/conflicts.c (find_state_item_number, report_state_counterexamples):
	Move to...
	* src/counterexample.h, src/counterexample.c (find_state_item_number)
	(counterexample_report_state): this.
	Add support for `out` as an argument.
	(counterexample_report_reduce_reduce, counterexample_report_shift_reduce):
	Accept an `out` argument, and be static.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

style: more uses of const
	* src/print.c, src/state.h, src/state.c: here.

2020-06-16  Akim Demaille  <akim.demaille@gmail.com>

Merge 'maint'
	* upstream/maint:
	  maint: post-release administrivia
	  version 3.6.4
	  glr.cc: don't leak glr.c/glr.cc scaffolding to the user

Some fixes were needed to adjust to recent changes in glr.cc and
	glr.c.

* data/skeletons/glr.cc: Stop messing with the user's epilogue to
	insert glr.cc code.  We need that code to be inserted _before_ the
	user's epilogue, not after.  So define b4_glr_cc_pre_epilogue.
	* data/skeletons/glr.c: Use it.

2020-06-15  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-06-15  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.4
	* NEWS: Record release date.

2020-06-15  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: don't leak glr.c/glr.cc scaffolding to the user
	Until we have a decent reimplementation of glr.cc, we have to use
	tricks to shoehorn C++ symbols to the C engine of glr.c.  Some of them
	are done via #define.  Unfortunately in Bison 3.6 some of these we
	done in the header file, which broke valid user code.

Reported by Egor Pugin.
	https://lists.gnu.org/r/bug-bison/2020-06/msg00003.html

* data/skeletons/glr.cc: Stop playing tricks with b4_pre_epilogue.
	(b4_glr_cc_setup, b4_glr_cc_cleanup): New.
	Much cleaner way to instal glr.cc's scaffolding around glr.c.
	* data/skeletons/glr.c: Adjust to use them.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

reports: the column width differs from the byte count
	From

"number"          shift, and go to state 1
	    "Ñùṃéℝô"  shift, and go to state 2

"number"  shift, and go to state 1
	    "Ñùṃéℝô"  shift, and go to state 2

* src/print.c: Use mbswidth, not strlen, to compute visual columns.
	* tests/report.at: Adjust.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

reports: don't escape the labels
	Currently we use "quotearg" to escape the strings output in Dot.  As a
	result, if the user's locale is C for instance, all the non-ASCII are
	escaped.  Unfortunately graphviz does not interpret this style of
	escaping.

For instance:

5 -> 2 [style=solid label="\"\303\221\303\271\341\271\203\303\251\342\204\235\303\264\""]

was displayed as a sequence of numbers.  We now output:

5 -> 2 [style=solid label="\"Ñùṃéℝô\""]

independently of the user's locale.

* src/system.h (obstack_backslash): New.
	* src/graphviz.h, src/graphviz.c (escape): Remove, use
	obstack_backslash instead.
	* src/print-graph.c: Likewise.
	* tests/report.at: Adjust.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

parser: keep string aliases as the user wrote it
	Currently our scanner decodes all the escapes in the strings, and we
	later reescape the strings when we emit them.

This is troublesome, as we do not respect the user input.  For
	instance, when the user writes in UTF-8, we destroy her string when we
	write it back.  And this shows everywhere: in the reports we show the
	escaped string instead of the actual alias:

0 $accept: . exp $end
	    1 exp: . exp "\342\212\225" exp
	    2    | . exp "+" exp
	    3    | . exp "+" exp
	    4    | . "number"
	    5    | . "\303\221\303\271\341\271\203\303\251\342\204\235\303\264"

"number"                                                    shift, and go to state 1
	    "\303\221\303\271\341\271\203\303\251\342\204\235\303\264"  shift, and go to state 2

This commit preserves the user's exact spelling of the string aliases,
	instead of interpreting the escapes and then reescaping.  The report
	now shows:

0 $accept: . exp $end
	    1 exp: . exp "⊕" exp
	    2    | . exp "+" exp
	    3    | . exp "+" exp
	    4    | . "number"
	    5    | . "Ñùṃéℝô"

"number"          shift, and go to state 1
	    "Ñùṃéℝô"  shift, and go to state 2

Likewise, the XML (and therefore HTML) outputs are fixed.

* src/scan-gram.l (STRING, TSTRING): Do not interpret the escapes in
	the resulting string.
	* src/parse-gram.y (unquote, parser_init, parser_free, unquote_free)
	(handle_defines, handle_language, obstack_for_unquote): New.
	Use them to unquote where needed.
	* tests/regression.at, tests/report.at: Update.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

tests: check reports with conflicts and UTF-8
	This is to record the current state of the report, which escapes the
	UTF-8 characters (as parse.error="verbose" does), but shouldn't (as
	parse.error="detailed" does).

* tests/report.at: here.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

style: factor common bits about string scanning
	* src/scan-gram.l: here.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

style: introduce & use STRING_1GROW
	* src/flex-scanner.h (STRING_1GROW): New.
	* src/scan-gram.l, src/scan-skel.l: Use it.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes
	* src/scan-gram.l (STRING_GROW_ESCAPE): Move the static_assert about
	type sizes here.

2020-06-13  Akim Demaille  <akim.demaille@gmail.com>

style: prefer 'FOO ()' to 'FOO' for function-like macros
	* src/flex-scanner.h (STRING_GROW, STRING_FINISH, STRING_FREE):
	Make them function-like macros.
	Adjust dependencies.

2020-06-11  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-06-10  Akim Demaille  <akim.demaille@gmail.com>

cex: suggest -Wcounterexamples when there are unexpected conflicts
	Suggesting -Wcounterexamples when there are conflicts is probably not
	what the user wants.  If she knows her conflicts and has set
	%expect/%expect-rr appropriately, we shouldn't warn.

The commit also swaps the counterexamples and the report of conflicts,
	into, IMHO, a more natural order: from

Shift/reduce conflict on token B:
	    1:    3 a: A .
	    1:    8 y: A . B
	    Example              A • B C
	    First derivation     s ::=[ a ::=[ A • ] x ::=[ B C ] ]
	    Example              A • B C
	    Second derivation    s ::=[ y ::=[ A • B ] c ::=[ C ] ]

input.y: warning: 1 shift/reduce conflict [-Wconflicts-sr]
	    input.y:4.4: warning: rule useless in parser due to conflicts [-Wother]

input.y: warning: 1 shift/reduce conflict [-Wconflicts-sr]
	    Shift/reduce conflict on token B:
	    1:    3 a: A .
	    1:    8 y: A . B
	    Example              A • B C
	    First derivation     s ::=[ a ::=[ A • ] x ::=[ B C ] ]
	    Example              A • B C
	    Second derivation    s ::=[ y ::=[ A • B ] c ::=[ C ] ]

input.y:4.4: warning: rule useless in parser due to conflicts [-Wother]

* src/conflicts.c (rule_conflicts_print): Rename as...
	(report_rule_expectation_mismatches): this.
	Move the handling of report_counterexamples to...
	(conflicts_print): Here.
	Display this warning when applicable.

2020-06-10  Akim Demaille  <akim.demaille@gmail.com>

cex: rename -Wcounterexample as -Wcounterexamples, and support -Wcex
	Plural vs. singular is always a problem...

But we already have conflicts-sr and conflicts-rr, so counterexamples
	makes more sense than counterexample.  Besides, -Wcounterexample will
	still be accepted as an unambiguous prefix of -Wcounterexamples.

Add -Wcex as a convenient alias.

While at it, use only "counterexample", never "counter example".

* src/complain.h, src/complain.c
	(Wcounterexample, warning_counterexample): Rename as...
	(Wcounterexamples, warning_counterexamples): these.
	(argmatch_warning_docs): Rename -Wcounterexample as -Wcounterexamples.
	(argmatch_warning_args): Likewise.
	Add support for -Wcex.
	Adjust dependencies.

2020-06-09  Akim Demaille  <akim.demaille@gmail.com>

api.header.include: document it, and fix its default value
	While defining api.header.include worked as expected, its default
	value was incorrectly defined.  As a result, by default, the generated
	parsers still duplicated the content of the generated header instead
	of including it.

* data/skeletons/yacc.c (api.header.include): Fix its default value.
	* tests/output.at: Check it.
	* doc/bison.texi (%define Summary): Document api.header.include.
	While at it, move the definition of api.namespace at the proper
	place.

2020-06-07  Akim Demaille  <akim.demaille@gmail.com>

cex: color the counterexamples
	Use colors to show the counterexamples and the derivations in color,
	to highlight their structure.  Align the outputs, and add i18n
	support. Reduce width by using a one-space separator instead of
	two-space.

From

Example  A  •  B  C
	    First  derivation  s ::=[ a ::=[ A  • ]  x ::=[ B  C ] ]
	    Second derivation  s ::=[ y ::=[ A  •  B ]  c ::=[ C ] ]

Example              A • B C
	    First derivation     s ::=[ a ::=[ A • ] x ::=[ B C ] ]
	    Example              A • B C
	    Second derivation    s ::=[ y ::=[ A • B ] c ::=[ C ] ]

with colors.

* data/bison-default.css (cex-dot, cex-0, cex-1, cex-2, cex-3, cex-4)
	(cex-5, cex-6, cex-7, cex-step, cex-leaf): New.
	* src/derivation.c (derivation_print_styled_impl): New.
	(derivation_print, derivation_print_leaves): Use it.
	* src/counterexample.c: Reformat the output.
	* tests/counterexample.at: Adjust.

2020-06-07  Akim Demaille  <akim.demaille@gmail.com>

cex: enforce case for tokens/nonterminals
	It's unfortunate that the traditions between formal language theory
	and Yacc differs, but here, tokens should be upper case, and
	nonterminals should be lower case.

* tests/counterexample.at: Comply with this.

2020-06-07  Akim Demaille  <akim.demaille@gmail.com>

cex: reformat the s/r and r/r reports
	In Bison we refer to "shift/reduce" conflicts, not "shift-reduce" (in
	Bison 3.6.3 186 occurrences vs 15).  Enforce consistency on this.

Instead of "spending" a second line for each conflict to report the
	lookaheads, put that on the same line as the type of conflict.  Also,
	prefer "token" to "symbol".  Maybe we should even prefer "lookahead".
	While at it, enable internationalization, with plurals where
	appropriate.

As a consequence, instead of

Shift-Reduce Conflict:
	    6:    3 b: . %empty
	    6:    6 d: c . A
	    On Symbol: A

display

Shift/reduce conflict on token A:
	    6:    3 b: . %empty
	    6:    6 d: c . A

* NEWS, doc/bison.texi, src/conflicts.c: Spell it "shift/reduce", not
	"shift-reduce".
	* src/counterexample.c (counterexample_report_shift_reduce)
	(counterexample_report_reduce_reduce): Reformat and internationalize
	output.
	* tests/counterexample.at: Adjust expectations.

2020-06-07  Akim Demaille  <akim.demaille@gmail.com>

style: fix syntax-check issues
	* src/counterexample.c, src/files.c, src/files.h, src/lssi.c,
	* src/state-item.c: here.

2020-06-07  Akim Demaille  <akim.demaille@gmail.com>

all: show the rules in comments before the user actions
	For instance, in the case of Bison's own parser:

-  case 40:
	    +  case 40:  /* grammar_declaration: "%code" "identifier" "{...}"  */
	         {
	           muscle_percent_code_grow ((yyvsp[-1].ID), (yylsp[-1]),
	                                     translate_code_braceless ((yyvsp[0].BRACED_CODE), (yylsp[0])),
	                                     (yylsp[0]));
	           code_scanner_last_string_free ();
	         }
	         break;

* data/skeletons/c.m4: Modified.
	* data/skeletons/d.m4: Modified.
	* data/skeletons/java.m4: Modified.
	* src/output.c (output_escaped): New.
	(quoted_output): Use it, and rename as...
	(output_quoted): this.
	Adjust dependencies.
	(rule_output): New.
	(user_actions_output): Use it.
	* data/skeletons/c.m4, data/skeletons/d.m4, data/skeletons/java.m4
	(b4_case): Add support for $3, an optional comment.

2020-06-06  Akim Demaille  <akim.demaille@gmail.com>

CI: use GCC10 on ppc too
	We were still using GCC9, because GCC10 was failing.

* .travis.yml (PPC64le): Use GCC10.
	While at it, use -O2 instead of -O3: it's certainly nicer for the
	CPUs, and allows to test different sets of compiler flags (we use -O3
	in several other configurations).

2020-06-06  Akim Demaille  <akim.demaille@gmail.com>

examples: fix missing includes
	* examples/c/bistromathic/parse.y: Use abort rather than assert so
	that the "unused result" warning is silenced even with -DNDEBUG.

2020-06-03  Akim Demaille  <akim.demaille@gmail.com>

warnings: fix -Wmissing-prototypes issues
	* src/counterexample.c, src/lssi.c, src/parse-simulation.c,
	* src/state-item.c:
	Here.

2020-06-03  Akim Demaille  <akim.demaille@gmail.com>

Merge maint into HEAD
	* upstream/maint:
	  maint: post-release administrivia
	  version 3.6.3
	  build: check -Wmissing-prototypes
	  tests: show logs
	  c++: fix printing of state number on streams

2020-06-03  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-06-03  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.3
	* NEWS: Record release date.

2020-06-01  Akim Demaille  <akim.demaille@gmail.com>

build: check -Wstrict-aliasing
	* configure.ac (warn_common): Add -Wstrict-aliasing.

2020-06-01  Akim Demaille  <akim.demaille@gmail.com>

doc: using asan
	* README-hacking.md: here.

2020-06-01  Akim Demaille  <akim.demaille@gmail.com>

style: fix includes
	* src/fixits.c: Follow our usual pattern.
	* src/scan-code.l, src/scan-gram.l, src/scan-skel.l: Prefer "" to
	include src/ headers.
	* README-hacking.md: Document the pattern.

2020-06-01  Akim Demaille  <akim.demaille@gmail.com>

lists: fix various issues with the use of gnulib's list
	First, we should avoid code such as

gl_list_iterator_t it = gl_list_iterator (deriv->children);
	    derivation *child = NULL;
	    while (gl_list_iterator_next (&it, (const void **) &child, NULL))
	      {
	        derivation_print (child, f);

because of -Wstrict-aliasing (whose job is to catch type-punning
	issues).  See https://lists.gnu.org/r/bug-bison/2020-05/msg00039.html.

Rather we need

gl_list_iterator_t it = gl_list_iterator (deriv->children);
	    const void **p = NULL;
	    while (gl_list_iterator_next (&it, &p, NULL))
	      {
	        derivation *child = (derivation *) p;
	        derivation_print (child, f);

Second, list iterators actually have destructors.  Even though they
	are noop in the case of linked-lists, we should use them.

Let's address both issues with typed wrappers (such as
	derivation_list_next) that take care of both issues, and besides allow
	to scope the iterators within the loop:

derivation *child;
	    for (gl_list_iterator_t it = gl_list_iterator (deriv->children);
	         derivation_list_next (&it, &child);
	         )
	      {
	        derivation_print (child, f);

* src/derivation.h, src/derivation.c (derivation_list_next): New.
	Use it where appropriate.
	* src/counterexample.c (search_state_list_next): New.
	Use it where appropriate.
	* src/parse-simulation.h, src/parse-simulation.c
	* src/state-item.h (state_item_list_next): New.
	Use it where appropriate.

2020-06-01  Akim Demaille  <akim.demaille@gmail.com>

build: check -Wmissing-prototypes
	pstate_clear is lacking a prototype.
	Reported by Ryan
	https://lists.gnu.org/r/bug-bison/2020-05/msg00101.html

Besides, none of the C examples were compiled with the warning flags.

* configure.ac (warn_c): Add -Wmissing-prototypes.
	* data/skeletons/yacc.c (pstate_clear): Make it static.
	* examples/local.mk (TEST_CFLAGS): New.
	* examples/c/bistromathic/local.mk, examples/c/calc/local.mk,
	* examples/c/lexcalc/local.mk, examples/c/mfcalc/local.mk,
	* examples/c/pushcalc/local.mk, examples/c/reccalc/local.mk,
	* examples/c/rpcalc/local.mk:
	Use it.

GCC's warn_unused_result is not silenced by a cast to void, so we have
	to "use" scanf's result.
	https://gcc.gnu.org/bugzilla/show_bug.cgi?id=25509
	https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

Flex generated code produces too many warnings, including things such
	as, with ICC:

examples/c/lexcalc/scan.c(1088): error #1682: implicit conversion
	              of a 64-bit integral type to a smaller integral type (potential portability problem)
	    2259                YY_INPUT( (&YY_CURRENT_BUFFER_LVALUE->yy_ch_buf[number_to_move]),
	    2260                ^
	    2261
	    2262

I am tired of trying to fix Flex's output.  The project does not seem
	maintained.  We ought to avoid it.  So, for the time being, don't try
	to enable warnings with Flex.

* examples/c/bistromathic/parse.y, examples/c/reccalc/scan.l: Fix
	warnings.
	* doc/bison.texi: Discard scanf's return value to defeat
	-Werror=unused-result.

2020-05-31  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-25  Akim Demaille  <akim.demaille@gmail.com>

style: make item_index a truly different type from item_number
	See previous commit.

* src/gram.h (item_index): Make it unsigned.
	Fix remaiming issues.

2020-05-25  Vincent Imbimbo  <vmi6@cornell.edu>

style: decouple different uses of item_number
	item_number is used for elements of ritem as well as indices into
	ritem which is fairly confusing.  Introduce item_index to represent
	indices into ritem.

* src/gram.h (item_index): Introduce it for ritem indices.
	* src/closure.h, src/closure.c, src/ielr.c, src/lr0.c,
	* src/print-graph.c, src/state.h, src/state.h:
	Replace uses of item_number with item_index where appropriate.

2020-05-24  Joshua Watt  <jpewhacker@gmail.com>

bison: add command line option to map file prefixes
	Teaches bison about a new command line option, --file-prefix-map OLD=NEW
	(based on the -ffile-prefix-map option from GCC) which causes it to
	replace and file path of OLD in the text of the output file with NEW,
	mainly for header guards and comments. The primary use of this is to
	make builds reproducible with different input paths, and in particular
	the debugging information produced when the source code is compiled. For
	example, a distro may know that the bison source code will be located at
	"/usr/src/bison" and thus can generate bison files that are reproducible
	with the following command:

bison --output=/build/bison/parse.c -d --file-prefix-map=/build/bison/=/usr/src/bison/ parse.y

Importantly, this will change the header guards and #line directives
	from:

#ifndef YY_BUILD_BISON_PARSE_H
	    #line 100 "/build/bison/parse.h"

#ifndef YY_USR_SRC_BISON_PARSE_H
	    #line 100 "/usr/src/bison/parse.h"

which is reproducible.

See https://lists.gnu.org/r/bison-patches/2020-05/msg00016.html

* src/files.h, src/files.c (spec_mapped_header_file)
	(mapped_dir_prefix, map_file_name, add_prefix_map): New.
	* src/getargs.c (-M, --file-prefix-map): New option.
	* src/output.c (prepare): Define b4_mapped_dir_prefix and
	b4_spec_header_file.
	* src/scan-skel.l (@ofile@): Output the mapped file name.
	* data/skeletons/glr.c, data/skeletons/glr.cc,
	* data/skeletons/lalr1.cc, data/skeletons/location.cc,
	* data/skeletons/yacc.c:
	Adjust.
	* doc/bison.texi: Document.
	* tests/input.at, tests/output.at: Check.

2020-05-24  Akim Demaille  <akim.demaille@gmail.com>

tests: fix expectations
	Should have been part of 1ec93ca2a2b4718b5d94871475520a2688b4c5c8.

* tests/counterexample.at: here.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

cex: clean the display of conflicted symbols
	Instead of `On Symbols: {b,c,}`, display `On Symbols: b, c`.

* src/counterexample.c (counterexample_report_reduce_reduce): We don't
	need braces.
	Use commas as a separator, not a terminator.
	* tests/counterexample.at: Adjust.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

tests: show logs
	* examples/c/bistromathic/bistromathic.test, examples/test: here.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

c++: fix printing of state number on streams
	Avoid this kind of display:

LAC: checking lookahead identifier: R4 R3 G^B S5

* data/skeletons/lalr1.cc: Convert state_t to int before printing it.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

c++: fix printing of state number on streams
	Avoid this kind of display:

LAC: checking lookahead identifier: R4 R3 G^B S5

* data/skeletons/lalr1.cc: Convert state_t to int before printing it.

2020-05-23  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix pruning crash
	Fixes a crash on Cim's grammar.
	https://lists.gnu.org/r/bison-patches/2020-05/msg00107.html

* src/state-item.c (prune_disabled_paths): Prune forward and backwards
	paths in seperate passes.
	(prune_forward, prune_backward): New.
	(disable_state_item): Change function argument from state_item_number
	to state_item.
	(state_items_report): Add disabling to graph print-out.
	* src/conflicts.c (find_state_item_number,
	report_state_counterexamples): Add SI_DISABLED checks.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

kinds: use the symbol kinds where applicable
	Instead of generating switch statements with numbers, let's use the
	symbol kinds.  Not only is this more readable, it also makes reading
	diff easier, as a change in symbol numbers won't have such a large
	effect on the implementation of symbol actions.

* data/skeletons/bison.m4 (_b4_symbol_case): Use the symbol kind
	rather than its number.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

kinds: also define the possibly qualified symbol kinds
	* data/skeletons/bison.m4 (b4_symbol_kind): Rename as...
	(b4_symbol_kind_base): this.
	(b4_symbol_kind): New, for fully qualified kind name.
	* data/skeletons/lalr1.cc (b4_symbol_kind): New.
	Adjust to use b4_symbol_kind where appropriate.
	* src/parse-gram.h, src/parse-gram.c: regen.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

m4: simplify useless quotation
	* data/skeletons/bison.m4: The result of b4_symbol is "quoted"
	already, no need for m4_expand.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

m4: use m4_shift2 etc.
	* data/skeletons/bison.m4 (m4_shift4): New.
	Use them where applicable.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

tests: show logs
	* examples/c/bistromathic/bistromathic.test, examples/test: here.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

style: spell fixes
	* Makefile.am (codespell): New.
	* doc/bison.texi: Fixes.
	Use @option for options.
	* src/lssi.c, src/lssi.h, src/parse-simulation.h, src/state-item.c:
	Fix spellos.

2020-05-23  Akim Demaille  <akim.demaille@gmail.com>

style: rename user_token_number as code
	This should have been done in 3.6, but I wanted to avoid introducing
	conflicts into Vincent's work on counterexamples.  It turns out it's
	completely orthogonal.

* data/README.md, data/skeletons/bison.m4, data/skeletons/c++.m4,
	* data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/java.m4,
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java,
	* data/skeletons/variant.hh, data/skeletons/yacc.c, src/conflicts.c,
	* src/derives.c, src/gram.c, src/gram.h, src/output.c,
	* src/parse-gram.c, src/parse-gram.y, src/print-xml.c, src/print.c,
	* src/reader.c, src/symtab.c, src/symtab.h, tests/input.at,
	* tests/types.at:
	s/user_token_number/code/g.
	Plus minor changes.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

Merge maint into master
	* upstream/maint:
	  fix generated comments
	  traces: provide a means to get short m4 traces
	  traces: show the full m4 invocation

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: replace state-item data structures
	* src/state-item.h: Add trans, prods, and revs edges to state-item
	struct.
	(si_trans, si_revs, si_prods_lookup): Remove.
	* src/state-item.c, src/lssi.c, src/parse-simulation.c,
	* src/counterexample.c: Update state-item API usage accordingly.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix bad reference counting
	* src/counterexample.c (si_bfs_free): Fix reference_count
	decrementing.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix miscellaneous leaks
	* src/counterexample.c (unifying_counterexample): Always free
	stage3result when it exists.
	* src/conflicts.c (report_state_counterexamples): free leaked bitset.
	* src/state-item.c (prune_disabled_paths): free leaked queue.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix counterexample leak
	* src/counterexample.c (free_counterexample): New.
	Free counterexamples after printing.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix lssi leaks
	* src/lssi.c (shortest_path_from_start): Free the root of
	shortest_path_from_start search.
	Free eligible bitset.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix parse state leaks
	* src/parse_simulation.c: Fix bug in parse_state_free.
	Free new_root when simulate_reduction generates zero states.

* src/parse-simulation.c, src/parse-simulation.h
	(parse_state_list, parse_state_list_append): New.
	* src/parse-simulation.c, src/parse-simulation.h,
	* src/counterexample.c: Replace all uses of lists of parse states and
	appends to parse_state_lists with the new API.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: derivation reference counting
	* src/derivation.h, src/derivation.c: Make derivation struct opaque.
	Add derivation_list type for clarity.
	(derivation_list_new): New.
	(derivation_list_append): New.
	(derivation_list_prepend): New.
	(derivation_new_leaf): New constructor for derivations with no
	children.
	* src/counterexample.c, src/parse-simulation.c,
	* src/parse-simulation.h: Replace uses of gl_list_t containing
	derivations with derivation_list and its API.
	Replace calls of dervation_new using null children with
	derivation_new_leaf.
	* src/parse-simulation.c: replace ps_chunk and its API with typed
	versions si_chunk and deriv_chunk.
	* src/parse-simlation.h, src/parse-simulation.c: Remove
	parse_state_retain_deriv in favor of derivation reference counting.
	* src/counterexample.c: Remove search_state_retain_deriv.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: style changes in parse-simulation
	* src/parse-simulation.c: Formatting changes.
	(parse_state_list_new): New.
	Use it.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: style: prefer res for returned value
	* src/lssi.c, src/parse-simulation.c: here.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: fix memory leaks when there are conflicts
	* src/counterexample.c (production_step, reduction_step): Release
	memory of temporary objects.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: be sure to always reclaim memory put in hashes
	One call to hash_initialize did not provide a function to free memory.

* src/state-item.c (hash_pair_table_create): New.
	Use it.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: properly reclaim hash's allocated memory
	* src/state-item.c: Use hash_free where appropriate.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: avoid gratuitous heap allocations
	There's no need to go for the heap when using gnulib's hash module.

* src/state-item.c (hash_pair_lookup, hash_pair_remove,
	state_sym_lookup): Use the heap Luke.
	That removes a leak from hash_pair_lookup.
	(init_prods): Use hash_pair_insert instead of duplicating it.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix leaks
	* src/state-item.c: Various functions were using heap allocated locals
	and not freeing them.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

style: use hash_xinsert
	* gnulib: Update to get hash_xinsert.
	Use it where appropriate.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: style changes in state-item
	* src/state-item.h, src/state-item.c (state_item): Make the state
	const.
	(state_item_set): Make it clearer that it works in the state_items
	global array.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: stylistic changes
	* src/counterexample.c: Use 'res' as a variable name for returned
	value, as elsewhere.
	Avoid uninitialized variables, especially pointers.
	Avoid assignment where possible.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: avoid uninitialized variables
	* src/counterexample.c (item_rule_bounds): Split into...
	(item_rule_start, item_rule_end): these.
	Adjust dependencies.
	* src/conflicts.c (find_state_item_number): New.
	Use it to avoid uninitialized variables.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: isolate missing API from gl_list
	* src/counterexample.c (list_get_end): New.
	Use it.
	Reduce scopes.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: tests: be robust to variations in time limit reports
	The CI has "failures" such as (253, "Null nonterminals"):

@@ -21,7 +21,7 @@
	     3:    3 b: . %empty
	     3:    4 c: . %empty
	     On Symbols: {A,}
	    -time limit exceeded: 6.000000
	    +time limit exceeded: 11.000000
	     First  Example c  •  c  A  A  $end
	     First  derivation  $accept ::=[ a ::=[ c  d ::=[ a ::=[ b ::=[ • ]  d ::=[ c  A  A ] ] ] ]  $end ]
	     Second Example c  •  A  $end

* tests/counterexample.at (AT_BISON_CHECK_CEX): New.
	Use it to neutralize differences in timeout values.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix stack overflow
	* src/parse-simulation.c: Replace reference counting with
	parse_state_retain everywhere.
	(free_parse_state): Make this function iterative instead of
	recursive. Long parse_state chains were causing stack exhaustion.

* tests/counterexample.at: Fix expectations.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: fix crash from zombie result
	Fixes the SEGV in test 247 (counterexample.at:195): "S/R after first
	token".

* src/counterexample.c: here.
	* tests/counterexample.at: Fix expectations.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: fixes, and enable tests
	* src/counterexample.c, src/derivation.c:
	Do not output diagnostics on stdout, that's the job of stderr, and the
	testsuite heavily depend on this.
	Do not leave trailing spaces in the output.
	* tests/counterexample.at: Use AT_KEYWORDS.
	Specify the expected outputs.
	* tests/local.mk: Add counterexample.at.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: fix a crash
	* src/state-item.c (init_state_items): If the rule has no reductions
	at all, don't read at all in its list of reduced rules.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: add tests
	* tests/counterexample.at: New.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: bind counterexample generation
	* src/complain.h, src/complain.c: Add support for -Wcounterexample.
	* src/conflicts.c (report_counterexamples): New.
	(rule_conflicts_print): Use it when -Wcounterexample is given.
	* src/getargs.h, src/getargs.c: Add support for --trace=cex.
	* src/main.c (main): Init and deinit counterexample generation.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: introduce counterexample search
	* src/counterexample.h, src/counterexample.c: New.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: introduce the parse simulator
	* src/derivation.h, src/derivation.c,
	* src/parse-simulation.h, src/parse-simulation.c: New.

2020-05-22  Vincent Imbimbo  <vmi6@cornell.edu>

cex: add support for state-item pair graph generation
	* src/lssi.h, src/lssi.c, src/state-item.h, src/state-item.c: New.

2020-05-22  Akim Demaille  <akim.demaille@gmail.com>

cex: add gnulib dependencies
	* bootstrap.conf (gnulib_modules): Add linked-list.

2020-05-21  Akim Demaille  <akim.demaille@gmail.com>

fix generated comments
	In Bison 3.6.2, the comments with brackets lose their brackets, for
	improper m4 quotation.

* data/skeletons/bison.m4 (b4_gsub): New.
	* data/skeletons/c-like.m4 (_b4_comment): Use it.
	* tests/m4.at: Check b4_gsub.

2020-05-21  Akim Demaille  <akim.demaille@gmail.com>

traces: provide a means to get short m4 traces
	Let --trace=m4-early dump all the logs from the start (as --trace=m4
	used to do), and have --trace=m4 now start traces only when actually
	working of the user's grammar.

Can make a big difference in the case of small inputs.  E.g.

$ bison -S tests/testsuite.dir/001/input.m4 tests/testsuite.dir/001/input.y --trace=m4 |& wc
	       3952   19446  251068
	    $ bison -S tests/testsuite.dir/001/input.m4 tests/testsuite.dir/001/input.y --trace=m4-early |& wc
	      19491  131904 1830495

* data/skeletons/traceon.m4: New.
	* src/getargs.h, src/getargs.c: Introduce --trace=m4-early.
	* src/output.c (output_skeleton): Adjust for --trace=m4 and --trace=m4-early.

2020-05-21  Akim Demaille  <akim.demaille@gmail.com>

traces: show the full m4 invocation
	Unfortunately the effect of -dV is still position independent.

* src/output.c (output_skeleton): here.

2020-05-20  Thomas Petazzoni  <thomas.petazzoni@bootlin.com>

src: make path to m4 relocatable
	Commit a4ede8f85b0c9a254fcb01e5888cee1983095669 ("package: make bison
	a relocatable package") made Bison relocatable, but in fact it still
	contains one absolute reference: the M4 variable, which points to the
	M4 program. Let's fix that by using relocate(), see if an M4 binary is
	available at the relocated location, and otherwise fallback to the
	original M4 location.

See https://lists.gnu.org/r/bison-patches/2020-05/msg00078.html,
	and https://lists.gnu.org/r/bison-patches/2020-05/msg00087.html.

* src/files.h, src/files.c (m4path): New.
	* src/output.c: Use it.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

CI: fix PPC recipe

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

Merge branch 'maint'
	* upstream/maint:
	  maint: post-release administrivia
	  version 3.6.2
	  tests: improve update-test
	  CI: add GCC 10 and Clang 10
	  fix: do not emit nested comments
	  todo: update
	  examples: use markdown hyperlinks
	  tests: don't use == to compare const char *...
	  gnulib: update

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

c: more fixes for _Noreturn
	The previous fix was insufficient.

tests/types.at:366: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o test test.cc $LIBS
	    ++ ccache clang++-mp-9.0 -Qunused-arguments -ggdb -Wall -Wextra -Wcast-align -Wchar-subscripts -fparse-all-comments -Wdocumentation -Wformat -Wimplicit-fallthrough -Wnull-dereference -Wno-sign-compare -Wno-tautological-constant-out-of-range-compare -Wpointer-arith -Wshadow -Wwrite-strings -Wextra-semi -Wold-style-cast -Wundefined-func-template -Wweak-vtables -Wunreachable-code -Wundef -pedantic -Wconversion -Wdeprecated -Wsign-compare -Wsign-conversion -Wtautological-constant-out-of-range-compare -fno-color-diagnostics -Wno-keyword-macro -Werror -std=c++98 -I/Users/akim/src/gnu/bison/tests -isystem /opt/gostai/include -isystem /opt/local/include -L/opt/gostai/lib -L/opt/local/lib -o test test.cc /Users/akim/src/gnu/bison/_build/c9d/lib/libbison.a -lintl -Wl,-framework -Wl,CoreFoundation
	    stderr:
	    test.cc:955:1: error: _Noreturn functions are a C11-specific feature [-Werror,-Wc11-extensions]
	    _Noreturn static void
	    ^
	    test.cc:963:1: error: _Noreturn functions are a C11-specific feature [-Werror,-Wc11-extensions]
	    _Noreturn static void
	    ^
	    2 errors generated.

* data/skeletons/c.m4 (b4_attribute_define): Do not use _Noreturn at
	all in C++, clang or not.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.2
	* NEWS: Record release date.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

tests: improve update-test
	* build-aux/update-test: When given a directory, use the testsuite.log
	which it contains.
	Do not accept empty "from"s, as substituting the empty string with
	something is rarely a good idea.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

CI: add GCC 10 and Clang 10
	* .travis.yml: Here.
	* tests/input.at, tests/regression.at: Beware of clang's -Wdocumentation.

2020-05-17  Akim Demaille  <akim.demaille@gmail.com>

fix: do not emit nested comments
	With input such as

%token<fl> yVL_CLOCK "/*verilator sc_clock*/"

we generate

yVL_CLOCK = 610,      /* "/*verilator sc_clock*/"  */

which is invalid since the comment will actually be closed on the
	first "*/".  Let's turn "*/" into "*\/" to avoid this.  But GCC will
	also warn about "/*" inside a comment, so let's "escape" it too.

Reported by Huang Rui.
	https://github.com/akimd/bison/issues/38

* data/skeletons/c-like.m4 (_b4_comment): Escape comment delimiters in
	comments.
	* tests/input.at (Torturing the Scanner): Check thes cases.
	* tests/m4.at: New.

2020-05-16  Akim Demaille  <akim.demaille@gmail.com>

c: restore definition of _Noreturn as [[noreturn]] in C++
	c.m4 contains a definition of _Noreturn which is modeled after
	gnulib's lib/_Noreturn.h.  The latter was recently
	changed (b61bf2f0e8bdc1e522ae8e97d57d5625163b42ea) to not using
	[[noreturn]] at all, because the uses of _Noreturn in gnulib are
	sometimes incompatible with the rules of [[noreturn]].

As a result glr.cc started to use _Noreturn in C++, which clang
	refuses (all the glr.cc tests currently fail with Clang++).

* data/skeletons/c.m4 (b4_attribute_define): Restore the definition of
	_Noreturn as [[noreturn]] in modern C++.
	The generated code uses _Noreturn in places where [[noreturn]] is
	valid.

2020-05-16  Akim Demaille  <akim.demaille@gmail.com>

examples: don't promote unchecked function calls
	* etc/bench.pl.in, examples/c/bistromathic/parse.y,
	* examples/c/calc/calc.y, examples/c/pushcalc/calc.y: Check scanf's
	return value.
	* doc/bison.texi: Likewise, but only for the second example, to avoid
	cluttering the very simple case.

2020-05-16  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-15  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-05-14  Akim Demaille  <akim.demaille@gmail.com>

examples: use markdown hyperlinks
	* examples/c++/README.md, examples/c++/calc++/README.md,
	* examples/c/README.md: here.

2020-05-14  Akim Demaille  <akim.demaille@gmail.com>

tests: don't use == to compare const char *...
	Reported by Dagobert Michelsen.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00091.html

* tests/c++.at: here.

2020-05-14  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-13  Akim Demaille  <akim.demaille@gmail.com>

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

Merge branch maint
	* maint:
	  news: update
	  maint: post-release administrivia
	  version 3.6.1
	  c++: style: reorder generated code
	  c++: provide yy::parser::symbol_type::name
	  c++: make parser::symbol_name public
	  examples: beware of ~/.inputrc
	  build: also provide lzip compressed tarballs
	  style: minor fixes
	  yacc.c: restore ansi-c compatibility

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

news: update

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

version 3.6.1
	* NEWS: Record release date.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

bench: add support to randomize the order of execution
	It's amazing how much the order matters.  To a point that many of
	these benches are meaningless.  For instance (some of the benches
	where run with `make -C benches/latest rand
	BENCHFLAGS=--benchmark_min_time=3`):

compiler: g++ -std=c++11 -O2
	      0. %define nofinal
	      1.
	    --------------------------------------------------
	    Benchmark           Time           CPU Iterations
	    --------------------------------------------------
	    BM_y0            1543 ns       1541 ns     441660
	    BM_y1            1521 ns       1520 ns     456535
	    --------------------------------------------------
	    BM_y0            1531 ns       1530 ns     440584
	    BM_y1            1512 ns       1511 ns     457591
	    --------------------------------------------------
	    BM_y0            1539 ns       1538 ns    2749330
	    BM_y1            1516 ns       1515 ns    2771500
	    --------------------------------------------------
	    BM_y0            1571 ns       1570 ns    2600782
	    BM_y1            1542 ns       1541 ns    2708349
	    --------------------------------------------------
	    BM_y0            1530 ns       1529 ns    2670363
	    BM_y1            1519 ns       1518 ns    2764096

--------------------------------------------------
	    Benchmark           Time           CPU Iterations
	    --------------------------------------------------
	    BM_y1            1529 ns       1528 ns     451937
	    BM_y0            1508 ns       1507 ns     453944
	    --------------------------------------------------
	    BM_y1            1525 ns       1524 ns    2750684
	    BM_y0            1516 ns       1515 ns    2794034
	    --------------------------------------------------
	    BM_y1            1526 ns       1525 ns    2749620
	    BM_y0            1515 ns       1514 ns    2808112
	    --------------------------------------------------
	    BM_y1            1524 ns       1523 ns    4475844
	    BM_y0            1502 ns       1501 ns    4611665

* etc/bench.pl.in: here.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

bench: use a Makefile
	This makes it much easier to toy with the benchs.

* etc/bench.pl.in: Generate a Makefile instead of directly compiling
	the files.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

examples: use markdown hyperlinks
	* examples/c++/README.md, examples/c++/calc++/README.md,
	* examples/c/README.md: here.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

don't use stdnoreturn
	Reported by Paul Eggert.

* src/getargs.c: We don't need it anyway, since we use _Noreturn.
	* data/skeletons/c.m4: While at it, update the definition of _Noreturn
	stolen from gnulib.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

c++: style: reorder generated code
	The implementation of yy::parser::symbol_name is emitted even before
	the implementation of yy::parser::parser.  This makes little sense.

* data/skeletons/lalr1.cc (symbol_name): Move its implementation in
	the same place as in the class definition: after "error" and before
	"context".

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

c++: provide yy::parser::symbol_type::name
	* data/skeletons/c++.m4 (yy::parser::basic_symbol::name): New.
	* data/skeletons/lalr1.cc (yy_print_): Use it.
	* doc/bison.texi: Document.
	* tests/c++.at: Check.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

c++: make parser::symbol_name public
	Reported by Martin Blais <blais@furius.ca>.
	https://lists.gnu.org/r/help-bison/2020-05/msg00005.html

* data/skeletons/lalr1.cc (symbol_name): Make it public.
	Add a private hidden hook to enable testing of private parts.
	* tests/local.at (AT_DATA_GRAMMAR_PROLOGUE): Help Emacs find the right
	language mode.
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): Check that we
	can read symbol_name.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of ~/.inputrc
	* examples/c/bistromathic/bistromathic.test: here.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

build: also provide lzip compressed tarballs
	Suggested by Matias Fonzo <selk@dragora.org>.

* cfg.mk: Post announcements to bison-announce.
	* configure.ac: Build lzip packages.
	* .travis.yml: Build only xz, we don't care about the other formats
	here.

2020-05-10  Akim Demaille  <akim.demaille@gmail.com>

style: minor fixes
	* examples/c/README.md: here.

2020-05-09  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: restore ansi-c compatibility
	Reported by neok-m4700.
	https://github.com/akimd/bison/issues/37

* data/skeletons/yacc.c: Don't use // comments.

2020-05-09  Akim Demaille  <akim.demaille@gmail.com>

bench: use *.cc for C++
	Using *.c is simpler, but triggers annoying warnings with Clang++.

* etc/bench.pl.in: Please the dictator.

2020-05-09  Akim Demaille  <akim.demaille@gmail.com>

style: minor fixes
	* examples/c/README.md: here.

2020-05-09  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update, and use the attribute module
	* gnulib: Update.
	* bootstrap.conf: Use attribute.
	* src/system.h: Remove macros for attributes.
	Adjust dependencies.
	* src/scan-gram.l (DEPRECATED): Rename as...
	(DEPRECATED_DIRECTIVE): this, to avoid the clash with the DEPRECATED macro.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

build: also provide lzip compressed tarballs
	Suggested by Matias Fonzo <selk@dragora.org>.

* cfg.mk: Post announcements to bison-announce.
	* configure.ac: Build lzip packages.
	* .travis.yml: Build only xz, we don't care about the other formats
	here.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

version 3.6
	* NEWS: Record release date.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of portability issue on Windows
	Reported by Jannick.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00040.html
	https://lists.gnu.org/r/bug-bison/2020-05/msg00066.html

* examples/test (diff_opts): Use --strip-trailing-cr if supported, to
	avoid \n vs. \r\n issues.
	* examples/c/bistromathic/bistromathic.test: When on MSYS, don't try
	to check autocompletion.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

doc: fix the generation of the man page
	When there is no bison.1 at all, the procedure fails.

* doc/local.mk (bison.1): Be robust to cold starts.

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

news: prepare for 3.6

2020-05-08  Akim Demaille  <akim.demaille@gmail.com>

doc: complete the table of symbols
	* doc/bison.texi: Add YYEMPTY, YYEOF and YYUNDEF.

2020-05-07  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify the glossary item about kinds
	* doc/bison.texi (Glossary): here.

2020-05-06  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-06  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.94
	* NEWS: Record release date.

2020-05-06  Akim Demaille  <akim.demaille@gmail.com>

doc: document yypstate_expected_tokens
	* doc/bison.texi (Push Parser Interface): Here.

2020-05-06  Akim Demaille  <akim.demaille@gmail.com>

doc: restructure the push parser documentation
	I don't think it's fair to have yypstate_new, yypstate_delete,
	yypush_parse and yypull_parse to have their own section, on par with
	yyparse and yylex.  Let them be in a single section about push
	parsers.  And show new/delete first.

* doc/bison.texi (Push Parser Interface): New.
	Fuse the aforementioned sections into it.

2020-05-06  Akim Demaille  <akim@lrde.epita.fr>

all: fix the interface of yyexpected_tokens
	The user gives yyexpected_tokens a limit: the max number of tokens she
	wants to hear about.  That's because an error message that reports a
	bazillion of possible tokens is useless.

In that case yyexpected_tokens returned 0, so the user would not know
	if there are too many expected tokens or none (yes, that's possible).

There are several ways to tell the user in which situation she's in:

- return some E2MANY, a negative value.  Then it makes the pattern

int argsize = yypcontext_expected_tokens (ctx, arg, ARGS_MAX);
	    if (argsize < 0)
	      return argsize;

no longer valid, as for E2MANY (i) the user must generate the error
	  message anyway, and (ii) she should not return E2MANY

- return ARGS_MAX + 1.  Then it makes it dangerous for the user, as
	  she has to iterate update `min (ARGS_MAX, argsize)`.

Returning 0 is definitely simpler and safer for the user, as it tells
	her "this is not an error, just generate your message without a list
	of expecting tokens".  So let's still return 0, but set arg[0] to the
	empty token when the list is really empty.

* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.java
	* data/skeletons/yacc.c (yyexpected_tokens): Put the empty symbol
	first if there are no possible tokens at all.
	* examples/c/bistromathic/parse.y: Demonstrate how to use that.

2020-05-05  Akim Demaille  <akim.demaille@gmail.com>

examples: fix handling of syntax errors
	The shell grammar does not allow empty statements in then/else part of
	an if, but examples/test failed to catch the syntax errors from the
	script it ran.  So exited with success anyway.

You would expect 'set -e' to suffice, but with bash 3.2 actually it
	does not.  As a matter of fact, I could find a way to have this behave
	properly:

$ cat test.sh
	    set -e
	    cleanup ()
	    {
	      status=$?
	      echo "cleanup: $status"
	      exit $status
	    }
	    trap cleanup 0 1 2 13 15
	    . $1
	    s=$?
	    echo "test.sh: $s"
	    exit $s

$ cat bistro.test
	    if true; then
	    fi

$ /bin/sh ./test.sh ./bistro.test
	    ./bistro.test: line 2: syntax error near unexpected token `fi'
	    cleanup: 0
	    $ echo $?
	    0

Remove the set -e (or the trap), and tada, it works...  So we have to
	deal with the error by hand.

* examples/test ($exit): Replace with...
	($status): this.
	Preserve the exit status of the test case.
	* examples/c/bistromathic/bistromathic.test: Fix syntax error.

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

doc: beware of timestamp issues on Haiku
	On Haiku, help2man is fired on a freshly extracted tarball.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00055.html

* doc/local.mk (bison.1): Be robust to a missing help2man.

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of wchar_t portability issues on AIX
	https://lists.gnu.org/r/bug-bison/2020-05/msg00050.html
	Reported by Bruno Haible.

* tests/diagnostics.at: here.

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

glr.c: beware of portability issues with PTRDIFF_MAX
	For instance test 386, "glr.cc api.value.type={double}":

types.at:366: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o test test.cc $LIBS
	    stderr:
	    test.cc: In function 'ptrdiff_t yysplitStack(yyGLRStack*, ptrdiff_t)':
	    test.cc:490:4: error: 'PTRDIFF_MAX' was not declared in this scope
	       (PTRDIFF_MAX < SIZE_MAX ? PTRDIFF_MAX : YY_CAST (ptrdiff_t, SIZE_MAX))
	        ^
	    test.cc:1805:37: note: in expansion of macro 'YYSIZEMAX'
	           ptrdiff_t half_max_capacity = YYSIZEMAX / 2 / state_size;
	                                         ^~~~~~~~~
	    test.cc:490:4: note: suggested alternative: '__PTRDIFF_MAX__'
	       (PTRDIFF_MAX < SIZE_MAX ? PTRDIFF_MAX : YY_CAST (ptrdiff_t, SIZE_MAX))
	        ^
	    test.cc:1805:37: note: in expansion of macro 'YYSIZEMAX'
	           ptrdiff_t half_max_capacity = YYSIZEMAX / 2 / state_size;
	                                         ^~~~~~~~~

The failing tests are using glr.cc only, which I don't understand, the
	problem is rather in glr.c, so I would expect glr.c tests to also fail.

Reported by Bruno Haible.
	https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00053.html

* data/skeletons/yacc.c: Move the block that defines
	YYPTRDIFF_T/YYPTRDIFF_MAXIMUM, YYSIZE_T/YYSIZE_MAXIMUM, and
	YYSIZEOF to...
	* data/skeletons/c.m4 (b4_sizes_types_define): Here.
	(b4_c99_int_type): Also take care of the #undefinition of short.
	* data/skeletons/yacc.c, data/skeletons/glr.c: Use
	b4_sizes_types_define.
	* data/skeletons/glr.c: Adjust to use YYPTRDIFF_T/YYPTRDIFF_MAXIMUM,
	YYSIZE_T/YYSIZE_MAXIMUM.

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of strnlen portability issues
	One function missing on Solaris 10 Sparc:

CCLD     examples/c/bistromathic/bistromathic
	    Undefined                       first referenced
	    symbol                             in file
	    strnlen                             examples/c/bistromathic/bistromathic-parse.o
	    ld: fatal: symbol referencing errors. No output written to examples/c/bistromathic/bistromathic

Reported by Dagobert Michelsen.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00048.html

* examples/c/bistromathic/parse.y (xstrndup): Don't use strnlen.
	xstrndup is assembled from gnulib's xstrndup, strndup and strnlen...

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of portability issues with sh's trap
	On AIX 7.2, when invoking "exit 77", we actually exit with 127.  The
	"cleanup" function, called via trap, received an incorrect exit
	status, something described in Autoconf's doc.
	Reported by Bruno Haible.
	https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00029.html
	https://lists.gnu.org/archive/html/bug-bison/2020-05/msg00047.html

* examples/test (skip): New.
	* examples/c/bistromathic/bistromathic.test,
	* examples/c/reccalc/reccalc.test: Use it, to ensure $? is set to 77
	when the trap is called.

2020-05-04  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of portability issues with diff -u
	AIX 7.1 supports diff -u, but its output does not match the expected
	one.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00049.html

* tests/atlocal.in (DIFF_U_WORKS): New.
	* tests/local.at (AT_DIFF_U_CHECK): New.
	* tests/existing.at (_AT_TEST_EXISTING_GRAMMAR): Use AT_DIFF_U_CHECK.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.93
	* NEWS: Record release date.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

news: update for 3.5.93

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

tests: really skip tricky multichar test on Cygwin
	In Autotest, anything outside AT_SETUP/AT_CLEANUP is discarded.

* tests/diagnostics.at (AT_TEST): Accept a skip-if test.
	Use it to skip on cygwin.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: beware of portability issues of readline on AIX
	Readline may emit escape sequences before the prompt.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/platform-testers/2020-05/msg00001.html.

* examples/c/bistromathic/bistromathic.test: Trust readline _only_ if
	we get what we expect on some reference computation.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of portability issues with cmp
	As someone wrote nearly 20 years ago in Autoconf's documentation,
	don't use cmp to compare text files, but diff.
	https://git.savannah.gnu.org/cgit/autoconf.git/commit/?id=abad4f0576a7dc361e5385e19c7681449103cdb1
	Reported by Jannick.

* examples/test: Use diff, not cmp.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

build: fix warnings (shown on IRIX)
	Appearing on IRIX with gcc -mabi=n32.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00039.html

* examples/c++/variant-11.yy, examples/c/bistromathic/parse.y: Don't
	give chars to isdigit, cast them to unsigned char before.
	* src/complain.c: Use c_isdigit.
	* src/fixits.c (fixits_run): Avoid casts.
	* src/lalr.c (goto_print): Use %zu for a size_t.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

c++: be compatible with the pre-3.6 way to get a symbol's kind
	Reported by Pramod Kumbhar.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00025.html

* data/skeletons/c++.m4: here.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: beware of portability issues with strndup
	Reported by Dagobert Michelsen.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00026.html

* examples/c/bistromathic/parse.y (xstrndup): New.
	Use it.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

flex: fix incorrect use of Automake conditional
	AM_CONDITIONAL does _not_ define a shell variable...
	Reported privately by Denis Excoffier.

* configure.ac (LEX_CXX_WORKS): Fix its definition.

2020-05-03  Bruno Haible  <bruno@clisp.org>

package: fix a link error on IRIX
	https://lists.gnu.org/r/bug-bison/2020-05/msg00035.html

* src/local.mk (src_bison_LDADD): Mention libbison.a before, not after, the
	system libraries.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: beware of portability of readline
	Don't try to build bistromathic if we don't have readline.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00028.html

* configure.ac (ENABLE_BISTROMATHIC): New.
	* examples/c/bistromathic/local.mk: Use it.
	* examples/c/bistromathic/bistromathic.test: Exit 77 for skip.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of portability issues of sh
	"foo || bar" does not invoke bar on AIX 7.2 when foo does not exist.
	It just dies.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00029.html

* examples/c/reccalc/reccalc.test: Check for seq in a subshell.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.92
	* NEWS: Record release date.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

news: update for 3.5.92

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

java: demonstrate push parsers
	* data/skeletons/lalr1.java (Location): Make it a static class.
	(Lexer.yylex, Lexer.getLVal, Lexer.getStartPos, Lexer.getEndPos):
	These are not needed in push parsers.
	* examples/java/calc/Calc.y: Demonstrate push parsers in the Java.
	* doc/bison.texi: Push parsers have been supported for a long time,
	remove incorrect statements stating the opposite.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify what a location is
	Reported by Arthur Schwarz <aschwarz1309@att.net>
	https://lists.gnu.org/r/help-bison/2013-12/msg00009.html

* doc/bison.texi (Location Type): here.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of mbswidth portability issues
	Shy away from these issues on Cygwin.
	Reported Denis Excoffier.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00003.html

* tests/diagnostics.at (Tabulations and multibyte characters): Split
	in two.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of intl portability issues
	Reported by Horst von Brand.
	https://lists.gnu.org/r/bug-bison/2020-04/msg00033.html

* examples/c/bistromathic/Makefile: libintl might not be needed, but
	libm probably is.
	* examples/c/bistromathic/parse.y: Include locale.h.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of portability issues with readline
	On OpenBSD 6.5, the prompt is repeated, but not the actual command
	line...  Don't try to cope with that.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00015.html

* examples/c/bistromathic/bistromathic.test: Skip when readline behave
	this way.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of the portability of flex --header-file
	The option --header was introduced in version 2.5.6.
	The option --header-file was introduced in version 2.6.4.
	Reported by Bruno Haible.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00013.html

So use --header, and do bother with versions that don't support it.

* m4/flex.m4: Check whether flex supports --header.
	* configure.ac (FLEX_WORKS, FLEX_CXX_WORKS): Set to false if it doesn't.
	* * examples/c/reccalc/local.mk, examples/c/reccalc/Makefile:
	Use --header rather than --header-file.

2020-05-03  Akim Demaille  <akim.demaille@gmail.com>

c++: provide backward compatibility on by_type
	To write unit tests for their scanners, some users depended on
	symbol_type::token():

Lexer lex("12345");
	    symbol_type t = lex.nextToken();
	    assert(t.token() == token::INTLIT);
	    assert(t.value.as<int>() == 12345);

But symbol_type::token() was removed in Bison 3.5 because it relied on
	a conversion table.  So users had to find other patterns, such as

assert(t.type_get() == by_type(token::INTLIT).type_get());

which relies on several private implementation details.

As part of transitioning from "token type" to "token kind", and making
	this a public and documented interface, "by_type" was renamed
	"by_kind" and "type_get()" was renamed as "kind()".  The latter had
	backward compatibility mechanisms, not the former.

In Bison 3.6 none of this should be used, but rather

assert(t.kind() == symbol_kind::S_INTLIT);

Reported by Pramod Kumbhar.
	https://lists.gnu.org/r/bug-bison/2020-05/msg00012.html

* data/skeletons/c++.m4 (by_type): Make it an alias to by_kind.

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: improve formatting of the generated code
	* data/skeletons/yacc.c (yy_reduce_print): here.

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

doc: java supports push parsers since 3.0 (2013-07-25)
	* doc/bison.texi: Clarify this.

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

java: fix coding style
	I don't plan to fix everything in one go.  But this was in the way of
	the next commit.

* data/skeletons/lalr1.java: Avoid space before parens.
	* tests/java.at: Adjust.

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* tests/java.at: here.

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

todo: more

2020-05-02  Akim Demaille  <akim.demaille@gmail.com>

style: more documentation about errs
	Suggested by Angelo Borsotti.
	https://lists.gnu.org/r/bug-bison/2014-02/msg00003.html

* src/state.h: here.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

doc: document the exit status
	Suggested by Alexandre Duret-Lutz.
	https://lists.gnu.org/r/bug-bison/2013-09/msg00015.html

* doc/bison.texi (Invocation): Here.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

java: add missing i18n requests
	* data/skeletons/lalr1.java (reportSyntaxError): Here.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

java: style: fix coding style of yyerror/reportSyntaxError
	* data/skeletons/lalr1.java: here.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

java: avoid useless work
	* data/skeletons/lalr1.java (yySymbolPrint): Avoid the computation of
	the argument if useless.
	While at it, fix Java coding style.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

java: comment changes
	* data/skeletons/lalr1.java, examples/java/calc/Calc.y: here.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

news: make it more consistent
	* NEWS: Use the same pattern for titles.

2020-05-01  Akim Demaille  <akim.demaille@gmail.com>

c++: use modern idioms to make classes non-copyable
	Reported by Don Macpherson.
	https://lists.gnu.org/r/bug-bison/2019-05/msg00015.html
	https://github.com/akimd/bison/issues/36

* data/skeletons/lalr1.cc, data/skeletons/stack.hh,
	* data/skeletons/variant.hh: Delete the copy-ctor and the copy operator.

2020-04-30  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: avoid the use of a temporary
	* data/skeletons/yacc.c: Use YYLLOC_DEFAULT directly with the final
	destination.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.91
	* NEWS: Record release date.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

build: fix syntax-check issues
	* cfg.mk: We do want to gettextize the examples.
	* po/POTFILES.in: Adjust.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

doc: document YYEOF, YYUNDEF and YYerror
	* doc/bison.texi (Special Tokens): New.
	* examples/c/bistromathic/parse.y: Formatting changes.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

package: fix distcheck
	Bison emits strings to translate in the generated code, for builtin
	tokens.  So they appear only in generated parsers, which are not
	shipped, so they are not in the src tree, so we cannot use them in our
	POTFILE.

Except src/parse-gram.c, which is in the source tree.  And even in the
	git repo.  But to avoid useless diffs in the repo, we do not keep the
	src/parse-gram.c _with_ the #lines.  This is done in a dist-hook which
	regenerates src/parse-gram.c when we run "make dist".

Unfortunately, then, update-po traverses the whole tree and sees that
	the location of the strings to translate in src/parse-gram.c have
	changed, so the bison.pot is to be updated.  And that is not possible
	in the "make dist" which is run within "make distcheck"
	(not the one preparing the dist for distcheck, the one run by
	distcheck to check that a distributed tarball can build a tarball)
	because then the src tree is read-only.

So let's not put src/parse-gram.c in the POTFILE, and expose these
	strings to gettextize by hand.

* src/i18n-strings.c: New.
	* po/POTFILES.in: Add it, and remove src/parse-gram.c.

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

style: avoid gettextize warnings
	* src/scan-gram.l, src/scan-skel.l: Help it see the start and end of
	"character literals".

2020-04-29  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of portability of readline
	* examples/test: here.

2020-04-28  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: install backward compatibility for YYERRCODE
	Some people have been using that symbol.  Some even have #defined it
	themselves.
	https://lists.gnu.org/r/bison-patches/2020-04/msg00138.html

Let's provide backward compatibility, having it point to YYUNDEF, so
	that an error message is generated.

* data/skeletons/yacc.c (YYERRCODE): New, at the exact same location
	it was defined before.

2020-04-28  Akim Demaille  <akim.demaille@gmail.com>

style: c++: s/type/kind/ where appropriate
	These are internal details.  `type_get ()` is still there to ensure
	backward compatibility, `kind ()` being the modern way.

* data/skeletons/c++.m4 (by_type, by_type::type): Rename as...
	(by_kind, by_kind::kind_): this.
	Adjust dependencies.

2020-04-28  Akim Demaille  <akim.demaille@gmail.com>

java: clean up the definition of token kinds
	From

public interface Lexer {
	      /* Token kinds.  */
	      /** Token number, to be returned by the scanner.  */
	      static final int YYEOF = 0;
	      /** Token number, to be returned by the scanner.  */
	      static final int YYERRCODE = 256;
	      /** Token number, to be returned by the scanner.  */
	      static final int YYUNDEF = 257;
	      /** Token number, to be returned by the scanner.  */
	      static final int BANG = 258;
	    ...
	      /** Deprecated, use b4_symbol(0, id) instead.  */
	      public static final int EOF = YYEOF;

public interface Lexer {
	      /* Token kinds.  */
	      /** Token "end of file", to be returned by the scanner.  */
	      static final int YYEOF = 0;
	      /** Token error, to be returned by the scanner.  */
	      static final int YYerror = 256;
	      /** Token "invalid token", to be returned by the scanner.  */
	      static final int YYUNDEF = 257;
	      /** Token "!", to be returned by the scanner.  */
	      static final int BANG = 258;
	    ...
	      /** Deprecated, use YYEOF instead.  */
	      public static final int EOF = YYEOF;

* data/skeletons/java.m4 (b4_token_enum): Display the symbol's tag in
	comment.
	* data/skeletons/lalr1.java: Address overquotation issue.
	* examples/java/calc/Calc.y, examples/java/simple/Calc.y: Use YYEOF,
	not EOF.

2020-04-28  Akim Demaille  <akim.demaille@gmail.com>

error: rename the error token from YYERRCODE to YYerror
	See https://lists.gnu.org/r/bison-patches/2020-04/msg00162.html.

* data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.cc,
	* data/skeletons/lalr1.java, doc/bison.texi,
	* examples/c/bistromathic/parse.y, src/scan-gram.l, src/symtab.c
	(YYERRCODE): Rename as...
	(YYerror): this.
	Adjust dependencies.

2020-04-27  Akim Demaille  <akim.demaille@gmail.com>

dogfooding: use YYERRCODE in our scanner
	* src/scan-gram.l: Use it.
	* tests/input.at: Adjust.

2020-04-27  Akim Demaille  <akim.demaille@gmail.com>

scanner: avoid spurious errors about empty character literals
	On an invalid character literal such as "'\777'" we used to produce
	two errors:

input.y:2.9-12: error: invalid number after \-escape: 777
	    input.y:2.8-13: error: empty character literal

Get rid of the second one.

* src/scan-gram.l (STRING_GROW_ESCAPE): New.
	* tests/input.at: Adjust.

2020-04-27  Akim Demaille  <akim.demaille@gmail.com>

scanner: bad character literals are errors
	* src/scan-gram.l: These are errors, not warnings.
	* tests/input.at: Adjust.

2020-04-27  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

all: don't emit an error message when the scanner returns YYERRCODE
	I'm quite pleased to see that the tricky case of glr.c was already
	prepared by the changes to support syntax_error exceptions.  Better
	yet, it is actually syntax_error that becomes a special case of the
	general pattern: make yytoken be YYERRCODE.

* data/skeletons/glr.c (YYFAULTYTOK): Remove the now useless (Basil)
	Faulty token.
	Instead, use the error token.
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java: When computing
	the action, first check the case of the error token.

* tests/calc.at: Check cases for the error token symbols before and
	after it.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

c: don't emit an error message when the scanner returns YYERRCODE
	* data/skeletons/yacc.c (yyparse): When the scanner returns YYERRCODE,
	go directly to error recovery (yyerrlab1).
	However, don't keep the error token as lookahead, that token is too
	special.
	* data/skeletons/lalr1.cc: Likewise.

* examples/c/bistromathic/parse.y (yylex): Use that feature to report
	nicely invalid characters.
	* examples/c/bistromathic/bistromathic.test: Check that.
	* examples/test: Neutralize gratuitous differences such as rule
	position.

* tests/calc.at: Check that case in C only.
	The other case seem to be working, but that's an illusion that the
	next commit will address (in fact, they can enter endless loops, and
	report the error several times anyway).

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: demonstrate error recovery
	* examples/c/bistromathic/parse.y: here.
	* examples/c/bistromathic/bistromathic.test: Check it.
	Included a stupid case where the error is actually ignored.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: when quitting, close the current line
	When the user ctrl-d the line, we left the cursor not at col 0.
	Let's fix that.
	This revealed a few short-comings in the testing framework.

* examples/test (run): Also display the diffs.
	And support -n.
	* examples/c/bistromathic/bistromathic.test
	* examples/c/bistromathic/parse.y

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: comment changes
	* examples/c/bistromathic/parse.y: here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

doc: hacking tricks
	* README-hacking.md: Here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

c++: make valid to print the empty symbol
	* data/skeletons/lalr1.cc (yy_print_): here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

c++: always define symbol_name
	* data/skeletons/lalr1.cc (symbol_name): Always define it, even when
	it's actually yytname which is used.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

c++: fix a few style issues
	* data/skeletons/lalr1.cc (yystack_print_, yy_reduce_print_): Add
	missing const.
	(yystack_print_): Rename as...
	(yy_stack_print_): this.
	* data/skeletons/glr.cc (yy_symbol_value_print_, yy_symbol_print_):
	Add missing const.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

all: prefer YYERRCODE to YYERROR
	We will not keep YYERRCODE anyway, it causes backward compatibility
	issues.  So as a first step, let all the skeletons use that name,
	until we have a better one.

* data/skeletons/bison.m4, data/skeletons/glr.c,
	* data/skeletons/glr.cc, data/skeletons/lalr1.cc,
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java,
	* data/skeletons/yacc.c, doc/bison.texi, tests/headers.at,
	* tests/input.at:
	here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

style: glr.c: clarify
	* data/skeletons/glr.c: Make the code a bit clearer.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

style: prefer b4_has_translations_if
	* data/skeletons/glr.c, data/skeletons/yacc.c: here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

style: glr.c: fix indentation issue
	* data/skeletons/glr.c (yyparse): here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

style: fix a few remaining 'type' instead of 'kind'
	* data/skeletons/glr.c, data/skeletons/yacc.c (YY_SYMBOL_PRINT):
	Here.

2020-04-26  Akim Demaille  <akim.demaille@gmail.com>

skeletons: make the warning about implementation details clearer
	* data/skeletons/bison.m4 (b4_disclaimer): Here.
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java: Use it.

2020-04-25  Akim Demaille  <akim.demaille@gmail.com>

style: c: fix a few minor issues about indentation of cpp directives
	* README-hacking.md: More about cpp.
	* data/skeletons/c.m4, data/skeletons/yacc.c: Style changes.

2020-04-25  Akim Demaille  <akim.demaille@gmail.com>

bench: store in benches/012 rather than in benches/12
	* etc/bench.pl.in ($basedir): New.
	Format $count with a least three digits.

2020-04-25  Akim Demaille  <akim.demaille@gmail.com>

bench: minor improvements
	* etc/bench.pl.in: Don't force parse.error=detailed
	Use a simpler way to display the pseudo %bison directive.
	(&bench_with_gbenchmark): Give details about the compiler.

2020-04-25  Akim Demaille  <akim.demaille@gmail.com>

style: clarify #endif
	We could try to avoid the weird "#if 1", but then the indentation of
	the inner #if would be wrong.  Let' keep it this way.

* data/skeletons/yacc.c: here.
	Also, avoid sticking the comment to the directive.

2020-04-25  Akim Demaille  <akim.demaille@gmail.com>

style: minor fixes
	* data/skeletons/bison.m4, doc/bison.texi: Spell check.
	* examples/c/bistromathic/parse.y (N_): Remove, now useless.

2020-04-24  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: shorten token description
	* examples/c/bistromathic/parse.y: "number" is enough.
	* doc/bison.texi: Likewise.

2020-04-24  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: demonstrate internationalization
	Currently it was only using stubs.  Let's actually translate the
	strings using gettext.

* examples/c/bistromathic/local.mk: Define LOCALEDIR, BISON_LOCALEDIR
	and link with libintl.
	* examples/c/bistromathic/parse.y: Use them.
	Remove useless includes.
	Take ENABLE_NLS into account.
	(error_format_string): New.
	(yyreport_syntax_error): Rewrite to rely on a format string, which is
	more appropriate for internationalization.
	* examples/c/bistromathic/Makefile: We no longer use Flex.
	We need readline and intl.

* doc/bison.texi: Point to bistromathic for a better option for
	internationalization.
	* po/POTFILES.in: Add bistromathic.

2020-04-24  Akim Demaille  <akim.demaille@gmail.com>

todo: update for YYERRCODE

2020-04-24  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix a typo
	* src/complain.c: here.

2020-04-20  Akim Demaille  <akim.demaille@gmail.com>

c, c++: provide a default definition for N_
	In C/C++, N_ is a no-op.  Define it if the user didn't.
	Suggested by Frank Heckenbach.
	https://lists.gnu.org/r/bug-bison/2020-04/msg00010.html

* src/output.c (prepare_symbol_names): Rename has_translations as
	has_translations_flag.
	* data/skeletons/bison.m4 (b4_has_translations_if): New.
	* data/skeletons/java.m4 (b4_trans): Use it.

* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/yacc.c
	(N_): Provide a default definition.

2020-04-19  Akim Demaille  <akim.demaille@gmail.com>

i18n: also look in src/parse-gram.c
	Some strings appears in the generated file only, e.g., translation of
	"end of file".  So don't forget to go and see there.

2020-04-19  Akim Demaille  <akim.demaille@gmail.com>

style: fix comments
	* data/skeletons/glr.c, data/skeletons/lalr1.cc,
	* data/skeletons/yacc.c: here.

2020-04-19  Akim Demaille  <akim.demaille@gmail.com>

tokens: clean up the translation of special symbols
	* src/output.c (prepare_symbol_names): Don't play tricks with the
	symbols, it's quite too late.
	(has_translations): Move to...
	* src/symtab.c: here.
	(symbols_pack): Use it to enable translation for special symbols.

2020-04-19  Akim Demaille  <akim.demaille@gmail.com>

news: fix typo
	Reported by Frank Heckenbach.

2020-04-19  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of portability issues with wc
	On macOS, wc -l always prepends the result with a tab, even when fed
	by stdin.  But anyway, we should have used `grep -c -v`, which appears
	to be portable according to Autoconf's "Limitations of Usual Tools"
	section.
	Reported by Denis Excoffier.
	https://lists.gnu.org/r/bug-bison/2020-04/msg00009.html

* tests/calc.at (_AT_CHECK_CALC): Use grep's -c instead.

2020-04-18  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-04-18  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.90
	* NEWS: Record release date.

2020-04-18  Akim Demaille  <akim.demaille@gmail.com>

examples: beware of readline on macOS
	macOS' version of readline does not repeat stdin on stdout in
	non-interactive mode, contrary to the current version of GNU readline.

* examples/test: Add support for strip_prompt.
	* examples/c/bistromathic/bistromathic.test (strip_prompt): Set it
	when needed.
	Early exit when needed.

2020-04-18  Akim Demaille  <akim.demaille@gmail.com>

c++: give public access to the symbol kind
	symbol_type::token () was removed: it returned the token kind of a
	symbol.  To do that, one needs to convert from the symbol kind to the
	token kind, which requires a table.

This broke some users' unit tests for scanners, see
	https://lists.gnu.org/r/bug-bison/2020-01/msg00001.html
	https://lists.gnu.org/r/bug-bison/2020-03/msg00020.html
	https://lists.gnu.org/r/help-bison/2020-04/msg00005.html

Instead of making this possible again, let's check the symbol's kind
	instead.  So give proper access to a symbol's kind.

That feature existed, undocumented, as 'type_get()'.  Let's rename
	this as 'kind()'.

* data/skeletons/c++.m4, data/skeletons/glr.cc,
	* data/skeletons/lalr1.cc (type_get): Rename as...
	(kind): This.
	(type_get): Install a backward compatibility alias.
	* doc/bison.texi (Complete Symbols): Document symbol_type and
	symbol_type::kind.

2020-04-17  Akim Demaille  <akim.demaille@gmail.com>

doc: token_kind_type in C++
	* data/skeletons/c++.m4: Define the old names in terms on the new
	ones, instead of the converse.
	* doc/bison.texi (C++ Parser Interface): Be more extensive about
	token_kind_type.

2020-04-16  Akim Demaille  <akim.demaille@gmail.com>

doc: updates for 3.6
	* doc/bison.texi: More s/token type/token kind/.
	* NEWS: Update.

2020-04-16  Akim Demaille  <akim.demaille@gmail.com>

skeletons: use symbol(-2, kind)
	Not all the symbols have a fixed symbol code.  UNDEF's one is fixed:
	-2.

* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/yacc.c: here.

2020-04-16  Akim Demaille  <akim.demaille@gmail.com>

style: comments changes about error handling
	* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/lalr1.java, data/skeletons/yacc.c: here.
	* data/skeletons/lalr1.cc: Reduce scope.

2020-04-14  Akim Demaille  <akim.demaille@gmail.com>

examples: bistro: don't be lazy with switch
	* examples/c/bistromathic/parse.y (yylex): Use the switch to
	discriminate all the cases.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check
	* doc/bison.texi, NEWS, README-hacking.md: here.
	And elsewhere.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

doc: more about the coding style
	* README-hacking.md: here.
	(Troubleshooting): New.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

java: promote YYEOF rather that Lexer.EOF
	* doc/bison.texi: here.
	* data/skeletons/lalr1.java: Use YYEOF.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

java: fix names
	* data/skeletons/lalr1.java (yySymbolPrint): There are no pointers
	here, remove the `p` suffix.
	Use the appropriate type for locations.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

doc: java: SymbolKind, etc.
	Why didn't I think about this before???  symbolName should be a method
	of SymbolKind.

* data/skeletons/lalr1.java (YYParser::yysymbolName): Move as...
	* data/skeletons/java.m4 (SymbolKind::getName): this.
	Make the table a static final table, not a local variable.
	Adjust dependencies.
	* doc/bison.texi (Java Parser Interface): Document i18n.
	(Java Parser Context Interface): Document SymbolKind.
	* examples/java/calc/Calc.y, tests/local.at: Adjust.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

style: java: get closer to the Java style
	* examples/java/calc/Calc.y, examples/java/simple/Calc.y: here.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

doc: c++: document parser::context
	* doc/bison.texi (C++ Parser Context): New.

* data/skeletons/lalr1.cc (parser::yysymbol_name): Rename as...
	(parser::symbol_name): this.
	(A Complete C++ Example): Promote LAC, now that we have it.
	Promote parse.error detailed over verbose.
	* examples/c++/calc++/calc++.test, tests/local.at: Adjust.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

doc: promote YYEOF
	* NEWS (Deep overhaul of the symbol and token kinds): New.
	* doc/bison.texi: Promote YYEOF over "0" in scanners.
	(Token Decl): No longer show YYEOF here, it now works by default.
	(Token I18n): More details about YYEOF here.
	(Calc++): Just use YYEOF.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

d: put YYEMPTY in the TokenKind
	* data/skeletons/d.m4, data/skeletons/lalr1.d (b4_token_enums): Rename
	YYTokenType as TokenKind.
	Define YYEMPTY.
	* examples/d/calc.y, tests/calc.at, tests/scanner.at: Adjust.

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-13  Akim Demaille  <akim.demaille@gmail.com>

c, c++: also define YYEMPTY in yytoken_kind_t
	I have been hesitating a lot before doing it ---after all the user
	must not use this kind, so what's the point of showing it in
	yytoken_kind_t.  And eventually I chose to play it safe with the
	typing system and make it possible to use yytoken_kind_t for all the
	tokens, even the "empty token".

* data/skeletons/c.m4: Give an id and a tag to YYEMPTY.
	(b4_token_enums): Define YYEMPTY.
	* data/skeletons/c++.m4 (b4_token_enums): Define YYEMPTY.
	* data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/yacc.c:
	(YYEMPTY): Remove.
	Use b4_symbol(-2, id) instead.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

doc: use "code", not "number", for token (and symbol) kinds
	"Number" is too much about arithmethics.  "Code" conveys better the
	"enum" nature of token kinds.  And of symbol kinds.

* doc/bison.texi: Here.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

doc: promote yytoken_kind_t, not yytokentype
	* data/skeletons/c.m4 (yytoken_kind_t): New.
	* data/skeletons/c++.m4, data/skeletons/lalr1.cc (yysymbol_kind_type):
	New.
	* examples/c/lexcalc/parse.y, examples/c/reccalc/parse.y,
	* tests/regression.at:
	Use them.
	* doc/bison.texi: Replace "enum yytokentype" by "yytoken_kind_t".
	(api.token.raw): Explain that it forces "yytoken_kind_t" to coincide
	with "yysymbol_kind_t".
	(Calling Convention): Mention YYEOF.
	(Table of Symbols): Add entries for "yytoken_kind_t" and
	"yysymbol_kind_t".
	(Glossary): Add entries for "Kind", "Token kind" and "Symbol kind".

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

doc: document yypcontext_t, and api.symbol.prefix
	* doc/bison.texi (%define Summary): Document api.symbol.prefix.
	(Syntax Error Reporting Function): Document yypcontext_t,
	yypcontext_location, yypcontext_token, yypcontext_expected_tokens, and
	yysymbol_kind_t.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

c: rename yyexpected_tokens as yypcontext_expected_tokens
	The user should think of yypcontext fields as accessible only via
	yypcontext_* functions.  So let's rename yyexpected_tokens to reflect
	that.

Let's _not_ rename yyreport_syntax_error, as the user may define this
	function, and is not allowed to access directly the fields of
	yypcontext_t: she *must* use the "accessors".  This is comparable to
	the case of C++/Java where the user defines
	parser::report_syntax_error, not parser::context::report_syntax_error.

* data/skeletons/glr.c, data/skeletons/yacc.c (yyexpected_tokens):
	Rename as...
	(yypcontext_expected_tokens): this.
	Adjust dependencies.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

skeletons: clarify the tag of special tokens
	From

GRAM_EOF = 0,                  /* $end  */
	    GRAM_ERRCODE = 1,              /* error  */
	    GRAM_UNDEF = 2,                /* $undefined  */

GRAM_EOF = 0,                  /* "end of file"  */
	    GRAM_ERRCODE = 1,              /* error  */
	    GRAM_UNDEF = 2,                /* "invalid token"  */

* src/output.c (symbol_tag): New.
	Use it to pass the token names and the symbol tags to the skeletons.

* tests/input.at: Adjust.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

skeletons: use "invalid token" instead of "$undefined"
	* src/output.c (prepare_symbol_names): Also handle undeftoken.
	* tests/actions.at, tests/calc.at, tests/regression.at: Adjust.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

skeletons: make the eof token translatable if i18n is enabled
	* src/output.c (has_translations): New.
	(prepare_symbol_names): Translate endtoken if the user already
	translated tokens.

* examples/c/bistromathic/parse.y, src/parse-gram.y: Simplify.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

skeletons: use "end of file" instead of "$end"
	The name "$end" is nice in the report, in particular it avoids that
	pointed-rules (aka items) be too long.  It also helps keeping them
	"standard".

But it is bad in error messages, we should report "end of file" (or
	maybe "end of input", this is debatable).  So, unless the user already
	defined the alias for the error token herself, make it "end of file".
	It should even be translated if the user already translated some
	tokens, so that there is now no strong reason to redefine the $end
	token.

* src/output.c (prepare_symbol_names): Issue "end of file" instead of
	"$end".

* data/skeletons/lalr1.java (yytnamerr_): Remove the renaming hack.

* build-aux/update-test: Accept files with names containing a "+",
	such as c++.at.
	* tests/actions.at, tests/c++.at, tests/conflicts.at,
	* tests/glr-regression.at, tests/regression.at, tests/skeletons.at:
	Adjust.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: replace "user token number" by "token code"
	Yet, don't change the structure identifier to avoid introducing
	conflicts in Vincent Imbimbo's PR (which, amusingly enough, is about
	conflicts).

* src/symtab.c: here.
	* tests/diagnostics.at, tests/input.at: Adjust.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

c++: remove the yy prefix from some functions
	yy::parser features a parse() function, not a yyparse() one.

* data/skeletons/lalr1.cc (yyreport_syntax_error)
	(context::yyexpected_tokens): Rename as...
	(report_syntax_error, context::expected_tokens): these.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

tokens: properly define the YYEOF token kind
	Currently EOF is handled in an adhoc way, with a #define YYEOF 0 in
	the implementation file.  As a result, the user has to define her own
	EOF token if she wants to use it, which is a pity.

Give the $end token a visible kind name, YYEOF.  Except that in C,
	where enums are not scoped, we would have collisions between all the
	definitions of YYEOFs in the header files, so in C, make it
	<api.PREFIX>EOF.

* data/skeletons/c.m4 (YYEOF): Override its name to avoid collisions.
	Unless the user already gave it a different name.
	* data/skeletons/glr.c (YYEOF): Remove.
	Use ]b4_symbol(0, [id])[ instead.
	Add support for "pre_epilogue", for glr.cc.
	* data/skeletons/glr.cc: Remove dead code (never emitted #undefs).
	* data/skeletons/yacc.c
	* src/parse-gram.c
	* src/reader.c
	* src/symtab.c
	* tests/actions.at
	* tests/input.at

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

tokens: define the "$undefined" token kind
	* data/skeletons/bison.m4 (b4_symbol_token_kind): Give a definition to
	$undefined.
	(b4_token_visible_if): $undefined has an id.
	* src/output.c (prepare_symbol_definitions): Stop lying: $undefined
	_is_ a token.
	* tests/input.at: Adjust.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

tokens: properly define the "error" token kind
	There are people out there that do use YYERRCODE (the token kind of
	the error token).  See for instance
	https://github.com/borbolla-automation/SPC_Machines/blob/3812012bb782bfdfe7b325950a35cd337925fcad/unixODBC-2.3.2/Drivers/nn/yylex.c.

Currently, YYERRCODE is defined by yacc.c in an adhoc way as a #define
	in the *.c file only.  It belongs with the other token kinds.

YYERRCODE is not a nice name, it does not fit in our naming scheme.
	YYERROR would be more logical, but it collides with the YYERROR macro.
	Shall we keep the same name in all the skeletons?  Besides, to avoid
	collisions in C, we need to apply the api prefix: YYERRCODE is
	actually <PREFIX>ERRCODE.  This is not needed in the other languages.

* data/skeletons/bison.m4 (b4_symbol_token_kind): New.
	Map the error token to "YYERRCODE".
	* data/skeletons/yacc.c (YYERRCODE): Don't define it, it's handled by...
	* src/output.c (prepare_symbol_definitions): this.
	* tests/input.at (Redefining the error token): Check it.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

tokens: style: minor fixes
	* data/skeletons/bison.m4 (b4_symbol_kind): Dispatch on the UNDEF
	token number rather than its name.
	* data/skeletons/c++.m4, data/skeletons/c.m4, data/skeletons/java.m4:
	Comment changes.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: remove dead code
	* data/skeletons/glr.cc: here.

2020-04-12  Akim Demaille  <akim.demaille@gmail.com>

c++: fix generated headers
	A forthcoming commit (tokens: properly define the "error" token kind)
	revealed a problem in the C++ generated headers: they are not
	self-contained.  With this file:

%language "c++"
	    %define api.value.type variant

%code {
	      static int yylex (yy::parser::semantic_type *lvalp);
	    }

%token <int> X

exp:
	      X { printf ("x\n"); }
	    ;

void
	    yy::parser::error (const std::string& m)
	    {
	      std::cerr << m << '\n';
	    }

static
	    int yylex (yy::parser::semantic_type *lvalp)
	    {
	      static int const input[] = {yy::parser::token::X, 0};
	      static int toknum = 0;
	      return input[toknum++];
	    }

int
	    main (int argc, char const* argv[])
	    {
	      yy::parser p;
	      return p.parse ();
	    }

the generated header fails to compile cleanly (foo.cc just #includes
	the generated header):

$ clang++-mp-9.0 -c -Wundefined-func-template foo.cc
	    In file included from foo.cc:1:
	    bar.tab.hh:550:12: warning: instantiation of function 'yy::parser::basic_symbol<yy::parser::by_type>::basic_symbol' required here, but no definition is available
	          [-Wundefined-func-template]
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    bar.tab.hh:436:7: note: forward declaration of template entity is here
	          basic_symbol (basic_symbol&& that);
	          ^
	    bar.tab.hh:550:12: note: add an explicit instantiation declaration to suppress this warning if 'yy::parser::basic_symbol<yy::parser::by_type>::basic_symbol' is explicitly instantiated
	          in another translation unit
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    1 warning generated.

* data/skeletons/c++.m4 (b4_public_types_define): Move the
	implementation of the basic_symbol move-ctor to...
	(b4_public_types_define): here, its declaration.
	* tests/headers.at (Sane headers): Use a declared token so that the
	corresponding token constructor is declared.  Which triggers the
	aforementioned issue.

2020-04-10  Akim Demaille  <akim.demaille@gmail.com>

style: rename YYNOMEM as YYENOMEM
	This is clearer.

* data/skeletons/glr.c, data/skeletons/yacc.c (YYNOMEM): Rename as...
	(YYENOMEM): here.

2020-04-10  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-04-10  Akim Demaille  <akim.demaille@gmail.com>

c++: improvements on symbol kinds
	Instead of

/// (Internal) symbol kind.
	    enum symbol_kind_type
	    {
	      YYNTOKENS = 5, ///< Number of tokens.
	      YYSYMBOL_YYEMPTY = -2,
	      YYSYMBOL_YYEOF = 0,                      // END_OF_FILE
	      YYSYMBOL_YYERROR = 1,                    // error
	      YYSYMBOL_YYUNDEF = 2,                    // $undefined
	      YYSYMBOL_TEXT = 3,                       // TEXT
	      YYSYMBOL_NUMBER = 4,                     // NUMBER
	      YYSYMBOL_YYACCEPT = 5,                   // $accept
	      YYSYMBOL_result = 6,                     // result
	      YYSYMBOL_list = 7,                       // list
	      YYSYMBOL_item = 8                        // item
	    };

generate

/// Symbol kinds.
	    struct symbol_kind
	    {
	      enum symbol_kind_type
	      {
	        YYNTOKENS = 5, ///< Number of tokens.
	        S_YYEMPTY = -2,
	        S_YYEOF = 0,                             // END_OF_FILE
	        S_YYERROR = 1,                           // error
	        S_YYUNDEF = 2,                           // $undefined
	        S_TEXT = 3,                              // TEXT
	        S_NUMBER = 4,                            // NUMBER
	        S_YYACCEPT = 5,                          // $accept
	        S_result = 6,                            // result
	        S_list = 7,                              // list
	        S_item = 8                               // item
	      };
	    };

* data/skeletons/c++.m4 (api.symbol.prefix): Define to S_.
	Adjust all the uses.
	(b4_public_types_declare): Nest the enum inside 'struct symbol_kind'.
	* data/skeletons/glr.cc, data/skeletons/lalr1.cc,
	* tests/headers.at, tests/local.at: Adjust.

2020-04-10  Akim Demaille  <akim.demaille@gmail.com>

d: improvements on symbol kinds
	    public enum SymbolKind
	    {
	      S_YYEMPTY = -2,    /* No symbol.  */
	      S_YYEOF = 0,       /* $end  */
	      S_YYERROR = 1,     /* error  */
	      S_YYUNDEF = 2,     /* $undefined  */
	      S_EQ = 3,          /* "="  */

* data/skeletons/d.m4 (api.symbol.prefix): Default to S_.
	Output the symbol kind definitions with a comment.

2020-04-10  Akim Demaille  <akim.demaille@gmail.com>

symbols: minor fixes
	* data/skeletons/bison.m4 (b4_symbol_kind): Series of _ are useless,
	one is enough.
	* data/skeletons/c.m4 (b4_token_enum): Fix overquoting.

2020-04-07  Akim Demaille  <akim.demaille@gmail.com>

skeletons: introduce api.symbol.prefix
	* data/skeletons/bison.m4 (b4_symbol_prefix): New.
	(b4_symbol_kind): Use it.
	* data/skeletons/c++.m4, data/skeletons/c.m4, data/skeletons/d.m4
	* data/skeletons/java.m4 (api.symbol.prefix): Provide a default value.

* data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/lalr1.cc,
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java, data/skeletons/yacc.c:
	Adjust: use b4_symbol_prefix instead of YYSYMBOL_.

2020-04-07  Akim Demaille  <akim.demaille@gmail.com>

java: also emit documenting comments for symbol kinds
	* data/skeletons/java.m4 (b4_symbol_enum): here.
	And stop defined YYSYMBOL_YYEMPTY, we no longer use it.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

todo: update
	* TODO (YYERRCODE): Remove, handled by YYSYMBOL_ERROR.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

skeletons: beware not to use yyarg when it's null
	Reported by Adrian Vogelsgesang.

* data/skeletons/glr.c, data/skeletons/lalr1.cc,
	* data/skeletons/lalr1.java, data/skeletons/yacc.c: Here.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

java: document new features
	* data/skeletons/lalr1.java: More comments.
	(Context.EMPTY): Remove.
	* doc/bison.texi (Java Parser Context Interface): New.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

java: prefer null to YYSYMBOL_YYEMPTY
	That's one nice benefit from using enums.

* data/skeletons/lalr1.java (YYSYMBOL_YYEMPTY): No longer define it.
	Use 'null' instead.
	* examples/java/calc/Calc.y, tests/local.at: Adjust.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

java: rename Lexer.yyreportSyntaxError as reportSyntaxError
	* data/skeletons/lalr1.java: here.
	* examples/java/calc/Calc.y, tests/local.at: Adjust.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

java: use getExpectedTokens, not yyexpectedTokens
	* data/skeletons/lalr1.java, examples/java/calc/Calc.y, tests/local.at:
	here.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

java: style: fix coding style
	* data/skeletons/java.m4: Indent by two.
	* data/skeletons/lalr1.java (yynnts_): Remove.
	(yyfinal_, yyntokens_, yylast_, yyempty_): Rename as...
	(YYFINAL_, YYNTOKENS_, YYLAST_, YYEMPTY_): these, they are constants.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

c: make the symbol kind definition nicer to read
	From

enum yysymbol_kind_t
	    {
	      YYSYMBOL_YYEMPTY = -2,
	      YYSYMBOL_YYEOF = 0,
	      YYSYMBOL_YYERROR = 1,
	      YYSYMBOL_YYUNDEF = 2,

enum yysymbol_kind_t
	    {
	      YYSYMBOL_YYEMPTY = -2,
	      YYSYMBOL_YYEOF = 0,                      /* "end of file"  */
	      YYSYMBOL_YYERROR = 1,                    /* error  */
	      YYSYMBOL_YYUNDEF = 2,                    /* $undefined  */

* data/skeletons/bison.m4 (b4_last_symbol): New.
	(b4_symbol_enum, b4_symbol_enums): Reformat the output.
	* data/skeletons/c.m4

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

c: make the token kind definition nicer to read
	From

enum gram_tokentype
	    {
	      GRAM_EOF = 0,
	      STRING = 3,
	      TSTRING = 4,
	      PERCENT_TOKEN = 5,

enum gram_tokentype
	    {
	      GRAM_EOF = 0,                  /* "end of file"  */
	      STRING = 3,                    /* "string"  */
	      TSTRING = 4,                   /* "translatable string"  */
	      PERCENT_TOKEN = 5,             /* "%token"  */

* data/skeletons/bison.m4 (b4_last_enum_token): New.
	* data/skeletons/c.m4 (b4_token_enum, b4_token_enums): Show the
	corresponding symbol.

2020-04-06  Akim Demaille  <akim.demaille@gmail.com>

c: make the generated YYSTYPE nicer to read
	From

union GRAM_STYPE
	    {
	      /* precedence_declarator  */
	      assoc precedence_declarator;
	      /* "string"  */
	      char* STRING;
	      /* "translatable string"  */
	      char* TSTRING;
	      /* "{...}"  */
	      char* BRACED_CODE;
	      /* "%?{...}"  */

union GRAM_STYPE
	    {
	      assoc precedence_declarator;             /* precedence_declarator  */
	      char* STRING;                            /* "string"  */
	      char* TSTRING;                           /* "translatable string"  */
	      char* BRACED_CODE;                       /* "{...}"  */

* data/skeletons/c.m4 (b4_symbol_type_register): Use m4_format to
	align the comments.
	* src/parse-gram.h: Regen.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

bison: use consistently "token kind", not "token type"
	* src/output.c, src/reader.c, src/scan-gram.l, src/tables.c: here.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

skeletons: use consistently "kind" instead of "type" in the code
	* data/skeletons/bison.m4, data/skeletons/c++.m4, data/skeletons/c.m4,
	* data/skeletons/glr.cc, data/skeletons/lalr1.cc,
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java:
	Refer to the "kind" of a symbol, not its "type", where appropriate.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

doc: refer to the token kind rather than the token type
	* doc/bison.texi: Replace occurrences of "token type" with "token
	kind".
	Stop referring to the "macro definitions" of the token kinds, just
	name them "definitions".

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

m4: we don't need undef_token_number
	It's replaced by YYSYMBOL_YYUNDEF.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

m4: rename b4_symbol_sid as b4_symbol_kind
	* data/skeletons/bison.m4, data/skeletons/c++.m4, data/skeletons/c.m4,
	* data/skeletons/d.m4, data/skeletons/java.m4
	(b4_symbol_sid): Rename as...
	(b4_symbol_kind): this.
	Adjust dependencies.
	* data/README.md: Document the kind.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

d, java: rename SymbolType as SymbolKind
	See https://lists.gnu.org/r/bison-patches/2020-04/msg00031.html.

* data/skeletons/d.m4, data/skeletons/lalr1.d,
	* data/skeletons/java.m4, data/skeletons/lalr1.java
	(SymbolType): Rename as...
	(SymbolKind): this.
	Adjust dependencies.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

c, c++: rename yysymbol_type_t as yysymbol_kind_t
	See https://lists.gnu.org/r/bison-patches/2020-04/msg00031.html

* data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/yacc.c
	(yysymbol_type_t): Rename as...
	(yysymbol_kind_t): this.
	Adjust dependencies.
	* data/skeletons/c++.m4, data/skeletons/glr.cc, data/skeletons/lalr1.cc
	(symbol_type_type): Rename as...
	(symbol_kind_type): this.
	Adjust dependencies.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

doc: remove obsolete release instructions
	* README-hacking.md: here.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

Merge branch 'maint'
	* maint:
	  maint: post-release administrivia
	  version 3.5.4
	  examples: reccalc: really compile cleanly in C99
	  news: announce that Bison 3.6 drops YYERROR_VERBOSE
	  news: update for 3.5.4
	  style: fix spellos
	  typo: succesful -> successful
	  package: improve the readme
	  java: check and fix support for api.token.raw
	  java: style: prefer 'int[] foo' to 'int foo[]'
	  build: fix syntax-check issues
	  tests: recheck: work properly when the test suite was interrupted
	  doc: c++: promote api.token.raw
	  build: fix compatibility with old compilers
	  examples: reccalc: compile cleanly in C99

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.4
	* NEWS: Record release date.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

news: update the yyreport_syntax_error example
	* examples/c/bistromathic/parse.y, tests/local.at
	(yyreport_syntax_error): Fix use of YYSYMBOL_YYEMPTY.
	* NEWS: Update.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

style: rename yysyntax_error_arguments as yy_syntax_error_arguments
	It's a private implementation detail.

* NEWS, data/skeletons/glr.c, data/skeletons/lalr1.cc,
	* data/skeletons/yacc.c, doc/bison.texi: here.

2020-04-05  Akim Demaille  <akim.demaille@gmail.com>

examples: reccalc: really compile cleanly in C99
	The previous fix does not suffice, and actually managed to make things
	worse by defining yyscan_t twice in parse.y...

Reported by kencu.
	https://trac.macports.org/ticket/59927#comment:29

* examples/c/reccalc/parse.y (yyscan_t): Define it with the same
	guards as used by Flex.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

c: rename yyparse_context_t as yypcontext_t
	The first name is too long.  We already have `yypstate`, so
	`yypcontext` is ok.  We are also migrating to using `*_t` for our
	types.

* NEWS, data/skeletons/glr.c, data/skeletons/yacc.c, doc/bison.texi,
	* examples/c/bistromathic/parse.y, src/parse-gram.y, tests/local.at:
	(yyparse_context_t, yyparse_context_location, yyparse_context_token):
	Rename as...
	(yypcontext_t, yypcontext_location, yypcontext_token): these.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

readme: more about the coding style

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

java: fixes in SymbolType
	Reported by Paolo Bonzini.
	https://github.com/akimd/bison/pull/34#issuecomment-609029634

* data/skeletons/java.m4 (SymbolType): Use 'final' where possible.
	(get): Rewrite on top of an array instead of a switch.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

java: use SymbolType
	The Java enums are very different from the C model.  As a consequence,
	one cannot "build" an enum directly from an integer, we must retrieve
	it.  That's the purpose of the SymbolType.get class method.

* data/skeletons/java.m4 (b4_symbol_enum, b4_case_code_symbol)
	(b4_declare_symbol_enum): New.
	* data/skeletons/lalr1.java: Use SymbolType,
	SymbolType.YYSYMBOL_YYEMPTY, etc.
	* examples/java/calc/Calc.y, tests/local.at: Adjust.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

examples: java: use explicit token identifiers
	* examples/java/calc/Calc.y: Declare all the tokens, so that we are
	compatibile with api.token.raw.
	* examples/java/calc/Calc.test: Adjust.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

news: announce that Bison 3.6 drops YYERROR_VERBOSE
	* NEWS: here.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

news: update for 3.5.4

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

style: fix spellos
	* src/complain.c, src/print.c, src/print-xml.c, src/symtab.h: here.

2020-04-04  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

typo: succesful -> successful
	* tests/calc.at: Here.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

package: improve the readme
	* README: Describe what Bison is.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

java: check and fix support for api.token.raw
	* tests/local.at (AT_LANG_MATCH, AT_YYERROR_DECLARE(java))
	(AT_YYERROR_DECLARE_EXTERN(java), AT_PARSER_CLASS): New.
	(AT_MAIN_DEFINE(java)): Use AT_PARSER_CLASS.
	* tests/scanner.at: Add a test for Java.
	* data/skeletons/lalr1.java (yytranslate_): Cast the result.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

d: use the SymbolType enum for symbol kinds
	* data/skeletons/d.m4 (b4_symbol_enum, b4_declare_symbol_enum): New.
	* data/skeletons/lalr1.d: Use them.
	Use SymbolType, SymbolType.YYSYMBOL_YYEMPTY etc. where appropriate.
	(undef_token_, token_number_type, yy_error_token_): Remove.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

java: style: prefer 'int[] foo' to 'int foo[]'
	* data/skeletons/java.m4 (b4_typed_parser_table_define): Here.

2020-04-04  Akim Demaille  <akim.demaille@gmail.com>

build: fix syntax-check issues
	* src/system.h, tests/local.mk: Fix indentation.

2020-04-02  Akim Demaille  <akim.demaille@gmail.com>

tests: recheck: work properly when the test suite was interrupted
	* tests/local.mk (recheck): Look at the per-test logs, not the overall
	log, which, when interrupted, contains only information about... the
	tests that passed.

2020-04-02  Akim Demaille  <akim.demaille@gmail.com>

doc: c++: promote api.token.raw
	* doc/bison.texi (Calc++ Parser): Here.

2020-04-02  Akim Demaille  <akim.demaille@gmail.com>

build: fix compatibility with old compilers
	GCC 4.2 dies with

src/InadequacyList.c: In function 'InadequacyList__new_conflict':
	    src/InadequacyList.c:37: error: #pragma GCC diagnostic not allowed inside functions
	    src/InadequacyList.c:37: error: #pragma GCC diagnostic not allowed inside functions
	    src/InadequacyList.c:40: error: #pragma GCC diagnostic not allowed inside functions

Reported by Evan Lavelle.
	See https://lists.gnu.org/r/bug-bison/2020-03/msg00021.html
	and https://trac.macports.org/ticket/59927.

* src/system.h (GCC_VERSION): New.
	Use it to control IGNORE_TYPE_LIMITS_BEGIN and
	IGNORE_TYPE_LIMITS_END.

2020-04-02  Akim Demaille  <akim.demaille@gmail.com>

examples: reccalc: compile cleanly in C99
	See https://trac.macports.org/ticket/59927.

* examples/c/reccalc/parse.y: C99 does not allow multiple typedefs.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

c++: replace symbol_number_type with symbol_type_type
	* data/skeletons/c++.m4, data/skeletons/glr.cc,
	* data/skeletons/lalr1.cc: here.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

c++: also use symbol_type_type
	Because of the insane current implementation of glr.cc, things are a
	bit nasty.  We will rename symbol_number_type as symbol_type_type
	later, to keep this commit small.

* data/skeletons/c++.m4 (b4_declare_symbol_enum): New.
	Also define YYNTOKENS to avoid type clashes when yyntokens_ was
	actually defined in another enum.
	Use it.
	(symbol_number_type): Be an alias of symbol_type_type.
	Use YYSYMBOL_YYEMPTY and the like.
	Use symbol_number_type where appropriate.
	(empty_symbol): Remove.
	(yytranslate_): Use symbol_number_type, not token_number_type.
	* data/skeletons/lalr1.cc: Use symbol_number_type where appropriate.
	Adjust to the replacement of empty_symbol by YYSYMBOL_YYEMPTY.
	(yy_error_token_, yy_undef_token_, yyeof_, yyntokens_): Remove.
	Adjust dependencies.

* data/skeletons/glr.cc: Use symbol_number_type where appropriate.
	Forward definitions of YYSYMBOL_YYEMPTY, etc. to glr.c.

* tests/headers.at: Accept YYNTOKENS and other YYSYMBOL_*.
	* tests/local.at (AT_YYERROR_DEFINE(c++)): Use symbol_number_type.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

glr.c: remove the yySymbol alias
	* data/skeletons/glr.c: Use yysymbol_type_t only.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

glr.c, yacc.c: propagate yysymbol_type_t
	Now that yacc.c and glr.c both know yysymbol_type_t, convert the
	common routines.

* data/skeletons/c.m4 (yydestruct, yy_symbol_value_print)
	(yy_symbol_print): Use yysymbol_type_t instead of int.
	* data/skeletons/glr.c: Use yySymbol where appropriate.
	* data/skeletons/yacc.c (YY_ACCESSING_SYMBOL): New wrapper around
	yystos.
	Use it.
	* tests/local.at (yyreport_syntax_error): Use yysymbol_type_t where
	appropriate.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

glr.c: use yysymbol_type_t, YYSYMBOL_YYEOF etc.
	Apply the same changes as in yacc.c.  Now yySymbol and yysymbol_type_t
	are aliases.  We will remove the former later, to avoid cluttering
	this commit.

* data/skeletons/glr.c: Use b4_declare_symbol_enum.
	Use YYSYMBOL_YYEOF etc. where appropriate.
	(YYUNDEFTOK, YYTERROR): Remove.
	(YYTRANSLATE, yySymbol, yyexpected_tokens, yysyntax_error_arguments):
	Adjust.
	(yy_accessing_symbol): New.
	Use it where appropriate.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: fix more errors from make maintainer-check-g++
	* data/skeletons/yacc.c (yyexpected_tokens): Use casts where needed.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: revert to not using yysymbol_type_t in the yytranslate table
	This triggers warnings with several compilers.  For instance ICC fills
	the logs with pages and pages of

input.c(477): error: a value of type "int" cannot be used to initialize an entity of type "const yysymbol_type_t={yysymbol_type_t}"
	             0,     2,     2,     2,     2,     2,     2,     2,     2,     2,
	             ^

And so does G++9 when compiling yacc.c's (C) output

input.c:545:8: error: invalid conversion from 'int' to 'yysymbol_type_t' [-fpermissive]
	      545 |        0,     5,     9,     2,     2,     2,     2,     2,     2,     2,
	          |        ^
	          |        |
	          |        int
	    input.c:545:15: error: invalid conversion from 'int' to 'yysymbol_type_t' [-fpermissive]
	      545 |        0,     5,     9,     2,     2,     2,     2,     2,     2,     2,
	          |               ^
	          |               |
	          |               int

Clang++ is no exception

input.c:545:8: error: cannot initialize an array element of type 'const yysymbol_type_t' with an rvalue of type 'int'
	           0,     5,     9,     2,     2,     2,     2,     2,     2,     2,
	           ^
	    input.c:545:15: error: cannot initialize an array element of type 'const yysymbol_type_t' with an rvalue of type 'int'
	           0,     5,     9,     2,     2,     2,     2,     2,     2,     2,
	                  ^

At some point we could use yysymbol_type_t's enumerators to define
	yytranslate.  Meanwhile...

* data/skeletons/yacc.c (yytranslate): Use the original integral type
	to define it.
	(YYTRANSLATE): Cast the result into yysymbol_type_t.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yysymbol_type_t: always assign an enumerator
	Currently we define enumerators only for symbols that have an
	identifier.  That rules out tokens such as '+', and nonterminals such
	as foo-bar and foo.bar.  As a consequence we are taking chances: the
	compiler might compile yysymbol_type_t as too small an integral type
	for some symbol codes.

* data/skeletons/bison.m4 (b4_symbol_sid): Forge a unique symbol
	identifier for symbols that don't have an ID.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: use symbol numbers instead of YYTRANSLATE
	* examples/c/bistromathic/parse.y: here.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: prefer YYSYMBOL_YYERROR to YYSYMBOL_error
	* data/skeletons/bison.m4 (b4_symbol_sid): Map "error" to YYSYMBOL_YYERROR.
	* data/skeletons/yacc.c: Adjust.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: also define a symbol number for the empty token
	This is not only cleaner, it also protects us from mixing signed
	values (YYEMPTY is #defined as -2) with unsigned types (the
	yysymbol_type_t enum is typically compiled as a small unsigned).
	For instance GCC 9:

input.c: In function 'yyparse':
	    input.c:1107:7: error: conversion to 'unsigned int' from 'int'
	                           may change the sign of the result
	                           [-Werror=sign-conversion]
	     1107 |   yyn += yytoken;
	          |       ^~
	    input.c:1107:10: error: conversion to 'int' from 'unsigned int'
	                            may change the sign of the result
	                            [-Werror=sign-conversion]
	     1107 |   yyn += yytoken;
	          |          ^~~~~~~
	    input.c:1108:47: error: comparison of integer expressions of
	                            different signedness:
	                            'yytype_int8' {aka 'const signed char'} and
	                            'yysymbol_type_t' {aka 'enum yysymbol_type_t'}
	                            [-Werror=sign-compare]
	     1108 |   if (yyn < 0 || YYLAST < yyn || yycheck[yyn] != yytoken)
	          |                                               ^~
	    input.c:702:25: error: operand of ?: changes signedness from 'int'
	                           to 'unsigned int' due to unsignedness of
	                           other operand [-Werror=sign-compare]
	      702 | #define YYEMPTY         (-2)
	          |                         ^~~~
	    input.c:1220:33: note: in expansion of macro 'YYEMPTY'
	     1220 |   yytoken = yychar == YYEMPTY ? YYEMPTY : YYTRANSLATE (yychar);
	          |                                 ^~~~~~~
	    input.c:1220:41: error: unsigned conversion from 'int' to
	                            'unsigned int' changes value
	                            from '-2' to '4294967294'
	                            [-Werror=sign-conversion]
	     1220 |   yytoken = yychar == YYEMPTY ? YYEMPTY : YYTRANSLATE (yychar);
	          |                                         ^

Eventually, it might be interesting to move away from -2 (which is the
	only possible negative symbol number) and use the next available
	number, to save bits.  We could actually even simply use "0" and shift
	the rest, which would allow to write "!yytoken" to mean really
	"yytoken != YYEMPTY".

* data/skeletons/c.m4 (b4_declare_symbol_enum): Define YYSYMBOL_YYEMPTY.
	* data/skeletons/yacc.c: Use it.

* src/parse-gram.y (yyreport_syntax_error): Use YYSYMBOL_YYEMPTY, not
	YYEMPTY, when dealing with a symbol.

* tests/regression.at: Adjust.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: use yysymbol_type_t instead of int for yytoken
	Now that we have a proper type for internal symbol numbers, let's use
	it.  More code needs conversion, e.g., printers and destructors, but
	they are shared with glr.c, which is not ready yet for this change.

It will also help us deal with warnings such as (GCC9 on GNU/Linux):

input.c: In function 'int yyparse()':
	    input.c:475:37: error: enumeral and non-enumeral type in conditional expression [-Werror=extra]
	      475 |   (0 <= (YYX) && (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYSYMBOL_YYUNDEF)
	          |    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	    input.c:1024:17: note: in expansion of macro 'YYTRANSLATE'
	     1024 |       yytoken = YYTRANSLATE (yychar);
	          |                 ^~~~~~~~~~~

* data/skeletons/yacc.c (yytranslate, yysymbol_name)
	(yyparse_context_t, yyexpected_tokens, yypstate_expected_tokens)
	(yysyntax_error_arguments):
	Use yysymbol_type_t instead of int.

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-04-01  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: introduce an enum that defines the symbol's number
	There's a number of advantage in exposing the symbol (internal)
	numbers:

- custom error messages can use them to decide how to represent a
	  given symbol, or a set of symbols.

- we need something similar in uses of yyexpected_tokens.  For
	  instance, currently, bistromathic's completion() reads:

int ntokens = expected_tokens (line, tokens, YYNTOKENS);
	    [...]
	    for (int i = 0; i < ntokens; ++i)
	      if (tokens[i] == YYTRANSLATE (TOK_VAR))
	      [...]
	      else if (tokens[i] == YYTRANSLATE (TOK_FUN))
	      [...]
	      else
	      [...]

- now that it's a compile-time expression, we can easily build static
	  tables, switch, etc.

- some users depended on the ability to get the token number from a
	  symbol to write test cases for their scanners.  But Bison 3.5
	  removed the table this feature depended upon (a reverse
	  yytranslate).  Now they can check against the actual symbol number,
	  without having pay (space and time) a conversion.
	  See https://lists.gnu.org/r/bug-bison/2020-01/msg00001.html, and
	  https://lists.gnu.org/archive/html/bug-bison/2020-03/msg00015.html.

- it helps us clearly separate the internal symbol numbers from the
	  external token numbers, whose difference is sometimes blurred in the
	  code when values coincide (e.g. "yychar = yytoken = YYEOF").

- it allows us to get rid of ugly macros with inconsistent names such
	  as YYUNDEFTOK and YYTERROR, and to group related definitions
	  together.

- similarly it provides a clean access to the $accept symbol (which
	  proves convenient in a current experimentation of mine with several
	  %start symbols).

Let's declare this type as a private type (in the *.c file, not
	the *.h one).  So it does not need to be influenced by the api prefix.

* data/skeletons/bison.m4 (b4_symbol_sid): New.
	(b4_symbol): Use it.
	* data/skeletons/c.m4 (b4_symbol_enum, b4_declare_symbol_enum): New.
	* data/skeletons/yacc.c: Use b4_declare_symbol_enum.
	(YYUNDEFTOK, YYTERROR): Remove.
	Use the corresponding symbol enum instead.

2020-03-30  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes about token numbers
	* data/skeletons/bison.m4, data/skeletons/c.m4: here.

2020-03-30  Akim Demaille  <akim.demaille@gmail.com>

java: move away from _ for internationalization
	The "_" is becoming a keyword in Java, which causes tons of warnings
	currently in our test suite.  GNU Gettext is now using "i18n" instead
	of "_"
	(https://git.savannah.gnu.org/gitweb/?p=gettext.git;a=commitdiff;h=e89fea36545f27487d9652a13e6a0adbea1117d0).

* data/skeletons/java.m4: Use "i18n", not "_".
	* examples/java/calc/Calc.y, tests/calc.at: Adjust.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

c: use YYNOMEM instead of -2
	See 84b1972c96060866b4bd94a33b97711f8f7d0b6c.

* data/skeletons/glr.c, data/skeletons/yacc.c (YYNOMEM): New.
	Use it.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

todo: update
	* TODO (Token Number): We have to clean this.
	(Naming conventions, Symbol numbers): New.
	(Bad styling): Addressed in e21ff47f5d0b64da693a47b7dd200a1a44a5bbeb.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

java: make yysyntaxErrorArguments a private detail
	* data/skeletons/lalr1.java (yysyntaxErrorArguments): Move it from the
	context, to the parser object.
	Generate only for detailed and verbose error messages.
	* tests/local.at (AT_YYERROR_DEFINE(java)): Use yyexpectedTokens
	instead.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

skeletons: make yysyntax_error_arguments a private detail
	We could just "inline yysyntax_error_arguments back" in the routines
	it was originally extracted from, but I think the code is nicer to
	read this way.

* data/skeletons/glr.c (yysyntax_error_arguments): Generate only for
	detailed and verbose error messages.
	* data/skeletons/yacc.c: Likewise.
	* data/skeletons/lalr1.cc (parser::context::yysyntax_error_arguments):
	Move as...
	(parser::yysyntax_error_arguments_): this.
	And only for detailed and verbose error messages.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: avoid using yysyntax_error_arguments
	* data/skeletons/lalr1.cc (context::token): New.
	* tests/local.at (yyreport_syntax_error): Don't use
	yysyntax_error_arguments.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

bison: avoid using yysyntax_error_arguments
	* src/parse-gram.y (yyreport_syntax_error): Use yyparse_context_token
	and yyexpected_tokens.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

tests: yacc.c: avoid yysyntax_error_arguments
	Because glr.c shares the same testing routines, we also need to
	convert it.

* data/skeletons/glr.c (yyparse_context_token): New.
	* tests/local.at (yyreport_syntax_error): here.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

examples: don't use yysyntax_error_arguments
	Suggested by Adrian Vogelsgesang.
	https://lists.gnu.org/archive/html/bison-patches/2020-02/msg00069.html

* data/skeletons/lalr1.java (Context.EMPTY, Context.getToken): New.
	(Context.yyntokens): Rename as...
	(Context.NTOKENS): this.
	Because (i) all the Java coding styles recommend upper case for
	constants, and (ii) the Java Skeleton exposes Lexer.EOF, not
	Lexer.YYEOF.
	* data/skeletons/yacc.c (yyparse_context_token): New.
	* examples/c/bistromathic/parse.y (yyreport_syntax_error): Don't use
	yysyntax_error_arguments.
	* examples/java/calc/Calc.y (yyreportSyntaxError): Likewise.

2020-03-28  Akim Demaille  <akim.demaille@gmail.com>

skeletons: fix incorrect type for translatable tokens
	* data/skeletons/glr.c, data/skeletons/lalr1.c, data/skeletons/yacc.c:
	Fix confusion between the "translatable" and the "translate" tables.

2020-03-23  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: use negative numbers for errors in auxiliary functions
	yyparse returns 0, 1, 2 since ages (accept, reject, memory exhausted).
	Some of our auxiliary functions such as yy_lac and
	yyreport_syntax_error also need to return error codes and also use 0,
	1, 2.  Because it uses yy_lac, yyexpected_tokens also needs to return
	"problem", "memory exhausted", but in case of success, it needs to
	return the number of tokens, so it cannot use 1 and 2 as error code.
	Currently it uses -1 and -2, which is later converted into 1 and 2 as
	yacc.c expects it.

Let's simplify this and use consistently -1 and -2 for auxiliary
	functions that are not exposed (or not yet exposed) to the user.  In
	particular this will save the user from having to convert
	yyexpected_tokens's -2 into yyreport_syntax_error's 2: both return -1
	or -2.

* data/skeletons/yacc.c (yy_lac, yyreport_syntax_error)
	(yy_lac_stack_realloc): Return -1, -2 for errors instead of 1, 2.
	Adjust callers.
	* examples/c/bistromathic/parse.y (yyreport_syntax_error): Do take
	error codes into account.
	Issue a syntax error message even if we ran out of memory.
	* src/parse-gram.y, tests/local.at (yyreport_syntax_error): Adjust.

2020-03-23  Akim Demaille  <akim.demaille@gmail.com>

style: reduce length of private constant
	* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/yacc.c
	(YYERROR_VERBOSE_ARGS_MAXIMUM): Rename as...
	(YYARGS_MAX): this.
	* src/parse-gram.y (YYERROR_VERBOSE_ARGS_MAXIMUM): Rename as...
	(ARGS_MAX): this.

2020-03-23  Akim Demaille  <akim.demaille@gmail.com>

doc: c++: promote api.token.raw
	* doc/bison.texi (Calc++ Parser): Here.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: calc: no need for super long inputs
	* etc/bench.pl.in ($iterations): Restore initial value, -1, meaning
	"at least one second".
	($calc_input): There is no need to generate 400 lines.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: calc: work on a string instead of a file
	The cost of the file layer is large and makes benchmarks too coarse,
	as seen for in following example, first with a file, then with a
	literal string:

0. %skeleton "yacc.c" %define parse.lac full
	    1. %skeleton "yacc-v1.c" %define nofinal %define parse.lac full
	    2. %skeleton "yacc-v2.c" %define nofinal %define parse.lac full
	    3. %skeleton "yacc-v3.c" %define nofinal %define parse.lac full
	    4. %skeleton "yacc.c"
	    5. %skeleton "yacc-v1.c" %define nofinal
	    6. %skeleton "yacc-v2.c" %define nofinal
	    7. %skeleton "yacc-v3.c" %define nofinal
	    --------------------------------------------------
	    Benchmark           Time           CPU Iterations
	    --------------------------------------------------
	    BM_y0           32558 ns      32537 ns      21228
	    BM_y1           32400 ns      32369 ns      21233
	    BM_y2           33485 ns      33464 ns      20625
	    BM_y3           32139 ns      32125 ns      21446
	    BM_y4           31343 ns      31329 ns      21747
	    BM_y5           31344 ns      31317 ns      22035
	    BM_y6           31287 ns      31255 ns      22039
	    BM_y7           31387 ns      31373 ns      22178
	    --------------------------------------------------
	    Benchmark           Time           CPU Iterations
	    --------------------------------------------------
	    BM_y0           10642 ns      10634 ns      63601
	    BM_y1           10657 ns      10654 ns      63625
	    BM_y2           10441 ns      10432 ns      65957
	    BM_y3           10558 ns      10554 ns      64546
	    BM_y4            9521 ns       9516 ns      72011
	    BM_y5            9179 ns       9157 ns      75028
	    BM_y6            9360 ns       9356 ns      73770
	    BM_y7            9365 ns       9359 ns      72609

Of course, at the same time it is less realistic: most users read
	files rather that strings, so it might lead to us to pay attention to
	costs most people don't see.

* etc/bench.pl.in (&calc_input): Output into a file given as argument.
	Output in C syntax.
	(&generate_grammar_calc): Use it.
	Simplify the grammar: remove operators we don't care about.
	Rewrite the scanner to work on a char* instead of a FILE*.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: add a "latest" symlink
	* etc/bench.pl.in: here.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: use the same prefix in both bench methods
	* etc/bench.pl.in (&bench_with_timethese): Also use y$i, as in
	&bench_with_gbenchmark.
	(&generate_grammar_calc): Don't add a prefix, let the callers do it.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: use a C++-11 compiler
	See https://github.com/google/benchmark#a-faster-keeprunning-loop.

* etc/bench.pl.in ($cxx): Be C++11.
	(&bench_with_gbenchmark): Adjust.

2020-03-22  Akim Demaille  <akim.demaille@gmail.com>

bench: create a README file with benches
	* etc/bench.pl.in (&bench_with_gbenchmark): Here.

2020-03-21  Akim Demaille  <akim.demaille@gmail.com>

bench: calc: add support for google benchmark
	* etc/bench.pl.in (&compiler): New, extracted from...
	(&compile): here.
	Don't link when using gbm.
	(&calc_input): Don't make massive input for micro
	benchmarks.
	(&generate_grammar_calc): When using gbm, use api.prefix to avoid name
	collisions.
	Be ready to issue BENCHMARKS instead of a main.
	(&bench): Rename as...
	(&bench_with_timethese): this.
	(&bench_with_gbenchmark): New.
	(&bench): New.
	Dispatch on these two.

2020-03-21  Akim Demaille  <akim.demaille@gmail.com>

bench: better error messages on invalid input
	* etc/bench.pl.in: here.

2020-03-21  Akim Demaille  <akim.demaille@gmail.com>

bench: simplify the calc grammar
	* etc/bench.pl.in (generate_grammar_calc): We don't need global_result
	etc.

2020-03-21  Akim Demaille  <akim.demaille@gmail.com>

bench: die clearly on incorrect --grammar arguments
	* etc/bench.pl.in (getopt): here.

2020-03-17  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-03-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: style: prefer switch to if
	* data/skeletons/yacc.c: Prefer switch to decode yy_lac's return value.

2020-03-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: yypstate_expected_tokens
	In push parsers, when asking for the list of expected tokens at some
	point, it makes no sense to build a yyparse_context_t: the yypstate
	alone suffices (the only difference being the lookahead).  Instead of
	forcing the user to build a useless shell around yypstate, let's offer
	yypstate_expected_tokens.

See https://lists.gnu.org/r/bison-patches/2020-03/msg00025.html.

* data/skeletons/yacc.c (yypstate): Declare earlier, so that we can
	use it for...
	(yypstate_expected_tokens): this new function, when in push parsers.
	Adjust dependencies.
	* examples/c/bistromathic/parse.y: Simplify: use
	yypstate_expected_tokens.
	Style fixes.
	Reduce scopes (reported by Joel E. Denny).

2020-03-09  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: simplify
	* examples/c/bistromathic/parse.y (expected_tokens): Remove useless "break".

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

merge branch 'maint'
	* upstream/maint:
	  maint: post-release administrivia
	  version 3.5.3
	  news: update for 3.5.3
	  yacc.c: make sure we properly propagated the user's number for error
	  diagnostics: don't crash because of repeated definitions of error
	  style: initialize some struct members
	  diagnostics: beware of zero-width characters
	  diagnostics: be sure to close the styling when lines are too short
	  muscles: fix incorrect decoding of $
	  code: be robust to reference with invalid tags
	  build: fix typo
	  doc: update recommandation for libtextstyle
	  style: comment changes
	  examples: use consistently the GFDL header for readmes
	  style: remove useless declarations
	  typo: succesful -> successful
	  README: point to tests/bison, and document --trace
	  gnulib: update
	  maint: post-release administrivia

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.3
	* NEWS: Record release date.

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

news: update for 3.5.3

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: make sure we properly propagated the user's number for error
	* data/skeletons/yacc.c (YYERRCODE): Be truthful.
	* tests/input.at (Redefining the error token): Check that.

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: don't crash because of repeated definitions of error
	According to https://www.unix.com/man-page/POSIX/1posix/yacc/, the
	user is allowed to specify her user number for the error token:

The token error shall be reserved for error handling. The name
	    error can be used in grammar rules. It indicates places where the
	    parser can recover from a syntax error. The default value of error
	    shall be 256. Its value can be changed using a %token
	    declaration. The lexical analyzer should not return the value of
	    error.

I think this feature is useless, the user should not have to deal with
	that.  The intend is probably to give the user a means to use 256 if
	she wants to, but provided "error" cleared the path first by being
	assigned another number.  In the case of Bison, 256 is assigned to
	"error" at the end if the user did not use it for a token of hers.  So
	this feature is useless.

Yet it is valid, and if the user assigns twice a token number to
	"error", then the second time we want to complain about it and want to
	show the original definition.  At this point, we try to display the
	built-in definition of "error", whose location is NULL, and we crash.

Rather, the location of the first user definition of "error" should
	become its defining location.

Reported byg Ahcheong Lee.
	https://lists.gnu.org/r/bug-bison/2020-03/msg00007.html

* src/symtab.c (symbol_class_set): If this is a declaration and the
	symbol was not declared yet, keep this as defining location.
	* tests/input.at (Redefining the error token): New.

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

style: initialize some struct members
	* src/symtab.c (sym_content_new): Initialize all the location members.
	Not needed by the code, but disturbing values when using a debugger.

2020-03-08  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: beware of zero-width characters
	Currenly we rely on (visual) width of the characters to decide where
	to open and close the styling of the quoted lines.  This breaks when
	we deal with zero-width characters: we cannot just rely on (visual)
	columns, we need to know whether we are before, inside, or after the
	highlighted portion.

* src/location.c (location_caret): col_end: no longer add 1, "regular"
	characters have a width of 1, only 0-width characters have 0-width.
	opened: replace with 'state', a three-valued enum.
	Don't reopen the style if we already did.
	* tests/diagnostics.at (Zero-width characters): New.

2020-03-07  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: be sure to close the styling when lines are too short
	     bar.y:4.12-17: <error>error:</error> redefining user token number of foo
	    -    4 | %token foo <error>123
	    +    4 | %token foo <error>123</error>
	           |            <error>^~~~~~</error>

* src/location.c (location_caret): Be sure to close.
	* tests/diagnostics.at (Line is too short, and then you die): New.

2020-03-07  Akim Demaille  <akim.demaille@gmail.com>

muscles: fix incorrect decoding of $
	Bug introduced in 458171e6df5a0110a35ee45ad8b2e9f6fb426f1d.
	https://lists.gnu.org/archive/html/bison-patches/2013-11/msg00009.html

Reported by Ahcheong Lee.
	https://lists.gnu.org/r/bug-bison/2020-03/msg00010.html

* src/muscle-tab.c (COMMON_DECODE): "$" is coded as "$][", not "$[][".
	* tests/input.at ("%define" enum variables): Check that case.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

code: be robust to reference with invalid tags
	Because we want to support $<a->b>$, we must accept -> in type tags,
	and reject $<->$, as it is unfinished.
	Reported by Ahcheong Lee.

* src/scan-code.l (yylex): Make sure "tag" does not end with -, since
	-> does not close the tag.
	* tests/input.at (Stray $ or @): Check this.

2020-03-06  Akimn Demaille  <akim.demaille@gmail.com>

build: fix typo
	* build-aux/cross-options.pl: here.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

doc: update recommandation for libtextstyle
	* README: here.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/symtab.h, src/lr0.c: here.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

examples: use consistently the GFDL header for readmes
	* examples/c++/README.md, examples/c++/calc++/README.md,
	* examples/c/calc/README.md, examples/c/lexcalc/README.md,
	* examples/c/reccalc/README.md:
	Prefer the GFDL banner to the GPL one.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

style: remove useless declarations
	* src/reader.h: Don't duplicate what parse-gram.h already exposes.
	* src/lr0.h: Remove useless include.

2020-03-06  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

typo: succesful -> successful
	* data/skeletons/lalr1.cc: here
	* etc/bench.pl.in: here
	* src/location.c: and here.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

README: point to tests/bison, and document --trace
	Reported by Victor Morales Cayuela.

* README, README-hacking.md: here.

2020-03-06  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-03-05  Akim Demaille  <akim.demaille@gmail.com>

README: point to tests/bison, and document --trace
	Reported by Victor Morales Cayuela.

* README, README-hacking.md: here.

2020-03-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: simplify yyparse_context_t member names
	* data/skeletons/yacc.c (yyparse_context_t): Rename yyes_p and
	yyes_capacity_p as...
	(yyes, yyes_capacity): These.

2020-03-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: yyerror_range does not need to be preserved accross calls
	* data/skeletons/yacc.c (b4_parse_state_variable_macros): Don't define
	yyerror_range.
	(yyparse): Add yyerror_range as local variable.

2020-03-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: push: undefine the pstate macros for the epilogue
	* data/skeletons/yacc.c (b4_macro_define, b4_macro_undef)
	(b4_pstate_macro_define, b4_parse_state_variable_macros):
	New.
	Use them.
	* examples/c/bistromathic/parse.y: Remove now useless undefs.

2020-03-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: push: initialize the pstate variables in pstate_new
	Currently pstate_new does not set up its variables, this task is left
	to yypush_parse.  This was probably to share more code with usual pull
	parsers, where these (local) variables are indeed initialized by
	yyparse.

But as a consequence yyexpected_tokens crashes at the very beginning
	of the parse, since, for instance, the stacks are not even set up.
	See https://lists.gnu.org/r/bison-patches/2020-03/msg00001.html.

The fix could have very simple, but the documentation actually makes
	it very clear that we can reuse a pstate for several parses:

After yypush_parse returns a status other than YYPUSH_MORE, the
	    parser instance yyps may be reused for a new parse.

so we need to restore the parser to its pristine state so that (i) it
	is ready to run the next parse, (ii) it properly supports
	yyexpected_tokens for the next run.

* data/skeletons/yacc.c (b4_initialize_parser_state_variables): New,
	extracted from the top of yyparse/yypush_parse.
	(yypstate_clear): New.
	(yypstate_new): Use it when push parsers are enabled.
	Define after the yyps macros so that we can use the same code as the
	regular pull parsers.
	(yyparse): Use it when push parsers are _not_ enabled.

* examples/c/bistromathic/bistromathic.test: Check the completion on
	the beginning of the line.

2020-03-04  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* data/skeletons/yacc.c, tests/torture.at: here.

2020-03-04  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: properly compute the lcp, as expected by readline
	Currently completion on "at" proposes only "atan", but does not
	actually complete "at" into "atan".

* examples/c/bistromathic/parse.y (completion): Install the lcp in
	matches[0].
	* examples/c/bistromathic/bistromathic.test: Check that case.

2020-03-04  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: don't require spaces after operators for completion
	Currently "(1+<TAB>" does not work as expected, because "+" is not a
	word breaking character.

* examples/c/bistromathic/parse.y (init_readline): Specify our word
	breaking characters.
	* examples/c/bistromathic/bistromathic.test: Avoid trailing spaces.

2020-03-02  Akim Demaille  <akim.demaille@gmail.com>

bistromathic: check completion
	* examples/c/bistromathic/bistromathic.test: here.
	* examples/c/bistromathic/parse.y (expected_tokens): Fix a memory
	leak.

2020-03-02  Akim Demaille  <akim.demaille@gmail.com>

m4: remove b4_function_define and b4_function_declare
	* data/skeletons/c.m4: here.

2020-03-02  Akim Demaille  <akim.demaille@gmail.com>

m4: decommission b4_function_declare
	* data/skeletons/glr.c, data/skeletons/glr.cc, data/skeletons/yacc.c:
	Stop using b4_function_declare.

2020-03-02  Akim Demaille  <akim.demaille@gmail.com>

m4: decommission function generating macro
	These macros have been extremely useful when we had to support K&R C,
	which we dropped long ago.  Now, they merely make the code uselessly
	hard to read.

* data/skeletons/c.m4, data/skeletons/glr.c, data/skeletons/glr.cc,
	* data/skeletons/yacc.c:
	Stop using b4_function_define.

2020-03-01  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: demonstrate the use of yyexpected_tokens
	Let's use GNU readline and its TAB autocompletion to demonstrate the
	use of yyexpected_tokens.

This shows a number of weaknesses in our current approach:

- some macros (yyssp, etc.) from push parsers "leak" in user code, we
	  need to undefine them

- the context needed by yyexpected_tokens does not need the token,
	  yypstate actually suffices

- yypstate is not properly setup when first allocated, which results
	  in a crash of yyexpected_tokens if fired before a first token was
	  read.  We should move initialization from yypush_parse into
	  yypstate_new.

* examples/c/bistromathic/parse.y (yylex): Take input as a string, not
	a file.
	(EXIT): New token.
	(input): Adjust to work only on a line.
	(line): Remove.
	(symbol_count, process_line, expected_tokens, completion)
	(init_readline): New.
	* examples/c/bistromathic/bistromathic.test: Adjust expectations.

2020-03-01  Akim Demaille  <akim.demaille@gmail.com>

examples: use consistently the GFDL header for readmes
	* examples/c++/README.md, examples/c++/calc++/README.md,
	* examples/c/calc/README.md, examples/c/lexcalc/README.md,
	* examples/c/pushcalc/README.md, examples/c/reccalc/README.md:
	Prefer the GFDL banner to the GPL one.

2020-03-01  Akim Demaille  <akim.demaille@gmail.com>

gnulib: use readline

2020-02-29  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: don't use Flex
	This example will soon use GNU readline, so its scanner should be easy
	to use (concurrently) on strings, not streams.  This is not a place
	where Flex shines, and anyway, these are examples of Bison, not Flex.
	There's already lexcalc and reccalc that demonstrate the use of Flex.

* examples/c/bistromathic/scan.l: Remove.
	* examples/c/bistromathic/parse.y (yylex): New.
	Adjust dependencies.

2020-02-29  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: strengthen tests
	* examples/c/bistromathic/bistromathic.test: here.
	* examples/test: Be clearer on failing tests.

2020-02-29  Akim Demaille  <akim.demaille@gmail.com>

examples: lexcalc: demonstrate location tracking
	The bistromathic example should not use Flex, it makes it too complex.
	But it was the only example to show location tracking with Flex.

* examples/c/lexcalc/lexcalc.test, examples/c/lexcalc/parse.y,
	* examples/c/lexcalc/scan.l: Demonstrate location tracking as is done
	in bistromathic.

2020-02-27  Akim Demaille  <akim.demaille@gmail.com>

c++: don't copy the lookahead
	The current implementation of parser::context keeps a copy of the
	lookahead.  This is troublesome since we support move-only types.
	Besides, while GCC is happy with the current implementation, Clang
	complains that the ctor it needs to build the copy of the lookahead is
	not yet available.

461. calc.at:1120: testing Calculator C++ %defines %locations parse.error=verbose %name-prefix "calc" %verbose  ...
	    calc.at:1120: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret -Wno-deprecated -o calc.cc calc.y
	    calc.at:1120: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o calc calc.cc calc-lex.cc calc-main.cc $LIBS
	    stderr:
	    In file included from calc-lex.cc:7:
	    calc.hh:351:12: error: instantiation of function 'calc::parser::basic_symbol<calc::parser::by_type>::basic_symbol' required here, but no definition is available [-Werror,-Wundefined-func-template]
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    calc.hh:273:7: note: forward declaration of template entity is here
	          basic_symbol (const basic_symbol& that);
	          ^
	    calc.hh:351:12: note: add an explicit instantiation declaration to suppress this warning if 'calc::parser::basic_symbol<calc::parser::by_type>::basic_symbol' is explicitly instantiated in another translation unit
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    1 error generated.
	    In file included from calc-main.cc:7:
	    calc.hh:351:12: error: instantiation of function 'calc::parser::basic_symbol<calc::parser::by_type>::basic_symbol' required here, but no definition is available [-Werror,-Wundefined-func-template]
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    calc.hh:273:7: note: forward declaration of template entity is here
	          basic_symbol (const basic_symbol& that);
	          ^
	    calc.hh:351:12: note: add an explicit instantiation declaration to suppress this warning if 'calc::parser::basic_symbol<calc::parser::by_type>::basic_symbol' is explicitly instantiated in another translation unit
	        struct symbol_type : basic_symbol<by_type>
	               ^
	    1 error generated.
	    stdout:
	    calc.at:1120: exit code was 1, expected 0
	    461. calc.at:1120: 461. Calculator C++ %defines %locations parse.error=verbose %name-prefix "calc" %verbose  (calc.at:1120): FAILED (calc.at:1120)

* data/skeletons/lalr1.cc (context::yyla_): Make it a const-ref.
	Move the implementation out of the declaration.

2020-02-27  Akim Demaille  <akim.demaille@gmail.com>

c++: minor fixes
	Address compiler warnings such as

warning: declaration of 'yyla' shadows a member of 'yy::parser::context' [-Wshadow]

* data/skeletons/lalr1.cc (context): Don't use the same names for
	variables and members.
	Use foo_ for private members, as in parser.
	Also, use the + trick in array accesses to please ICC and provide it
	with an int.

2020-02-27  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

c++: add support for parse.error=custom
	* data/skeletons/lalr1.cc: added support here
	* tests/calc.at: added test cases
	* tests/local.at: added yyreport_syntax_error implementation
	   for C++ test cases

2020-02-27  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

c++: add parser::context for syntax error handling
	* data/skeletons/lalr1.cc: here

2020-02-27  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

c++: add support for parse.error=detailed
	* data/skeletons/lalr1.cc: added support here
	* tests/calc.at: added a test case

2020-02-27  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

skeletons: prefer b4_parse_error_{case,bmatch} over manual solution
	Prefer b4_parse_error_case over the adhoc solution
	`m4_case + b4_percent_define_get`. Same for b4_parse_error_bmatch.

* data/skeletons/glr.c: here
	* data/skeletons/yacc.c: here

2020-02-27  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

typo: succesful -> successful
	* data/skeletons/lalr1.cc: here
	* etc/bench.pl.in: here
	* src/location.c: here
	* tests/calc.at: and here

2020-02-24  Akim Demaille  <akim.demaille@gmail.com>

bench.pl: clean up the dust
	* etc/bench.pl.in: Adjust to the current use of %define's values.
	Don't use %error-verbose.
	Prefer Bison to CPP (e.g., api.value.type).
	Avoid returning characters directly, so that %define api.token.raw works.

2020-02-23  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/symtab.h, src/lr0.c: here.

2020-02-23  Akim Demaille  <akim.demaille@gmail.com>

style: avoid using 'this' as an identifier
	LLDB insists on parsing 'this' as a C++ keyword, even when debugging a
	C program.

* src/symtab.c: Please the dictator.

2020-02-23  Akim Demaille  <akim.demaille@gmail.com>

style: remove useless declarations
	* src/reader.h: Don't duplicate what parse-gram.h already exposes.
	* src/lr0.h: Remove useless include.

2020-02-19  Akim Demaille  <akim.demaille@gmail.com>

examples: fix c/calc
	* examples/c/calc/calc.y: Remove experiment traces.

2020-02-19  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-02-15  Akim Demaille  <akim.demaille@gmail.com>

doc: update recommandation for libtextstyle
	* README: here.

2020-02-15  Akimn Demaille  <akim.demaille@gmail.com>

build: fix typo
	* build-aux/cross-options.pl: here.

2020-02-15  Akim Demaille  <akim.demaille@gmail.com>

doc: simplify the cross references
	* doc/bison.texi: here.

2020-02-15  Akim Demaille  <akim.demaille@gmail.com>

doc: document token internationalization
	* doc/bison.texi (Parser Internationalization): Move most of its
	content into...
	(Enabling I18n): this new node.
	(Token I18n): New.
	(Token Decl): Refer to token internationalization.
	(Error Reporting Function): Promote parse.error detailed.

2020-02-15  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-02-15  Victor Morales Cayuela  <victor.morales_cayuela@nokia-sbell.com>

diagnostics: modernize the display of submessages
	Since Bison 2.7, output was indented four spaces for explanatory
	statements.  For example:

input.y:2.7-13: error: %type redeclaration for exp
	    input.y:1.7-11:     previous declaration

Since the introduction of caret-diagnostics, it became less clear.
	Remove the indentation and display submessages as in GCC:

input.y:2.7-13: error: %type redeclaration for exp
	        2 | %type <float> exp
	          |       ^~~~~~~
	    input.y:1.7-11: note: previous declaration
	        1 | %type <int> exp
	          |       ^~~~~

* src/complain.h (SUB_INDENT): Remove.
	(warnings): Add "note" to the enum.
	* src/complain.h, src/complain.c (complain_indent): Replace by...
	(subcomplain): this.
	Adjust all dependencies.
	* tests/actions.at, tests/diagnostics.at, tests/glr-regression.at,
	* tests/input.at, tests/named-refs.at, tests/regression.at:
	Adjust expectations.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

doc: simplify uses of references
	This reverts "doc: work around problems with PDF generation", commit
	d810aa3d8f76b1a4d7d402072f45a0662152ffd4.  Upstream issue is fixed.
	https://lists.gnu.org/r/bug-texinfo/2020-02/msg00006.html

* gnulib: Update.
	* doc/bison.texi: Simplify.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

java: provide a Context ctor
	This is really a private auxiliary inner class, so it should not
	matter.  But it's better style.

* data/skeletons/lalr1.java: here.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

Merge tag 'v3.5.2'
	bison 3.5.2

* tag 'v3.5.2':
	  version 3.5.2
	  news: 3.5.2
	  gnulib: update
	  doc: update Doxygen template
	  java: avoid trailing white spaces
	  m4: fix b4_token_format
	  doc: clearly state that %yacc only makes sense with yacc.c
	  doc: spell check
	  examples: be more robust to spaces in paths
	  larlr1.cc: Reject unsupported values for parse.lac

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.2
	* NEWS: Record release date.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

news: 3.5.2
	* NEWS: Update.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

doc: update Doxygen template
	* Doxyfile.in: Run doxygen -u on it.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

java: avoid trailing white spaces
	* data/skeletons/java.m4 (b4_maybe_throws): Issue a space before when needed.
	* data/skeletons/lalr1.java: Avoid trailing spaces.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

m4: fix b4_token_format
	We used to emit:

/** Token number,to be returned by the scanner.  */
	    static final int NUM = 258;
	    /** Token number,to be returned by the scanner.  */
	    static final int NEG = 259;

with no space after the comma.  Fix that.

* data/skeletons/bison.m4 (b4_token_format): Quote where appropriate.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

doc: clearly state that %yacc only makes sense with yacc.c
	* doc/bison.texi: here.

2020-02-13  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check
	* doc/bison.texi: here.

2020-02-12  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: demonstrate named references
	* examples/c/bistromathic/parse.y: here.

2020-02-12  Akim Demaille  <akim.demaille@gmail.com>

c++: simplify
	* data/skeletons/stack.hh (ssize): Remove, same as size.

2020-02-12  Akim Demaille  <akim.demaille@gmail.com>

tests: check calls to yyerror from the user actions
	This revealed a number of things I had not realized:

- the Java location tracking was aliasing the same pair of positions
	  for all the symbols (see previous commit).

- in impure parsers, it's quite easy to use incorrect locations for
	  diagnostics, since yyerror uses yylloc, which is the location of the
	  lookahead, not that of the current lhs.  So we need something like

{
	      YYLTYPE old_yylloc = yylloc;
	      yylloc = @$;
	      yyerror (]AT_PARAM_IF([result, count, nerrs, ])[buf);
	      yylloc = old_yylloc;
	    }

Maybe we should do that little yylloc dance in the skeleton instead
	  of leaving it to the user?  It might be costly...  But that's only
	  for users of the impure parsers, which are asking for trouble
	  anyway.

- in glr.cc invoking yyerror is somewhat cumbersome: the C++ interface
	  is not available as we are in yyparse (which in C), and yyerror is
	  used by glr.cc itself to bind it to the user's parser::error.  If we
	  call yyerror, we need:

yyerror (]AT_LOCATION_IF([[&@$, ]])[yyparser, ]AT_PARAM_IF([result, count, nerrs, ])[msg);

However calling yy::parser::error is easier, once we know that the
	  current parser object is available as 'yyparser'.  Which also saves
	  us from having to pass the parse-params ourselves:

yyparser.error (]AT_LOCATION_IF([[@$, ]])[msg);

* tests/calc.at: Invoke yyerror by hand, instead of using fprintf etc.
	Adjust expectations.

2020-02-11  Akim Demaille  <akim.demaille@gmail.com>

java: beware not to alias the locations of the various symbols
	* examples/java/calc/Calc.y, tests/calc.at, tests/local.at
	(getStartPos, getEndPos): Always return a new object.
	* doc/bison.texi: Clarify this.

2020-02-11  Akim Demaille  <akim.demaille@gmail.com>

java: check that parse.error custom|detailed work with push parsers
	* tests/calc.at: here.

2020-02-11  Akim Demaille  <akim.demaille@gmail.com>

java: don't expose the Context's members
	* data/skeletons/lalr1.java (Context): Make data members private.
	(Context.getLocation): New.
	* examples/java/calc/Calc.y, tests/java.at, tests/local.at: Adjust.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

build: pacify syntax-check
	* src/complain.c: Fix indentation.
	* cfg.mk: Using strcmp is ok in the tests.
	Test cases and examples don't need Bison's PO support.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

build: prefer %D% and %C% to hard coded values
	* doc/local.mk: here.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

parse.error: document and diagnose the incompatibility with %token-table
	* doc/bison.texi (Tokens from Literals): Move to code using
	%token-table to...
	(Decl Summary: %token-table): here.
	* data/skeletons/bison.m4: Implement mutual exclusion.
	* tests/input.at: Check it.
	* doc/local.mk: Be robust to the removal of doc/.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check
	* doc/bison.texi: here.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: here.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: work around problems with PDF generation
	With texinfo.tex 2019-09-24.13, node names with + are not properly
	handled.
	https://lists.gnu.org/r/bug-texinfo/2020-02/msg00004.html

* doc/bison.texi: Always use the three-argument form for references to
	node with a + in the name.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

java: revert "style: avoid useless initializers"
	This reverts commit ebab1ffca8a728158051481795ae798231cfd93d.
	This commit removed "useless" initializers, going from

/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing
	       STATE-NUM.  */
	    private static final byte yypact_[] = yypact_init ();
	    private static final byte[] yypact_init ()
	    {
	      return new byte[]
	      {
	        25,    -7,    -8,    37,    -8,    40,    -8,    20,    -8,    61,
	        -8,    -8,     3,     9,    51,    -8,    -8,    -2,    -2,    -2,
	        -2,    -2,    -2,    -8,    -8,    -8,     1,    66,    66,     3,
	         3,     3
	      };
	    }

/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing
	       STATE-NUM.  */
	    private static final byte[] yypact_ =
	    {
	        25,    -7,    -8,    37,    -8,    40,    -8,    20,    -8,    61,
	        -8,    -8,     3,     9,    51,    -8,    -8,    -2,    -2,    -2,
	        -2,    -2,    -2,    -8,    -8,    -8,     1,    66,    66,     3,
	         3,     3
	    };

But it turns out that this was on purpose, to work around the 64KB
	limitation in JVM methods.  It was introduced on the 2008-11-10 by
	Di-an Jan in 09ccae9b18a7c09ebf7bb8df2a18c8c4a6def248: "Work around
	Java's ``code too large'' problem for parser tables".  See
	https://lists.gnu.org/r/help-bison/2008-11/msg00004.html.  A real
	test, where we would hit the JVM limitation, would be nice.

To avoid further regressions, add comments.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

skeletons: avoid b4_error_verbose_if, which is confusing
	parse.error has more than two possible values.

* data/skeletons/bison.m4 (b4_error_verbose_if, b4_error_verbose_flag):
	Remove.
	(b4_parse_error_case, b4_parse_error_bmatch): New.
	Adjust dependencies.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

skeletons: decorelate %token-table from verbose error messages
	Reported by Adrian Vogelsgesang.

* data/skeletons/bison.m4: Here.
	* data/skeletons/lalr1.cc: Adjust.

2020-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: here.

2020-02-09  Akim Demaille  <akim.demaille@gmail.com>

doc: clearly state that %yacc only makes sense with yacc.c
	* doc/bison.texi: here.
	* tests/calc.at: Stop testing %yacc with non yacc.c skeletons.

2020-02-09  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

style: stylistic cleanups in the C skeletons
	* data/skeletons/glr.c, data/skeletons/yacc.c:
	Avoid duplicated declaration of yysymbol_name.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: provide Context with a more OO interface
	* data/skeletons/lalr1.java (yyexpectedTokens)
	(yysyntaxErrorArguments): Make them methods of Context.
	(Context.yysymbolName): New.
	* tests/local.at: Adjust.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: add support for parse.error custom
	* data/skeletons/lalr1.java: Add support for custom parse errors.
	(yyntokens_): Make it public.  Under...
	(yyntokens): this name.
	(Context): Capture the location too.
	* examples/c/bistromathic/parse.y,
	* examples/c/bistromathic/bistromathic.test:
	Improve error message.
	* examples/java/calc/Calc.test, examples/java/calc/Calc.y: Use custom
	error messages.
	* tests/calc.at, tests/local.at: Check custom error messages.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: let the Context give access to yyntokens
	* data/skeletons/lalr1.java (Context.yytokens): New.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: make the syntax error format string translatable
	The error format should be translated, but contrary to the case of
	C/C++, we cannot just depend on macros to adapt on the
	presence/absence of '_'.  Let's consider that the message format is to
	be translated iff there are some internationalized tokens.

* src/output.c (prepare_symbol_names): Define b4_has_translations.
	* data/skeletons/java.m4 (b4_trans): New.
	* data/skeletons/lalr1.java: Use it to emit translatable or not the
	format string.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: introduce yyexpectedTokens
	And allow syntax error messages for 'detailed' and 'verbose' to be
	translated.

* data/skeletons/lalr1.java (Context, yyexpectedTokens)
	(yysyntaxErrorArguments): New.
	(yysyntax_error): Use it.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

java: add support for parse.error=detailed
	In Java there is no need for N_ and yytranslate_.  So instead of
	hard-coding the use of N_ in the table of the symbol names, rely on
	b4_symbol_translate.

* src/output.c (prepare_symbol_names): Use b4_symbol_translate instead
	of N_.
	* data/skeletons/c.m4 (b4_symbol_translate): New.
	* data/skeletons/lalr1.java (yysymbolName): New.
	Use it.
	* examples/java/calc/Calc.y: Use parse.error=detailed.
	* tests/calc.at: Check parse.error=detailed.

2020-02-08  Akim Demaille  <akim.demaille@gmail.com>

m4: fix b4_token_format
	We used to emit:

/** Token number,to be returned by the scanner.  */
	    static final int NUM = 258;
	    /** Token number,to be returned by the scanner.  */
	    static final int NEG = 259;

with no space after the comma.  Fix that.

* data/skeletons/bison.m4 (b4_token_format): Quote where appropriate.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: tests: remove now redundant tests
	* tests/javapush.at: here.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: tests: check push parsers like the others
	Currently in javapush.at.

* tests/calc.at: Here.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: tests: remove now redundant tests
	* tests/java.at: Calculator tests are now in calc.at.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: tests: check location tracking in the calculator
	Unfortunately in the Java skeleton the user cannot override the way
	locations are displayed, and locations don't know the structure of the
	positions.  So they cannot implement the tricks used in the C/C++
	skeletons to display "1.1" instead of "1.1-1.2".

* tests/local.at (Java): Add support for column tracking in the
	locations, as we did in examples/java/calc.
	* tests/calc.at: Use AT_CALC_YYLEX.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: tests: prepare the replacement of calculator tests
	Soon calculator tests for Java will move from java.at to calc.at.
	Which implies improving the Java testing infrastructure in
	local.at (for instance really tracking columns in positions, not just
	token number).  Detach java.at from local.at.

* tests/java.at (AT_JAVA_POSITION_DEFINE_OLD): New.
	Use it.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: style: avoid useless initializers
	Instead of

/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing
	       STATE-NUM.  */
	      private static final byte yypact_[] = yypact_init ();
	      private static final byte[] yypact_init ()
	      {
	        return new byte[]
	        {
	          25,    -7,    -8,    37,    -8,    40,    -8,    20,    -8,    61,
	          -8,    -8,     3,     9,    51,    -8,    -8,    -2,    -2,    -2,
	          -2,    -2,    -2,    -8,    -8,    -8,     1,    66,    66,     3,
	           3,     3
	        };
	      }

generate

/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing
	       STATE-NUM.  */
	      private static final byte[] yypact_ =
	      {
	          25,    -7,    -8,    37,    -8,    40,    -8,    20,    -8,    61,
	          -8,    -8,     3,     9,    51,    -8,    -8,    -2,    -2,    -2,
	          -2,    -2,    -2,    -8,    -8,    -8,     1,    66,    66,     3,
	           3,     3
	      };

I have no idea what motivated the previous approach.

* data/skeletons/java.m4 (b4_typed_parser_table_define): Here.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: style: prefer putting the square brackets on the type
	* examples/java/calc/Calc.y, examples/java/simple/Calc.y,
	* tests/calc.at, tests/local.at: here.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: examples: fix the tracking of locations
	* examples/java/calc/Calc.y: The StreamTokenizer cannot "peek" for the
	next character, it reads it, and keeps it for the next call.  So the
	current location is one passed the end of the current token.  To avoid
	this, keep the previous position, and use it to end the current token.
	* examples/java/calc/Calc.test: Adjust.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: examples: prefer switch to chains of else-if
	* examples/java/calc/Calc.y, examples/java/simple/Calc.y: here.
	* examples/java/simple/Calc.y: Use the tokenizer's handling of blanks.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

java: examples: split in two
	* examples/java: Split in...
	* examples/java/simple, examples/java/calc: these.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

traces: don't print the stack before the gotos
	The C, C++ and D skeletons used to show the stack right after popping
	the stack during the reduction.  Now that the stack is printed after
	reaching a new state, that has become useless:

Entering state 1
	    Stack now 0 1
	    Reducing stack by rule 5 (line 83):
	       $1 = token "number" (1)
	    -> $$ = nterm exp (1)
	    Stack now 0
	    Entering state 8
	    Stack now 0 8

Remove the "Stack now 0" line.

* data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/lalr1.java, data/skeletons/yacc.c:
	Here.

2020-02-05  Akim Demaille  <akim.demaille@gmail.com>

traces: show the stack after reading a token
	Currently, if we have long rules and series of shift, we stack states
	without showing stack.  Let's be more incremental, and do how the Java
	skeleton does.

* data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/yacc.c:
	Here.
	Adjust test cases.
	* tests/torture.at (AT_DATA_STACK_TORTURE): Disable stack traces: this
	test produces a very large stack, and showing the stack each time we
	shift a token goes quadatric.

2020-02-04  Akim Demaille  <akim.demaille@gmail.com>

traces: write the "Reading a token" alone on its line
	The Java skeleton displays

Reading a token:
	    Next token is token "number" (1)

while the other display

Reading a token: Next token is token "number" (1)

When generating logs in the scanner, the first part is separated from
	the second, and the end of the scanner logs have the second part
	pasted in.  So let's propagate the Java way, but with the colon.

* data/skeletons/glr.c, data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/lalr1.java, data/skeletons/yacc.c: Do it.
	Adjust test cases and doc.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: use the same calc tests as the other skeletons
	* tests/local.at (AT_LANG_MATCH): New.
	(AT_YYERROR_DECLARE(java), AT_YYERROR_DECLARE_EXTERN(java)): New.
	* tests/calc.at: The grammar file for Java is quite different for the
	others, and continuing to assemble it from pieces makes the grammar
	file hard to understand.  Let's also dispatch on the language to
	assemble it, and isolate Java from the others.
	Most of this comes from java.at.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: add access to the number of errors
	* data/skeletons/lalr1.java (yynewrrs, getNumberOfErrors): New.
	Formatting changes.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: formatting changes
	* data/skeletons/java.m4, data/skeletons/lalr1.java: here.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: avoid trailing white spaces
	* data/skeletons/java.m4 (b4_maybe_throws): Issue a space before when needed.
	* data/skeletons/lalr1.java: Avoid trailing spaces.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: example: properly track the locations
	This example, so far, was tracking the current token number, not the
	current column number.  This is not nice for an example...

* examples/java/Calc.y (PositionReader): New.
	Use it.
	* examples/java/Calc.test: Check the output.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: example: improve
	* examples/java/Calc.y: Propagate the exit status.
	Support -p.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

java: example: rely on autoboxing
	AFAICT, autoboxing/unboxing was added in Java 5 (September 30, 2004).
	I think we can afford to use it.  It should help us merge some Java
	tests with the main ones.

However, beware that != does not unbox: it compares the object
	addresses.

* examples/java/Calc.y, tests/java.at: Simplify.
	* examples/java/Calc.test, tests/java.at: Improve tests.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

tests: comment changes
	* tests/calc.at: Shorten titles and reduce redundancy.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

skeletons: add support for %code epilogue
	When building the test cases, emitting code in the epilogue is very
	constraining.  Let's make it simpler thanks to %code epilogue.

However, I don't want to document this: it is bad style to use it (we
	should avoid having too many ways to write the same thing,
	TI!MTOWTDI), just put your code in the true epilogue section.

* data/skeletons/glr.c, data/skeletons/lalr1.d, data/skeletons/lalr1.java,
	* data/skeletons/yacc.c: Implement support for %code epilogue.
	Remove useless comments.
	* tests/calc.at, tests/java.at: Simplify.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

examples: bistromathic: fix location tracking
	* examples/c/bistromathic/scan.l (LOCATION_STEP): New.
	Use to properly ignore blanks.
	* examples/c/bistromathic/bistromathic.test: Check that case.

2020-02-02  Akim Demaille  <akim.demaille@gmail.com>

doc: document new features of parse.error
	* doc/bison.texi (Error Reporting): Rename as...
	(Error Reporting Function): this.
	Adjust dependencies.
	Make it a subsection of this...
	(Error Reporting): new section.
	(Syntax Error Reporting Function): New.
	(parse.error): Update description.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

glr.c: add support for parse.error=custom
	* data/skeletons/glr.c (yyreportSyntaxError): Call the user's
	yyreport_syntax_error in custom mode.
	* tests/calc.at: Check it.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

glr.c: add support for parse.error=detailed
	* data/skeletons/glr.c (yystrlen, yysymbol_name): New.
	Implement parse.error detailed.
	* tests/calc.at: Check it.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

glr.c: introduce yyexpected_tokens and yysyntax_error_arguments
	Modeled after what was done in yacc.c, yet somewhat different since
	yyGLRStack play the role of the full yyparse_context_t.  It will
	probably be defined as a alias, to make interfaces compatible.

LAC is not supported here.  And is somewhat tricky.

* data/skeletons/glr.c (yyexpected_tokens, yysyntax_error_arguments): New.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

glr.c: move code around
	* data/skeletons/glr.c: Move the handling of error messages after the
	definition of types.  Especially after yyGLRStack, which we will need
	in forthcoming patches.
	Simplify: yyGLRStack is defined at this point.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

glr.c: rename yyStateNum as yy_state_t
	* data/skeletons/glr.c: here.
	For consistency with yacc.c.

2020-01-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: fix misleading indentation
	* data/skeletons/yacc.c: here.

2020-01-27  Akim Demaille  <akim.demaille@gmail.com>

doc: simplify uses of @ref
	The PDF output is more consistent: some nodes were not pointed to with
	their title.  The HTML output becomes "see section Foo" instead of
	"see Foo", but this should be addressed in the next Texinfo release.
	Info output is simplified, as it uses only the node name and not its
	title.  But it's considered easier to read this way.
	See https://lists.gnu.org/r/help-texinfo/2020-01/msg00031.html.

* doc/bison.texi: Set @xrefautomaticsectiontitle on.
	Simplify all uses of ref.

2020-01-27  Akim Demaille  <akim.demaille@gmail.com>

examples: be more robust to spaces in paths
	Reported by Nikki Valen.
	https://lists.gnu.org/r/bug-bison/2020-01/msg00032.html

* examples/test ($prog): Remove, replaced by...
	(prog): This new function, which pays attention to quoting shell
	variables.

2020-01-27  Akim Demaille  <akim.demaille@gmail.com>

doc: don't pretend trigonometry is part of arithmetics
	* doc/bison.texi (arith_funs): Rename as...
	(funs): this.

2020-01-27  Akim Demaille  <akim.demaille@gmail.com>

doc: update Doxygen template
	* Doxyfile.in: Run doxygen -u on it.

2020-01-27  Akim Demaille  <akim.demaille@gmail.com>

examples: add a complete example with all the bells and whistles
	* examples/c/bistromathic/Makefile,
	* examples/c/bistromathic/README.md,
	* examples/c/bistromathic/bistromathic.test,
	* examples/c/bistromathic/local.mk,
	* examples/c/bistromathic/parse.y,
	* examples/c/bistromathic/scan.l:
	New.

* Makefile.am (AM_YFLAGS_WITH_LINES): Add -Wdangling-alias.
	* examples/test: Make failure errors easier to read.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

examples: add an example of a push parser
	Add an example to demonstrate the use of push parser.  I'm pleasantly
	surprised: parse.error=detailed works like a charm with push parsers.

* examples/c/local.mk, examples/c/pushcalc/Makefile
	* examples/c/pushcalc/README.md, examples/c/pushcalc/calc.test,
	* examples/c/pushcalc/calc.y, examples/c/pushcalc/local.mk:
	New.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

examples: more tests
	* examples/c/mfcalc/mfcalc.test: here.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

examples: clean up
	* examples/c/calc/calc.y: Restore to its original state, with
	parse.error=detailed instead of parse.error=custom (this example
	should be simple).
	* examples/c/calc/calc.test: Check syntax errors.
	* examples/c/lexcalc/parse.y: Add comments.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

tests: check custom error messages and push parsers
	* tests/local.at (AT_LAC_IF): New.
	* tests/calc.at: And also check the suppot for LAC.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* data/skeletons/lalr1.java: here.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

bison: pretend to 3.6 already
	* src/parse-gram.y: here.

2020-01-26  Akim Demaille  <akim.demaille@gmail.com>

git: update ignores
	* lib/.gitignore: here.

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: modernize bison's syntax errors
	We used to display the unexpected token first:

$ bison foo.y
	    foo.y:1.8-13: error: syntax error, unexpected %token, expecting character literal or identifier or <tag>
	        1 | %token %token
	          |        ^~~~~~

GCC uses a different format:

$ gcc-mp-9 foo.c
	    foo.c:1:5: error: expected identifier or '(' before ')' token
	        1 | int()()()
	          |     ^

and so does Clang:

$ clang-mp-9.0 foo.c
	    foo.c:1:5: error: expected identifier or '('
	    int()()()
	        ^
	    1 error generated.

They display the unexpected token last (or not at all).  Also, they
	don't waste width with "syntax error".  Let's try that.  It gives, for
	the same example as above:

$ bison foo.y
	    foo.y:1.8-13: error: expected character literal or identifier or <tag> before %token
	        1 | %token %token
	          |        ^~~~~~

* src/complain.h, src/complain.c (syntax_error): New.
	* src/parse-gram.y (yyreport_syntax_error): Use it.

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: report syntax errors in color
	* src/parse-gram.y (parse.error): Set to 'custom'.
	(yyreport_syntax_error): New.
	* data/bison-default.css (.expected, .unexpected): New.
	* tests/diagnostics.at: Adjust.

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: translate bison's own tokens
	As a test case, support translations in Bison itself.

* src/parse-gram.y: Mark the translatable tokens.
	While at it, use clearer names.
	* tests/input.at: Adjust expectations.

2020-01-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: handle -fno-caret in the called functions
	Don't force callers of location_caret to have to deal with flags that
	disable it.

* src/location.h, src/location.c (location_caret)
	(location_caret_suggestion): Early return if disabled.
	* src/complain.c: Simplify.

2020-01-22  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: fixes
	* data/skeletons/yacc.c: Avoid warnings about unused functions.
	Fix typo.

2020-01-21  Akim Demaille  <akim.demaille@gmail.com>

examples: be more robust to spaces in paths
	Reported by Nikki Valen.
	https://lists.gnu.org/r/bug-bison/2020-01/msg00032.html

* examples/test ($prog): Remove, replaced by...
	(prog): This new function, which pays attention to quoting shell
	variables.

2020-01-21  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

larlr1.cc: Reject unsupported values for parse.lac
	Just as the yacc.c skeleton, the lalr1.cc skeleton should reject
	invalid values for parse.lac.

* data/skeletons/lalr1.cc: check validity of parse.lac
	* tests/input.at: new test cases

2020-01-21  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

larlr1.cc: Reject unsupported values for parse.lac
	Just as the yacc.c skeleton, the lalr1.cc skeleton should reject
	invalid values for parse.lac.

* data/skeletons/lalr1.cc: check validity of parse.lac
	* tests/input.at: new test cases

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

parsers: issue tname with i18n markup
	Some users would like to avoid having to "parse" the *.y file to find
	the strings to translate.  Let's issue the translatable tokens with N_
	to allow "parsing" the generated parsers instead.

See
	https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00015.html

* src/output.c (prepare_symbol_names): Issue symbol_names with N_()
	markup.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

tests: check token internationalization
	* tests/calc.at: Check it.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

parsers: support translatable token aliases
	In addition to

%token NUM "number"

%token NUM _("number")

in which case the token will be translated in error messages.
	Do not use _() in the output if there are no translatable tokens.

* src/symtab.h, src/symtab.c (symbol): Add a 'translatable' member.
	* src/parse-gram.y (TSTRING): New token.
	(string_as_id.opt): Replace with...
	(alias): this.
	Use it.
	* src/scan-gram.l (SC_ESCAPED_TSTRING): New start conditions, to match
	TSTRINGs.
	* src/output.c (prepare_symbols): Define b4_translatable if there are
	translatable strings.

* data/skeletons/glr.c, data/skeletons/lalr1.cc,
	* data/skeletons/yacc.c (yytnamerr): Receive b4_translatable, and use it.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

tests: check that detailed error messages preserve UTF-8 characters
	* tests/regression.at: here.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: escape trigraphs in detailed parse.error
	* src/output.c (escape_trigraphs, xescape_trigraphs): New.
	(prepare_symbol_names): Use it.
	* tests/regression.at: Check the handling of trigraphs with
	parse.error = detailed.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

bison: use detailed error messages
	* #: .

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: tests: check detailed error messages
	* tests/local.at (AT_ERROR_DETAILED_IF): New.
	(AT_ERROR_SIMPLE_IF): Adjust.
	* tests/calc.at: Check parse.error=detailed.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: add support for parse.error detailed
	"detailed" error messages are almost like "verbose", except that we
	don't double escape them, they don't get inner quotes, we don't use
	yytnamerr, and we hide the table.

"custom" is exposed with the "detailed" tokens, not the "verbose"
	ones: they are not double-quoted.

Because there's a risk that some people use yytname even without
	"verbose", let's keep yytname (instead of yys_name) in "simple"
	parse.error.

* src/output.c (prepare_symbol_names): Be ready to output symbol names
	unquoted.
	(prepare_symbol_names): Output both the old tname table, and the new
	symbol_names one.
	* data/skeletons/bison.m4: Accept 'detailed'.
	* data/skeletons/yacc.c: When parse.error is 'detailed', don't emit
	yytname and yytnamerr, just yysymbol_name with the table inside.
	* tests/calc.at: Adjust.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

c: use yysymbol_name in traces
	Only parse.error verbose and simple will get the original yytname: the
	other options will rely on a different table.  So let's move on top of
	the yysymbol_name function.

* data/skeletons/c.m4 (yy_symbol_print): Use yysymbol_name.
	* data/skeletons/glr.c (yytokenName): Rename as...
	(yysymbol_name): this.
	The change of naming scheme is unfortunate, but it's definitely glr.c
	which is "wrong".

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

glr.c: move some functions after the definition of types
	Currently yy_symbol_print is defined before yytokenName, although it
	should use it instead of read yytname directly.  Move blocks around to
	avoid this.

* data/skeletons/glr.c (yy_symbol_print): Move its definition after
	that of yytokenName.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

Merge branch 'maint'
	* maint:
	  maint: post-release administrivia
	  version 3.5.1
	  news: update
	  CI: use ICC again
	  warnings: pacify ICC in lalr1.cc
	  test: report.at: avoid tiny new failure
	  git: update ignores

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

version 3.5.1
	* NEWS: Record release date.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

news: update

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

CI: use ICC again
	See https://github.com/nemequ/icc-travis/issues/15.
	Thanks to Jeff Hammond and Evan Nemerson for their help.

* configure.ac (warn_common): Disable dubious warnings.
	* .travis.yml: Use ICC again.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

warnings: pacify ICC in lalr1.cc
	See 139d0655947c87f90af08718618feaaca0e558d7.

* data/skeletons/yacc.c: If I might be a char, write a[+I] instead of
	a[I], so that ICC does not complain.

2020-01-19  Jim Meyering  <meyering@fb.com>

test: report.at: avoid tiny new failure
	Be robust to newer versions of Autoconf where the package URL defaults
	to https instead of http.

* configure.ac (AC_INIT): Use https.
	* tests/report.at: Adjust expected output s/http/https/
	to match updated URL.

2020-01-19  Akim Demaille  <akim.demaille@gmail.com>

git: update ignores

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: portability to G++ 4.8
	Currently we get warnings with GCC 4.8 when running the
	maintainer-check-g++ tests:

143. skeletons.at:85: testing Installed skeleton file names ...
	    ../../tests/skeletons.at:120: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret --skeleton=yacc.c -o input-cmd-line.c input-cmd-line.y
	    ../../tests/skeletons.at:121: $CC $CFLAGS $CPPFLAGS  $LDFLAGS -o input-cmd-line input-cmd-line.c $LIBS
	    stderr:
	    input-cmd-line.c: In function 'int yysyntax_error(long int*, char**, const yyparse_context_t*)':
	    input-cmd-line.c:977:52: error: conversion to 'int' from 'long int' may alter its value [-Werror=conversion]
	                                       YYSIZEOF (yyarg) / YYSIZEOF (*yyarg));
	                                                        ^
	    cc1plus: all warnings being treated as errors
	    stdout:
	    ../../tests/skeletons.at:121: exit code was 1, expected 0

and

429. calc.at:823: testing Calculator parse.error=custom %locations api.prefix={calc}  ...
	    ../../tests/calc.at:823: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret -Wno-deprecated -o calc.c calc.y
	    ../../tests/calc.at:823: $CC $CFLAGS $CPPFLAGS  $LDFLAGS -o calc calc.c $LIBS
	    stderr:
	    calc.y: In function 'int yyreport_syntax_error(const yyparse_context_t*)':
	    calc.y:157:58: error: conversion to 'int' from 'long unsigned int' may alter its value [-Werror=conversion]
	       int n = yysyntax_error_arguments (ctx, arg, sizeof arg / sizeof *arg);
	                                                              ^
	    cc1plus: all warnings being treated as errors
	    stdout:
	    ../../tests/calc.at:823: exit code was 1, expected 0

We could use a cast to avoid the warning, but it becomes too
	cluttered.  We can also use YYPTRDIFF_T, but that forces the user to
	use YYPTRDIFF_T too, although this is an array of tokens, which is
	limited by YYNTOKENS, an int.  So let's completely avoid this warning.

* data/skeletons/yacc.c, tests/local.at (yyreport_syntax_error): Avoid
	relying on sizeof to compute the array capacity.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: pass the parse-params to yyreport_syntax_error
	Enhance the calculator tests: show that passing arguments to yyerror
	works.

* tests/calc.at: Add a new parse-param, nerrs, which counts the number
	of syntax errors in a run.
	* tests/local.at: Adjust to handle the new 'nerrs' argument, when
	present.

The custom error reporting function show sees the user's additional
	arguments.  Let's experiment with passing them as arguments to
	yyreport_syntax_error, but maybe storing them in the context would be
	a bettter alternative.

* data/skeletons/yacc.c (yyreport_syntax_error): Handle the
	parse-params.
	* tests/calc.at, tests/local.at: Adjust.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

tests: a clearer test for parse-params
	Currently the parse-params are tested in calc.at by checking that the
	global variable and the parse-params have the same value.  But it does
	not check that value, that could remain being 0 just as well.

* tests/calc.at: Don't define the params when they are not used.
	Check the final value of result and count.
	Also, do count the number of line of logs.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: check custom error messages with parse-params
	* tests/calc.at: Check with prefix and parse-params.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: let custom error messages see the location
	* data/skeletons/yacc.c (yyparse_context_t): Add yylloc when
	applicable.
	(yyparse_context_location): New.
	* tests/local.at (AT_YYERROR_DEFINE(c)): Handle the location.
	* tests/calc.at: Check it.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: isolate yyexpected_tokens
	Provide users with a means to query for the currently allowed tokens.
	Could be used for autocompletion for instance.

* data/skeletons/yacc.c (yyexpected_tokens): New, extracted from
	yysyntax_error_arguments.
	* examples/c/calc/calc.y (PRINT_EXPECTED_TOKENS): New.
	Use it.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

tests: compute verbose error messages from the custom ones
	We use a different format to check parse.error custom.  Compute the
	"verbose" one from it instead of forcing the test author to provide
	the various formats of expected error messages.

* tests/calc.at (_AT_CHECK_CALC_ERROR): Handle this transformation
	when needed.
	Simplify callers.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: check custom error messages
	* tests/local.at (AT_ERROR_CUSTOM_IF, AT_ERROR_VERBOSE_IF)
	(AT_ERROR_SIMPLE_IF): New.
	(AT_YYERROR_DEFINE(c)): Generate yyreport_syntax_error.
	* tests/calc.at (_AT_CHECK_CALC_ERROR): Accept custom error messages
	as additional test case.
	Use it.
	Add a new test case for %define parse.error custom.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: add custom error message generation
	When parse.error is custom, let users define a yyreport_syntax_error
	function, and use it.

* data/skeletons/bison.m4 (b4_error_verbose_if): Accept 'custom'.
	* data/skeletons/yacc.c: Implement it.
	* examples/c/calc/calc.y: Experiment with it.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: style: avoid macros
	* data/skeletons/yacc.c (YYSYNTAX_ERROR): Remove, the call is now
	sufficiently small so that we can afford to duplicate it.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: store token numbers, not token strings
	That allows users to cover more cases, such as easily filtering some
	arguments they don't want to expose.  But they now have to call
	yysymbol_name explicitly.

* data/skeletons/yacc.c (yysyntax_error_arguments, yysyntax_error):
	Deal with symbol numbers instead of symbol names.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: extract yyerror_message_arguments
	Isolate a function that returns the list of expected and unexpected
	tokens.  It will be exposed to users willing to customize their error
	messages.

* data/skeletons/yacc.c (yyparse_context_t): New.
	(yyerror_message_arguments): New, extracted from yysyntax_error.

2020-01-17  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

tests: make AT_PARSE_PARAMS usable at the end of arguments
	When not empty, AT_PARSE_PARAMS was guaranteed to end with a comma.
	Remove the trailing comma, so that we can use AT_PARSE_PARAMS at the
	end of the arguments, not only at the beginning.

* tests/local.at: here.
	Unfortunately, m4_append relies on the macro not being defined whereas
	we would have preferred it to check for emptiness.  So use
	m4_define/m4_undefine instead of m4_pushdef/m4_popdef.

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

tests: fix AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS pairs
	* tests/glr-regression.at, tests/java.at, tests/javapush.at:
	Close properly what is opened.
	Do not nest.

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

d, java: use traces more alike that of C
	Same order, same places, same content.

* data/skeletons/lalr1.d, data/skeletons/lalr1.java: here.

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

c++: report the stack at the same places as in C
	Let's have C be the reference, and match it elsewhere.  Maybe C is too
	verbose and some adjustments are needed, but then that would be done
	in another batch of patches.

* data/skeletons/lalr1.cc: Print the stack once we popped after
	YYERROR, and before emptying the stack at the end of parsing.

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

c++: display the stack in the same order as in C
	Currently the C and C++ parse traces differ in the order in which the
	stack is displayed: bottom up in C, top down in C++.  Let's stick to
	the C order.

* data/skeletons/stack.hh (stack::iterator, stack::const_iterator)
	(begin, end): Be forward, not backward.

2020-01-15  Akim Demaille  <akim.demaille@gmail.com>

style: avoid redundancy in the tests
	* tests/local.at (m4_rpatsubst): New.
	Use it to handle %parse-params.
	* tests/calc.at: Use %parse-params with several arguments.

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: comment changes
	In particular, import Adrian Vogelsgesang's comments about LAC from
	lalr1.cc.

* data/skeletons/yacc.c: here.

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: style: double-quote the argument of b4_percent_define_get
	* data/skeletons/yacc.c: Here, for consistency.

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: introduce yysymbol_name
	Provide the users with a public API to get the name of the tokens.  A
	thin wrapper around yytname.

* data/skeletons/yacc.c (yysymbol_name): New.
	Use it.

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

CI: check on PPC64le, ARM64 and s390x
	I was hoping it would help us catch warnings when char is
	unsigned (see 78bb152a63f711af65364881c434af4c198e1ee0), but it does
	not seem to help.  It's a pity that the compiler is the same all over
	the place, I would have preferred testing others.

* .travis.yml: here.

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

todo: update

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

Merge branch 'maint' into HEAD
	* maint:
	  gnulib: update
	  lalr1.cc: avoid static_cast
	  glr.c: add missing cast
	  regen
	  package: bump copyrights to 2020
	  gitignore: update

2020-01-11  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: avoid static_cast
	Reported by donmac703.
	Fixes https://github.com/akimd/bison/issues/20.

* data/skeletons/lalr1.cc: here.

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

glr.c: add missing cast
	Reported by psjo.
	Fixes https://github.com/akimd/bison/issues/19.

* data/skeletons/glr.c (yyprocessOneStack): Here.

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

package: bump copyrights to 2020
	Run 'make update-copyright'.

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

gitignore: update

2020-01-10  Akim Demaille  <akim.demaille@gmail.com>

regen

2020-01-09  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: simplify use of YYDPRINTF
	* data/skeletons/yacc.c (YYDPRINTF): Expand to no-op (instead of
	nothing) when disabled.
	Simplify callers.

2020-01-05  Akim Demaille  <akim.demaille@gmail.com>

package: bump copyrights to 2020
	Run 'make update-copyright'.

2020-01-05  Akim Demaille  <akim.demaille@gmail.com>

gitignore: update

2020-01-04  Akim Demaille  <akim.demaille@gmail.com>

doc: YYERROR_VERBOSE is no longer supported
	* doc/bison.texi (Table of Symbols): Remove last reference to it.
	* NEWS: Be clear about that.

2020-01-04  Akim Demaille  <akim.demaille@gmail.com>

glr.c: no longer support YYERROR_VERBOSE
	* data/skeletons/glr.c: Rather, dispatch directly on parse.error's
	value.

2020-01-04  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: no longer support YYERROR_VERBOSE
	Supporting YYERROR_VERBOSE via cpp is a nuisance: m4 is in charge of
	handling alternatives.  When adding more options for %define
	parse.error, supporting both CPP and M4 is too complex.  Anyway,
	YYERROR_VERBOSE was deprecated long ago.

* data/skeletons/yacc.c: Use m4 only to handle verbose/simple error
	messages.

2020-01-03  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: avoid negations
	* data/skeletons/yacc.c (yyerrlab): here.

2019-12-31  Akim Demaille  <akim.demaille@gmail.com>

glr.c: clarify yyreportSyntaxError
	See the previous commit.

* data/skeletons/glr.c (yyreportSyntaxError): First compute the
	arguments of the error message, _then_ th error message size.

2019-12-31  Akim Demaille  <akim.demaille@gmail.com>

yacc: restructure and fix yysyntax_error
	I would like to offer new ways to build the error message.  As a first
	step, let's simplify yysyntax_error whose first loop does two things
	at the same time: (i) collect the tokens to be reported in the error
	message, and (ii) accumulate their sizes and possibly return
	"overflow".  Let's pull (ii) in a second step.

Then test 525 (regression.at:1193: parse.error=verbose overflow)
	failed.  This test checks that we correctly report "memory overflow"
	when the error message is too large.  However the test is mistaken: it
	is triggered in a place where there are five (large) expected tokens,
	so anyway we would not display them, so there is no (memory) overflow
	here!  Transform this test to (i) check that indeed there is no
	overflow, and (ii) create syntax_error3 which does check the intended
	behavior, but with four expected tokens.

* data/skeletons/yacc.c (yysyntax_error): First compute the list of
	arguments, then compute yysize.
	* tests/regression.at (parse.error=verbose overflow): Enhance and fix.

2019-12-31  Akim Demaille  <akim.demaille@gmail.com>

tests: also check -Wchar-subscripts
	GCC's -Wchar-subscripts may report issues on platforms where char is
	unsigned.  Unfortunately the current CI does not reproduce the
	problem.  But that would allow contributors to report issues if the
	warning appears somewhere.

See 139d0655947c87f90af08718618feaaca0e558d7.
	Problem reported by Andy Fiddaman in:
	https://lists.gnu.org/r/bug-bison/2019-12/msg00021.html

* configure.ac (warn_common): Add -Wchar-subscripts.

2019-12-30  Akim Demaille  <akim.demaille@gmail.com>

CI: do not specify the language
	When we give travis the langugage, it overrides our envvars.  Instead
	of the MATRIX_EVAL trick, just stop specifying the language.

2019-12-30  Akim Demaille  <akim.demaille@gmail.com>

CI: remove ICC support, we can no longer use it
	https://github.com/nemequ/icc-travis/issues/15

2019-12-29  Akim Demaille  <akim.demaille@gmail.com>

doc: clean up the description of YYDEBUG
	* doc/bison.texi: Make it clearer that %define parse.trace is the
	preferred options.
	Fix a typo about api.prefix.

2019-12-29  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: avoid compiler warnings
	    381. types.at:366: testing glr.cc api.value.type={double} ...
	    test.cc:207:57: error: "__clang_major__" is not defined, evaluates to 0 [-Werror=undef]
	      207 | #if defined __APPLE__ && YY_CPLUSPLUS < 201103L && 4 <= __clang_major__
	          |                                                         ^~~~~~~~~~~~~~~

* data/skeletons/glr.cc: Check __clang_major__ before using it.

2019-12-18  Paul Eggert  <eggert@cs.ucla.edu>

warnings: pacify ‘gcc -Wchar-subscripts’ in yacc.c
	Problem reported by Andy Fiddaman in:
	https://lists.gnu.org/r/bug-bison/2019-12/msg00021.html
	* data/skeletons/yacc.c (yy_reduce_print, yy_lac, yysyntax_error)
	(yyreturn): If I might be a char, write a[+I] instead of a[I],
	so that ‘gcc -Wchar-subscripts’ does not complain.

2019-12-14  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: No output changes.

2019-12-14  Akim Demaille  <akim.demaille@gmail.com>

tests: don't fail if seq is no available
	As is the case on Solaris.
	Reported by Dennis Clarke.
	https://lists.gnu.org/archive/html/bug-bison/2019-12/msg00011.html

* examples/c/reccalc/reccalc.test: Skip if there is no seq.

2019-12-11  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-12-11  Akim Demaille  <akim.demaille@gmail.com>

version 3.5
	* NEWS: Record release date.

2019-12-10  Akim Demaille  <akim.demaille@gmail.com>

news: prepare for 3.5

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: disable warnings from Clang on macOS
	    $ cat test.cc
	    #include <stddef.h>
	    #include <stdint.h>

ptrdiff_t half_max_capacity = PTRDIFF_MAX;
	    $ clang++-mp-9.0 -pedantic -std=c++98 /tmp/test.cc -c
	    /tmp/test.cc:4:31: warning: 'long long' is a C++11 extension [-Wc++11-long-long]
	    ptrdiff_t half_max_capacity = PTRDIFF_MAX;
	                                  ^
	    /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/usr/include/stdint.h:149:23:
	            note: expanded from macro 'PTRDIFF_MAX'
	    #define PTRDIFF_MAX       INT64_MAX
	                              ^
	    /Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/usr/include/stdint.h:75:26:
	            note: expanded from macro 'INT64_MAX'
	    #define INT64_MAX        9223372036854775807LL
	                             ^
	    1 warning generated.

* data/skeletons/glr.cc: here.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: fix it in C++
	Another breakage revealed by vcsn.

* data/skeletons/c++.m4 (yytranslate_): Do not hard code "yy" and
	"parser", both can be changed by the user.
	Actually, since we are in the parser itself, there's really no need to
	qualify the type.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

c++: fix comments for %code blocks
	In a project of mine, vcsn, this commit fixes the following comments.

--- /tmp/parse.hh	2019-12-08 15:51:24.792934703 +0100
	    +++ lib/vcsn/rat/parse.hh	2019-12-08 16:00:59.137107503 +0100
	    @@ -43,7 +43,7 @@

#ifndef YY_YY_USERS_AKIM_SRC_LRDE_2_LIB_VCSN_RAT_PARSE_HH_INCLUDED
	     # define YY_YY_USERS_AKIM_SRC_LRDE_2_LIB_VCSN_RAT_PARSE_HH_INCLUDED
	    -// //                    "%code requires" blocks.
	    +// "%code requires" blocks.
	     #line 20 "/Users/akim/src/lrde/2/lib/vcsn/rat/parse.yy"

#include <iostream>
	    @@ -1851,7 +1851,7 @@

-// //                    "%code provides" blocks.
	    +// "%code provides" blocks.
	     #line 60 "/Users/akim/src/lrde/2/lib/vcsn/rat/parse.yy"

#define YY_DECL_(Class) \

* data/skeletons/bison.m4 (b4_percent_code_get): Pass an expanded
	string to b4_comment.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

parser: pretend we are Bison 3.5
	* src/parse-gram.y: Accept we're Bison 3.5.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

c++: fix spello
	* data/skeletons/lalr1.cc: here.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

todo: update
	* TODO: Schedule some features for 3.6.
	Remove obsolete stuff.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

version 3.4.92
	* NEWS: Record release date.

2019-12-08  Akim Demaille  <akim.demaille@gmail.com>

news: fixes
	Reported by Paul Eggert.
	https://lists.gnu.org/archive/html/bison-patches/2019-12/msg00014.html

* NEWS: here.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

doc: minor changes
	* README-hacking.md: here.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

doc: clearly deprecate YYPRINT
	* doc/bison.texi (Prologue): Stop using YYPRINT as an example.
	(The YYPRINT Macro): Clearly show this macro is deprecated.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: here.
	No change in content.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

news: update

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

d: obey parse.error
	* data/skeletons/lalr1.d (yysyntax_error): Let the dispatch be
	bison-time, not runtime.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

c++: also prefer YY_ASSERT to YYASSERT
	Like the other skeletons.

* data/skeletons/variant.hh: here.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

glr.c: obey the parse.assert %define variable
	* data/skeletons/glr.c (YYASSERT): Rename as...
	(YY_ASSERT): this, for consistency with yacc.c, and also to emphasize
	the fact that this is not for the end user (YY_ prefix).
	* tests/glr-regression.at: Define parse.assert.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

c++: beware of short ranges for state numbers
	Now that we use small integral types, possibly unsigned (e.g.,
	unsigned char), to store state numbers, using -1 to denote an empty
	state (i.e., a state that stores no semantical value) is very
	dangerous: it will be confused with state 255, which might be
	non-empty.

Rather than allocating a larger range of state numbers to keep the
	empty-state apart, let's use the number of a state known to store no
	value.  The initial state, numbered 0, seems to fit perfectly the job.

Reported by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html

* data/skeletons/lalr1.cc (empty_state): Be 0.

2019-12-07  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: check it against api.token.constructor
	* tests/scanner.at: here.

2019-12-06  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-12-06  Akim Demaille  <akim.demaille@gmail.com>

warnings: enable -Wuseless-cast, and eliminate warnings
	Prompted by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html.

* configure.ac (warn_cxx): Add -Wuseless-cast.
	* data/skeletons/c.m4 (b4_attribute_define): Define
	YY_IGNORE_USELESS_CAST_BEGIN and YY_IGNORE_USELESS_CAST_END.
	* data/skeletons/glr.c (YY_FPRINTF): New, replaces YYFPRINTF, wrapped
	with YY_IGNORE_USELESS_CAST_BEGIN and YY_IGNORE_USELESS_CAST_END.
	(YY_DPRINTF): Likewise.
	* tests/actions.at: Remove useless cast.
	* tests/headers.at: Adjust.

2019-12-02  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style changes
	* src/complain.h, src/complain.c: Comment changes.
	* src/scan-skel.l: Reduce scopes.
	* data/skeletons/bison.m4: Factor diagnostic functions.

2019-12-02  Akim Demaille  <akim.demaille@gmail.com>

glr.c: style changes
	* data/skeletons/glr.c (yysplitStack): Reduce scopes.
	* tests/atlocal.in: Formatting changes.

2019-12-01  Akim Demaille  <akim.demaille@gmail.com>

c++: get rid of symbol_type::token ()
	It is not used.  And its implementation was wrong when api.token.raw
	was defined, as it was still mapping to the external token numbers,
	instead of the internal ones.  Besides it was provided only when
	api.token.constructor is defined, yet always declared.

* data/skeletons/c++.m4 (by_type::token): Remove, useless.

2019-12-01  Akim Demaille  <akim.demaille@gmail.com>

c++: remove useless cast about user_token_number_max_
	Reported by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html

The cast is needed when yytranslate_'s argument type is token_type,
	i.e., when api.token.constructor is defined.

373. types.at:138: testing lalr1.cc api.value.type=variant api.token.constructor ...
	    ======== Testing with C++ standard flags: ''
	    ../../tests/types.at:138: bison --color=no -fno-caret  -o test.cc test.y
	    ../../tests/types.at:138: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o test test.cc $LIBS
	    stderr:
	    test.cc:966:16: error: result of comparison of constant 257 with
	                    expression of type 'yy::parser::token_type'
	                   (aka 'yy::parser::token::yytokentype') is always true
	                   [-Werror,-Wtautological-constant-out-of-range-compare]
	        else if (t <= user_token_number_max_)
	                 ~ ^  ~~~~~~~~~~~~~~~~~~~~~~
	    1 error generated.

It is because it is expected that when api.token.constructor is
	defined, only symbol constructors will be used, that yytranslate_ then
	takes a token_type.  But it is wrong: we still allow literal
	characters in this case, as demonstrated by test 373 for instance.

%define api.value.type variant
	    %define api.token.constructor
	    %token <std::pair<int, int>> '1' '2';
	    [...]
	    static yy::parser::symbol_type yylex ()
	    {
	      static char const input[] = "12";
	      int res = input[toknum++];
	      typedef yy::parser::symbol_type symbol;
	      if (res)
	        return symbol (res, std::make_pair (res - '0', res - '0' + 1));
	      else
	        return symbol (res);
	    }

So let yytranslate_ always take an int, which makes the cast truly
	useless.

* data/skeletons/c++.m4, data/skeletons/lalr1.cc (yytranslate_): here.

2019-12-01  Akim Demaille  <akim.demaille@gmail.com>

c++: clean a few issues wrt special tokens
	The C++ implementation of LAC did not skip the $undefined token,
	probably because it was not exposed.  Expose it, and use clearer
	names.

* data/skeletons/c++.m4: Don't define undef_token_ in yytranslate_,
	but...
	* data/skeletons/lalr1.cc (yy_undef_token_): here.
	Use a more precise type to define yy_undef_token_ and yy_error_token_.
	Unfortunately we move from a compile-time value defined via an enum to
	a static const member.  Eventually we should make it constexpr.
	Make LAC implementation more alike yacc.c's one.

2019-12-01  Akim Demaille  <akim.demaille@gmail.com>

d, java: improve yytranslate and neighbors
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java: Don't expose
	yyuser_token_number_max_ and yyundef_token_.  Do as in C++: scope them
	into yytranslate_, and only when api.token.raw is not defined.
	(yyterror_): Rename as...
	(yy_error_token_): this.
	* data/skeletons/lalr1.d (token_number_type): New.
	Use it.
	Can't be done in the Java backend, as Java does not have type aliases.

2019-12-01  Akim Demaille  <akim.demaille@gmail.com>

d, java: get rid of a useless table
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java (yytoken_number_):
	Remove, useless.
	Was used in ancient C skeletons to support YYPRINT, long obsoleted by
	%printer.

2019-11-30  Akim Demaille  <akim.demaille@gmail.com>

c++, d, java: remove yyerrcode
	It is not used at all.  We will remove it also from yacc.c, but
	later (see TODO).

* data/skeletons/lalr1.cc, data/skeletons/lalr1.d,
	* data/skeletons/lalr1.java (yyerrcode_):
	Remove.

2019-11-30  Akim Demaille  <akim.demaille@gmail.com>

c++: improve typing
	* data/skeletons/lalr1.cc (yysyntax_error_): symbol_type::type_get
	returns a symbol_number_type (which is indeed an int).

2019-11-30  Akim Demaille  <akim.demaille@gmail.com>

c++: remove useless cast about yyeof_
	Reported by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00016.html

* data/skeletons/c++.m4 (b4_yytranslate_define): Don't use yyeof_ as
	if it had two different types.
	It is used once against the input argument, which is the value
	returned by yylex, which is an "external token number", typically an
	int.  It is also used as output type, an "internal symbol number".
	It turns out that in both cases we mean "0", but let's keep yyeof_
	only for the case "internal symbol number", i.e., _after_ conversion
	by yytranslate.
	This frees us from one cast.

2019-11-30  Akim Demaille  <akim.demaille@gmail.com>

glr: style change
	* data/skeletons/glr.c (YYDPRINTF): Expand into an empty statement,
	instead of nothing.
	Simplify callers.

2019-11-30  Akim Demaille  <akim.demaille@gmail.com>

glr: remove useless casts
	Reported by GCC's -Wuseless-cast.

* data/skeletons/glr.c: Don't cast to yybool, it's useless.

2019-11-29  Akim Demaille  <akim.demaille@gmail.com>

yacc.c, glr.c: fix crash when reporting errors in consistent states
	The current code for yysyntax_error for %define parse.error verbose is
	fishy (given that YYEMPTY is -2, invalid argument for yytname[]):

static int
	    yysyntax_error ([...])
	    {
	      YYPTRDIFF_T yysize0 = yytnamerr (YY_NULLPTR, yytname[yytoken]);
	    [...]
	      if (yytoken != YYEMPTY)

A nearby comment reports

The only way there can be no lookahead present (in yychar) is if
	    this state is a consistent state with a default action.  Thus,
	    detecting the absence of a lookahead is sufficient to determine
	    that there is no unexpected or expected token to report.  In that
	    case, just report a simple "syntax error".

So it _is_ possible to call yysyntax_error with yytoken == YYEMPTY,
	albeit quite difficult when meaning to, so virtually impossible by
	accident (after all, there was never a bug report about this).

I failed to produce a test case, but Joel E. Denny provided me with
	one (added to the test suite below).  The yacc.c skeleton fails on
	this, and once fixed dies on a second problem.  The glr.c skeleton was
	also dying, but immediately of this second problem.

Indeed we were not allocating space for the error message's final \0.
	This was hidden by the fact that we only had error messages with at
	least an unexpected token displayed, so with at least one "%s" in the
	format string, whose size (2) was included (incorrectly) in the final
	size of the message (where the %s have been replaced by the actual
	content).

* data/skeletons/glr.c, data/skeletons/yacc.c (yysyntax_error):
	Do not invoke yytnamerr on YYEMPTY.
	Clarify the computation of the length of the _final_ error message,
	with the NUL terminator but without the '%s's.
	* tests/conflicts.at (Syntax error in consistent error state):
	New, contributed by Joel E. Denny.

2019-11-26  Akim Demaille  <akim.demaille@gmail.com>

tests: avoid creating files whose name collide with standard headers
	Having a file named "exception" is risky: the compiler might use that
	file in #include.
	Reported by 马俊 <majun123@whu.edu.cn>.

* tests/local.at (AT_SKIP_IF_EXCEPTION_SUPPORT_IS_POOR): Generate
	'exceptions', not 'exception'.

2019-11-22  Akim Demaille  <akim.demaille@gmail.com>

doc: more details about the test suite
	* README-hacking.md: here.

2019-11-20  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-11-20  Akim Demaille  <akim.demaille@gmail.com>

version 3.4.91
	* NEWS: Record release date.

2019-11-20  Akim Demaille  <akim.demaille@gmail.com>

style: pacify syntax-check
	* cfg.mk: No need to translate *.md files.
	* data/skeletons/glr.c, data/skeletons/yacc.c: Fix space issues.

2019-11-19  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-11-18  Akim Demaille  <akim.demaille@gmail.com>

doc: don't promote dangling aliases
	String literals as tokens serve two distinct purposes: freeing from
	having to implement the keyword matching in the scanner, and improving
	error messages.  Most of the time both can be achieved at the same
	time, but on occasions, it does not work so well.

We promote their use for error messages.  We will also still support
	the former case, but it is _not_ the recommended approach.

* doc/bison.texi (Tokens from Literals): Clearly state that we don't
	recommend looking up the token types in the list of token names.

2019-11-17  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: complain about undeclared string tokens
	String literals, which allow for better error messages, are (too)
	liberally accepted by Bison, which might result in silent errors.  For
	instance

%type <exVal> cond "condition"

does not define “condition” as a string alias to 'cond' (nonterminal
	symbols do not have string aliases).  It is rather equivalent to

%nterm <exVal> cond
	    %token <exVal> "condition"

i.e., it gives the type 'exVal' to the "condition" token, which was
	clearly not the intention.

Introduce -Wdangling-alias to catch this.

* src/complain.h, src/complain.c: Add support for -Wdangling-alias.
	(argmatch_warning_args): Sort.
	* src/symtab.c (symbol_check_defined): Complain about dangling
	aliases.
	* doc/bison.texi: Document it.
	* tests/input.at (Dangling aliases): New test.

2019-11-17  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: yacc reserves %type to nonterminals
	On

%token TOKEN1
	    %type  <ival> TOKEN1 TOKEN2 't'
	    %token TOKEN2
	    %%
	    expr:

bison -Wyacc gives

input.y:2.15-20: warning: POSIX yacc reserves %type to nonterminals [-Wyacc]
	        2 | %type  <ival> TOKEN1 TOKEN2 't'
	          |               ^~~~~~
	    input.y:2.29-31: warning: POSIX yacc reserves %type to nonterminals [-Wyacc]
	        2 | %type  <ival> TOKEN1 TOKEN2 't'
	          |                             ^~~
	    input.y:2.22-27: warning: POSIX yacc reserves %type to nonterminals [-Wyacc]
	        2 | %type  <ival> TOKEN1 TOKEN2 't'
	          |                      ^~~~~~

The messages appear to be out of order, but they are emitted when the
	error is found.

* src/symtab.h (symbol_class): Add pct_type_sym, used to denote
	symbols appearing in %type.
	* src/symtab.c (complain_pct_type_on_token): New.
	(symbol_class_set): Check that %type is not applied to tokens.
	(symbol_check_defined): pct_type_sym also means undefined.
	* src/parse-gram.y (symbol_decl.1): Set the class to pct_type_sym.
	* src/reader.c (grammar_current_rule_begin): pct_type_sym also means
	undefined.
	* tests/input.at (Yacc's %type): New.

2019-11-16  Akim Demaille  <akim.demaille@gmail.com>

doc: promote %nterm over %type
	As an extension to POSIX Yacc, Bison's %type accepts tokens.
	Unfortunately with string literals as implicit tokens, this is
	misleading, and led some users to write

%type <exVal> cond "condition"

believing that "condition" would be associated to the 'cond'
	nonterminal (see https://github.com/apache/httpd/pull/72).

* doc/bison.texi: Promote %nterm rather than %type to declare the type
	of nonterminals.

2019-11-16  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: No visible changes.

2019-11-16  Akim Demaille  <akim.demaille@gmail.com>

doc: work around warnings when Flex C output is compiled in C++
	* doc/bison.texi (calc++/scanner.ll): here.
	While at it, clarify clang vs. warnings.

2019-11-16  Akim Demaille  <akim.demaille@gmail.com>

tests: be robust to old Perl versions on Cygwin
	Reported by Denis Excoffier.
	https://lists.gnu.org/archive/html/bug-bison/2019-11/msg00008.html.

* tests/output.at: Be sure to remove back up files.

2019-11-16  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-11-12  kaneko y  <spiketeika@gmail.com>

gram.c: Fix condition of aver
	* src/gram.c (grammar_dump): Fix condition of aver.
	What we want to check is that rhs is followed by its rule.

2019-11-11  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify build instructions
	* README: A few fixes.
	Explain how to install color support.
	* README-hacking: Rename as...
	* README-hacking.md: this, and convert to Markdown.
	Improve typography.
	Improve explanations about update-test.

2019-11-11  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-11-11  Yuichiro Kaneko  <spiketeika@gmail.com>

gram.c: also print terminals in grammar_dump
	* src/gram.c (grammar_dump): Print terminals likewise non terminals.
	* tests/sets.at (Reduced Grammar): Update test case to catch up the
	change and add a test case where prec and assoc are used.

2019-11-10  Akim Demaille  <akim.demaille@gmail.com>

doc: work around Texinfo 6.7 bug
	When @code is used in a @deftype... definition, it issues quotes.
	Remove them.
	See https://lists.gnu.org/archive/html/help-texinfo/2019-11/msg00004.html.

* doc/local.mk: here.

2019-11-09  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: Wrap lines.
	No semantical difference.

2019-11-09  Akim Demaille  <akim.demaille@gmail.com>

doc: use upper case for tokens
	* doc/bison.texi: here.

2019-11-07  Akim Demaille  <akim.demaille@gmail.com>

doc: type-face fixes
	* doc/bison.texi: Use @code for types in function definitions.

2019-11-06  Akim Demaille  <akim.demaille@gmail.com>

c++: expose the type used to store line and column numbers
	* data/skeletons/location.cc (position::counter_type)
	(location::counter_type): New.
	Use them.
	* doc/bison.texi (C++ position, C++ location): Adjust.

2019-11-03  Akim Demaille  <akim.demaille@gmail.com>

tests: fix comment and adjust to locale names on GNU/Linux
	Reported by Denis Excoffier.

* tests/diagnostics.at: here.

2019-11-03  Akim Demaille  <akim.demaille@gmail.com>

tests: really check complaints from m4
	* tests/diagnostics.at (Locations from M4, Tabulations and multibyte
	characters from M4): These tests are actually checking a message
	coming from C, not from M4.  Replace with...
	(Complaints from M4): This.

2019-11-03  Akim Demaille  <akim.demaille@gmail.com>

tests: simplify prologue
	* tests/testsuite.h: We no longer load gnulib in the tests.

2019-11-03  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: add missing translation
	* src/muscle-tab.c (muscle_percent_define_check_kind): Here.

2019-11-02  Akim Demaille  <akim.demaille@gmail.com>

c++: fix old cast warnings
	We still have a few old C casts in lalr1.cc, let's get rid of them.
	Reported by Frank Heckenbach.

Actually, let's monitor all our casts using easy to grep macros.
	Let's use these macros to use the C++ standard casts when we are in
	C++.

* data/skeletons/c.m4 (b4_cast_define): New.
	* data/skeletons/glr.c, data/skeletons/glr.cc,
	* data/skeletons/lalr1.cc, data/skeletons/stack.hh,
	* data/skeletons/yacc.c:
	Use it and/or its casts.

* tests/actions.at, tests/cxx-type.at,
	* tests/glr-regression.at, tests/headers.at, tests/torture.at,
	* tests/types.at:
	Use YY_CAST instead of C casts.

* configure.ac (warn_cxx): Add -Wold-style-cast.
	* doc/bison.texi: Disable it.

2019-11-01  Akim Demaille  <akim.demaille@gmail.com>

tests: be robust to tput errors
	Reported by Denis Excoffier.

* tests/bison.in: here.

2019-11-01  Akim Demaille  <akim.demaille@gmail.com>

git: update ignores
	I don't understand what happened in
	10acc148bb90fac8a52a5d35f2bd18bd824c1639.

2019-10-29  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-10-29  Akim Demaille  <akim.demaille@gmail.com>

version 3.4.90
	* NEWS: Record release date.

2019-10-29  Akim Demaille  <akim.demaille@gmail.com>

C++: finish propagating the unsigned->signed conversion in locations
	* data/skeletons/location.cc: Remove the u (for unsigned) suffix from
	the initial line and column.
	* NEWS: AFAICT, only C++ backends have their location types changed.

2019-10-29  Akim Demaille  <akim.demaille@gmail.com>

style: fix cpp indentation
	Reported by syntax-check.

* src/system.h: here.

2019-10-29  Akim Demaille  <akim.demaille@gmail.com>

style: glr.c: comment changes
	* data/skeletons/glr.c: here.

2019-10-26  Akim Demaille  <akim.demaille@gmail.com>

CI: pass -O1 to GCC8 with sanitizers
	This build never finishes in the 50min credit given by Travis.  See if
	with optimizations it works better.

* .travis.yml: here.

2019-10-26  Akim Demaille  <akim.demaille@gmail.com>

reader: reduce the "scope" of global variables
	We have too many global variables, adding structure would help.  For a
	start, let's hide some of the variables closer to their usage.

* src/getargs.c, src/files.h (current_file): Move to...
	* src/scan-gram.c: here.
	* src/scan-gram.h (gram_in, gram__flex_debug): Remove, make them
	private to the scanner.
	* src/reader.h, src/reader.c (reader): Take a grammar file as argument.
	Move the handling of scanner variables to...
	* src/scan-gram.l (gram_scanner_open, gram_scanner_close): here.
	(gram_scanner_initialize): Remove, replaced by gram_scanner_open.
	* src/main.c: Adjust.

2019-10-26  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-10-26  Akim Demaille  <akim.demaille@gmail.com>

parser: use grammar_file instead of current_file
	* src/parse-gram (%initial-action): here.
	(handle_skeleton): Don't depend on the current file name to look for
	"local" skeletons (subject to changes coming from "#lines"): depend
	only on the initial file name, the one given on the command line.

2019-10-26  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use grammar_file instead of current_file
	Currently there are two globals denoting the input file: grammar_file
	is the one from the command line, and current_file which might change
	because of #line.  Use only the former.

* src/complain.c (error_message): here.
	* tests/diagnostics.at: Adjust.

2019-10-25  Akim Demaille  <akim.demaille@gmail.com>

reader: let symtab deal with the symbols
	* src/reader.c (reader): Move the setting up of the builtin symbols to...
	* src/symtab.c (symbols_new): here.

2019-10-25  Akim Demaille  <akim.demaille@gmail.com>

style: remove incorrect comment
	Reported by Paul Eggert.

* src/system.h: here.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: fix previous commit: printing of state numbers
	* data/skeletons/lalr1.cc: Printing a char prints... a char.
	Print ints instead.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: use computed state types
	This skeleton uses a single stack of state structures, so it is less
	likely to benefit from a stack size reduction than yacc.c (which uses
	several stacks: state number, value and location).  But it will reduce
	the size of the LAC stack.

This skeleton was already using int for state numbers, so, contrary to
	yacc.c, this brings nothing for large automata.

Overall, it is still nicer to make the skeletons alike.

* data/skeletons/lalr1.cc (state_type): Here.

2019-10-24  kaneko y  <spiketeika@gmail.com>

README: Fix a typo
	* README: Fix a typo. Git command name is submodule.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

examples: fix missing dependencies
	Reported by Thomas Petazzoni.
	https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00000.html

* examples/c/reccalc/local.mk: Complete dependencies, including for
	earlier versions of Automake (for sake of our CI, on top of Ubuntu
	Xenial/Bionic, which feature only Automake 1.15).
	(%D%/scan.c %D%/scan.h): Upgrade to the full version provided in
	Automake's documentation.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: simplify location handling
	Locations start at line 1.  Don't accept line 0.

* src/location.c (location_print): Don't print locations with line 0.
	(location_caret): Simplify.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

build: reenable -Wtype-limits
	See https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html
	to https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00073.html.

Paul Eggert's changes in gnulib do fix the issue for modern GCCs (7,
	8, 9) on macOS.  Unfortunately these warnings are back on the
	CI (GNU/Linux) with GCC 4.6, 4.7, (not 4.8) and 4.9.

Disable the warning locally.

* configure.ac (warn_common, warn_tests): Remove -Wtype-limits.
	* src/system.h (IGNORE_TYPE_LIMITS_BEGIN, IGNORE_TYPE_LIMITS_END): New.
	* src/InadequacyList.c, src/parse-gram.c, src/parse-gram.y,
	* src/symtab.c: Use it.

2019-10-24  Akim Demaille  <akim.demaille@gmail.com>

build: remove dmalloc support
	Today sanitizers are a better alternative.

* m4/dmalloc.m4: Remove.
	* configure.ac, src/system.h: Adjust.

2019-10-23  Akim Demaille  <akim.demaille@gmail.com>

gitignore: update

2019-10-23  Paul Eggert  <eggert@cs.ucla.edu>

build: update gnulib submodule to latest

2019-10-23  Yuichiro Kaneko  <spiketeika@gmail.com>

style: update comment in reader.c
	rrhs and rlhs were removed by b2ed6e5826e772162719db595446b2c58e4ac5d6.

* src/reader.c (packgram): Update comment.

2019-10-22  kaneko y  <spiketeika@gmail.com>

yacc.c: fix a typo
	* data/skeletons/yacc.c (yysetstate): fix comment.

2019-10-22  Akim Demaille  <akim.demaille@gmail.com>

style: pacify syntax-check
	* doc/.gitignore, src/complain.c, src/getargs.c,
	* src/output.c: here.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

main: also free memory on errors
	* src/derives.c (derives_free): Beware of NULL.
	* src/main.c (main): Let the 'finish' label include memory release.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	To get bitset_free accept NULL.  See
	https://lists.gnu.org/archive/html/bug-gnulib/2019-10/msg00054.html

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scope in derives
	* src/derives.c: here.
	And prefer prefix to postfix increment.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

build: disable -Wtautological-constant-out-of-range-compare
	Also see e31f92495ce14a5d924b148c8ea1470003cc47c1 and
	https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html

* configure.ac (warn_common): Disable
	-Wtautological-constant-out-of-range-compare.
	(warn_tests): Restore it.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

CI: formatting changes
	* .travis.yml: Use the single line form of lists, when reduced to a
	singletons.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

CI: rename jobs
	* .travis.yml (compile, test): Rename as...
	(dist, check): these, which are more traditional for GNU projects.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

doc: update README
	* README: Be clearer that README-hacking _must_ be read.
	Convert to Markdown.

2019-10-21  Akim Demaille  <akim.demaille@gmail.com>

bootstrap: relieve developpers from Gettext version mismatch issues
	* .travis.yml (compile): Move the workaround from here...
	* bootstrap.conf (bootstrap_epilogue): to there.

2019-10-20  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of GCC9 warnings in push mode
	This is really weird: GCC points to the LHS of the assignment...

260. headers.at:184: testing Sane headers: api.pure api.push-pull=both ...
	    tests/headers.at:184: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret -d -o input.c input.y
	    tests/headers.at:184: $CC $CFLAGS $CPPFLAGS  -c -o input.o input.c
	    stderr:
	    input.c: In function 'yyparse':
	    input.c:1276:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized]
	     1276 |         yylval = *yypushed_val;
	          |         ~~~~~~~^~~~~~~~~~~~~~~
	    input.c: In function 'yypull_parse':
	    input.c:1276:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized]
	     1276 |         yylval = *yypushed_val;
	          |         ~~~~~~~^~~~~~~~~~~~~~~
	    cc1: all warnings being treated as errors
	    stdout:
	    tests/headers.at:184: exit code was 1, expected 0

See also d87c8ac79ab844d6a7a4f5103dcf7a842d18b611
	and 9645a2b20ee7cbfa8bb4ac2237f87d598afe349c.

* tests/headers.at (Several parsers, Several parsers): Disable these
	warnings when in push parser.

2019-10-20  Akim Demaille  <akim.demaille@gmail.com>

CI: try GCC9 and Clang9
	The logs show:

Disallowing sources: llvm-toolchain-bionic-8, ubuntu-toolchain-r-test
	    To add unlisted APT sources, follow instructions in
	    https://docs.travis-ci.com/user/installing-dependencies#Installing-Packages-with-the-APT-Addon

* .travis.yml: Remove a few apt sources which are ignored in
	Bionic (e.g., see
	https://github.com/travis-ci/apt-source-safelist/issues/410).
	Where needed, use sources/sourceline instead.
	Also, don't use -DNDEBUG with older builds.

2019-10-20  Akim Demaille  <akim.demaille@gmail.com>

parser: clarify version checking
	* src/parse-gram.y: Use the same conventions for gnulib as elsewhere:
	<header.h>.
	(str_to_version): New.
	(handle_require): Use it.
	Prefer < to >.

2019-10-20  Akim Demaille  <akim.demaille@gmail.com>

build: disable -Wtype-limits, except in the test suite
	The current implementation of lib/intprops.h results in "unsigned < 0"
	comparisons, which triggers warnings.  See

https://lists.gnu.org/archive/html/bug-bison/2019-10/msg00061.html

* configure.ac (warn_common): Disable -Wtype-limits.
	(warn_tests): Restore it.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

c++: port to Sun C++ 5.12
	The documentation for Oracle Solaris Studio 12.3 (Sun C++ 5.12
	2011/11/16) says it supports C++03.  This compiler rejects the
	location.cc use of std::max for some reason; I don’t know why
	since I don’t use C++ as a rule.  The simplest workaround is to
	open-code ‘max’.
	* data/skeletons/location.cc (add_):
	Do max by hand rather than relying on std::max.
	Don’t include <algorithm.h>; no longer needed.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

regen

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

tests: port to Solaris 10 grep
	* tests/scanner.at (Token numbers: $1): Use $EGREP, not grep -E.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

tests: port to Solaris 10 sed
	As documented in the Autoconf manual, Solaris 10 sed rejects
	script labels contianing more than 7 characters.  POSIX requires
	support for at least 8 characters, but we might as well be portable
	to Solaris 10 which is still supported.
	* tests/local.at (AT_SETS_CHECK): Use only the first 7 characters
	in sed labels.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

bison: check for int overflow in token numbers
	* src/symtab.c: Include intprops.h
	(symbol_user_token_number_set): Don’t allow user_token_number ==
	INT_MAX because too much other code adds 1 to the user token number.
	(symbols_token_translations_init): Complain on integer overflow
	instead of indulging in undefined behavior.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

bison: check for int overflow when scanning
	* src/scan-gram.l: Include errno.h, for errno.
	(scan_integer, handle_syncline): Check for integer overflow.
	* tests/input.at (too-large.y): Adjust to match new diagnostics.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

bison: check version numbers more carefully
	* src/parse-gram.y: Include intprops.h.
	(handle_require): Don’t indulge in undefined behavior if the major
	or minor number is out of range.  Instead, check that the
	resulting value is nonnegative, fits in int, and that the minor
	number is less than 100.  Also, check that a number was parsed.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

c: port YY_ATTRIBUTE_UNUSED to Sun C 5.12
	Sun C 5.12 defines __SUNPRO_C to 0x5120 but diagnoses
	‘__attribute__ ((__unused__))’.  Change the ifdefs to use
	the same method as Gnulib in this area.
	* data/skeletons/c.m4 (YY_ATTRIBUTE): Remove, since
	not all attributes were added in the same compiler version.
	(YY_ATTRIBUTE_PURE, YY_ATTRIBUTE_UNUSED):
	Use specific GCC version for each attribute.
	Pay no attention to __SUNPRO_C.
	* tests/headers.at (Several parsers): Tighten tests accordingly.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

c: improve port of stdint.h usage to pre-C99
	Oracle Solaris Studio 12.3 (Sun C 5.12 2011/11/16) by default does
	not conform to C99; it defines __STDC_VERSION__ to be 199409L, so
	the Bison code does not include <stdint.h> (not required by C89
	amendment 1) even though this compiler does have <stdint.h>.  On
	this platform <limits.h> defines INT_LEAST8_MAX (POSIX allows
	this) so the skeleton got confused and thought that <stdint.h> had
	been included even though it wasn’t.
	* data/skeletons/c.m4 (b4_c99_int_type_define) [!__PTRDIFF_MAX__]:
	Always include <limits.h>.
	(YY_STDINT_H): Define when <stdint.h> was included.
	All uses of expressions like ‘defined INT_LEAST8_MAX’ changed to
	‘defined YY_STDINT_H’, since Sun C 5.12 <limits.h> defines macros
	like INT_LEAST8_MAX but does not declare types like int_least8_t.

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

gnulib:update

2019-10-17  Paul Eggert  <eggert@cs.ucla.edu>

autoconf:update

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

TODO: more updates

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

TODO: update

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

yacc: rename types for states
	* data/skeletons/yacc.c (yy_state_num): Rename as...
	(yy_state_t): this.
	(yy_state_fast_t): New.
	Use it.

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

glr: style changes
	* data/skeletons/glr.c (yytnamerr): here.
	(yyprocessOneStack): Initialize variables.

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

yacc: style changes
	* data/skeletons/yacc.c: Move call to lac discard to clarify the
	shifting of the token.
	Like in lalr1.cc.

2019-10-15  Akim Demaille  <akim.demaille@gmail.com>

tests: avoid $(...)
	Reported by Paul Eggert.

* tests/local.at (AT_DATA_NO_FINAL_EOL): here.

2019-10-14  Akim Demaille  <akim.demaille@gmail.com>

tests: use a portable 'truncate' implementation
	Suggested by Paul Eggert.
	https://lists.gnu.org/archive/html/bison-patches/2019-10/msg00044.html

* tests/local.at (AT_DATA_NO_FINAL_EOL): Use dd instead of perl.

2019-10-13  Akim Demaille  <akim.demaille@gmail.com>

tests: factor the generation of files without the final eol
	AFAICT Autotest 2.69 still does not support AT_DATA without the final
	eol.

* tests/local.at (AT_DATA_NO_FINAL_EOL): New.
	* tests/input.at: Use it.

2019-10-13  Akim Demaille  <akim.demaille@gmail.com>

tests: refactor the handling of Perl
	Let's make a difference between places where Perl is required for the
	test (AT_PERL_REQUIRE), and the places where it's used to run the
	test, but it's not not to run the test (AT_PERL_CHECK).

* tests/local.at (AT_REQUIRE): New.
	(AT_PERL_CHECK, AT_PERL_REQUIRE): New.
	Use them where appropriate.

* tests/local.mk ($(TESTSUITE)): Beware not to start the line with
	'-pi' if Perl is empty, as Make understands this as "it's ok to fail".
	Which it is not.

2019-10-12  Akim Demaille  <akim.demaille@gmail.com>

d: comment changes
	* data/skeletons/lalr1.d: Here.

2019-10-12  Akim Demaille  <akim.demaille@gmail.com>

i18n: don't push too hard for '…'
	Suggested by Paul Eggert.

* src/location.c (ellipsis): Clarify comment for translators.

2019-10-11  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-10-11  Akim Demaille  <akim.demaille@gmail.com>

glr: display line numbers in traces
	Suggested by Lars Maier.

* data/skeletons/glr.c: Also display rule locations when rules are
	deferred, and rejected.

2019-10-11  Akim Demaille  <akim.demaille@gmail.com>

tests: be really robust to Perl missing
	My previous tests (with ./configure PERL=false) have been fooled by
	configure, that managed to find perl anyway.  This time, I ran this on
	a Fedora in Docker, without Perl.

* tests/calc.at, tests/diagnostics.at, tests/headers.at,
	* tests/input.at, tests/local.at, tests/named-refs.at,
	* tests/output.at, tests/regression.at, tests/skeletons.at,
	* tests/synclines.at, tests/torture.at: Don't require Perl.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

configure: perl is not required
	But it's used in various places, including in some tests.

* configure.ac: here.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

news: update

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: prefer "…" to "..." if the locale supports it
	* src/location.c (ellipsis, ellipsize): New.
	Use them.

2019-10-10  Paul Eggert  <eggert@cs.ucla.edu>

c: improve patch for UCHAR_MAX etc. problem
	* data/skeletons/c.m4 (b4_c99_int_type_define): Reorder to put the
	signed types first, since they’re simpler and this keeps similar
	code closer.  For signed types, don’t bother checking whether the
	type promotes to int since the type must be signed anyway.  For
	unsigned types, protect a test like ‘UCHAR_MAX <= INT_MAX’ with
	‘!defined __UINT_LEAST8_MAX__’, as otherwise the logic is wrong
	for oddball platforms; and once we do that, there should no need
	for ‘defined INT_MAX’ so remove that.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

tests: do not depend on config.h
	Currently we face test suite failures in different environments,
	because of a conflict between the definitions of isnan by gnulib, and
	by the C++ library:

262. headers.at:186: testing Sane headers: %locations %debug c++ ...
	    ./headers.at:186: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret -d -o input.cc input.y
	    ./headers.at:186: $CXX $CXXFLAGS $CPPFLAGS  -c -o input.o input.cc
	    stderr:
	    In file included from /usr/include/c++/4.8.2/cmath:44:0,
	                     from /usr/include/c++/4.8.2/random:38,
	                     from /usr/include/c++/4.8.2/bits/stl_algo.h:65,
	                     from /usr/include/c++/4.8.2/algorithm:62,
	                     from location.hh:41,
	                     from input.hh:90,
	                     from input.cc:50:
	    /u/cs/fac/eggert/src/gnu/bison/lib/math.h: In function 'bool isnan(double)':
	    /u/cs/fac/eggert/src/gnu/bison/lib/math.h:2849:1: error: new declaration 'bool isnan(double)'
	     _GL_MATH_CXX_REAL_FLOATING_DECL_2 (isnan, isnan, bool)
	     ^
	    In file included from /usr/include/features.h:375:0,
	                     from /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/os_defines.h:39,
	                     from /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/c++config.h:2097,
	                     from /usr/include/c++/4.8.2/cstdlib:41,
	                     from input.hh:48,
	                     from input.cc:50:
	    /usr/include/bits/mathcalls.h:235:1: error: ambiguates old declaration 'int isnan(double)'
	     __MATHDECL_1 (int,isnan,, (_Mdouble_ __value)) __attribute__ ((__const__));
	     ^

There might be something to do in gnulib about this, but I believe
	that gnulib should not be used in the test suite in the first place.

The test suite should work with other compilers than the one used to
	compile the package.  For a start, Bison sources are more
	demanding (C99) than the generated parsers.  Last time I tried, tcc
	for example, was not able to compile Bison, yet our generated parsers
	should compile cleanly with it.

Besides the problem at hand is with the C++ compiler, with is not the
	one used to set up gnulib at configuration-time (config.h is mainly
	built from probing the C compiler).

We should really not depend on gnulib in tests.

This was introduced in 2001 to check whether including
	stdlib.h/string.h is safe thanks to STDC_HEADERS
	(2ce1014469742b5c6618daf8506b69e38787c7d5).  Today, we assume at least
	a C90 compiler, it should be safe enough.

* tests/local.at, tests/testsuite.h: Do not include config.h.
	* tests/atlocal.in (conftest.cc): Likewise.
	(CPPFLAGS): Do not expose lib/, as because of this we might picked up
	gnulib replacement headers for system headers.

* tests/input.at: Use int instead of ptrdiff_t, for easier portability
	(some machine on the CI did not find ptrdiff_t).
	* tests/c++.at: Add missing include for getchar.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check
	* doc/bison.texi: Remove the index about yyoutput, it is no longer
	documented.
	Spell check.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

tests: style changes
	* tests/actions.at: Prefer printf to fprintf.
	Prefer yyo to yyoutput in %printer.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

tests: formatting changes
	* tests/actions.at, tests/local.at: here.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

tests: add missing includes
	* tests/actions.at, tests/c++.at, tests/headers.at,
	* tests/regression.at: here.

2019-10-10  Akim Demaille  <akim.demaille@gmail.com>

c: don't assume that UCHAR_MAX, etc. are defined
	A number of portability issues with GCC 4.6 .. 4.9 (inclusive):

input.c:184:7: error: "UCHAR_MAX" is not defined [-Werror=undef]
	     #elif UCHAR_MAX <= INT_MAX
	           ^
	    input.c:184:20: error: "INT_MAX" is not defined [-Werror=undef]
	     #elif UCHAR_MAX <= INT_MAX
	                        ^
	    input.c:202:7: error: "USHRT_MAX" is not defined [-Werror=undef]
	     #elif USHRT_MAX <= INT_MAX
	           ^
	    input.c:202:20: error: "INT_MAX" is not defined [-Werror=undef]
	     #elif USHRT_MAX <= INT_MAX
	                        ^

* data/skeletons/c.m4 (b4_c99_int_type_define): Don't assume they are
	defined.

2019-10-09  Akim Demaille  <akim.demaille@gmail.com>

configure: don't require Flex
	Flex should not be required to build Bison or run the test suite (of
	course it is needed for maintaining Bison).  Yet the Automake
	conditional FLEX_WORKS does not work.

* m4/flex.m4 (_AC_PROG_LEX_YYTEXT_DECL): Since this is called
	conditionally, don't define LEX_IS_FLEX here, but rather...
	(AC_PROG_LEX): here.
	* configure.ac: Be more cautious about possibly undefined variables.

2019-10-07  Paul Eggert  <eggert@cs.ucla.edu>

Move the integer-type selection into c.m4
	That way, glr.c can use it too.
	* data/skeletons/c.m4 (b4_int_type):
	Do not special-case ‘char’; it’s not worth the trouble,
	as clang complains about char subscripts.
	(b4_c99_int_type, b4_c99_int_type_define): New macros,
	taken from yacc.c.
	* data/skeletons/glr.c: Use b4_int_type_define.
	* data/skeletons/yacc.c (b4_int_type): Remove, since there’s
	no longer any need to redefine it.
	Use b4_c99_int_type_define rather than its body.

2019-10-07  Paul Eggert  <eggert@cs.ucla.edu>

Use “least” types for integers in Yacc tables
	This changes the Yacc skeleton to use “least” integer types to
	keep tables smaller on some platforms, which should lessen cache
	pressure.  Since Bison uses the Yacc skeleton, it follows suit.
	* data/skeletons/yacc.c: Include limits.h and stdint.h if this
	seems to be needed.
	(yytype_uint8, yytype_int8, yytype_uint16, yytype_int16):
	If available, use GCC predefined macros __INT_MAX__ etc. to select
	a “least” type, as this avoids namespace hassles.  Otherwise, if
	available fall back on selecting a “least” type via the C99 macros
	INT_MAX, INT_LEAST8_MAX, etc.  Otherwise, fall further back on one of
	the builtin C99 types signed char, short, and int.  Make sure that
	any selected type promotes to int.  Ignore any macros YYTYPE_INT16,
	YYTYPE_INT8, YYTYPE_UINT16, YYTYPE_UINT8 defined by the user.
	(ptrdiff_t, PTRDIFF_MAX): Simplify in the light of the above.
	(yytype_uint8, yytype_uint16): Do not assume that unsigned char
	and unsigned short promote to int, as this isn’t true on some
	platforms (e.g., TI TMS320C55x).
	* src/parse-gram.y (YYTYPE_INT16, YYTYPE_INT8, YYTYPE_UINT16)
	(YYTYPE_UINT8): Remove, as these are no longer effective.

2019-10-06  Paul Eggert  <eggert@cs.ucla.edu>

Port better to C++ platforms
	* data/skeletons/yacc.c (YYPTRDIFF_T, YYPTRDIFF_MAXIMUM):
	Default to long, not int.
	(yy_lac_stack_realloc, yy_lac, yytnamerr, yyparse):
	Avoid casts to YYPTRDIFF_T that were masking the problem.

2019-10-06  Paul Eggert  <eggert@cs.ucla.edu>

Work around GCC 4.8 false alarms without casts
	* data/skeletons/yacc.c (yyparse):
	Initialize yyes_capacity with a signed expression.
	* tests/local.at (AT_YYLEX_DEFINE(c)):
	Use enum to avoid cast.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

tests: make recheck
	* tests/local.mk (recheck): New.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: also show suggested %empty
	* src/reader.c (grammar_rule_check_and_complete): Suggest to add %empty.
	* tests/actions.at, tests/diagnostics.at: Adjust expectations.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: sort symbols per location
	Because the checking of the grammar is made by phases after the whole
	grammar was read, we sometimes have diagnostics that look weird.  In
	some case, within one type of checking, the entities are not checked
	in the order in which they appear in the file.  For instance, checking
	symbols is done on the list of symbols sorted by tag:

foo.y:1.20-22: warning: symbol BAR is used, but is not defined as a token and has no rules [-Wother]
	        1 | %destructor {} QUX BAR
	          |                    ^~~
	    foo.y:1.16-18: warning: symbol QUX is used, but is not defined as a token and has no rules [-Wother]
	        1 | %destructor {} QUX BAR
	          |                ^~~

Let's sort them by location instead:

foo.y:1.16-18: warning: symbol 'QUX' is used, but is not defined as a token and has no rules [-Wother]
	        1 | %destructor {} QUX BAR
	          |                ^~~
	    foo.y:1.20-22: warning: symbol 'BAR' is used, but is not defined as a token and has no rules [-Wother]
	        1 | %destructor {} QUX BAR
	          |                    ^~~

* src/location.h (location_cmp): Be robust to empty file names.
	* src/symtab.c (symbol_cmp): Sort by location.
	* tests/input.at: Adjust expectations.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: suggest fixes for undeclared symbols
	From

input.y:1.17-19: warning: symbol baz is used, but is not defined as a token and has no rules [-Wother]
	         1 | %printer {} foo baz
	           |                 ^~~

input.y:1.17-19: warning: symbol 'baz' is used, but is not defined as a token and has no rules; did you mean 'bar'? [-Wother]
	        1 | %printer {} foo baz
	          |                 ^~~
	          |                 bar

* bootstrap.conf: We need fstrcmp.
	* src/symtab.c (symbol_from_uniqstr_fuzzy): New.
	(complain_symbol_undeclared): Use it.
	* tests/diagnostics.at (Suggestions): New.
	* data/bison-default.css (insertion): Rename as...
	(fixit-insert): this, as this is what GCC uses.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

style: isolate complain_symbol_undeclared
	* src/symtab.c (complain_symbol_undeclared): New.
	Use it.
	Use quote on the guilty symbol (like GCC does, and we also do
	elsewhere).
	* tests/input.at: Adjust.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

style: simplify the handling of symbol and semantic_type tables
	Both are stored in a hash, and back in the days, we used to iterate
	over these tables using hash_do_for_each.  However, the order of
	traversal was not deterministic, which was a nuisance for
	deterministic output (and therefore also a problem for tests).  So at
	some point (83b60c97ee1f98bb1f15ffa38acdc4cc765515f5) we generated a
	sorted list of these symbols, and symbols_do actually iterated on that
	list.  But we kept the constraints of using hash_do_for_each, which
	requires a lot of ceremonial code, and makes it hard/unnatural to
	preserve data between iterations (see the next commit).

Alas, this is C, not C++.

Let's remove this abstraction, and directly iterate on the sorted
	tables.

* src/symtab.c (symbols_do): Remove.
	Adjust callers to use a simple for-loop instead.
	(table_sort): New.
	(symbols_check_defined): Use it.
	(symbol_check_defined_processor, symbol_pack_processor)
	(semantic_type_check_defined_processor, symbol_translation_processor):
	Remove.
	Simplify the corresponding functions (that no longer need to return a
	bool).

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: display suggested update after the caret-info
	This commit adds the suggestion in green, on the line below the
	caret-and-tildes.

foo.y:1.1-14: warning: deprecated directive: '%error-verbose', use '%define parse.error verbose' [-Wdeprecated]
	        1 | %error-verbose
	          | ^~~~~~~~~~~~~~
	          | %define parse.error verbose

The current approach, with location_caret_suggestion, is fragile:
	there's a protocol of calls to the complain functions which is strict.
	We should rather have a richer structure describing the diagnostics,
	including with submessages such as the suggestions, passed in the end
	to the routines in charge of formatting and printing them.

* src/location.h, src/location.c (location_caret_suggestion): New.
	* src/complain.c (deprecated_directive): Use it.
	* tests/diagnostics.at, tests/input.at: Adjust expectations.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: isolate caret_set_column
	* src/location.c (caret_info): Add width and skip members.
	(caret_set_column): New.
	Use it.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: isolate caret_set_file
	* src/location.c (caret_set_file): New.
	Store the current line's length in caret_info.line_len.
	Pay attention to fseek's return value.
	Extracted from...
	(location_caret): here.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

tests: use tput to get the number of columns
	* tests/bison.in: here.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

TODO: update
	I no longer agree with that item, there are indeed two things to
	report: lack of definition, and being useless.  We could have either
	one without the other, they are not directly related.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: work around warnings from G++ 4.8
	input.c: In function 'int yyparse()':
	input.c: error: conversion to 'long int' from 'long unsigned int'
	                may change the sign of the result [-Werror=sign-conversion]
	   yyes_capacity = sizeof yyesa / sizeof *yyes;
	                                ^
	cc1plus: all warnings being treated as errors

* data/skeletons/yacc.c: here.

2019-10-06  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: work around warnings from Clang++ 3.3 and 3.4
	When we run the test suite with these C++ compilers to compile C code,
	we get:

239. synclines.at:440: testing syncline escapes: yacc.c ...
	    ../../tests/synclines.at:440: $CC $CFLAGS $CPPFLAGS \"\\\"\".c -o \"\\\"\" ||
	              exit 77
	    stderr:
	    stdout:
	    ../../tests/synclines.at:440: COLUMNS=1000; export COLUMNS;  bison --color=no -fno-caret  -o \"\\\"\".c \"\\\"\".y
	    ../../tests/synclines.at:440: $CC $CFLAGS $CPPFLAGS  $LDFLAGS -o \"\\\"\" \"\\\"\".c $LIBS
	    stderr:
	    "\"".c:1102:41: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
	          YYPTRDIFF_T yysize = yyssp - yyss + 1;
	                      ~~~~~~   ~~~~~~~~~~~~~^~~
	    1 error generated.

193. conflicts.at:545: testing parse.error=verbose and consistent errors: lr.type=canonical-lr parse.lac=full ...
	    input.c:737:75: error: implicit conversion loses integer precision: 'long' to 'int'
	                           [-Werror,-Wshorten-64-to-32]
	      YYPTRDIFF_T yysize_old = *yytop == yytop_empty ? 0 : *yytop - *yybottom + 1;
	                  ~~~~~~~~~~                               ~~~~~~~~~~~~~~~~~~~^~~
	    input.c:901:48: error: implicit conversion loses integer precision: 'long' to 'int'
	                           [-Werror,-Wshorten-64-to-32]
	                YYPTRDIFF_T yysize = yyesp - *yyes + 1;
	                            ~~~~~~   ~~~~~~~~~~~~~~^~~

* data/skeletons/yacc.c: Add more casts.

2019-10-05  Akim Demaille  <akim.demaille@gmail.com>

tests: avoid a GCC 4.8 warning
	GCC 4.8 reports:

input.y:57:33: error: conversion to 'int' from 'long unsigned int'
	                          may alter its value [-Werror=conversion]
	       int input_elts = sizeof input / sizeof input[0];
	                                     ^

* tests/local.at (AT_YYLEX_DEFINE(c)): Add a cast (sorry, Paul!).

2019-10-05  Paul Eggert  <eggert@cs.ucla.edu>

* data/skeletons/glr.c (yysplitStack): Pacify Clang 8.

2019-10-05  Paul Eggert  <eggert@cs.ucla.edu>

Avoid quiet conversion of pointer to bool
	* src/location.c (caret_set_file):
	* src/scan-code.l (contains_dot_or_dash):
	Do not quietly convert pointer to bool, as Oracle Developer Studio
	12.6 complains and it is arguably confusing style anyway.

2019-10-05  Paul Eggert  <eggert@cs.ucla.edu>

Port ARGMATCH_DEFINE_GROUP calls to C99
	* src/complain.c, src/getargs.c: Omit ‘;’ after call
	to ARGMATCH_DEFINE_GROUP, as C99 does not allow ‘;’ there.

2019-10-05  Paul Eggert  <eggert@cs.ucla.edu>

Port lexcalc scan.l to Solaris 10
	* examples/c/lexcalc/scan.l: Include errno.h.

2019-10-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: use casts instead of pragmas when losing integer width
	For instance with Clang 4, 8, etc.:

input.c:1166:12: error: implicit conversion loses integer precision: 'int' to 'yy_state_num' (aka 'signed char') [-Werror,-Wconversion]
	      *yyssp = yystate;
	             ~ ^~~~~~~

And GCC 8:

input.c:1166:12: error: implicit conversion loses integer precision: 'int' to 'yy_state_num' (aka 'signed char') [-Werror,-Wimplicit-int-conversion]
	      *yyssp = yystate;
	             ~ ^~~~~~~

* data/skeletons/yacc.c (YY_CONVERT_INT_BEGIN): Remove.
	Adjust callers.

2019-10-04  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: fix warnings about undefined macros
	For instance with GCC 4.9 and --enable-gcc-warnings:

25. input.at:1201: testing Torturing the Scanner ...
	    ../../tests/input.at:1344: $CC $CFLAGS $CPPFLAGS  -c -o input.o input.c
	    stderr:
	    input.c:239:18: error: "__STDC_VERSION__" is not defined [-Werror=undef]
	     # elif 199901 <= __STDC_VERSION__
	                      ^
	    input.c:256:18: error: "__STDC_VERSION__" is not defined [-Werror=undef]
	     # elif 199901 <= __STDC_VERSION__
	                      ^

* data/skeletons/yacc.c: Check that __STDC_VERSION__ is defined before
	using it.

2019-10-04  Akim Demaille  <akim.demaille@gmail.com>

tests: check more state numbers
	* tests/torture.at (State number type): Also check 128, 129 and
	32768.

2019-10-03  Paul Eggert  <eggert@cs.ucla.edu>

* doc/bison.texi (Table of Symbols): Mention memory exhaustion.

2019-10-03  Paul Eggert  <eggert@cs.ucla.edu>

Simplify mfcalc error handling
	* doc/bison.texi (Mfcalc Symbol Table, Mfcalc Lexer):
	Don’t abort on memory allocation failure or integer overflow.
	Instead, comment that these things aren’t checked for.

2019-10-03  Akim Demaille  <akim.demaille@gmail.com>

c++: fix comments suggesting to use %require
	* data/skeletons/location.cc, data/skeletons/stack.hh: Here.

2019-10-03  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: simplify uses of size_t
	* data/skeletons/stack.hh (stack::index_type): New type.
	(stack::size, stack::operator[]): Be about an index_type rather than a
	size_type and an int.

2019-10-03  Akim Demaille  <akim.demaille@gmail.com>

c++: fixes for old compilers
	On the CI with GCC 6:

examples/c++/calc++/parser.cc:845:5: error: 'ptrdiff_t' was not declared in this scope
	         ptrdiff_t yycount = 0;
	         ^~~~~~~~~
	    examples/c++/calc++/parser.cc:845:5: note: suggested alternatives:
	    /usr/include/x86_64-linux-gnu/c++/6/bits/c++config.h:202:28: note:   'std::ptrdiff_t'
	       typedef __PTRDIFF_TYPE__ ptrdiff_t;
	                                ^~~~~~~~~

* data/skeletons/lalr1.cc: Qualify ptrdiff_t and size_t with std::.

2019-10-03  Akim Demaille  <akim.demaille@gmail.com>

tests: be robust to -DNDEBUG
	input.y: In function 'yylex':
	input.y:67:7: error: unused variable 'input_elts' [-Werror=unused-variable]
	   int input_elts = sizeof input / sizeof input[0];
	       ^~~~~~~~~~
	cc1: all warnings being treated as errors

* tests/input.at, tests/local.at: Avoid that.

2019-10-03  Akim Demaille  <akim.demaille@gmail.com>

CI: remove the symlink before creating it
	Currently we fail if we rerun a job that succeeded to push the
	tarball.

2019-10-03  Paul Eggert  <eggert@cs.ucla.edu>

Adjust ‘Big horizontal’ test case
	* tests/torture.at (Big horizontal): Adjust to recent changes with
	integers.  If there are states 0..256, Bison now uses a signed
	rather than an unsigned 16-bit integer.

2019-10-03  Paul Eggert  <eggert@cs.ucla.edu>

regen

2019-10-03  Paul Eggert  <eggert@cs.ucla.edu>

Prefer signed to unsigned integers
	This patch contains more fixes to prefer signed to unsigned
	integer types, as modern tools like 'gcc -fsanitize=undefined'
	can check for signed integer overflow but not unsigned overflow.
	* NEWS: Document the API change.
	* boostrap.conf (gnulib_modules): Add intprops.
	* data/skeletons/glr.c: Include stddef.h and stdint.h,
	since this skeleton can assume C99 or later.
	(YYSIZEMAX): Now signed, and the minimum of SIZE_MAX and PTRDIFF_MAX.
	(yybool) [!__cplusplus]: Now signed (which is how bool behaves).
	(YYTRANSLATE): Avoid use of unsigned, and make the macro
	safe even for values greater than UINT_MAX.
	(yytnamerr, struct yyGLRState, struct yyGLRStateSet, struct yyGLRStack)
	(yyaddDeferredAction, yyinitStateSet, yyinitGLRStack)
	(yyexpandGLRStack, yymarkStackDeleted, yyremoveDeletes)
	(yyglrShift, yyglrShiftDefer, yy_reduce_print, yydoAction)
	(yyglrReduce, yysplitStack, yyreportTree, yycompressStack)
	(yyprocessOneStack, yyreportSyntaxError, yyrecoverSyntaxError)
	(yyparse, yy_yypstack, yypstack, yypdumpstack):
	* tests/input.at (Torturing the Scanner):
	Prefer ptrdiff_t to size_t.
	* data/skeletons/c++.m4 (b4_yytranslate_define):
	* src/AnnotationList.c (AnnotationList__computePredecessorAnnotations):
	* src/AnnotationList.h (AnnotationIndex):
	* src/InadequacyList.h (InadequacyListNodeCount):
	* src/closure.c (closure_new):
	* src/complain.c (error_message, complains, complain_indent)
	(complain_args, duplicate_directive, duplicate_rule_directive):
	* src/gram.c (nritems, ritem_print, grammar_dump):
	* src/ielr.c (ielr_compute_ritem_sees_lookahead_set)
	(ielr_item_has_lookahead, ielr_compute_annotation_lists)
	(ielr_compute_lookaheads):
	* src/location.c (columns, boundary_print, location_print):
	* src/muscle-tab.c (muscle_percent_define_insert)
	(muscle_percent_define_check_values):
	* src/output.c (prepare_rules, prepare_actions):
	* src/parse-gram.y (id, handle_require):
	* src/reader.c (record_merge_function_type, packgram):
	* src/reduce.c (nuseless_productions, nuseless_nonterminals)
	(inaccessable_symbols):
	* src/relation.c (relation_print):
	* src/scan-code.l (variant, variant_table_size, variant_count)
	(variant_add, get_at_spec, show_sub_message, show_sub_messages)
	(parse_ref):
	* src/scan-gram.l (<SC_ESCAPED_STRING,SC_ESCAPED_CHARACTER>)
	(scan_integer, convert_ucn_to_byte, handle_syncline):
	* src/scan-skel.l (at_complain):
	* src/symtab.c (complain_symbol_redeclared)
	(complain_semantic_type_redeclared, complain_class_redeclared)
	(symbol_class_set, complain_user_token_number_redeclared):
	* src/tables.c (conflict_tos, conflrow, conflict_table)
	(conflict_list, save_row, pack_vector):
	* tests/local.at (AT_YYLEX_DEFINE(c)):
	Prefer signed to unsigned integer.
	* data/skeletons/lalr1.cc (yy_lac_check_):
	* tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR):
	* tests/local.at (AT_YYLEX_DEFINE(c)):
	Omit now-unnecessary casts.
	* data/skeletons/location.cc (b4_location_define):
	* doc/bison.texi (Mfcalc Lexer, C++ position, C++ location):
	Prefer int to unsigned for line and column numbers.
	Change example to abort explicitly on memory exhaustion,
	and fix an off-by-one bug that led to undefined behavior.
	* data/skeletons/stack.hh (stack::operator[]):
	Also allow ptrdiff_t indexes.
	(stack::pop, slice::slice, slice::operator[]):
	Index arg is now ptrdiff_t, not int.
	(stack::ssize): New method.
	(slice::range_): Now ptrdiff_t, not int.
	* data/skeletons/yacc.c (b4_state_num_type): Remove.
	All uses replaced by b4_int_type.
	(YY_CONVERT_INT_BEGIN, YY_CONVERT_INT_END): New macros.
	(yylac, yyparse): Use them around conversions that -Wconversion
	would give false alarms about. 	Omit unnecessary casts.
	(yy_stack_print): Use int rather than unsigned, and omit
	a cast that doesn’t seem to be needed here any more.
	* examples/c++/variant.yy (yylex):
	* examples/c++/variant-11.yy (yylex):
	Omit no-longer-needed conversions to unsigned.
	* src/InadequacyList.c (InadequacyList__new_conflict):
	Don’t assume *node_count is unsigned.
	* src/output.c (muscle_insert_unsigned_table):
	Remove; no longer used.

2019-10-02  Paul Eggert  <eggert@cs.ucla.edu>

Prefer signed types for indexes in skeletons
	* NEWS: Mention this.
	* data/skeletons/c.m4 (b4_int_type):
	Prefer char if it will do, and prefer signed types to unsigned if
	either will do.
	* data/skeletons/glr.c (yy_reduce_print): No need to
	convert rule line to unsigned long.
	(yyrecoverSyntaxError): Put action into an int to
	avoid GCC warning of using a char subscript.
	* data/skeletons/lalr1.cc (yy_lac_check_, yysyntax_error_):
	Prefer ptrdiff_t to size_t.
	* data/skeletons/yacc.c (b4_int_type):
	Prefer signed types to unsigned if either will do.
	* data/skeletons/yacc.c (b4_declare_parser_state_variables):
	(YYSTACK_RELOCATE, YYCOPY, yy_lac_stack_realloc, yy_lac)
	(yytnamerr, yysyntax_error, yyparse): Prefer ptrdiff_t to size_t.
	(YYPTRDIFF_T, YYPTRDIFF_MAXIMUM): New macros.
	(YYSIZE_T): Fix "! defined YYSIZE_T" typo.
	(YYSIZE_MAXIMUM): Take the minimum of PTRDIFF_MAX and SIZE_MAX.
	(YYSIZEOF): New macro.
	(YYSTACK_GAP_MAXIMUM, YYSTACK_BYTES, YYSTACK_RELOCATE)
	(yy_lac_stack_realloc, yyparse): Use it.
	(YYCOPY, yy_lac_stack_realloc): Cast to YYSIZE_T to pacify GCC.
	(yy_reduce_print): Use int instead of unsigned long when int
	will do.
	(yy_lac_stack_realloc): Prefer long to unsigned long when
	either will do.
	* tests/regression.at: Adjust to these changes.

2019-09-30  Akim Demaille  <akim.demaille@gmail.com>

yacc: use the most appropriate integral type for state numbers
	Currently we properly use the "best" integral type for tables,
	including those storing state numbers.  However the variables for
	state numbers used in yyparse (and its dependencies such as
	yy_stack_print) still use int16_t invariably.  As a consequence, very
	large models overflow these variables.

Let's use the "best" type for these variables too.  It turns out that
	we can still use 16 bits for twice larger automata: stick to unsigned
	types.

However using 'unsigned' when 16 bits are not enough is troublesome
	and generates tons of warnings about signedness issues.  Instead,
	let's use 'int'.

Reported by Tom Kramer.
	https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00018.html

* data/skeletons/yacc.c (b4_state_num_type): New.
	(yy_state_num): Be computed from YYNSTATES.
	* tests/linear: New.
	* tests/torture.at (State number type): New.
	Use it.

2019-09-30  Akim Demaille  <akim.demaille@gmail.com>

yacc: introduce a type for states
	* data/skeletons/yacc.c (yy_state_num): New.
	Use it for arrays of states.

2019-09-30  Akim Demaille  <akim.demaille@gmail.com>

style: prefer symbolic values rather than litterals
	Instead of

#define YYPACT_NINF -130
	    #define yypact_value_is_default(Yystate) \
	      (!!((Yystate) == (-130)))

generate

#define YYPACT_NINF (-130)
	    #define yypact_value_is_default(Yyn) \
	      ((Yyn) == YYPACT_NINF)

* data/skeletons/c.m4 (b4_table_value_equals): Add support for $4.
	* data/skeletons/glr.c, data/skeletons/yacc.c: Use it.
	Also, use shorter macro argument names, the name of the macro is clear
	enough.

2019-09-30  Akim Demaille  <akim.demaille@gmail.com>

style: change misleading macro argument name
	* data/skeletons/glr.c, data/skeletons/yacc.c
	(yypact_value_is_default): It does not take a rule number as argument.

2019-09-28  Akim Demaille  <akim.demaille@gmail.com>

Merge remote-tracking branch 'upstream/maint'
	* upstream/maint:
	  c++: add copy ctors for compatibility with the IAR compiler
	  CI: show git status
	  CI: disable ICC
	  tests: pass -jN from Make to the test suite
	  quotearg: avoid leaks
	  maint: post-release administrivia

2019-09-27  Akim Demaille  <akim.demaille@gmail.com>

c++: add copy ctors for compatibility with the IAR compiler
	Reported by Andreas Damm.
	https://savannah.gnu.org/support/?110032

* data/skeletons/lalr1.cc (stack_symbol_type::operator=): New
	overload, const, to please the IAR C++ compiler (version ca 2013).

2019-09-23  Akim Demaille  <akim.demaille@gmail.com>

CI: show git status

2019-09-23  Akim Demaille  <akim.demaille@gmail.com>

CI: disable ICC
	It seems that Intel changed something in their license management.
	https://github.com/nemequ/icc-travis/issues/15

2019-09-23  Akim Demaille  <akim.demaille@gmail.com>

tests: pass -jN from Make to the test suite
	I am sooooo tired of typing "make -j5 TESTSUITEFLAGS=-j5"...
	Should have done this years ago.

* cfg.mk (TESTSUITEFLAGS): here.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

quotearg: avoid leaks
	Reported by Tomasz Kłoczko.
	https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00008.html

* src/main.c (main): Free quotearg's memory later.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: get the screen width from the terminal
	* bootstrap.conf: We need winsz-ioctl and winsz-termios.
	* src/location.c (columns): Use winsize to get the number of
	columns.
	Code taken from the GNU Coreutils.
	* src/location.h, src/location.c (caret_init): New.
	* src/complain.c (complain_init): Call it.
	* tests/bison.in: Export COLUMNS so that users of tests/bison can
	enjoy proper line truncation.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: don't print ellipsis on the caret line
	From

9 | ...TUVWXYZ  ABCDEFGHIJKLMNOPQRSTUVWXYZ  ABCDEFGHIJKL
	      | ...         ^~~~~~~~~~~~~~~~~~~~~~~~~~

9 | ...TUVWXYZ  ABCDEFGHIJKLMNOPQRSTUVWXYZ  ABCDEFGHI...
	      |             ^~~~~~~~~~~~~~~~~~~~~~~~~~

* src/location.c (location_caret): here.
	* tests/diagnostics.at: Adjust expectations.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: also show truncation at the end of line with "..."
	From

9 | ...TUVWXYZ  ABCDEFGHIJKLMNOPQRSTUVWXYZ  ABCDEFGHIJKL
	      | ...         ^~~~~~~~~~~~~~~~~~~~~~~~~~

9 | ...TUVWXYZ  ABCDEFGHIJKLMNOPQRSTUVWXYZ  ABCDEFGHI...
	      | ...         ^~~~~~~~~~~~~~~~~~~~~~~~~~

* src/location.c (location_caret): here.
	* tests/diagnostics.at: Adjust expectations.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: check that quoted lines are truncated
	* tests/diagnostics.at (Screen width: 60 columns, Screen width: 80
	columns, Screen width: 200 columns): New tests.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: truncate quoted sources to fit the screen
	* src/location.c (min_int, columns): New.
	(location_caret): Compute the line width.  Based on it, compute how
	many columns must be skipped before the quoted location and truncated
	after, to fit the sceen width.
	* tests/local.at (AT_QUELL_VALGRIND): Transform into...
	(AT_SET_ENV_IF, AT_SET_ENV): these.
	Define COLUMNS to protect the test suite from the user's environment.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: learn how to count column number with multibyte chars
	So far diagnostics were cheating: in addition to the 'column' field of
	locations (based on actual screen width per multibyte characters and
	on tabulation expansion), the scanner sets the 'byte' field.
	Diagnostics used this byte count to decide where to insert (color)
	style.

We want to be able to truncate the quoted lines when there are too
	wide to fit the screen.  This requires that the diagnostics learn how
	to count columns, the byte-in-boundary trick no longer works.

Bytes are still used for fix-its.

* bootstrap.conf: We need mbfile for mbf_getc.
	* src/location.c (caret_info): We need an mbfile.
	(caret_set_file): Initialize it.
	(caret_getc): Convert to mbfile.
	(location_caret): Instead of relying on the byte position to decide
	where to insert the color style, count the current column using
	boundary_compute.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style: rename member for clariy
	* src/location.c (caret_info): Now that we no longer have a 'file'
	member (see previous commit), rename 'source' as 'file'.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style: use a boundary to track the caret_info
	* src/location.c (caret_info): Replace file and line with pos, a
	boundary.  This will allow us to use features of the boundary type,
	such as boundary_compute.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: extract boundary_compute from location_compute
	The handling of the contributions of the tabulations in the columns is
	burried inside location_compute.  We will soon be willing to use the
	boundary part of the computation (to compute the current column number
	each time we read a multibyte char).

* src/location.c (boundary_compute): New, extracted from...
	(location_compute): here.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style: add caret_set_file
	To make the following commits easier to read.

* src/location.c (caret_set_file): New.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style: minor changes
	* src/location.c (location_caret): Factor two branches of an if.

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

CI: show git status

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

git: update ignores

2019-09-22  Akim Demaille  <akim.demaille@gmail.com>

git: update ignores

2019-09-21  Akim Demaille  <akim.demaille@gmail.com>

quotearg: avoid leaks
	Reported by Tomasz Kłoczko.
	https://lists.gnu.org/archive/html/bug-bison/2019-09/msg00008.html

* src/main.c (main): Free quotearg's memory later.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

tests: pass -jN from Make to the test suite
	I am sooooo tired of typing "make -j5 TESTSUITEFLAGS=-j5"...
	Should have done this years ago.

* cfg.mk (TESTSUITEFLAGS): here.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

java: handle eof in yytranslate
	* data/skeletons/lalr1.java (yytranslate_): Handle eof here, as is done
	in lalr1.cc.
	* tests/javapush.at: Adjust.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

d: handle eof in yytranslate
	This changes the traces from

Reading a token:
	    Now at end of input.

Reading a token:
	    Next token is token $end (7FFEE56E6474)

which is ok.  Actually it is even better, as it gives the location
	when locations are enabled, and is clearer when rules explicitly use
	the EOF token.

* data/skeletons/lalr1.d (yytranslate_): Handle eof here, as is done
	in lalr1.cc.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

parser: use api.token.raw
	* src/parse-gram.y: Here.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: document it
	* doc/bison.texi: here.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: cannot be used with character literals
	* src/parse-gram.y (CHAR): api.token.raw and character literals are
	mutually exclusive.
	* tests/input.at (Character literals and api.token.raw): New.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: apply to the other skeletons
	* data/skeletons/c++.m4, data/skeletons/glr.c,
	* data/skeletons/lalr1.c, data/skeletons/lalr1.java:
	Add support for api.token.raw.

* tests/scanner.at: Check them.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: check it
	* tests/local.at (AT_TOKEN_RAW_IF): New.
	* tests/local.mk: New.
	Use it.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

api.token.raw: implement
	Bison used to feature %raw, documented as follows:

@item %raw
	    The output file @file{@var{name}.h} normally defines the tokens with
	    Yacc-compatible token numbers.  If this option is specified, the
	    internal Bison numbers are used instead.  (Yacc-compatible numbers start
	    at 257 except for single character tokens; Bison assigns token numbers
	    sequentially for all tokens starting at 3.)

Unfortunately, as far as I can tell, it never worked: token numbers
	are indeed changed in the generated tables (from external token number
	to internal), yet the code was still applying the mapping from
	external token numbers to internal token numbers.

This commit reintroduces the feature as it was expected to be.

* data/skeletons/bison.m4 (b4_token_format): When api.token.raw is
	enabled, use the internal token number.
	* data/skeletons/yacc.c (yytranslate): Don't emit if api.token.raw is
	enabled.
	(YYTRANSLATE): Adjust.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

style: tidy yacc.c
	* data/skeletons/yacc.c: Include 'c.m4' first.
	Then sort the handling of %define variables.
	* tests/input.at: Adjust.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

CI: disable ICC
	It seems that Intel changed something in their license management.
	https://github.com/nemequ/icc-travis/issues/15

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix use of complain_indent
	* src/symtab.c (symbol_class_set): Here.
	* tests/diagnostics.at, tests/input.at, tests/regression.at: Adjust
	expectations.

2019-09-14  Akim Demaille  <akim.demaille@gmail.com>

input: stop treating lone CRs as end-of-lines
	We used to treat lone CRs (\r, aka ^M) as regular NLs (\n), probably
	to please Classic MacOS.  As of today, it makes more sense to treat \r
	like a plain white space character.

https://lists.gnu.org/archive/html/bison-patches/2019-09/msg00027.html

* src/scan-gram.l (no_cr_read): Remove.  Instead, use...
	(eol): this new abbreviation denoting end-of-line.
	* src/location.c (caret_getc): New.
	(location_caret): Use it.
	* tests/diagnostics.at (Carriage return): Adjust expectations.
	(CR NL): New.

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

Merge tag 'v3.4.2' into HEAD
	bison 3.4.2

* tag 'v3.4.2': (24 commits)
	  version 3.4.2
	  CI: always uninstall icc
	  news: more bug fixes thanks to Marc Schönefeld
	  diagnostics: beware of unexpected EOF when quoting the source file
	  gnulib: update
	  build: fix distcheck
	  tests: add noexcept to please GCC 9
	  news: update
	  fix: don't die when EOF token is defined twice
	  tests: check token redeclaration
	  yacc.c: beware of GCC's -Wmaybe-uninitialized
	  glr.c: initialize vector of bools
	  gnulib: update
	  check for memory exhaustion
	  diagnostics: avoid global variables
	  diagnostics: fix invalid error message indentation
	  git: ignore files generated in gnulib-po
	  c++: avoid duplicate definition of YYUSE
	  gnulib: update
	  CI: more compilers
	  ...

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

version 3.4.2
	* NEWS: Record release date.

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

CI: always uninstall icc

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

news: more bug fixes thanks to Marc Schönefeld

2019-09-12  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: beware of unexpected EOF when quoting the source file
	When the input file contains lone CRs (aka, ^M, \r), the locations see
	a new line.  Diagnostics look only at \n as end-of-line, so sometimes
	there is an offset in diagnostics.  Worse yet: sometimes we loop
	endlessly waiting for \n to come from a continuous stream of EOF.

Fix that:
	- check for EOF
	- beware not to call end_use_class if begin_use_class was not
	  called (which would abort).  This could happen if the actual
	  line is shorter that the expected one.

Prompted by a (private) report from Marc Schönefeld.

* src/location.c (location_caret): here.
	* tests/diagnostics.at (Carriage return): New.

2019-09-11  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	Contains the creation of the xhash module.
	https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00046.html

* src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c:
	Use hash_xinitialize.

2019-09-11  Akim Demaille  <akim.demaille@gmail.com>

build: fix distcheck
	* configure.ac (gl_LIBOBJS): Adjust so that the generated files are
	indeed the expected ones.

2019-09-10  Akim Demaille  <akim.demaille@gmail.com>

Fix that:
	- check for EOF
	- beware not to call end_use_class if begin_use_class was not
	  called (which would abort).  This could happen if the actual
	  line is shorter that the expected one.

Prompted by a (private) report from Marc Schönefeld.

* src/location.c (location_caret): here.
	* tests/diagnostics.at (Carriage return): New.

2019-09-10  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	Contains the creation of the xhash module.
	https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00046.html

* src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c:
	Use hash_xinitialize.

2019-09-10  Akim Demaille  <akim.demaille@gmail.com>

build: fix distcheck
	* configure.ac (gl_LIBOBJS): Adjust so that the generated files are
	indeed the expected ones.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

tests: add noexcept to please GCC 9
	    bison/tests/c++.at:552: bison --color=no -fno-caret  -o list.cc list.y
	    bison/tests/c++.at:552: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o list list.cc $LIBS
	    stderr:
	    gcc9/c++/ext/new_allocator.h: In instantiation of 'void __gnu_cxx::new_allocator<_Tp>::construct(_Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string]':
	    gcc9/c++/bits/alloc_traits.h:482:2:   required from 'static void std::allocator_traits<std::allocator<_CharT> >::construct(std::allocator_traits<std::allocator<_CharT> >::allocator_type&, _Up*, _Args&& ...) [with _Up = string; _Args = {string}; _Tp = string; std::allocator_traits<std::allocator<_CharT> >::allocator_type = std::allocator<string>]'
	    gcc9/c++/bits/stl_uninitialized.h:888:67:   required from 'void std::__relocate_object_a(_Tp*, _Up*, _Allocator&) [with _Tp = string; _Up = string; _Allocator = std::allocator<string>]'
	    gcc9/c++/bits/stl_uninitialized.h:920:47:   required from '_ForwardIterator std::__relocate_a_1(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator<string>]'
	    gcc9/c++/bits/stl_uninitialized.h:942:37:   required from '_ForwardIterator std::__relocate_a(_InputIterator, _InputIterator, _ForwardIterator, _Allocator&) [with _InputIterator = string*; _ForwardIterator = string*; _Allocator = std::allocator<string>]'
	    gcc9/c++/bits/stl_vector.h:430:35:   required from 'static constexpr bool std::vector<_Tp, _Alloc>::_S_nothrow_relocate(std::true_type) [with _Tp = string; _Alloc = std::allocator<string>; std::true_type = std::integral_constant<bool, true>]'
	    gcc9/c++/bits/stl_vector.h:446:28:   required from 'void std::vector<_Tp, _Alloc>::_M_realloc_insert(std::vector<_Tp, _Alloc>::iterator, _Args&& ...) [with _Args = {const string&}; _Tp = string; _Alloc = std::allocator<string>; std::vector<_Tp, _Alloc>::iterator = __gnu_cxx::__normal_iterator<string*, std::vector<string> >; typename std::_Vector_base<_Tp, _Alloc>::pointer = string*]'
	    gcc9/c++/bits/stl_vector.h:1195:4:   required from 'void std::vector<_Tp, _Alloc>::push_back(const value_type&) [with _Tp = string; _Alloc = std::allocator<string>; std::vector<_Tp, _Alloc>::value_type = string]'
	    list.y:126:110:   required from here
	    gcc9/c++/bits/vector.tcc:459:44:   in 'constexpr' expansion of 'std::vector<string>::_S_use_relocate()'
	    list.y:41:7: error: but 'string::string(string&&)' does not throw; perhaps it should be declared 'noexcept' [-Werror=noexcept]
	       41 |       string (string&& s)
	          |       ^~~~~~

* tests/c++.at (Variants): Add noexcept where appropriate.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

* tests/c++.at (Variants): Add noexcept where appropriate.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

news: update

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

fix: don't die when EOF token is defined twice
	With

%token EOF 0 EOF 0

we get

input.y:3.14-16: warning: symbol EOF redeclared [-Wother]
	        3 | %token EOF 0 EOF 0
	          |              ^~~
	    input.y:3.8-10: previous declaration
	        3 | %token EOF 0 EOF 0
	          |        ^~~
	    Assertion failed: (nsyms == ntokens + nvars), function check_and_convert_grammar,
	        file /Users/akim/src/gnu/bison/src/reader.c, line 839.

Reported by Marc Schönefeld.

* src/symtab.c (symbol_user_token_number_set): Register only the
	first definition of the end of input token.
	* tests/input.at (Symbol redeclared): Check that case.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

tests: check token redeclaration
	* src/symtab.c (symbol_class_set): Report previous definitions when
	redeclared.
	* tests/input.at (Symbol redeclared): New.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: beware of GCC's -Wmaybe-uninitialized
	Test 400 (calc.at:773: testing Calculator api.push-pull=both
	api.pure=full parse.error=verbose %debug %locations %defines
	api.prefix={calc} %verbose %yacc) fails on the CI with GCC 8 on
	Bionic:

400. calc.at:773: testing Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc  ...
	    ../../tests/calc.at:773: bison --color=no -fno-caret -Wno-deprecated -o calc.c calc.y
	    ../../tests/calc.at:773: $CC $CFLAGS $CPPFLAGS  $LDFLAGS -o calc calc.c calc-lex.c calc-main.c $LIBS
	    stderr:
	    calc.y: In function 'int calcpush_parse(calcpstate*, int, const CALCSTYPE*, CALCLTYPE*)':
	    calc.y:26:20: error: 'yylval.CALCSTYPE::ival' may be used uninitialized in this function [-Werror=maybe-uninitialized]
	     %printer { fprintf (yyo, "%d", $$); } <ival>;
	                        ^
	    calc.c:1272:9: note: 'yylval.CALCSTYPE::ival' was declared here
	     YYSTYPE yylval YY_INITIAL_VALUE (= yyval_default);
	             ^~~~~~
	    cc1plus: all warnings being treated as errors
	    stdout:
	    ../../tests/calc.at:773: exit code was 1, expected 0
	    400. calc.at:773: 400. Calculator api.push-pull=both api.pure=full parse.error=verbose %debug %locations %defines api.prefix={calc} %verbose %yacc  (calc.at:773): FAILED (calc.at:773)

* data/skeletons/c.m4 (yy_symbol_value_print): Disable the warning
	locally.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

glr.c: initialize vector of bools
	The CI, with CC='gcc-7 -fsanitize=undefined,address
	-fno-omit-frame-pointer', reports:

calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool'
	    ../../tests/calc.at:867: cat stderr
	    --- expout	2019-09-05 20:30:37.887257545 +0000
	    +++ /home/travis/build/bison-3.4.1.72-79a1-dirty/_build/tests/testsuite.dir/at-groups/438/stdout	2019-09-05 20:30:37.887257545 +0000
	    @@ -1 +1,2 @@
	     syntax error
	    +calc.cc:1652:50: runtime error: load of value 190, which is not a valid value for type 'bool'
	    438. calc.at:867: 438. Calculator glr.cc  (calc.at:867): FAILED (calc.at:867)

The problem is that yylookaheadNeeds is not initialized in
	yyinitStateSet, and when it is copied, the value is not 0 or 1.

* data/skeletons/glr.c (yylookaheadNeeds): Initialize yylookaheadNeeds.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	Contains a fix for
	https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00016.html.
	See
	https://lists.gnu.org/archive/html/bug-gnulib/2019-09/msg00005.html.
	Reported by 江 祖铭 (Zu-Ming Jiang).

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

check for memory exhaustion
	hash_initialize returns NULL when out of memory.  Check for it, and
	die cleanly instead of crashing.

Reported by 江 祖铭 (Zu-Ming Jiang).
	https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00015.html

* src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c:
	Check the value returned by hash_initialize.

2019-09-08  László Várady  <laszlo.varady93@gmail.com>

diagnostics: avoid global variables
	* src/complain.c (indent_ptr): Remove.
	(error_message, complains): Take indent as an argument.
	Adjust callers.

2019-09-08  László Várady  <laszlo.varady93@gmail.com>

diagnostics: fix invalid error message indentation
	https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00007.html

When Bison is started with a flag that suppresses warning messages, the
	error_message() function can produce a few gigabytes of indentation
	because of a dangling pointer.

* src/complain.c (error_message): Don't reset indent_ptr here, but...
	(complain_indent): here.
	* tests/diagnostics.at (Indentation with message suppression): Check
	this case.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

git: ignore files generated in gnulib-po
	Because of them, the CI generates "-dirty" tarballs.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

c++: avoid duplicate definition of YYUSE
	Reported by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-06/msg00009.html

* data/skeletons/lalr1.cc (b4_shared_declarations): Remove the
	duplicate definition of YYUSE, the other one coming from
	b4_attribute_define.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	This update brings file from Gettext 0.20, which is not available on
	the CI yet.

.travis.yml: Adjust.
	Use Bionic now that it's available.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

CI: more compilers
	* .travis.yml: Bionic is now available, with GCC8.
	GCC7 sanitizers work, but they are too longer: cover only part 1.
	Redefine part 1 and part 2 so that part 1 is really the core of the
	tests: not playing with POSIX and C++ compiler for C code.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

CI: fail fast

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

CI: propagate sftp failures
	* .travis.yml (stage: "compile"): here.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

CI: avoid useless git costs
	Travis answered favorably to my suggestion to provide a means to
	disable git clone on some jobs (issue 7542).  See
	https://docs.travis-ci.com/user/customizing-the-build/#disabling-git-clone.

* .travis.yml: Disable git globally, enable it for i. the compile job,
	and ii. the test job on ICC which needs the install-icc.sh script.

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

CI: factor
	* .travis.yml (Clang 7 libc++ and ASAN part 2): Reuse bits from "Clang
	7 libc++ and ASAN part 1".

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-09-08  Akim Demaille  <akim.demaille@gmail.com>

2019-09-07  Akim Demaille  <akim.demaille@gmail.com>

fix: don't die when EOF token is defined twice
	With

%token EOF 0 EOF 0

we get

Reported by Marc Schönefeld.

* src/symtab.c (symbol_user_token_number_set): Register only the
	first definition of the end of input token.
	* tests/input.at (Symbol redeclared): Check that case.

2019-09-07  Akim Demaille  <akim.demaille@gmail.com>

tests: check token redeclaration
	* src/symtab.c (symbol_class_set): Report previous definitions when
	redeclared.
	* tests/input.at (Symbol redeclared): New.

2019-09-06  Akim Demaille  <akim.demaille@gmail.com>

glr.c: initialize vector of bools
	The CI, with CC='gcc-7 -fsanitize=undefined,address
	-fno-omit-frame-pointer', reports:

The problem is that yylookaheadNeeds is not initialized in
	yyinitStateSet, and when it is copied, the value is not 0 or 1.

* data/skeletons/glr.c (yylookaheadNeeds): Initialize yylookaheadNeeds.

2019-09-06  Akim Demaille  <akim.demaille@gmail.com>

* data/skeletons/c.m4 (yy_symbol_value_print): Disable the warning
	locally.

2019-09-06  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: fix LAC support
	* data/skeletons/lalr1.cc (ctor): Initialize yy_lac_established_.
	This is quite painful to write, and ugly to read.

2019-09-06  Akim Demaille  <akim.demaille@gmail.com>

style: fix comment
	* tests/atlocal.in: here.

2019-09-06  Akim Demaille  <akim.demaille@gmail.com>

CI: fail fast

2019-09-04  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	This update brings file from Gettext 0.20, which is not available on
	the CI yet.

.travis.yml: Adjust.
	Use Bionic now that it's available.

2019-09-01  Akim Demaille  <akim.demaille@gmail.com>

check for memory exhaustion
	hash_initialize returns NULL when out of memory.  Check for it, and
	die cleanly instead of crashing.

Reported by 江 祖铭 (Zu-Ming Jiang).
	https://lists.gnu.org/archive/html/bug-bison/2019-08/msg00015.html

* src/muscle-tab.c, src/state.c, src/symtab.c, src/uniqstr.c:
	Check the value returned by hash_initialize.

2019-08-30  Akim Demaille  <akim.demaille@gmail.com>

news: LAC for C++

2019-08-29  Akim Demaille  <akim.demaille@gmail.com>

d: remove useless imports
	* examples/d/calc.y, tests/calc.at: here.

2019-08-18  László Várady  <laszlo.varady93@gmail.com>

diagnostics: avoid global variables
	* src/complain.c (indent_ptr): Remove.
	(error_message, complains): Take indent as an argument.
	Adjust callers.

2019-08-18  László Várady  <laszlo.varady93@gmail.com>

diagnostics: fix invalid error message indentation
	https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00007.html

When Bison is started with a flag that suppresses warning messages, the
	error_message() function can produce a few gigabytes of indentation
	because of a dangling pointer.

* src/complain.c (error_message): Don't reset indent_ptr here, but...
	(complain_indent): here.
	* tests/diagnostics.at (Indentation with message suppression): Check
	this case.

2019-08-18  Akim Demaille  <akim.demaille@gmail.com>

c++: use resize to shrink a vector
	Suggested by Adrian Vogelsgesang.
	https://lists.gnu.org/archive/html/bison-patches/2019-08/msg00009.html

* data/skeletons/lalr1.cc (yy_lac_check_): here.

2019-08-09  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: check LAC support
	* tests/conflicts.at, tests/input.at, tests/regression.at: here.

2019-08-09  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

lalr1.cc: add LAC support
	Implement lookahead correction (LAC) for the C++ skeleton.  LAC is a
	mechanism to make sure that we report the correct list of expected
	tokens if a syntax error occurs.  So far, LAC was only supported for
	the C skeleton "yacc.c".

* data/skeletons/lalr1.cc: Add LAC support.
	* doc/bison.texi: Update.

2019-08-09  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

style: readability improvements to yacc.c
	* data/skeletons/yacc.c (yysyntax_error): Change the nesting of `m4`
	conditions slightly to make it more readable.
	The generated C code stays unchanged.

2019-08-09  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

lalr1.cc: reduce "scope"
	* data/skeletons/lalr1.cc (yy_lr_goto_state_): Make it static.

2019-08-09  Adrian Vogelsgesang  <avogelsgesang@tableau.com>

lalr1.cc: fix indentation of table declarations in the header
	* data/skeletons/lalr1.cc: Fix indentation of table declarations in
	the generated header.

2019-08-09  Akim Demaille  <akim.demaille@gmail.com>

tests: prepare LAC tests for more languages
	* tests/regression.at: Use %expect to avoid warnings.
	Set the keywords to facilitate running specific tests.
	Use macros such as AT_YYLEX_DECLARE to facilitate tests for other
	languages.
	Likewise for AT_FULL_COMPILE.

2019-08-09  Akim Demaille  <akim.demaille@gmail.com>

git: ignore files generated in gnulib-po
	Because of them, the CI generates "-dirty" tarballs.

2019-07-26  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use the modern argmatch interface
	* src/complain.h (warnings): Remove Werror.
	Adjust dependencies.
	Sort.
	Remove useless comments (see the doc in argmatch group).
	* src/complain.c (warnings_args, warnings_types): Remove.
	(warning_argmatch): Use argmatch_warning_value.
	(warnings_print_categories): Use argmatch_warning_argument.

2019-07-19  Akim Demaille  <akim.demaille@gmail.com>

doc: avoid spurious empty lines in the option table
	In Texinfo. empty lines in multitable rows generate empty lines in the
	output.  Avoid them altogether.

With help from Gavin Smith.
	https://lists.gnu.org/archive/html/bug-texinfo/2019-07/msg00000.html

* build-aux/cross-options.pl: Separate rows with empty lines.
	So, to be more readable, generate a single line for each row.
	Use Perl format to this end.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

--fixed-output-files: detach from --yacc
	See the previous commit.  This option should be removed, -o suffices.

* src/getargs.c (FIXED_OUTPUT_FILES): New.
	Add support for it.
	(getargs): Define loc, and use it.
	This is safer when we need to pass a pointer to a location.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

%fixed-output-files: detach from %yacc
	The name fixed-output-files is pretty clear: generate y.tab.c, as Yacc
	does.  So let's detach this from %yacc which does more: it requires
	POSIX Yacc behavior.

This directive is obsolete since December 29th 2001
	8c9a50bee13474c1491df8f79f075f5214dda0d1.  It does not show in the
	doc.  I don't want to spend more time on improving its diagnostics, it
	could be removed just as well as far as I'm concerned.

* src/scan-gram.l, src/parse-gram.y (%fixed-output-files): Detach from
	%yacc.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

style: clarify control flow
	* src/getargs.c (language_argmatch): Initialize msg.
	Check it instead of relying on a return.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

remove MS-DOS support
	DJGPP support was dropped in Bison 3.3
	(c239e53bab8e87b30889ac446d3b513da9951a35).

AS_FILE_NAME was introduced in
	ae4048011562c250d2f3c96687422d99a38745ce.

* src/getargs.c (AS_FILE_NAME): Remove.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

style: declare options in the same order as in --help
	* src/getargs.c (long_options): here.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	Contains a fix for argmatch to get proper man pages.
	See https://lists.gnu.org/archive/html/bug-gnulib/2019-07/msg00038.html

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

style: comment change
	* src/getargs.c: here.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

doc: remove the --report=look-aheads alias
	Years ago we moved from 'look-ahead' to 'lookahead', and that alias
	was kept for backward compatibility.  But now that we use argmatch to
	generate the documentation, that value clutters the doc.

* src/getargs.c (argmatch_report_args): Remove the
	--report=look-aheads alias.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

doc: fix inaccuracies wrt --define and --force-define
	The doc says that -Dfoo=bar is the same as %define foo "bar".  It is
	not: the quotes are not added (and it makes a difference).

* doc/bison.texi (Tuning the Parser): Fix the definition of -D/-F
	* src/getargs.c (usage): Likewise.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

doc: put diagnostics related options together
	* doc/bison.texi (Diagnostics): New section.
	Move --warning, --color and --style there.
	* src/getargs.c (usage): Likewise.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

doc: move -y's documentation into "Tuning the Parser"
	Let's clarify --help: use clearer "section" names, as in the doc.
	Move --yacc to where it belongs.

* src/getargs.c (usage): Rename "Parser" as "Tuning the Parser", as in
	the doc.
	Rename "Output" as "Output Files"
	Move --yacc to "Tuning the Parser".
	* doc/bison.texi: Likewise.

2019-07-07  Akim Demaille  <akim.demaille@gmail.com>

doc: document colorized diagnostics
	* src/getargs.c (argmatch_color_group): New.
	(usage): Document --color and --style.
	* doc/bison.texi (Bison Options): Split into three subsections.
	Document --color and --style.

2019-07-03  Akim Demaille  <akim.demaille@gmail.com>

gnulib: use new features of the argmatch module
	It can now generate the usage message.

* src/complain.h (feature_fixit_parsable): Rename as...
	(feature_fixit): this, for column economy.
	Adjust dependencies.
	(warning_usage): New.
	Use it.
	* src/complain.h, src/complain.c, src/getargs.h, src/getargs.c:
	Use ARGMATCH_DEFINE_GROUP instead of the older interface.

2019-07-02  Akim Demaille  <akim.demaille@gmail.com>

preserve the indentation in the ouput
	Preserve the actions' initial indentation.  For instance, on

| %define api.value.type {int}
	    | %%
	    | exp: exp '/' exp { if ($3)
	    |                     $$ = $1 + $3;
	    |                   else
	    |                     $$ = 0; }

we used to generate

|     { if (yyvsp[0])
	    |                     yyval = yyvsp[-2] + yyvsp[0];
	    |                   else
	    |                    yyval = 0; }

now we produce

|                  { if (yyvsp[0])
	    |                     yyval = yyvsp[-2] + yyvsp[0];
	    |                   else
	    |                     yyval = 0; }

See https://lists.gnu.org/archive/html/bison-patches/2019-06/msg00012.html.

* data/skeletons/bison.m4 (b4_symbol_action): Output the code in
	column 0, leave indentation matters to the C code.
	* src/output.c (user_actions_output): Preserve the incoming
	indentation in the output.
	(prepare_symbol_definitions): Likewise for %printer/%destructor.
	* tests/synclines.at (Output columns): New.

2019-07-01  Akim Demaille  <akim.demaille@gmail.com>

style: prefer passing locations by pointer
	The code is inconsistent: sometimes we pass by value, sometimes by
	reference.  Let's stick to the last, more conventional for large
	values in C.

* src/scan-code.l: Pass locations by reference.

2019-06-30  Akim Demaille  <akim.demaille@gmail.com>

c++: avoid duplicate definition of YYUSE
	Reported by Frank Heckenbach.
	https://lists.gnu.org/archive/html/bug-bison/2019-06/msg00009.html

* data/skeletons/lalr1.cc (b4_shared_declarations): Remove the
	duplicate definition of YYUSE, the other one coming from
	b4_attribute_define.

2019-06-27  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* examples/c/lexcalc/local.mk, examples/c/reccalc/local.mk:
	Here.

2019-06-23  Akim Demaille  <akim.demaille@gmail.com>

tests: restructure for clarity
	* tests/calc.at (AT_CALC_MAIN, AT_CALC_LEX): Rewrite on top of
	AT_LANG_DISPATCH.

2019-06-23  Akim Demaille  <akim.demaille@gmail.com>

d: track locations
	* configure.ac (DCFLAGS): Pass -g.
	* data/skeletons/d.m4 (b4_locations_if): Remove, let bison.m4's one do
	its job.
	* data/skeletons/lalr1.d (position): Leave filename empty by default.
	(position::toString): Don't print empty file names.
	(location::this): New ctor.
	(location::toString): Match the implementations of C/C++.
	(yy_semantic_null): Leave undefined, the previous implementation does
	not compile.
	* tests/calc.at: Improve the implementation for D.
	Enable more checks, in particular using locations.
	* tests/local.at (AT_YYERROR_DEFINE(d)): Fix its implementation.

2019-06-23  Akim Demaille  <akim.demaille@gmail.com>

d: style changes
	* data/skeletons/lalr1.d: Use a more traditional quotation scheme.
	Formatting changes.

2019-06-23  Akim Demaille  <akim.demaille@gmail.com>

d: put internal details inside the parser
	Avoid name clashes, etc.

* data/skeletons/lalr1.d (YYStackElement, YYStack): Move inside the
	parser.

2019-06-22  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-06-22  Akim Demaille  <akim.demaille@gmail.com>

remove "experimental" warnings
	Sadly enough, AFAIK, there were never answers to the "More user
	feedback will help to stabilize it" sentences.  Remove them.

* src/getargs.c: IELR, canonical LR and XML output are here to stay,
	and they are no more experimental than some other features.
	* doc/bison.texi: Likewise.
	Also remove "experimental" warning for Java, LAC, LR tuning options,
	and named references.

2019-06-22  Akim Demaille  <akim.demaille@gmail.com>

CI: propagate sftp failures
	* .travis.yml (stage: "compile"): here.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

d: honor %define parse.trace
	* data/skeletons/lalr1.d: Don't generate debug code if parse.trace is
	not enabled.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

d: style changes
	* data/skeletons/lalr1.d: here.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

d: prefer delegation to duplication
	* data/skeletons/lalr1.d: Delegate the construction of the scanner.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

d: enable #line output
	* data/skeletons/d.m4 (b4_sync_start): New.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

d: style changes
	* data/skeletons/lalr1.d: here.
	* examples/d/calc.y: Remove incorrect support for decimal numbers.
	Formatting changes.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes in glr.c
	* data/skeletons/glr.c: here.

2019-06-20  Akim Demaille  <akim.demaille@gmail.com>

java: honor %define parse.trace
	* data/skeletons/lalr1.java: Don't generate debug code if parse.trace
	is not enabled.

2019-06-19  Akim Demaille  <akim.demaille@gmail.com>

java: fix support for api.prefix
	* data/skeletons/java.m4: here.
	* tests/java.at: Check it.

2019-06-19  Akim Demaille  <akim.demaille@gmail.com>

java: style changes
	* data/skeletons/lalr1.java: Use more conventional function names for
	Java.
	Prefer < and <= to => and >.
	Use the same approach for m4 quotation as in the other skeletons.
	Fix indentation issues.

* tests/calc.at, tests/java.at, tests/javapush.at: Fix quotation style.
	(main): Use 'args', not 'argv', the former seems more conventional and
	is used elsewhere in Bison.
	Prefer character literals to integers to denote characters.
	* examples/java/Calc.y: Likewise.

2019-06-15  Akim Demaille  <akim.demaille@gmail.com>

* .travis.yml: Disable git globally, enable it for i. the compile job,
	and ii. the test job on ICC which needs the install-icc.sh script.

2019-06-12  Akim Demaille  <akim.demaille@gmail.com>

style: simplify strings to translate
	* src/conflicts.c (log_resolution): Don't translate indentation.

2019-06-12  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes, propagate const
	* src/conflicts.c (conflicts_output): here.

2019-06-12  Akim Demaille  <akim.demaille@gmail.com>

style: use clearer types
	* src/conflicts.c (conflicts): Array of Booleans.

2019-06-11  Akim Demaille  <akim.demaille@gmail.com>

tests: prefer %empty
	* tests/regression.at: here.

2019-06-09  Akim Demaille  <akim.demaille@gmail.com>

CI: factor
	* .travis.yml (Clang 7 libc++ and ASAN part 2): Reuse bits from "Clang
	7 libc++ and ASAN part 1".

2019-06-09  Akim Demaille  <akim.demaille@gmail.com>

lr0: more debug traces
	* src/lr0.c (kernel_check): New.
	(new_itemsets, save_reductions): Add traces.

2019-06-09  Akim Demaille  <akim.demaille@gmail.com>

traces: add some colors
	This is an experiment.  Maybe more styles will be used (in which case
	a short-hand function will be useful), maybe it will be just reverted.
	* data/bison-default.css (.traces0): New.
	* src/lalr.c (lalr): Use it.

2019-06-09  Akim Demaille  <akim.demaille@gmail.com>

tests: make sure the default action properly works in C++
	See e3fdc370495ffdedadd6ac621e32e34a0e1a9de0: in C++ we generate
	explicitly the code for the default action instead of simply copying
	blindly the semantic value buffer.  This is important when copying
	raw memory is not enough, as exemplified by move-only types.

This is currently tested by examples/c++/variant.yy and variant-11.yy.
	But it is safer to also have a test in the main test suite.

* tests/local.at (AT_REQUIRE_CXX_STD): Fix.
	(AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Define/undefine
	AT_BISON_OPTIONS.
	* tests/c++.at (Default action): New.

2019-06-09  Akim Demaille  <akim.demaille@gmail.com>

tests: main: support -s and -p
	* tests/local.at (AT_MAIN_DEFINE(c), AT_MAIN_DEFINE(c++)): here.

2019-06-04  Akim Demaille  <akim.demaille@gmail.com>

tests: remove useless support of '.' in integers
	* tests/calc.at: here.
	* doc/bison.texi: Avoid uninitialized variables.

2019-05-29  Akim Demaille  <akim.demaille@gmail.com>

tests: refactor checks on sets
	It will be convenient to check sets elsewhere.

* tests/sets.at (AT_EXTRACT_SETS): Transform into...
	* tests/local.at (AT_SETS_CHECK): this.
	* tests/sets.at: Adjust.

2019-05-29  Akim Demaille  <akim.demaille@gmail.com>

update-test: some file names have dashes in them
	* build-aux/update-test (log): Rename as...
	(trace): this, to avoid clashes with the log variable.
	(getargs): Clarify the type of the arguments.

2019-05-26  Akim Demaille  <akim.demaille@gmail.com>

tests: take SHELL into account
	Reported by Dennis Clarke.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00053.html

* examples/local.mk, tests/local.mk: here.

2019-05-26  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update to get gnulib translations
	This update contains a fix needed for gnulib-po to work properly.
	https://lists.gnu.org/archive/html/bug-gnulib/2019-05/msg00146.html

2019-05-25  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify the purpose of symbol_type constructors
	Reported by Frank Heckenbach.
	http://lists.gnu.org/archive/html/bug-bison/2019-02/msg00006.html

* doc/bison.texi (Complete Symbols): Here.

2019-05-22  Akim Demaille  <akim.demaille@gmail.com>

thanks: fix an address

2019-05-22  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-05-22  Akim Demaille  <akim.demaille@gmail.com>

version 3.4.1
	* NEWS: Record release date.

2019-05-22  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2019-05-20  Akim Demaille  <akim.demaille@gmail.com>

CI: remove useless apt-get update
	The apt addons already ran it for us, it is not needed.

* .travis.yml: here.

2019-05-20  Akim Demaille  <akim.demaille@gmail.com>

c++: beware of to_string portability issues
	Reported by Bruno Haible.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00033.html

* m4/bison-cxx-std.m4 (_BISON_CXXSTD_11_snippet): Check it.

2019-05-20  Akim Demaille  <akim.demaille@gmail.com>

doc: avoid Texinfo portability issues
	Reported by Bruno Haible.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00024.html
	Fixed by Karl Berry.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00034.html

* doc/bison.texi: Don't specify the langage, rely on the default.
	Avoid blank pages.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

examples: don't run those that require f?lex when it's not available
	Reported by Bruno Haible.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00026.html

* configure.ac (FLEX_WORKS): New.
	* examples/c/lexcalc/local.mk, examples/c/reccalc/local.mk: Use it.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: don't crash when libtextstyle is installed
	Reported by neok m4700.
	https://lists.gnu.org/archive/html/bison-patches/2019-05/msg00025.html
	https://github.com/akimd/bison/pull/11

* src/complain.c (complain_init_color): style_file_prepare _needs_ a
	string as second argument.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

CI: avoid useless git costs
	The final gain is small: 2h2min instead 2h9min.  But that is still an
	improvement.

* .travis.yml (git.depth): Make the clone very shallow.
	(git.submodules): Don't clone gnulib in test jobs.
	(jobs.include.compile.script): Do it here.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

fix: copyable instead of copiable
	Reported by Frank Heckenbach.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00020.html

* data/skeletons/lalr1.cc, doc/bison.texi: here.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

version 3.4
	* NEWS: Record release date.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

fix: use copiable, not copyable
	Reported by Hans Åberg.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00017.html

* data/skeletons/lalr1.cc, doc/bison.texi: here.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update for 3.4

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

tests: adjust to GCC9 diagnostics with a margin
	* tests/synclines.at (_AT_SYNCLINES_COMPILE): Remove the margin.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: %pure-parser is obsolete
	Reported by Uxio Prego.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00029.html

* src/scan-gram.l, src/parse-gram.y (PERCENT_PURE_PARSER)
	(handle_pure_parser): New.
	Issue a deprecation/update notice for %pure-parser.
	* doc/bison.texi (Java Bison Interface): Don't mention %pure-parser.
	* tests/actions.at, tests/input.at: Adjust.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: clean up convention for colored diagnostics
	* data/diagnostics.css: Rename as...
	* data/bison-default.css: this.
	Add the GPL header.
	This is the convention followed by Bruno Haible in gettext.
	Adjust dependencies.
	* src/complain.c (complain_init_color): Use BISON_STYLE instead of
	BISON_DIAGNOSTICS_STYLE.

2019-05-19  Akim Demaille  <akim.demaille@gmail.com>

CI: use a pipeline: first build the tarball, then check it
	Build the tarball in one job, check it in many.
	Unfortunately no real gain in overall duration.
	With help from Clément Démoulins.

* .travis.yml: here.
	Remove all the tricks that were used to be able to boostrap on old
	distros.
	(before_install): Merge into 'script', because before_install applies
	to all the jobs, and we don't want to run it for the 'compile' job.

2019-05-18  Akim Demaille  <akim.demaille@gmail.com>

examples: fix srcdir/builddir issues
	* examples/d/local.mk, examples/java/local.mk: here.

2019-05-18  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	In preparation for Bison 3.4, revert to the version before this
	commit:

commit 03752516b21091cf3c4beea7e8b9bcad462d50ed
	    Author: John Darrington <john@darrington.wattle.id.au>
	    Date:   Sun May 12 00:42:36 2019 +0200

version-etc: Ease translation.

* lib/version-etc.c (version_etc_arn, emit_bug_reporting_address):
	    Move URLs and formatting newlines out of translatable string.

because it changes the messages for --version, translated in
	gnulib.po.  These changes are not yet available on the translation
	project, so we would have a regression in the set of translated
	strings.

2019-05-13  Akim Demaille  <akim.demaille@gmail.com>

build: do not use $< in plain rules
	It works only in implicit rules (or with GNU Make, but not with
	Solaris Make).

Reported by Bruno Haible.
	http://lists.gnu.org/archive/html/bug-bison/2019-05/msg00009.html
	Diagnosed thanks to Kiyoshi Kanazawa.

* examples/c/reccalc/local.mk, examples/d/local.mk,
	* examples/java/local.mk: Don't use $< in non implicit rules.

2019-05-12  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-05-12  Akim Demaille  <akim.demaille@gmail.com>

version 3.3.91
	* NEWS: Record release date.

2019-05-12  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2019-05-12  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-05-11  Akim Demaille  <akim.demaille@gmail.com>

style: remove incorrect comment
	* src/getargs.c: here.
	It's documented in getargs.h anyway.

2019-05-09  Akim Demaille  <akim.demaille@gmail.com>

doc: use colors for diagnostics in TeX too
	Thanks to Gavin Smith and Patrice Dumas.
	http://lists.gnu.org/archive/html/help-texinfo/2019-04/msg00015.html

* doc/bison.texi (@colorWarning, @colorError, @colorNotice)
	(@colorOff): Define for TeX and HTML.
	(@dwarning, @derror, @dnotice): Use them.

2019-05-08  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update to fix location tracking in UTF-8 on Solaris
	This update contains Bruno Haible's fix for the location tracking
	issue reported by Kiyoshi Kanazawa.

https://lists.gnu.org/archive/html/bug-gnulib/2019-05/msg00020.html
	https://lists.gnu.org/archive/html/bug-bison/2019-04/msg00020.html

2019-05-08  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: rename --style=debug as --color=debug
	It is more consistent with --color=html, --color=test, etc.

* src/getargs.h, src/getargs.c (style_debug): Rename as...
	(color_debug): this.
	(getargs_colors): Rename --style=debug as --color=debug.
	Adjust dependencies.

2019-05-08  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: support --color=html
	Based on a message from Bruno Haible.
	https://git.savannah.gnu.org/gitweb/?p=gettext.git;a=commitdiff;h=fe18e92743b7226791a5f28d7c786941a1bf8cc9

This does not generate proper HTML: special characters are not escaped
	for instance.  This is a hidden feature meant for Bison developers,
	not end users.

* src/complain.c (complain_init_color): Support --color=html.

2019-05-08  Akim Demaille  <akim.demaille@gmail.com>

tests: use %empty instead of comments
	* tests/c++.at, tests/glr-regression.at: here.

2019-05-08  Akim Demaille  <akim.demaille@gmail.com>

fixits: sort them before applying them
	An experimental commit introduced a fix-it hint that changes comments
	such as "/* empty */" into %empty.  But in some case, because
	diagnostics are not necessarily emitted in order, the fixits also come
	in disorder, which must never happen, as the fixes are installed in
	one pass.

* src/fixits.c (fixits_register): Insert them in order.

2019-05-04  Akim Demaille  <akim.demaille@gmail.com>

style: use warning_is_enabled instead of duplicating it
	* src/complain.c (deprecated_directive): Here.

2019-05-03  Akim Demaille  <akim.demaille@gmail.com>

fixits: be sure to preserve the action when adding %empty
	Currently we remove the rhs to install %empty instead.

* src/reader.c (grammar_rule_check_and_complete): Insert the missing
	%empty in front of the rhs, not in replacement thereof.
	* tests/actions.at (Add missing %empty): Check that.

2019-05-03  Akim Demaille  <akim.demaille@gmail.com>

tests: don't duplicate the portability prologue
	* tests/actions.at, tests/input.at: Don't repeat the prologue, skip it.
	* tests/diagnostics.at, tests/local.at: Comment changes.

2019-05-03  Akim Demaille  <akim.demaille@gmail.com>

style: use consistently *_loc for locations
	Some members are called foo_location, others are foo_loc.  Stick to
	the latter.

* src/gram.h, src/location.h, src/location.c, src/output.c,
	* src/parse-gram.y, src/reader.h, src/reader.c, src/reduce.c,
	* src/scan-gram.l, src/symlist.h, src/symlist.c, src/symtab.h,
	* src/symtab.c:
	Use _loc consistently, not _location.

2019-05-03  Akim Demaille  <akim.demaille@gmail.com>

style: clarify the use of symbol_lists' locations
	symbol_list features a 'location' and a 'sym_loc' member.  The former
	is expected to be set only for symbol_lists that denote a symbol (not
	a type name), and the latter should only denote the location of the
	symbol/type name.  Yet both are set, and the name "location" is too
	unprecise.

* src/symlist.h, src/symlist.c (symbol_list::location): Rename as
	rhs_loc for clarity.  Move it to the "section" of data valid only
	for rules.
	* src/reader.c, src/scan-code.l: Adjust.

2019-05-03  Akim Demaille  <akim.demaille@gmail.com>

maint: update gnulib-po/.gitignore

2019-04-29  Akim Demaille  <akim.demaille@gmail.com>

tests: don't require a D compiler
	Reported by Kiyoshi Kanazawa.
	http://lists.gnu.org/archive/html/bug-bison/2019-04/msg00018.html

* tests/atlocal.in (BISON_DC_WORKS): New.
	* tests/local.at (AT_COMPILE_D): Use it.

2019-04-29  Akim Demaille  <akim.demaille@gmail.com>

doc: use svg instead of png
	* doc/bison.texi, doc/local.mk: here.

2019-04-29  Akim Demaille  <akim.demaille@gmail.com>

doc: use colors
	* doc/bison.texi (dwarning, derror, dnotice): New.
	Use them in the diagnostics.
	* doc/local.mk (AM_MAKEINFOFLAGS): Pass customization variables.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

version 3.3.90
	* NEWS: Record release date.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

package: add missing CLEANFILES
	* examples: here.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

build: don't generate the graph reports
	Revert "build: also generate the graph
	reports" (4ec413da32760defe1bf382c048d1d2f67e0b58a).  The problem is
	Automake's ylwrap which does not rename y.dot with the appropriate
	name.  We should completely stop using Automake's support for Yacc,
	which is not something I will do right now.  So step back.

* Makefile.am (AM_YFLAGS_WITH_LINES): Don't pass --graph.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

package: don't regen the parser during dist if unneeded
	* Makefile.am (gen-synclines): New.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

package: don't ship the sources generated from the parser
	Because some of our examples use

%C%_reccalc_SOURCES = %D%/parse.y

Automake ships parse.y and parse.c, and possibly parse.h when it
	"understands" that there is one.  This is not what we want: ship only
	parser.y.  Yet we still want to use Automake to compile the sources
	from parser.y.  The easiest seems to use

nodist_%C%_reccalc_SOURCES = %D%/parse.y

together with

dist_reccalc_DATA = %D%/parse.y %D%/scan.l %D%/Makefile %D%/README.md

which guarantees that parse.y is indeed shipped.

* examples/c/calc/local.mk, examples/c/lexcalc/local.mk,
	* examples/c/reccalc/local.mk: Always use nodist_*SOURCES for parsers,
	let the dist_*_DATA rules do their job.

2019-04-28  Akim Demaille  <akim.demaille@gmail.com>

package: various fixes for syntax-check
	* cfg.mk: Disable checks where needed (e.g., we do want to check the
	behavior with tabs).
	(sc_at_parser_check): Remove.  Unfortunately since
	a11c144609255bc6e42c2aff83548e91cbd05425 we no longer use the './'
	prefix to run programs in the current directory.  That was so that we
	could run Java programs like the other, although they are no run with
	the `./` prefix (see 967a59d2c08a33f24708450561e2f8010b604523).
	As a consequence this sc check no longer makes sense.
	However, since now AT_PARSER_CHECK passes the `./` prefix itself, this
	sc-check was superfluous.
	* examples/c/reccalc/scan.l: Use memcpy, not strncpy.
	* src/ielr.c, src/reader.c: Obfuscate "lr(0)" so that the sc-check for
	"space before paren" does not fire.
	* tests/diagnostics.at: Avoid space-tab, use tab-tab.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

doc: clarify -fsyntax-error
	* NEWS, doc/bison.texi: here.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

traces: use colors for the semantic values
	This makes reading the trace slightly easier.  It would be very nice
	to highlight the "big steps", especially reductions.  But this is a
	private experiment: do not use it.

* data/diagnostics.css (value): New.
	* src/parse-gram.y: Use no delimiters and no c quotation for strings
	to facilitate debugging.
	(tron, troff, TRACE): New.
	Not very elegant, but until there is support for printf-formats in
	libtextstyle, it shall be enough.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: give m4 precise locations
	Currently we pass only the columns based on the screen-width, which is
	important for the carets.  But we don't pass the bytes-based columns,
	which is important for the colors.  Pass both.

* src/muscle-tab.c (muscle_boundary_grow): Also pass the byte-based column.
	* src/location.c (location_caret): Clarify.
	(boundary_set_from_string): Adjust to the new format.
	* tests/diagnostics.at (Tabulations and multibyte characters from M4): New.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix locations coming from M4
	Locations issued from M4 need the byte-based column for the
	diagnostics to work properly.  Currently they were unassigned, which
	typically resulted in partially non-colored diagnostics.

* src/location.c (boundary_set_from_string): Fix the parsed location.
	* src/muscle-tab.c (muscle_percent_define_default): Set the byte values.
	* tests/diagnostics.at (Locations from M4): New.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: show locations in full when debugging
	This is meant for developers, not end users, that's why I attached it
	to --trace.

* src/getargs.h, src/getargs.c (trace_locations): New.
	* src/location.c (location_print): Use it.

2019-04-27  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use flush, not fflush
	* src/complain.c: here.

2019-04-25  Akim Demaille  <akim.demaille@gmail.com>

build: use gettext-h
	We were using the gnulib's gettext module with tricks in
	bootstrap.conf to avoid useless files.  Instead, use gnulib's
	gettext-h module.

* .travis.yml: Force Gettext 0.18.3 on Trusty.
	* bootstrap.conf: Use gettext-h instead of gettext.
	(excluded_files): Remove.
	* configure.ac (AM_GNU_GETTEXT_VERSION): Bump to 0.19.

2019-04-25  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2019-04-25  Akim Demaille  <akim.demaille@gmail.com>

api.location.type: support it in C
	Reported by Balázs Scheidler.

* data/skeletons/c.m4 (b4_location_type_define): Use api.location.type
	if defined.
	* doc/bison.texi: Document it.
	* tests/local.at (AT_C_IF, AT_LANG_CASE): New.
	Support Span in C.
	* tests/calc.at (Span): Convert it to be usable in C and C++.
	Check api.location.type with yacc.c and glr.c.

2019-04-24  Akim Demaille  <akim.demaille@gmail.com>

updates: insert/remove %empty
	* src/reader.c (grammar_rule_check_and_complete): Generate fixits for
	adding/removing %empty.
	* tests/actions.at, tests/diagnostics.at, tests/existing.at: Adjust.

2019-04-24  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-04-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: better rule locations
	The "identifier and colon" of a rule is implemented as a single token,
	but whose location is only that of the identifier (so that messages
	about the lhs of a rule are accurate).  When reducing empty rules, the
	default location is the single point location on the end of the
	previous symbol.  As a consequence, when Bison parses a grammar, the
	location of the right-hand side of an empty rule is based on the
	lhs, *independently of the position of the colon*.  And the colon can
	be way farther, separated by comments, white spaces, including empty
	lines.

As a result, some messages look really bad.  For instance:

$ cat foo.y
	    %%
	    foo     : /* empty */
	    bar
	    : /* empty */

gives

$ bison -Wall foo.y
	    foo.y:2.4: warning: empty rule without %empty [-Wempty-rule]
	        2 | foo     : /* empty */
	          |    ^
	    foo.y:3.4: warning: empty rule without %empty [-Wempty-rule]
	        3 | bar
	          |    ^

The carets are not at the right column, not even the right line.

This commit passes the colon "again" after the "id colon" token, which
	gives more accurate locations for these messages:

$ bison -Wall foo.y
	    foo.y:2.10: warning: empty rule without %empty [-Wempty-rule]
	        2 | foo     : /* empty */
	          |          ^
	    foo.y:4.2: warning: empty rule without %empty [-Wempty-rule]
	        4 | : /* empty */
	          |  ^

* src/scan-gram.l (SC_AFTER_IDENTIFIER): Rollback the colon, so that
	we scan it again afterwards.
	(INITIAL): Scan colons.
	* src/parse-gram.y (COLON): New.
	(rules): Parse the colon after the rule's id_colon (and possible
	named reference).
	* tests/actions.at, tests/conflicts.at, tests/diagnostics.at,
	* tests/existing.at: Adjust.

2019-04-24  Akim Demaille  <akim.demaille@gmail.com>

fixits: track byte-columns, not character-columns
	Because the fix-its were ready the character-based columns, but were
	applied on byte-based columns, the result with multibyte characters or
	tabs could be "interesting".  For instance

%fixed-output_files
	            %fixed_output-files
	    %fixed-output-files
	    %define api.prefix {foo}
	    %no-default-prec

would give

%fixed-%fixed-output-files  %fixed_output-files
	    %fixed-orefix= "foo"
	    o_default-prec

* src/fixits.c (fixit_print, fixits_run): Work on byte-base columns.
	* tests/input.at: Check it.

2019-04-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: expose a means to know whether a warning is enabled
	* src/complain.h, src/complain.c (warning_is_enabled): New.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

gnulib: let it use its own PO domain
	See
	https://www.gnu.org/software/gnulib/manual/html_node/Localization.html.

* bootstrap.conf: Create gnulib-po.
	* Makefile.am, configure.ac: Use it.
	* po/POTFILES.in: Remove files now in gnulib.
	* src/main.c: Open the bison-gnulib domain.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: don't try to quote special files
	Based on a report by Todd Freed.
	http://lists.gnu.org/archive/html/bug-bison/2019-04/msg00000.html
	See also https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90034

* src/location.c (caret_info): Also track the file name.
	(location_caret): Don't quote special files.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: document the change of format
	* doc/bison.texiL Adjust output.
	Also, Graphviz has no uppercsae V.
	* NEWS: Explain the format change.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: copy GCC9's format
	Currently, when we quote the source file, we indent it with one space,
	and preserve tabulations, so there is a discrepancy and the visual
	rendering is bad.  One way out is to indent with a tab instead of a
	space, but then this space can be used for more information.  This is
	what GCC9 does.  Let's play copy cats.

See
	https://lists.gnu.org/archive/html/bison-patches/2019-04/msg00025.html
	https://developers.redhat.com/blog/2019/03/08/usability-improvements-in-gcc-9/
	https://gcc.gnu.org/onlinedocs/gccint/Guidelines-for-Diagnostics.html#Guidelines-for-Diagnostics

* src/location.c (location_caret): Prefix quoted lines with the line
	number and a pipe, fitting 8 columns.

* tests/actions.at, tests/c++.at, tests/conflicts.at,
	* tests/diagnostics.at, tests/input.at, tests/java.at,
	* tests/named-refs.at, tests/reduce.at, tests/regression.at,
	* tests/sets.at: Adjust expectations.
	Partly by "./build-aux/update-test tests/testsuite.dir/*/testsuite.log"
	repeatedly, and partly by hand.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix the handling of multibyte characters
	This is a pity: efforts were invested in computing correctly the
	number of screen columns consumed by multibyte characters, but the
	routines that do that were fed by single-byte inputs...

As a consequence Bison never displayed correctly locations when there
	are multibyte characters.

* src/scan-gram.l (mbchar): New.
	Use it instead of . in the catch-all clause.
	* tests/diagnostics.at (Tabulations): Enhance into...
	(Tabulations and multibyte characters): this.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: check the handling of tabulations
	* tests/diagnostics.at (Tabulations): here.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix styling issues
	Single point locations (equal boundaries) are troublesome, and we were
	incorrectly ending the style in their case.  Which results in an abort
	in libtextstyle.

There is also a confusion between columns as displayed on the
	screen (which take into account multibyte characters and tabulations),
	and the number of bytes.  Counting the screen-column
	incrementally (character by character) is uneasy (because of multibyte
	characters), and I don't want to maintain a buffer of the current line
	when displaying the diagnostic.  So I believe the simplest solution is
	to track the byte number in addition to the screen column.

* src/location.h, src/location.c (boundary): Add the byte-column.
	Adjust dependencies.
	* src/getargs.c, src/scan-gram.l: Adjust.
	* tests/diagnostics.at: Check zero-width locations.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: check the styling
	Enable checking of styles even when libtextstyle is not installed.

* src/getargs.h, src/getargs.c (style_debug): New.
	(getargs_colors): Set it when --style=debug.
	* src/complain.c (begin_use_class, end_use_class): Use it.
	* tests/diagnostics.at: New.

2019-04-23  Akim Demaille  <akim.demaille@gmail.com>

TODO: update
	Let's prepare 3.4 with more or less what we have.  Schedule some
	features for 3.5 and 3.6.  Remove obsolete stuff.

2019-04-19  Akim Demaille  <akim.demaille@gmail.com>

doc: sort the warning categories
	* doc/bison.texi, src/getargs.c: here.

2019-04-19  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* tests/actions.at, tests/calc.at, tests/input.at: here.

2019-04-19  Akim Demaille  <akim.demaille@gmail.com>

graphviz: move constant computation out of a loop
	* src/graphviz.c (output_red): here.

2019-04-18  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix memory leak in libtextstyle
	* src/complain.h, src/complain.c (complain_free): New.
	* src/main.c: Use it.

2019-04-17  Akim Demaille  <akim.demaille@gmail.com>

tests: remove useless feature
	* tests/calc.at (read_signed_integer): Rename as...
	(read_integer): this.
	We never read signs here.

2019-04-14  Akim Demaille  <akim.demaille@gmail.com>

traces: make closure() less verbose
	* src/getargs.h, src/getargs.c (trace_closure): New.
	* src/closure.c (closure): Use it.

2019-04-14  Akim Demaille  <akim.demaille@gmail.com>

build: also generate the graph reports
	* Makefile.am (AM_YFLAGS_WITH_LINES): here.

2019-04-12  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: minor style change
	* data/skeletons/yacc.c: To improve consistency with other similar
	pieces of code.

2019-04-12  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in lalr.c
	* src/lalr.c (initialize_goto_follows): here.

2019-04-12  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/closure.h, src/closure.c, src/lalr.c: here.

2019-04-12  Akim Demaille  <akim.demaille@gmail.com>

traces: improve logs
	* src/lalr.c: Move logs to a better place to understand the chronology
	of events.
	* src/symlist.c (symbol_list_syms_print): Don't dump core on type
	elements.

2019-04-07  Akim Demaille  <akim.demaille@gmail.com>

doc: minor fixes
	* doc/bison.texi: Use consistently $ and @kbd in shell examples.
	Prefer sticking to English words: output and file instead of outfile
	and infile.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

bison: use no-lines
	The 'regen' commit in Bison's history are a nuisance.  They are
	especially big because of the #lines.  Let's generate our parse
	without these lines in the repository, but generate them in the
	tarball.

* Makefile.am (AM_YFLAGS_WITH_LINES): New.
	(AM_YFLAGS): Use it.
	(dist-hook): Regenerate the parser with #lines.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

no-lines: avoid leaving an empty line instead of the syncline
	Currently, with --no-lines, instead of "#line file line\n", we emit
	"\n".  Let's emit nothing.

* data/skeletons/bison.m4 (b4_syncline): Emit at end-of-line when enabled.
	* data/skeletons/bison.m4, data/skeletons/c.m4, data/skeletons/glr.cc,
	* data/skeletons/lalr1.cc, src/output.c: Use dnl after b4_syncline to
	avoid spurious empty lines.

* tests/synclines.at (Sync Lines): Make sure that --no-lines is like
	grep -v #line.
	* tests/calc.at: Make sure that a rich grammar file behaves properly
	with %no-lines.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

java: use full locations for diagnostics about destructors
	Currently we use the syncline to report errors about a symbol's
	destructor/printer.  This is not accurate (only file and line), and
	this is incorrect: the file name is double quotes (a recent change,
	needed to make sure we escape properly double quotes in it).  And
	worst of all: with --no-line, b4_syncline expands to nothing.

Rather, push the locations into the backend, and use them.

* src/muscle-tab.h, src/muscle-tab.c (muscle_location_grow): Make it
	public.
	* src/output.c (prepare_symbol_definitions): Use it to pubish the
	location of the printer and destructor.
	* data/skeletons/lalr1.java: Use complain_at instead of complain.
	* tests/java.at (Java invalid directives): Adjust expectations.
	* data/skeletons/bison.m4 (b4_symbol_action_location): Remove.
	We should not use b4_syncline this way.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

java: prefer errors to fatal errors
	Fatal errors are inconvenient, and should be reserved to cases where
	we cannot continue.  Here, it could even be warnings actually: these
	directives will simply be ignored.

* data/skeletons/lalr1.java: Prefer error (b4_complain) to fatal
	errors (b4_fatal).
	* tests/java.at (Java invalid directives): New.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

tests: formatting changes
	* tests/javapush.at: here.

2019-04-03  Akim Demaille  <akim.demaille@gmail.com>

lalr: offer more flexibility in debugging routines
	* src/state.h, src/state.c (state_transitions_print): New, extracted
	from...
	(state_transitions_set): here.

2019-03-31  Akim Demaille  <akim.demaille@gmail.com>

lalr: don't overbook memory
	I never understood why we book ngotos+1 slots for relations between
	gotos: there are at most ngotos images, not ngotos+1 (and "includes"
	does have cases where a goto is in relation with itself, so it's not
	ngotos-1).

Maybe bbf37f2534a8e5a6b4e28047f0a10903e6dc73f9 explains the +1: a bug
	left us register a goto several times on occasion, and the +1 might
	have been a means to avoid this problem in most cases.  Now that this
	bug is addressed, we should no longer overbook memory, if only for the
	clarity of the code ("why ngotos+1 instead of ngotos?").

* src/lalr.c: A goto has at most ngotos images, not ngotos+1.
	While at it, avoid useless repeated call to map_goto introduced in
	bbf37f2534a8e5a6b4e28047f0a10903e6dc73f9.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

lalr: show lookback for debug
	* src/lalr.c (lookback_print): New.
	(build_relations): Use it.
	Also show edges.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: don't crash when declaring the error token as an nterm
	Reported by wcventure.
	http://lists.gnu.org/archive/html/bug-bison/2019-03/msg00008.html

* src/symtab.c (complain_class_redeclared): Don't print empty
	locations.
	There can only be empty locations for predefined symbols.  And the
	only symbol that is lexically available is the error token.  So this
	appears to be the only possible way to have an error involving an
	empty location.
	* tests/input.at (Symbol class redefinition): Check it.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

lalr: fix segmentation violation
	The "includes" relation [DeRemer 1982] is between gotos, so of course,
	for a given goto, there cannot be more that ngotos (number of gotos)
	images.  But we manipulate the set of images of a goto as a list,
	without checking that an image was not already introduced.  So we can
	"register" way more images than ngotos, leading to a crash (heap
	buffer overflow).

Reported by wcventure.
	http://lists.gnu.org/archive/html/bug-bison/2019-03/msg00007.html

For the records, this bug is present in the first committed version of
	Bison.

* src/lalr.c (build_relations): Don't insert the same goto several
	times.
	* tests/sets.at (Build Relations): New.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

state: more debug traces
	* src/state.c (state_transitions_set): Show the transitions.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

style: rename variables for consistency
	* src/lalr.c: Use trans for transitions, and reds for reductions, as
	elsewhere in the code.
	* src/state.h: Comment changes.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

gram: fix and improve log message
	It seems that not many people read these logs: the error was
	introduced in 2001 (3067fbef531832df1e43bbd28787655808361eed),

* src/gram.c (grammar_dump): Fix the headers of the table: remove
	duplicate display of "Ritem Range".
	While at it, remove duplicate display of the rule number (and remove
	an incorrect comment about it: these numbers _are_ equal).
	* tests/sets.at (Reduced Grammar): Use useless rule, nterm and token
	in the example.

2019-03-30  Akim Demaille  <akim.demaille@gmail.com>

tests: add a tool for mass updates
	When we update some output format, too many adjustements must be made
	by hand.  This script updates most tests based on the actual output
	made during the tests.

* build-aux/update-test: New.

2019-03-25  Akim Demaille  <akim.demaille@gmail.com>

style: remove now useless _GL_UNUSED
	* src/getargs.c (getargs_colors): Here.
	Useless since 4d34b06fb3a38eb050439084476a6b3e292c5680.

2019-03-24  Theophile Ranquet  <ranquet@lrde.epita.fr>

tables: use bitsets for a performance boost
	Suggested by Yuri at
	<http://lists.gnu.org/archive/html/bison-patches/2012-01/msg00000.html>.

The improvement is marginal for most grammars, but notable for large
	grammars (e.g., PosgreSQL's postgre.y), and very large for the
	sample.y grammar submitted by Yuri in
	http://lists.gnu.org/archive/html/bison-patches/2012-01/msg00012.html.
	Measured with --trace=time -fsyntax-only.

parser action tables    postgre.y     sample.y
	Before                 0,129 (44%)  37,095 (99%)
	After                  0,117 (42%)   5,046 (93%)

* src/tables.c (pos): Replace this set of integer coded as an unsorted
	array or integers with...
	(pos_set): this bitset.

2019-03-24  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: don't suggest api.header.include when --defines is not used
	See 4e19ab9fcd28c9361ff08f46e5e353effb0a0520: the suggestion to
	include the header file should not be emitted when the header is not
	generated.

* data/skeletons/yacc.c: Here.

2019-03-24  Akim Demaille  <akim.demaille@gmail.com>

reader: clarify variable names
	* src/reader.c (grammar_rule_check_and_complete): When 'p' and 'lhs'
	are aliases, prefer the latter, for clarity and consistency.
	(grammar_current_rule_begin): Avoid 'p', current_rule suffices.
	* src/gram.h, src/gram.c: Comment changes.

ptdr#	calc.tab.c

2019-03-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style changes
	* src/location.c (location_caret): Clarify a bit.

2019-03-24  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use gnulib's libtextstyle-optional
	Bruno Haible just added a default implementation of libtextstyle's
	interface when the library is not available.
	https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00025.html

* gnulib: Update.
	* bootstrap.conf: Replace libtextstyle with libtextstyle-optional.
	* src/complain.c, src/getargs.c: Remove now useless cpp guards.

2019-03-23  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: fix handling of style in limit cases
	* src/location.c (location_caret): Beware of the cases where the start
	and end columns are the same, or when the location is multilines.

2019-03-23  Akim Demaille  <akim.demaille@gmail.com>

warnings: don't use _Noreturn with G++ 4.7 in C++98 mode
	The timevar and bitset modules now use the c99 module which causes
	$CXX to now include -std=gnu++11 when possible.  Unfortunately, G++
	4.7 does not implement [[noreturn]] in C++11 mode, so our tests of
	glr.cc (which uses _Noreturn) fail with

input.cc:954:1: error: expected unqualified-id before '[' token

right before [[noreturn]].  4.8 works fine.

* data/skeletons/c.m4 (b4_attribute_define): Do not use [[noreturn]]
	with GCC 4.7.

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

d: tests: use more a natural approach for the scanner
	See f8408562f8439654261418406296e4108d2a995f.

* tests/calc.at: Stop imitating the C API.
	Prepare more tests to run in the future.
	%verbose works as expected (what a surprise, it's unrelated to the
	skeleton...).

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

style: rename spec_defines_file as spec_header_file
	The variable spec_defines_file denotes the name of the generated
	header.  Its name is derived from --defines/%defines, whose name in
	turn is derived from the fact that the header, in Yacc, contained the

Not only does the header now contain a lot more than just the token
	definitions, but we no longer even generate macros, but an enum...

Let's modernize our vocabulary.

* src/files.h, src/files.c (spec_defines_file): Rename as...
	(spec_header_file): this.

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: provide a means to include the header in the implementation
	Currently when --defines is used, we generate a header, and paste an
	exact copy of it into the generated parser implementation file.  Let's
	provide a means to #include it instead.

We don't do it by default because of the Autotools' ylwrap.  This
	program wraps invocations of yacc (that uses a fixed output name:
	y.tab.c, y.tab.h, y.output) to support a more modern naming
	scheme (dir/foo.y -> dir/foo.tab.c, dir/foo.tab.h, etc.).  It does
	that by renaming the generated files, and then by running sed to
	propagate these renamings inside the files themselves.

Unfortunately Automake's Makefiles uses Bison as if it were Yacc (with
	--yacc or with -o y.tab.c) and invoke bison via ylwrap.  As a
	consequence, as far as Bison is concerned, the output files are
	y.tab.c and y.tab.h, so it emits '#include "y.tab.h"'.  So far, so
	good.  But now ylwrap processes this '#include "y.tab.h"' into
	'#include "dir/foo.tab.h"', which is not guaranteed to always work.

So, let's do the Right Thing when the output file is not y.tab.c, in
	which case the user should %define api.header.include.  Binding this
	behavior to --yacc is tempting, but we recently told people to stop
	using --yacc (as it also enables the Yacc warnings), but rather to use
	-o y.tab.c.

Yacc.c is the only skeleton concerned: all the others do include their
	header.

* data/skeletons/yacc.c (b4_header_include_if): New.
	(api.header.include): Provide a default value when the output is not
	y.tab.c.
	* src/parse-gram.y (api.header.include): Define.

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

d: don't link against LIBS
	* tests/local.at (AT_COMPILE_D): Don't pass LIBS, dmd does not like
	being given -lintl.

2019-03-17  Akim Demaille  <akim.demaille@gmail.com>

address warnings from GCC's UB sanitizer
	Running with CC='gcc-mp-8 -fsanitize=undefined' revealed Undefined
	Behaviors.
	https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00008.html

* src/state.c (errs_new): Don't call memcpy with NULL as source.
	* src/location.c (add_column_width): Don't assume that the column
	argument is nonnegative: the scanner sometimes "backtracks" (e.g., see
	ROLLBACK_CURRENT_TOKEN and DEPRECATED) in which case we can have
	negative column numbers (temporarily).
	Found in test 3 (Invalid inputs).

2019-03-16  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: use libtextstyle for colored output
	Bruno Haible released libtextstyle, a library for colored output based
	on CSS.  Let's use it to generate colored diagnostics, provided
	libtextstyle is available.

See
	https://lists.gnu.org/archive/html/bug-gnulib/2019-01/msg00176.html
	https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00073.html
	https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00084.html
	https://lists.gnu.org/archive/html/bison-patches/2019-03/msg00007.html

* bootstrap.conf (gnulib_modules): Use libtextstyle when possible.
	* data/diagnostics.css: New.
	* src/complain.c (begin_use_class, end_use_class, flush)
	(severity_style, complain_init_color): New.
	Use them.
	* src/getargs.c (getargs_colors): New.
	(getargs): Use it.
	Skip --color and --style.
	* src/location.h, src/location.c (location_print): Use a style.

* tests/bison.in: Force --color=yes when stderr is a tty.
	* tests/local.at: Disable colors during the test suite.
	* tests/input.at: Adjust expectations to the extra options passed on
	the command line.

2019-03-16  Akim Demaille  <akim.demaille@gmail.com>

style: clean up complain.c
	* src/complain.c (severity_prefix): New.
	(error_message): Take the severity as argument, instead of the prefix.

2019-03-16  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: emit the header before the implementation file
	* data/skeletons/yacc.c: here.
	This is more logical for the time stamps, but it's also required by
	following patches: the shared declarations are also in charge of
	handling api.value.type=union.  So far, they are run in the
	implementation file in both cases (with or without header).  But if we
	run them only in the header, then the implementation file is emited
	with incorrect support for api.value.type=union.
	Arguably we should not have such dependencies.  This is because we
	have side-effects in our backend (redefining the symbols' type and
	type_tag).  In the future we should find a better solution for this,
	without sacrificing the independence of the backend from bison
	itself (i.e., I don't think we should handle api.value.type=union in
	bison, leave it to m4).

2019-03-16  Akim Demaille  <akim.demaille@gmail.com>

simplify the generated #line
	Currently we generate things like:

#line 683 "src/parse-gram.y" /* yacc.c:316  */

The first part is of course very important: compilers point the users
	to their grammar file rather than into the generated parser.  The
	second part points to the place in the skeletons that generated this
	piece of code.

This dependency on the Bison skeletons generates lots of useless 'git
	diff'.  This location is useless for the regular user (who does not
	care about the skeletons) and is actually not useful for Bison
	developpers too (I never used this to locate the code in skeletons
	that generated output).  So disable it completely.  If someone thinks
	this was actually useful, a %define variable should be provided to
	control the level of verbosity of '#line', in replacement of
	--no-lines.

So now, generate:

#line 683 "src/parse-gram.y"

* data/skeletons/bison.m4 (b4_sync_end): Emit nothing.

2019-03-13  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-03-13  Akim Demaille  <akim.demaille@gmail.com>

tests: remove duplicates
	* tests/regression.at (Invalid inputs, Invalid inputs with {}):
	Remove, there are exact copies of them in input.at.

2019-03-02  Akim Demaille  <akim.demaille@gmail.com>

d: simplify the API to build the scanner of the example
	* examples/d/calc.y (calcLexer): Add an overload for File.
	Use it.

2019-03-01  H. S. Teoh  <hsteoh@quickfur.ath.cx>

d: modernize the scanner of the example
	https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00121.html

* examples/d/calc.y (CalcLexer): Stop shoehorning C's API into D: use
	a range based approach in the scanner, rather than some imitation of
	getc/ungetc.
	(main): Adjust.

2019-03-01  Akim Demaille  <akim.demaille@gmail.com>

d: tests: use fewer global variables
	* tests/calc.at: Move 'input' into the scanner.

2019-02-28  Akim Demaille  <akim.demaille@gmail.com>

lalr: clarify the count of lookaheads
	* src/lalr.c (state_lookahead_tokens_count): Remove wierd `+=` that is
	actually an `=`.

2019-02-28  Akim Demaille  <akim.demaille@gmail.com>

lalr: clarify the API
	* src/state.h, src/state.c (state_reduction_find): Clarify.
	Die on errors.
	* src/lalr.c (goto_list_new): New.
	Use it.

2019-02-28  Akim Demaille  <akim.demaille@gmail.com>

lalr: improve traces
	* src/lalr.c (follows_print): Just print the symbol tag.
	Take and print a title.
	Indent the output.
	Use it to print the various steps of the computation.
	(lookahead_tokens_print): Fix a lie: the number displayed is not the
	number of tokens.
	Don't display states that don't even have reductions.

2019-02-27  Akim Demaille  <akim.demaille@gmail.com>

lalr: print the 'reads' relation
	* src/relation.h, src/relation.c (relation_print): Accept and use a
	title.
	Don't print empty rows.
	Indent the output.
	Adjust dependencies.
	* src/lalr.c (initialize_goto_follows): Print 'reads' in traces.

2019-02-27  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/lr0.c: here.

2019-02-26  Akim Demaille  <akim.demaille@gmail.com>

dlang: initial changes to run the calc tests on it
	* configure.ac (DCFLAGS): Define.
	* tests/atlocal.in: Receive it.
	* data/skeletons/d.m4 (api.parser.class): Remove spurious YY.
	* data/skeletons/lalr1.d (yylex): Return an int instead of a
	YYTokenType, so that we can use characters as tokens.
	* examples/d/calc.y: Adjust.
	* tests/local.at: Initial support for D.
	(AT_D_IF, AT_DATA_GRAMMAR(D), AT_YYERROR_DECLARE(d))
	(AT_YYERROR_DECLARE_EXTERN(d), AT_YYERROR_DEFINE(d))
	(AT_MAIN_DEFINE(d), AT_COMPILE_D, AT_LANG_COMPILE(d), AT_LANG_EXT(d)):
	New.
	* tests/calc.at: Initial support for D.
	* tests/headers.at

2019-02-26  Akim Demaille  <akim.demaille@gmail.com>

d: improve the example
	* examples/d/calc.y: Exit with failure on errors.
	Remove useless operators (=, !) meant for the test suite.
	Add unary + for symmetry.
	* examples/d/calc.test: Adjust expectations.

2019-02-26  Akim Demaille  <akim.demaille@gmail.com>

tests: style changes
	* tests/local.at AT_YYERROR_DEFINE(java): Use more consistent names.

2019-02-25  Akim Demaille  <akim.demaille@gmail.com>

style: eliminate useless indirection
	* src/relation.h, src/relation.c (relation_digraph): Don't take the
	biteetv as a pointer, it is already a pointer (as it's an array).

2019-02-25  Akim Demaille  <akim.demaille@gmail.com>

style: rename function for clarity
	Commit db34f7988941444bdc5f2b6adcf7fb83648f9a18 renames the variable F
	as goto_follows, but forgot to rename this function.

* src/lalr.c (initialize_F): Rename as...
	(initialize_goto_follows): this.

2019-02-25  Akim Demaille  <akim.demaille@gmail.com>

lalr: more debug traces
	I need to be able to read includes and goto_follows.

* src/relation.h, src/relation.c (relation_print): Provide a means to
	pretty-print the nodes of the relation.
	* src/lalr.c (goto_print, follows_print): New.
	(set_goto_map): Use goto_print.
	(build_relations): Show INCLUDES.
	(compute_FOLLOWS): Rename as...
	(compute_follows): this.
	Show FOLLOWS.

2019-02-24  Akim Demaille  <akim.demaille@gmail.com>

style: minor changes
	* examples/c/calc/calc.y, src/lalr.c: Reduce scope.
	* src/gram.c: Prefer < to >.

2019-02-24  Akim Demaille  <akim.demaille@gmail.com>

style: clarify the computation of the lookback edges
	* src/lalr.c (build_relations): Reduce the scopes.
	Instead of keeping rp alive in two different loops, clarify the second
	one by having an index on the path we traverse (i.e., use that index
	to compute the source state _and_ the symbol that labels the
	transition).
	This allows to turn an obscure 'while'-loop in a clearer (IMHO)
	'for'-loop.  We also consume more variables (by introducing p instead
	of making more side effects on length), but we're in 2019, I don't
	think this matters.  What does matter is that (IMHO again), this is
	now clearer.
	Also, use clearer names.

2019-02-24  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in tables.c
	* src/tables.c: here.
	* src/lalr.c: Prefer < to >.

2019-02-24  Akim Demaille  <akim.demaille@gmail.com>

d: formatting changes
	* data/skeletons/d.m4, data/skeletons/lalr1.d: Avoid trailing spaces.

2019-02-23  Akim Demaille  <akim.demaille@gmail.com>

examples: remove stray examples
	* examples/c/reentrant-calc: Remove.
	I did not mean to include this example, it was replaced by
	examples/c/reccalc.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: factor the execution of Java parsers
	* tests/local.at (AT_MAIN_DEFINE(java)): Exit failure on failure.
	(AT_PARSER_CHECK): If in Java, run AT_JAVA_PARSER_CHECK.
	* tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Simplify.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: fix a Java tests
	* tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Fix quotation error.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: simplify AT_PARSER_CHECK usage
	Currently the caller must specify the ./ prefix to its command.  Let's
	avoid that: it will be nicer to read, make it easier to have a version
	that works for Java and C/C++.

* tests/local.at (AT_PARSER_CHECK): Prefix the command with ./.
	Adjust callers.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: java: factor the definition of Position
	* tests/local.at (AT_JAVA_POSITION_DEFINE): New.
	* tests/java.at, tests/javapush.at: Use it.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: dispatch per lang on AT_DATA_GRAMMAR
	* tests/java.at: Do that.
	* tests/conflicts.at: Simplify.

* tests/actions.at, tests/c++.at, tests/input.at, tests/local.at,
	* tests/named-refs.at:
	Use AT_BISON_OPTION_PUSHDEFS/AT_BISON_OPTION_POPDEFS.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: dispatch per lang on the definition of yylex
	* tests/local.at (AT_YYLEX_DEFINE): Dispatch on the language.
	(AT_YYLEX_DEFINE(java)): New.
	* tests/conflicts.at, tests/java.at: Use it.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: de-duplicate
	* tests/conflicts.at (AT_YYLEX_PROTOTYPE): Don't define it, leave that
	task to AT_DATA_GRAMMAR.
	But be honest: tell AT_BISON_OPTION_PUSHDEFS all the options we use.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

tests: formatting changes
	* tests/local.at: here.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

graph: prefer *.gv to *.dot
	Reported by Hans Åberg.
	https://lists.gnu.org/archive/html/help-bison/2019-02/msg00064.html

* src/files.c (spec_graph_file): Use `*.gv` when 3.4 or better,
	otherwise `*.dot`.
	* src/parse-gram.y (handle_require): Pretend we are already 3.4.
	* doc/bison.texi: Adjust.
	* tests/local.at, tests/output.at: Exercise this.

2019-02-21  Akim Demaille  <akim.demaille@gmail.com>

doc: fixes to please older versions of Texinfo
	Currently, we have errors:

./doc/bison.texi:13005: node `History' lacks menu item for `Yacc' despite being its Up target

* doc/bison.texi (History): Add the local menu.
	While at it, add a few index entries.

2019-02-17  Akim Demaille  <akim.demaille@gmail.com>

doc: style changes
	* doc/bison.texi (Bibliography): Add [Corbett 1984] and [Johnson
	1978].
	(History): Use them.
	And other minor changes.

2019-02-17  Eric S. Raymond  <esr@thyrsus.com>

doc: a history section
	* bison.texi (A Brief History of the Greater Ungulates): New section.

2019-02-17  Akim Demaille  <akim.demaille@gmail.com>

examples: add an example with a reentrant parser in Flex+Bison
	Suggested by Eric S. Raymond.
	https://lists.gnu.org/archive/html/bison-patches/2019-02/msg00066.html

* examples/c/reentrant-calc/Makefile, examples/c/reentrant-calc/README.md,
	* examples/c/reentrant-calc/parse.y, examples/c/reentrant-calc/scan.l
	* examples/c/reentrant-calc/lexcalc.test,
	* examples/c/reentrant-calc/local.mk:
	New.

2019-02-16  Akim Demaille  <akim.demaille@gmail.com>

examples: fixes in lexcalc
	* examples/c/lexcalc/parse.y: Formatting/comment changes.
	(line): Don't return a value.
	Print the result here, which avoids printing a value for lines with an
	error.
	(yyerror): Be sure to increment the pointed, not the pointer...
	* examples/c/lexcalc/lexcalc.test: Check errors.
	* examples/c/lexcalc/local.mk: Fix a dependency.

2019-02-16  Akim Demaille  <akim.demaille@gmail.com>

build: fix distcheck
	Regression introduced in 05a80977798abf472bfc11c477751303ad604733.

* doc/local.mk (bison.help): Don't depend on the path of the bison
	executable.

2019-02-16  Akim Demaille  <akim.demaille@gmail.com>

style: move pkgdatadir to files.*
	Let's move it to a more logical place.

* src/output.h, src/output.c (pkgdatadir): Move to...
	* src/files.h, src/files.c: here.

2019-02-14  Akim Demaille  <akim.demaille@gmail.com>

style: rename cleanup_caret as caret_free
	* src/location.c, src/location.h, src/main.c: here.

2019-02-14  Akim Demaille  <akim.demaille@gmail.com>

style: avoid default in switch on enums
	* src/assoc.c (assoc_to_string): here.

2019-02-14  Akim Demaille  <akim.demaille@gmail.com>

doc: run tests/bison, not src/bison
	* doc/local.mk: Don't run src/bison, as it expects to find all its
	files installed.  Instead, run tests/bison which is ready to run in
	builddir.

2019-02-13  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-02-12  Akim Demaille  <akim.demaille@gmail.com>

style: comment and names changes in map_goto
	* src/lalr.h, src/lalr.c: Use clearer names.

2019-02-12  Akim Demaille  <akim.demaille@gmail.com>

yacc: support parse.assert
	While hacking on the computation of the automaton, I had yystate being
	equal to -1, and the parser loops.  Let's catch this when
	parser.assert is enabled.

* data/skeletons/yacc.c (YY_ASSERT): New.
	Use it.
	Not using the name YYASSERT, to make it clear that this is private.
	glr.c should probably move to YY_ASSERT too.
	Also, while at it, report 'Entering state...' even before growing the
	stacks.

2019-02-12  Akim Demaille  <akim.demaille@gmail.com>

examples: depend on Bison's sources
	* examples/c/calc/local.mk, examples/c/lexcalc/local.mk,
	* examples/c/mfcalc/local.mk, examples/c/rpcalc/local.mk:
	Regenerate the files if dependencies have changed.

2019-02-12  Eric S. Raymond  <esr@thyrsus.com>

README: point to README-hacking
	* README (Build from git): New.
	* README-hacking: Describe easier submodule update.

2019-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: a single space before closing comments
	I don't think this style:

/* If buffer is full, make it bigger.        */
	    if (i == length)
	      {
	        length *= 2;
	        symbuf = (char *) realloc (symbuf, length + 1);
	      }
	    /* Add this character to the buffer.         */
	    symbuf[i++] = c;
	    /* Get another character.                    */
	    c = getchar ();

or more readable than

/* If buffer is full, make it bigger. */
	    if (i == length)
	      {
	        length *= 2;
	        symbuf = (char *) realloc (symbuf, length + 1);
	      }
	    /* Add this character to the buffer. */
	    symbuf[i++] = c;
	    /* Get another character. */
	    c = getchar ();

Actually, I think the latter is more readable, and helps with width
	issues in the PDF.  As a matter of fact, I would happily move to //
	only for single line comments.

* doc/bison.texi: A single space before closing comments.

2019-02-10  Akim Demaille  <akim.demaille@gmail.com>

doc: modernize the examples
	* doc/bison.texi: Prefer 'fun' to 'fnct'.
	Reduce local variable scopes.
	Prefer strdup to malloc + strcpy.
	Avoid gratuitous casts.
	Use simpler names (e.g., 'name' instead of 'fname').
	Avoid uses of 0 for NULL.
	Avoid using NULL when possible (e.g., 'p' instead of 'p != NULL').
	Prefer union names to casts (e.g. 'yylval.VAR = s' instead of
	'*((symrec**) &yylval) = s').
	Give arguments a name in fun declarations.
	Use our typedefs instead of duplicating them (func_t).
	Stop promoting an explicit $$ = $1;, it should be implicit (Bison
	might be able to eliminate useless chain rules).
	Help a bit Texinfo by making smaller groups.
	Rely on the C compiler to call function pointers (prefer
	'$1->value.fun ($3)' to (*($1->value.fnctptr))($3)').

2019-02-10  Akim Demaille  <akim.demaille@gmail.com>

examples: add a simple infix calculator in C
	Currently we have no simple example: rpcalc in reverse Polish, mfcalc
	has functions, and lexcalc is using lex.

* examples/c/calc/Makefile, examples/c/calc/calc.y,
	* examples/c/calc/calc.test, examples/c/calc/local.mk: New.

2019-02-10  Akim Demaille  <akim.demaille@gmail.com>

examples: fix annoying off-by-one errors
	* examples/extexi: Since we issue #lines only at the beginning of
	@example, leave empty line when removing content (such as @comment
	lines), otherwise the lines that follow have incorrect source line
	location.  This leaves ugly empty lines, but they are removed when you
	tidy the output for the end user: sequences of \n are mapped to at
	most two sucessive \n.

2019-02-09  Akim Demaille  <akim.demaille@gmail.com>

style: factor printing of rules
	* src/gram.h, src/gram.c (rule_print): New.
	Use it.

2019-02-09  Akim Demaille  <akim.demaille@gmail.com>

style: use lower case for variable names
	* src/relation.c (INDEX, VERTICES): Rename as...
	(indexes, vertices): these.

2019-02-09  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in relation.c

2019-02-09  Akim Demaille  <akim.demaille@gmail.com>

report: stop counting uselessly
	* src/print.c (print_nonterminal_symbols): Replace left_count and
	right_count with on_left and on_right.

2019-02-09  Akim Demaille  <akim.demaille@gmail.com>

report: clean up its format
	The format is inconsistent.  For instance most sections are
	indented (including "Terminals unused in grammar" for instance), but
	the sections "Terminals, with rules where they appear" and
	"Nonterminals, with rules where they appear" are not.  Let's indent
	them.  Also, these two sections try to wrap the output to avoid lines
	too long.  Yet we don't do that in the rest of the file, for instance
	when listing the lookaheads of an item.

For instance in the case of Bison's parse-gram.output we go from:

Terminals, with rules where they appear

"end of file" (0) 0
	    error (256) 28 88
	    "string" <char*> (258) 9 13 16 17 20 23 24 109 116
	    [...]

Nonterminals, with rules where they appear

$accept (58)
	        on left: 0
	    input (59)
	        on left: 1, on right: 0
	    prologue_declarations (60)
	        on left: 2 3, on right: 1 3
	    prologue_declaration (61)
	        on left: 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 23 24
	        25 26 27 28 29, on right: 3
	    [...]

Terminals, with rules where they appear

"end of file" (0) 0
	    error (256) 28 88
	    "string" <char*> (258) 9 13 16 17 20 23 24 109 116
	    [...]

Nonterminals, with rules where they appear

$accept (58)
	            on left: 0
	        input (59)
	            on left: 1
	            on right: 0
	        prologue_declarations (60)
	            on left: 2 3
	            on right: 1 3
	        prologue_declaration (61)
	            on left: 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29
	            on right: 3
	    [...]

* src/print.c (END_TEST): Remove.
	(print_terminal_symbols): Don't try to wrap the output.
	(print_nonterminal_symbols): Likewise.
	Make two different lines for occurrences on the left, and occurrence
	on the rhs of the rules.
	Indent by 4 and 8, not 3.
	* src/reduce.c (reduce_output): Indent by 4, not 3.

* tests/conflicts.at, tests/existing.at, tests/reduce.at,
	* tests/regression.at, tests/report.at:
	Adjust.

2019-02-05  Akim Demaille  <akim.demaille@gmail.com>

add LR(0) output
	This should not be used to generate parsers.  My point is actually to
	facilitate debugging (when tweaking the generation of the LR(0)
	automaton for instance, not carying -yet- about lookaheads).

* src/reader.c (prepare_percent_define_front_end_variables): Add lr(0).
	* src/conflicts.c (set_conflicts): Be robust to reds not having
	lookaheads at all.
	* src/ielr.c (LrType, lr_type_get): Adjust.
	(ielr): Implement support for LR(0).
	* src/lalr.c (lalr_free): Don't free LA when it's not computed.

2019-02-05  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in derives.c
	* src/derives.c: here.

2019-02-05  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes and refactoring in state.c
	* src/state.h, src/state.c: Comment changes.
	(transitions_to): Take a state* as argument.
	* src/lalr.h, src/lalr.c: Comment changes.
	(initialize_F): Use clear variable names.

2019-02-05  Akim Demaille  <akim.demaille@gmail.com>

tests: fix typos
	* tests/reduce.at: here.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

Merge branch maint
	* maint:
	  maint: post-release administrivia
	  version 3.3.2
	  style: minor fixes
	  NEWS: named constructors are preferable to symbol_type ctors
	  gram: fix handling of nterms in actions when some are unused
	  style: rename local variable
	  CI: update the ICC serial number for travis-ci.org

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

version 3.3.2
	* NEWS: Record release date.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

style: minor fixes
	* NEWS, src/reduce.c, src/reduce.h: Use 'nonterminal'.
	Fix comments.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

NEWS: named constructors are preferable to symbol_type ctors
	Reported by Frank Heckenbach.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00043.html

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

gram: fix handling of nterms in actions when some are unused
	Since Bison 3.3, semantic values in rule actions (i.e., '$...') are
	passed to the m4 backend as the symbol number.  Unfortunately, when
	there are unused symbols, the symbols are renumbered _after_ the
	numbers were used in the rule actions.  As a result, the evaluation of
	the skeleton failed because it used non existing symbol numbers.
	Which is the happy scenario: we could use numbers of other existing
	symbols...

Reported by Balázs Scheidler.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00044.html

Translating the rule actions after the symbol renumbering moves too
	many parts in bison.  Relying on the symbol identifiers is more
	troublesome than it might first seem: some don't have an
	identifier (tokens with only a literal string), some might have a
	complex one (tokens with a literal string with characters special for
	M4).  Well, these are tokens, but nterms also have issues: "dummy"
	nterms (for midrule actions) are named $@32 etc. which is risky for
	M4.

Instead, let's simply give M4 the mapping between the old numbers and
	the new ones.  To avoid confusion between old and new numbers, always
	emit pre-renumbering numbers as "orig NUM".

* data/README: Give details about "orig NUM".
	* data/skeletons/bison.m4 (__b4_symbol, _b4_symbol): Resolve the
	"orig NUM".
	* src/output.c (prepare_symbol_definitions): Pass nterm_map to m4.
	* src/reduce.h, src/reduce.c (nterm_map): Extract it from
	nonterminals_reduce, to make it public.
	(reduce_free): Free it.
	* src/scan-code.l (handle_action_dollar): When referring to a nterm,
	use "orig NUM".
	* tests/reduce.at (Useless Parts): New, based Balázs Scheidler's
	report.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

tests: strengthen some of them
	* tests/reduce.at: Check that the generated parsers are proper C.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

package: rename data/README as data/README.md
	So that it is properly rendered by online git services.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes
	* src/symlist.c (symbol_list_free): New.

2019-02-03  Akim Demaille  <akim.demaille@gmail.com>

style: prefer snprintf to sprintf
	* src/symtab.c (dummy_symbol_get): There's no need for the buffer to
	be so big and static.
	Use snprintf for safety.

2019-02-02  Akim Demaille  <akim.demaille@gmail.com>

style: comment and name changes
	* src/output.c (prepare_symbol_names): here.
	* src/reader.c: Remove obsolete comment.
	* src/scan-code.l: Use || for Boolean or.

2019-02-02  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/reader.c, src/scan-code.l: here.

2019-02-02  Akim Demaille  <akim.demaille@gmail.com>

make: regenerate the example parsers when bison changes
	* Makefile.am (dependencies): Also depend on Bison's sources.

2019-02-02  Akim Demaille  <akim.demaille@gmail.com>

style: rename local variable
	* src/reduce.c (nonterminals_reduce): Rename nontermmap as nterm_map.
	We will expose it.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

gram: detect and report (in debug traces) useless chain rules
	A rule is a useless chain iff it's a chain (aka unit, or injection)
	rule (i.e., the RHS has length 1), and it's useless (it has no used
	defined semantic action).

* src/gram.h, src/gram.c (rule_useless_chain_p): New.
	(grammar_dump): Report useless chain rules.
	* tests/sets.at: Check the traces.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

lr(0): more debug traces
	* src/lr0.c (core_print, kernel_print): New.
	Use them.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

lr(0): remove useless conditional
	* src/lr0.c (new_itemsets): There's no harm in setting a Boolean
	several times.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

style: sort includes and avoid assignments
	* src/symtab.c: Sort includes.
	* src/gram.c (grammar_rules_print_xml): Avoid assignments to define
	'usefulness'.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

style: use item_rule
	* src/print-graph.c, src/print-xml.c: here.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

gram: factor the printing of items and the computation of their rule
	There are several places where we need to recover the rule from an
	item, let's factor that into item_rule.  We also want to print items
	in a nice way: we do it when generating the *output file, but it is
	also useful in debug messages.

* src/gram.h, src/gram.c (item_rule, item_print): New.
	* src/print.c (print_core): Use them.
	* src/state.h, src/state.c: Propagate constness.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in print-xml
	* src/print-xml.c: here.

2019-01-30  Akim Demaille  <akim.demaille@gmail.com>

tests: check XML and dot reports
	* tests/report.at: Here.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

CI: update the ICC serial number for travis-ci.org
	On travis-ci.org, there are five concurrent slaves, instead of three
	on travis-ci.com.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

CI: update the ICC serial number for travis-ci.org
	On travis-ci.org, there are five concurrent slaves, instead of three
	on travis-ci.com.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* src/lr0.c, src/state.c, src/state.h: here.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

closure: initialize it once for all
	The memory allocated by 'closure' (and some data such as 'fderives')
	is used to computed a state's full itemset from its core.  This is
	needed during the construction of the LR(0) automaton, and the memory
	is reclaimed immediately afterwards.

Unfortunately the reports (graph, text, xml) also need this
	information when describing the states with their full itemsets.  As a
	consequence the memory was allocated again, fderives computed again
	too, and more --trace reports are generated which only duplicate what
	was already reported.

Stop that.  It does mean that we release the memory later (hence the
	peak memory usage is higher now), but I don't think that's a problem
	today.

* src/lr0.c (generate_states): Don't call closure_free.
	* src/state.c (states_free): Do it here.
	(for symmetry with closure_new which is called in generate_states).
	* src/print-graph.c, src/print-xml.c, src/print.c: You can now expect
	the closure module to be functional.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

style: rename closure_* functions as closure_*
	This is more consistent with the other files.

* closure.h, closure.c (new_closure, free_closure): Rename as...
	(closure_new, closure_free): this.
	Adjust dependencies.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

lr0: use a bitset for the set of "shiftable symbols"
	This will make it easier to add new elements (that might already be
	part of shift_symbol) without having to worry about the size of
	shift_symbol (which is currently a fixed size vector).

I could not measure any significant differences in performances in the
	generation of LR(0) automaton (benched on gramamrs of Ruby, C, and C++).

* src/lr0.c (shift_symbol): Make it a bitset.

2019-01-28  Akim Demaille  <akim.demaille@gmail.com>

add -fsyntax-only
	When debugging Bison itself, this is very handy, especially when
	tweaking the frontend badly enough to break the backends. It can also
	be used to check a grammar.

* src/getargs.h, src/getargs.c (feature_syntax_only): New.
	(feature_args, feature_types): Adjust.
	* src/main.c (main): Use it.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

style: beware of collisions on status
	* src/symtab.h (status): Rename as...
	(declaration_status): this, to avoid colliding with status, the
	argument of 'usage'.
	'status' seems a tad too general to be used only here.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

usage: document -ffixit
	* src/getargs.c (usage): Document -ffixit.
	Document the aliases of -f.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes in state.c and ielr.c

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

Merge branch 'maint'
	* maint:
	  maint: post-release administrivia
	  version 3.3.1
	  yacc: issue warnings, not errors, for Bison extensions
	  style: formatting changes in NEWS and complain.c
	  tests: don't depend on the user's definition of SHELL

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

version 3.3.1
	* NEWS: Record release date.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

yacc: issue warnings, not errors, for Bison extensions
	Reported by Kiyoshi Kanazawa.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00029.html

* src/getargs.c (getargs): Let --yacc imply -Wyacc, not -Werror=yacc.
	* tests/input.at: Adjust.
	* doc/bison.tex (Bison Options): Document.

2019-01-27  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes in NEWS and complain.c

2019-01-27  Kiyoshi Kanazawa  <yoi_no_myoujou@yahoo.co.jp>

tests: don't depend on the user's definition of SHELL
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00031.html

* examples/test (SHELL): Set it to /bin/sh.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

traces: always print the reduced grammar and fix it
	* src/gram.c (grammar_dump): Print the effective number first instead
	of last.  And fix it (remove the incorrect "+1").
	Use t/f for Booleans.
	* src/reduce.c: When asked, always print the reduced grammar, even if
	there was nothing useless.
	* tests/sets.at (Reduced Grammar): Check that.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: rename LR0.* as lr0.*
	Let's stick to lower case for file names.

* src/LR0.h, src/LR0.c: Rename as...
	* src/lr0.h, src/lr0.c: these.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: rename print_graph.* as print-graph.*
	These are the only files with _.

* src/print_graph.h, src/print_graph.c: Rename as...
	* src/print-graph.h, src/print-graph.c: these.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: various fixes
	* src/gram.c: Use consistent variable names.
	Prefix prefix unary operators.
	(grammar_dump): Use rule_rhs_length instead of duplicating it.
	* src/reduce.c: Avoid useless variables.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes in gram.h
	* src/gram.h: Shorten comments.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

version 3.3
	* NEWS: Record release date.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

c++: fix comment
	* data/skeletons/c++.m4: here.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* data/skeletons/lalr1.cc: Add dnl.
	* data/skeletons/bison.m4: Comment the use of dnl.

2019-01-26  Akim Demaille  <akim.demaille@gmail.com>

tests: run the printer/destructor test on glr.cc
	* tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR): Adjust for
	glr.cc, and use it.

2019-01-24  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-01-22  Akim Demaille  <akim.demaille@gmail.com>

--update: when used, do not generate the output files
	It is inconvenient that we also generate the output files when we
	update the grammar file, and it's somewhat unexpected.  Let's not do
	that.

* src/main.c (main): Skip generation when --update is passed.
	* src/getargs.c (usage): Update the help message.
	* doc/bison.texi (Bison Options): Likewise.
	* tests/input.at: Check that we don't generate the output.

2019-01-22  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: let redundant definitions be only warnings
	After all, this is clearly harmless.

* src/muscle-tab.c (muscle_percent_define_insert): Let equal
	definitions of a %define variable be only a warning.
	Adjust test cases.

2019-01-22  Akim Demaille  <akim.demaille@gmail.com>

tests: improve check for updated variable names
	* tests/input.at ("%define" backward compatibility): Don't define
	twice "api.namespace", so that we don't get an error, which stops the
	process too soon to see an error about the value given to
	'lr.keep-unreachable-state'.

2019-01-21  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: remove redundancy
	Don't repeat the name of the warning in the sub messages.  E.g.,
	remove the second "[-Wother]" in the following message

foo.y:2.1-27: warning: %define variable 'parse.error' redefined [-Wother]
	     %define parse.error verbose
	     ^~~~~~~~~~~~~~~~~~~~~~~~~~~
	    foo.y:1.1-27:     previous definition [-Wother]
	     %define parse.error verbose
	     ^~~~~~~~~~~~~~~~~~~~~~~~~~~

* src/complain.c (error_message): Don't print the warning type when
	it's indented.
	Adjust test cases.

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

c++: better "scope" a workaround for GCC
	* data/skeletons/lalr1.cc: Enable it only for GCC 4.8 and before.

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

c++: address -Wweak-vtables warnings
	Reported by Derek Clegg
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00021.html

aux/parser-internal.h:429:12: error: 'syntax_error' has no out-of-line virtual
	         method definitions; its vtable will be emitted in every translation unit
	         [-Werror,-Wweak-vtables]
	       struct syntax_error : std::runtime_error

To avoid this warning, we need syntax_error to have a virtual function
	defined in a compilation unit.  Let it be the destructor.  To comply
	with C++98, this dtor should be 'throw()'.  Merely making YY_NOEXCEPT
	be 'throw()' in C++98 triggers
	errors (http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00022.html),
	so let's introduce YY_NOTHROW and flag only ~syntax_error with it.

Also, since we now have an explicit dtor, we need to provide an copy
	ctor.

* configure.ac (warn_cxx): Add -Wweak-vtables.
	* data/skeletons/c++.m4 (YY_NOTHROW): New.
	(syntax_error): Declare the dtor, and define the copy ctor.
	* data/skeletons/glr.cc, data/skeletons/lalr1.cc (~syntax_error):
	Define.

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

style: prefer bool to char
	* src/state.h, src/state.c (state::consistent): Make it a bool.
	Adjust dependencies.

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: be more alike lalr1.cc

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* data/skeletons/c++.m4: Un-remove an end-of-line.

2019-01-20  Akim Demaille  <akim.demaille@gmail.com>

NEWS: fixes

2019-01-19  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-01-19  Akim Demaille  <akim.demaille@gmail.com>

version 3.2.91
	* NEWS: Record release date.

2019-01-18  Akim Demaille  <akim.demaille@gmail.com>

style: various fixes
	Some reported by syntax-check.

* po/POTFILES.in: Add fixits.cc.
	* src/muscle-tab.c: Don't cast for free.
	* src/files.c: Reduce scopes.
	* cfg.mk: We need the cast for free in muscle_percent_define_insert.

2019-01-18  Akim Demaille  <akim.demaille@gmail.com>

doc: document -ffixit and --update
	* doc/bison.texi (Bison Options): here.

2019-01-18  Akim Demaille  <akim.demaille@gmail.com>

doc: style fixes
	* doc/bison.texi: Use @kbd where appropriate.
	Update ^~~~ marks for caret-errors.
	* build-aux/cross-options.pl: Do not add quotes to %define's argument.

2019-01-18  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update for fixits and --update

2019-01-17  Akim Demaille  <akim.demaille@gmail.com>

fixits: handle duplicates of %name-prefix
	The test case "Deprecated directives" (currently 56) no longer emits
	warnings after 'bison -u'!

* src/files.h, src/files.c (spec_name_prefix_loc): New.
	* src/parse-gram.y (handle_name_prefix): Emit fixits for duplicate
	%name-prefix.
	* tests/input.at (Deprecated directives): Adjust.

2019-01-17  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-01-17  Akim Demaille  <akim.demaille@gmail.com>

fixits: handle %file-prefix
	* src/files.h, src/files.c (spec_file_prefix_loc): New.
	* src/scan-gram.l (%file-prefix): Delegate diagnostics to...
	* src/parse-gram.y (handle_file_prefix): here.
	* src/complain.c (duplicate_directive): Quote the directive.
	* tests/input.at: Adjust.

2019-01-17  Akim Demaille  <akim.demaille@gmail.com>

fixits: handle per-rule duplicates
	* src/complain.c (duplicate_rule_directive): Here.
	* tests/actions.at (Invalid uses of %empty): Check it.

2019-01-17  Akim Demaille  <akim.demaille@gmail.com>

fixits: fix warnings about duplicates
	* src/complain.c (duplicate_directive): Fix the complaint level.
	* tests/input.at: Adjust.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: properly indent the "previous declaration" message
	* src/complain.c (duplicate_directive, duplicate_rule_directive):
	Here.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

style: rename some functions for consistency
	"handle_" is the prefix used in scan-code.l for instance.

* src/parse-gram.y (do_error_verbose, do_name_prefix, do_require)
	(do_skeleton, do_yacc):
	Rename as...
	(handle_error_verbose, handle_name_prefix, handle_require)
	(handle_skeleton, handle_yacc):
	these.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

fixits: report duplicate %yacc directives
	We should use -ffixit and --update to clean files with duplicate
	directives.  And we should complain only once about duplicate obsolete
	directives: keep only the "duplicate" warning.  Let's start with %yacc.

For instance on:

%fixed-output_files
	    %fixed-output-files
	    %yacc
	    %%
	    exp:

This run of bison:

$ bison /tmp/foo.y -u
	    foo.y:1.1-19: warning: deprecated directive, use '%fixed-output-files' [-Wdeprecated]
	     %fixed-output_files
	     ^~~~~~~~~~~~~~~~~~~
	    foo.y:2.1-19: warning: duplicate directive [-Wother]
	     %fixed-output-files
	     ^~~~~~~~~~~~~~~~~~~
	    foo.y:1.1-19: previous declaration
	     %fixed-output_files
	     ^~~~~~~~~~~~~~~~~~~
	    foo.y:3.1-5: warning: duplicate directive [-Wother]
	     %yacc
	     ^~~~~
	    foo.y:1.1-19: previous declaration
	     %fixed-output_files
	     ^~~~~~~~~~~~~~~~~~~
	    bison: file 'foo.y' was updated (backup: 'foo.y~')

gives:

%fixed-output-files
	    %%
	    exp:

* src/location.h, src/location.c (location_empty): New.
	* src/complain.h, src/complain.c (duplicate_directive): New.

* src/getargs.h, src/getargs.c (yacc_flag): Instead of a Boolean, be
	the location of the definition.
	Update dependencies.

* src/scan-gram.l (%yacc, %fixed-output-files): Move the handling of
	its warnings to...
	* src/parse-gram.y (do_yacc): This new function.

* tests/input.at (Deprecated Directives): Adjust expectations.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

style: rename duplicate_directive as duplicate_rule_directive
	* src/complain.h, src/complain.c: here.
	Adjust callers.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

fixits: suggest running --update if there are fixits
	* src/fixits.h, src/fixits.c (fixits_empty): New.
	* src/complain.c (deprecated_directive): Register the Wdeprecated
	fixits only if -Wdeprecated was enabled, so that we don't apply
	updates if the user didn't ask for them.
	* src/main.c (main): If there were fixits, issue a warning suggesting
	running with --update.
	Free uniqstrs after the fixits, since the latter use the former.
	* tests/headers.at, tests/input.at: Update expectations.

2019-01-16  Akim Demaille  <akim.demaille@gmail.com>

fixits: avoid generating empty lines
	* src/fixits.c (fixits_run): If erase the content of a line, also
	erase the following \n.
	* tests/input.at (Deprecated directives): Update expectations.

2019-01-15  Akim Demaille  <akim.demaille@gmail.com>

configure: don't try to run C++ warnings on C
	Reported by Derek Clegg.
	https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00066.html

* configure.ac: here.

2019-01-15  Akim Demaille  <akim.demaille@gmail.com>

c, c++: avoid implicit fall-throw
	Reported by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00004.html

* configure.ac (warn_common): Add -Wimplicit-fallthrough.
	This does trigger failures in the test suite.
	* data/skeletons/glr.c, data/skeletons/lalr1.cc,
	* data/skeletons/yacc.c, tests/c++.at:
	Make fall-throws explicit.

2019-01-15  Akim Demaille  <akim.demaille@gmail.com>

c++: avoid -Wundefined-func-template warnings from clang
	Reported by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00006.html

Clang does not like this:

template <typename D>
	    struct basic_symbol : D
	    {
	      basic_symbol();
	    };

struct by_type {};

struct symbol_type : basic_symbol<by_type>
	    {
	      symbol_type(){}
	    };

It gives:

$ clang++-mp-7.0 -Wundefined-func-template foo.cc -c
	    foo.cc:11:3: warning: instantiation of function 'basic_symbol<by_type>::basic_symbol'
	                 required here, but no definition is available [-Wundefined-func-template]
	      symbol_type(){}
	      ^
	    foo.cc:4:3: note: forward declaration of template entity is here
	      basic_symbol();
	      ^
	    foo.cc:11:3: note: add an explicit instantiation declaration to suppress this warning
	                 if 'basic_symbol<by_type>::basic_symbol' is explicitly instantiated in
	                 another translation unit
	      symbol_type(){}
	      ^
	    1 warning generated.

The same applies for the basic_symbol's destructor and `clear()`.

* configure.ac (warn_cxx): Add -Wundefined-func-template.
	This triggered one failure in the test suite:
	* tests/headers.at (Sane headers): here, where we check that we can
	compile the generated headers in other compilation units than the
	parser's.
	Add a variant type to make sure that basic_symbol and symbol_type are
	properly generated in this case.
	* data/skeletons/c++.m4 (basic_symbol): Inline the definitions of the
	destructor and of `clear` in the class definition.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

Revert the last two commits
	They should not have been pushed, sorry about that.

This reverts
	- commit 8575bd06ae6e65f3a30b21a3e022a968e4c7ae7a.
	- commit 55bf52860eac5c1394dc344a691220272df32b09.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

c++: avoid warnings about extraneous semi-colons
	Reported by Derek Clegg.

* configure.ac (warn_common): Add -Wextra-semi.
	* data/skeletons/c++.m4: Remove extraneous semi-colon.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

WIP

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: add fixit support for duplicate removal
	* src/muscle-tab.c (muscle_percent_define_insert): Register a fixit
	for duplicate removal.
	* tests/input.at: Adjust expectations.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: improve the accuracy for %error-verbose
	Avoid duplicate warnings about %error-verbose, once for deprecation,
	another for duplicate.  Keep only the duplicate warning for the second
	occurrence of %error-verbose.

This will help removal fixits.

* src/scan-gram.l (%error-verbose): Return as a PERCENT_ERROR_VERBOSE
	token.
	* src/parse-gram.y (do_error_verbose): New.
	Use it.
	* src/muscle-tab.c (muscle_percent_variable_update): Handle pseudo
	variables such as %error-verbose.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: avoid duplicate warnings for deprecated directives
	Currently, on

%define parser_class_name "Parser"
	    %define parser_class_name "Parser"
	    %%
	    exp:;

we issue:

foo.y:1.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated]
	     %define parser_class_name "Parser"
	             ^~~~~~~~~~~~~~~~~
	    foo.y:2.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated]
	     %define parser_class_name "Parser"
	             ^~~~~~~~~~~~~~~~~
	    foo.y:2.9-25: error: %define variable 'api.parser.class' redefined
	     %define parser_class_name "Parser"
	             ^~~~~~~~~~~~~~~~~
	    foo.y:1.9-25:     previous definition
	     %define parser_class_name "Parser"
	             ^~~~~~~~~~~~~~~~~

Let's get rid of the second warning about the deprecated variable
	parser_class_name.  This is noise, but it will also be a problem with
	fixits for removing duplicates, as we will first generate the update,
	and then it's too late to remove it: fixits do not edit the result of
	previous fixits.

So generate this instead:

foo.y:1.1-34: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated]
	     %define parser_class_name "Parser"
	     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	    foo.y:2.1-34: error: %define variable 'api.parser.class' redefined
	     %define parser_class_name "Parser"
	     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	    foo.y:1.1-34:     previous definition
	     %define parser_class_name "Parser"
	     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

* src/muscle-tab.c (muscle_percent_variable_update): Pass the warning
	to the caller, instead of issuing it.
	(muscle_percent_define_insert): Issue this warning only if we don't
	have to complain about a duplicate definition.
	* tests/input.at: Adjust expectations.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: update the grammar file
	Let's use the fixits to actually update the grammar files.

* src/getargs.h, src/getargs.c (update_flag): New.
	* src/fixits.h, src/fixits.c (fixits_run): New.
	* src/main.c (main): Invoke fixits_run when --update is passed.
	* tests/input.at (Deprecated directives): Check --update.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: improve accuracy for deprecated %define variables
	* src/parse-gram.y: Use the location of the whole definition to record
	the location of a %define variable, instead of just the name of the
	variable.
	Adjust tests.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: keep the fixits
	Introduce proper support for fixits, instead of just printing them on
	demand.

* bootstrap.conf: We need gnulib's xlists.
	* src/fixits.h, src/fixits.c: New.
	* src/complain.c (deprecated_directive): Use fixits_register.
	* src/main.c (main): Use fixits_free.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: add -ffixit support for deprecated features
	Issue directives for IDE/editors to fix the source file.
	http://clang.llvm.org/docs/UsersManual.html#cmdoption-fdiagnostics-parseable-fixits

Do it for deprecated features.  For instance:

$ cat foo.y
	    %error-verbose

%name-prefix = "foo"
	    %name-prefix="bar"
	    %define parser_class_name "Parser"

%%
	    exp:;

$ LC_ALL=C ./_build/8d/tests/bison -ffixit /tmp/foo.yy
	    /tmp/foo.yy:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated]
	     %error-verbose
	     ^^^^^^^^^^^^^^
	    fix-it:"/tmp/foo.yy":{1:1-1:15}:"%define parse.error verbose"
	    /tmp/foo.yy:3.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated]
	     %name-prefix = "foo"
	     ^^^^^^^^^^^^^^^^^^^^
	    fix-it:"/tmp/foo.yy":{3:1-3:21}:"%define api.prefix {foo}"
	    /tmp/foo.yy:4.1-18: warning: deprecated directive, use '%define api.prefix {bar}' [-Wdeprecated]
	     %name-prefix="bar"
	     ^^^^^^^^^^^^^^^^^^
	    fix-it:"/tmp/foo.yy":{4:1-4:19}:"%define api.prefix {bar}"
	    /tmp/foo.yy:5.9-25: warning: deprecated directive, use '%define api.parser.class {Parser}' [-Wdeprecated]
	     %define parser_class_name "Parser"
	             ^^^^^^^^^^^^^^^^^
	    fix-it:"/tmp/foo.yy":{5:9-5:26}:"%define api.parser.class {Parser}"
	    /tmp/foo.yy:5.9-25: error: %define variable 'api.parser.class' is not used
	     %define parser_class_name "Parser"
	             ^^^^^^^^^^^^^^^^^

* src/getargs.h, src/getargs.c (feature_fixit_parsable): New.
	(feature_types, feature_args): Use it.
	* src/complain.c (deprecated_directive): Use it.

* tests/input.at: Check it.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: prefer ^~~~ to ^^^^ to underline code
	That's what both GCC and Clang do, and it is indeed much nicer to
	read.  From:

foo.y:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated]
	     %error-verbose
	     ^^^^^^^^^^^^^^
	    foo.y:4.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated]
	     %name-prefix = "foo"
	     ^^^^^^^^^^^^^^^^^^^^

to:

foo.y:1.1-14: warning: deprecated directive, use '%define parse.error verbose' [-Wdeprecated]
	     %error-verbose
	     ^~~~~~~~~~~~~~
	    foo.y:4.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated]
	     %name-prefix = "foo"
	     ^~~~~~~~~~~~~~~~~~~~

* src/location.c (location_caret): Use ^~~~.
	Adjust tests expectations.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

regen

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: improve them for %name-prefix
	Currently the diagnostics for %name-prefix are not precise enough.  In
	particular, they does not show that braces must be used instead of
	quotes.

Before:

foo.y:3.1-14: warning: deprecated directive, use '%define api.prefix' [-Wdeprecated]
	     %name-prefix = "foo"
	     ^^^^^^^^^^^^^^

After:

foo.y:3.1-20: warning: deprecated directive, use '%define api.prefix {foo}' [-Wdeprecated]
	     %name-prefix = "foo"
	     ^^^^^^^^^^^^^^^^^^^^

To do this we need the value passed to %name-prefix, so move the
	warning from the scanner to the parser.

Accuracy will be very important for the forthcoming changes.

* src/parse-gram.y (do_name_prefix): New.
	(PERCENT_NAME_PREFIX): Have a semantic value: the raw source, with
	possibly underscores, equal sign, and spaces.  This is used to provide
	a more accurate message.  It does not take comments into account,
	but...
	* src/scan-gram.l (%name-prefix): Delegate the warnings to the parser.

* tests/headers.at, tests/input.at: Adjust expectations.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

diagnostics: style: avoid allocating memory when not needed
	* src/muscle-tab.c (muscle_percent_variable_update): Avoid allocating
	memory when it is not needed, which should be most of the time (when
	there's no update to perform).
	Adjust callers.

2019-01-14  Akim Demaille  <akim.demaille@gmail.com>

c++: avoid warnings about extraneous semi-colons
	Reported by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00005.html

* configure.ac (warn_cxx): Add -Wextra-semi.
	* data/skeletons/c++.m4: Remove extraneous semi-colon.

2019-01-13  Akim Demaille  <akim.demaille@gmail.com>

style: minor changes
	* src/muscle-tab.c: Sort alphabetically.
	* src/scan-gram.l: Reduce scopes.
	Initialize variables.

2019-01-13  Akim Demaille  <akim.demaille@gmail.com>

c++: beware of -Wshadow
	This line:

slice<stack_symbol_type, stack_type> slice (yystack_, yylen);

triggers warnings:

parse.h:1790:11: note: shadowed declaration is here

Reported by Frank Heckenbach.
	http://lists.gnu.org/archive/html/bug-bison/2019-01/msg00002.html

* configure.ac (warn_c): Move -Wshadow to...
	(warn_common): here.
	* data/skeletons/stack.hh (slice): Define as an inner class of stack.
	* data/skeletons/lalr1.cc: Adjust.
	Rename the variable as 'range' instead of 'slice'.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

version 3.2.90
	* NEWS: Record release date.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

yacc: fix relocatability
	* src/yacc.in (prefix): Define it, as it's typically needed for
	exec_prefix.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

syntax-check: adjust paths
	* cfg.mk: here.
	(require_config_h): New.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

style: formatting clean up
	* data/skeletons/d.m4, examples/d/calc.y, src/output.c,
	* src/parse-gram.y:
	No tab, no trailing spaces.
	Reported by syntax-check.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

po: remove bitset/stats.c
	* po/POTFILES.in: here.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

tests: fix usage of AT_PARSER_CHECK
	The parser must be the first command.  Caught by syntax-check.

* tests/c++.at: here.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

style: isolate the creation of tname in a function
	* src/output.c (prepare_symbol_names): New.
	Use it.

2019-01-12  Akim Demaille  <akim.demaille@gmail.com>

tests: formatting changes
	* tests/input.at, tests/regression.at: here.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

doc: use @option consistently
	* doc/bison.texi: Use @option, not @code, for options.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: avoid negated if
	* data/skeletons/yacc.c: Prefer a "direct" conditional.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

package: bump copyrights to 2019

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

java/d: rename some %define variables for consistency
	See 890ee8a1fd288b3cc1c21c49ea0ece696ef40565 and
	https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00024.html.

* data/skeletons/d.m4, data/skeletons/java.m4
	(abstract, annotations, extends, final, implements, public, strictfp):
	Rename as...
	(api.parser.abstract, api.parser.annotations, api.parser.extends)
	(api.parser.final, api.parser.implements, api.parser.public)
	(api.parser.strictfp):
	these.

* src/muscle-tab.c (muscle_percent_variable_update): Ensure backward
	compatibility.

* doc/bison.texi, examples/d/calc.y, examples/java/Calc.y,
	tests/input.at: Adjust.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

java/d: remove useless macros
	There are many macros that are defined and used just
	once (b4_public_if, b4_abstract_if, etc.).  That's overkill.  Rather,
	let's define a macro to build the "public class YYParser" line.

It appears that the same syntax with "extends", "abstract", etc. is
	implemented in the D parser, which looks very fishy...

* data/skeletons/d.m4, data/skeletons/java.m4 (b4_public_if)
	(b4_abstract_if, b4_final_if, b4_strictfp_if): Replace with
	(b4_parser_class_declaration): this.
	* data/skeletons/lalr1.d, data/skeletons/lalr1.java: Adjust.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

style: clean tests
	* tests/named-refs.at: here.

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update
	In particular, report uninitialized submodules instead of breaking
	symlinks.

For reports, see
	https://lists.gnu.org/archive/html/bug-bison/2011-05/msg00012.html
	https://lists.gnu.org/archive/html/help-bison/2018-12/msg00034.html

For a fix, see
	https://lists.gnu.org/archive/html/bug-gnulib/2019-01/msg00024.html

2019-01-05  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: fix the handling of syntax_error from the scanner
	Commit 90a8537e6287f92fb3d5be0258a69247a742f12e was right, but issued
	two error messages.  Commit 80ef7e7639f99618bee490b2dea02b5fd9ab28e5
	tried to address that by mapping yychar and yytoken to empty, but that
	completely breaks the invariants of glr.c.  In particular, yygetToken
	can be called repeatedly and is expected to return the latest result,
	unless yytoken is YYEMPTY.  Since the previous attempt was "recording"
	that the token was coming from an exception by setting it to YYEMPTY,
	instead of getting again the faulty token, we fetched another one.

Rather, revert to the first approach: map yytoken to "invalid token",
	but record in yychar the fact that we come from an exception thrown in
	the scanner.

* data/skeletons/glr.c (YYFAULTYTOK): New.
	(yygetToken): Use it to record syntax errors from the scanner.
	* tests/c++.at (Syntax error as exception): In addition to checking
	syntax_error with error recovery, make sure it also behaves as
	expected without.

2019-01-03  Akim Demaille  <akim.demaille@gmail.com>

clearly deprecate %name-prefix
	* src/scan-gram.l (%name-prefix): Issue a deprecation warning.

* tests/calc.at, tests/headers.at, tests/input.at, tests/java.at,
	* tests/javapush.at, tests/local.at: Adjust expectations.
	Or disable -Wdeprecated.

* doc/bison.texi: Document that %name-prefix is replaced by %define
	api.prefix.

2019-01-03  Akim Demaille  <akim.demaille@gmail.com>

examples: clean up the Java/D examples
	* examples/java/Calc.y: Fix indentation.
	Sort.
	Don't use %name-prefix, since api.parser.class is already defined.
	* examples/d/calc.y: Likewise.

2019-01-03  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: here.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

rename parser_class_name as api.parser.class
	The previous name was historical and inconsistent.

* src/muscle-tab.c (define_directive): Use the proper value passing
	syntax, based on the muscle kind.
	(muscle_percent_variable_update): Use the right value passing syntax.
	Migrate from parser_class_name to api.parser.class.

* data/skeletons: Migrate from parser_class_name to api.parser.class.

* doc/bison.texi (%define Summary): Document both parser_class_name
	and api.parser.class.
	Promote the latter over the former.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* src/scan-gram.l: Here.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

style: glr.c: prefer returning a value rather than passing pointers
	This is very debatable.  This function is not pure at all, so it could
	stick to returning void: that's a common coding style to tell the
	difference between "real" (pure) functions and side-effecting
	subroutines.  However, we already have this style elsewhere (e.g.,
	yylex), and I feel the callers are somewhat nice to read this way.

* data/skeletons/glr.c (yygetLRActions): Return the action rather than
	passing by pointer.
	While at it, fix type of yytoken.
	Adjust callers.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: don't issue two error messages when syntax_error is thrown
	Reported by Askar Safin.
	https://lists.gnu.org/archive/html/bison-patches/2019-01/msg00000.html

* data/skeletons/glr.c (yygetToken): Return YYEMPTY when an exception
	is thrown.
	* data/skeletons/lalr1.cc: Log when an exception is caught.
	* tests/c++.at (Syntax error as exception): Be sure to recover from
	error before triggering another error.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

skeletons: shorten b4_parser_class_name to b4_parser_class
	* skeletons/c++.m4, skeletons/d.m4, skeletons/glr.c, skeletons/glr.cc,
	* skeletons/java.m4, skeletons/lalr1.cc, skeletons/lalr1.d,
	* skeletons/lalr1.java: Here.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: remove duplicate definition of YYLLOC_DEFAULT
	It's already provided by glr.c.

* data/skeletons/glr.cc (b4_post_prologue): Here.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

style: sort includes in scanners
	* src/scan-code.l, src/scan-gram.l, src/scan-skel.l: Reorder includes.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes in the doc
	* doc/bison.texi: here.

2019-01-02  Akim Demaille  <akim.demaille@gmail.com>

style: remove stray empty lines
	* data/skeletons/glr.c, data/skeletons/glr.cc: here.
	* data/skeletons/bison.m4 (b4_glr_cc_if): Move it here.

2018-12-31  Akim Demaille  <akim.demaille@gmail.com>

glr.cc: support syntax_error exceptions
	Kindly requested by Аскар Сафин (Askar Safin).
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00033.html

* data/skeletons/glr.c (b4_glr_cc_if): New.
	Use it.
	(yygetToken): Catch syntax_errors.
	* data/skeletons/glr.cc (YY_EXCEPTIONS): New.
	* tests/c++.at: Check it.

2018-12-31  Akim Demaille  <akim.demaille@gmail.com>

glr.c: factor the calls to yylex
	The call protocol of yylex is quite complex, and repeated three
	times.  Let's factor it.

* data/skeletons/glr.c (yygetToken): New.
	Use it.

2018-12-31  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes in glr.c
	* data/skeletons/glr.c (yyrecoverSyntaxError): here.

2018-12-30  Akim Demaille  <akim.demaille@gmail.com>

tests: refactor

2018-12-30  Akim Demaille  <akim.demaille@gmail.com>

c++: inline the implementation of syntax_error in its definition
	This way, it is easier to make sure its implementation is available in
	glr.cc too, which is not the case currently.

* data/skeletons/c++.m4 (b4_public_types_define): Move the
	implementation of syntax_error...
	(b4_public_types_declare): here.

2018-12-29  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* tests/input.at: here.

2018-12-29  Akim Demaille  <akim.demaille@gmail.com>

symbol: don't crash on symbol without content
	When running with --trace=parse, we may crash.

* src/symtab.c (symbol_print): Avoid that.

2018-12-28  Akim Demaille  <akim.demaille@gmail.com>

README-hacking: update

2018-12-28  Akim Demaille  <akim.demaille@gmail.com>

reader: get rid of a useless function
	Useless since 58d7a1a1c7497ba51a35fcf991c5b047f692fe18 (2006).

* src/parse-gram.y, src/reader.h (token_name): Remove, unused.

2018-12-27  Akim Demaille  <akim.demaille@gmail.com>

parsers: fix minor stylistic issues
	* data/skeletons/variant.hh (b4_token_constructor_declare): Remove,
	unused since the previous commit.
	Fix indentation issues.
	* data/skeletons/c++.m4: Fix indentation issues.

2018-12-27  Akim Demaille  <akim.demaille@gmail.com>

c++: check several parsers in the same program
	* tests/local.at (AT_LOCATION_TYPE_IF): Turn into...
	(AT_LOCATION_TYPE_SPAN_IF): this.
	Adjust dependencies.
	* tests/headers.at (Several parsers): Add another C++ parser,
	which uses the first C++ parser's locations.

2018-12-26  Akim Demaille  <akim.demaille@gmail.com>

c++: variants: fuse declarations and definitions
	We used to create a short definition of yy::parser with all the
	implementations of its member functions outside.  But yy::parser is no
	longer short and simple to read.  Maintaining each function twice is
	painful: a lot of redundancy but different indentation levels, output
	which depends on whether we are in a header or not (see
	d132c2d5455135f63a7497c38358eadd6e3a6af8), etc.

Let's simplify this and put the implementations into the class
	definition itself.

Discussed in this monologue:
	https://lists.gnu.org/archive/html/bison-patches/2018-12/msg00058.html.

* data/skeletons/c++.m4, data/skeletons/lalr1.cc,
	* data/skeletons/variant.hh (b4_basic_symbol_constructor_define)
	(_b4_token_constructor_declare, b4_token_constructor_declare)
	Merge into...
	(b4_basic_symbol_constructor_define, _b4_token_constructor_define)
	(b4_token_constructor_define): these.

2018-12-26  Akim Demaille  <akim.demaille@gmail.com>

examples: fix dependencies
	Commit 112ccb5ed73ba5c64b0b5300d8b9b686f02f094c moved the skeletons
	from dist_pkgdata_DATA to dist_skeletons_DATA, hence broke the dependencies.

* Makefile.am (dependencies): New.
	Use it where appropriate.

2018-12-26  Akim Demaille  <akim.demaille@gmail.com>

c++: move stack<T> inside yy::parser
	We used to define such auxiliary structures outside the class, mainly
	as a matter of style to keep the definition of yy::parser short and
	simple.  However, now there's a lot more code generated inside the
	class definition (e.g., all the token constructors), so the
	readability no longer applies.

However, if we move stack (and slice) inside yy::parser, then it
	should no longer be needed to change the namespace to have multiple
	parsers: changing the class name should suffice.

One common argument against inner classes is that they code bloat.  It
	hardly applies here, since typically different parsers will have
	different semantic value types, hence different actual stack types.

* data/skeletons/lalr1.cc: Invoke b4_stack_define inside yy::parser.

2018-12-25  Akim Demaille  <akim.demaille@gmail.com>

package: make bison a relocatable package
	Suggested by David Barto
	https://lists.gnu.org/archive/html/help-bison/2015-02/msg00004.html
	and Victor Zverovich.
	https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00121.html

This is very easy to do, thanks to work by Bruno Haible in gnulib.
	See "Supporting Relocation" in gnulib's documentation.

* bootstrap.conf: We need relocatable-prog and relocatable-script (for yacc).

* src/yacc.in: New.
	* configure.ac, src/local.mk: Instantiate it.
	* src/main.c, src/output.c (main, pkgdatadir): Use relocatable2.

* doc/bison.texi (FAQ): Document it.

2018-12-25  Akim Demaille  <akim.demaille@gmail.com>

README: wrap paragraphs

2018-12-25  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-12-25  Akim Demaille  <akim.demaille@gmail.com>

package: move skeletons into data/skeletons
	* data/bison.m4, data/c++-skel.m4, data/c++.m4, data/c-like.m4,
	* data/c-skel.m4, data/c.m4, data/d-skel.m4, data/d.m4, data/glr.c,
	* data/glr.cc, data/java-skel.m4, data/java.m4, data/lalr1.cc,
	* data/lalr1.d, data/lalr1.java, data/location.cc, data/stack.hh,
	* data/variant.hh, data/yacc.c:
	Move to...
	* data/skeletons: here.
	Use b4_skeletonsdir instead of b4_pkgdatadir.

* data/local.mk, src/output.c: Adjust.

2018-12-24  Akim Demaille  <akim.demaille@gmail.com>

c++: style: use consistently this/that instead of this/other
	* data/lalr1.cc, data/variant.hh: here.

2018-12-24  Akim Demaille  <akim.demaille@gmail.com>

c++: also provide a copy constructor for symbol_type
	Suggested by Wolfgang Thaller.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00081.html

* data/c++.m4 (basic_symbol, by_type): Instead of provide either move
	or copy constructor, always provide the copy one.
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): Check it.

2018-12-24  Akim Demaille  <akim.demaille@gmail.com>

c++: fix double free when a symbol_type was moved
	Currently the following piece of code crashes (with parse.assert),
	because we don't record that s was moved-from, and we invoke its dtor.

{
	      auto s = parser::make_INT (42);
	      auto s2 = std::move (s);
	    }

Reported by Wolfgang Thaller.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00077.html

* data/c++.m4 (by_type): Provide a move-ctor.
	(basic_symbol): Be sure not to read a moved-from value.
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): Check this case.

2018-12-24  Akim Demaille  <akim.demaille@gmail.com>

c++: style: improve tests
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): Provide better
	assertions.
	Use them.
	Avoid useless Bison invocations.

2018-12-24  Akim Demaille  <akim.demaille@gmail.com>

c++: style: use consistently this/that instead of this/other
	* data/c++.m4: here.

2018-12-23  Akim Demaille  <akim.demaille@gmail.com>

tests: fixes
	* tests/c++.at: Use AT_YYLEX_PROTOTYPE etc.
	Which requires that we use the same argument names (lvalp, etc.).
	* tests/local.at (AT_NAME_PREFIX): Fix regex.

2018-12-22  Akim Demaille  <akim.demaille@gmail.com>

c++: style: rename a few macros for clarity
	* data/c++.m4, data/lalr1.cc, data/variant.hh:
	s/b4_symbol_constructor/b4_token_constructor/g, as this is really what
	is being defined.

2018-12-22  Akim Demaille  <akim.demaille@gmail.com>

c++: exhibit a safe symbol_type
	Instead of introducing make_symbol (whose name, btw, somewhat
	infringes on the user's "name space", if she defines a token named
	"symbol"), let's make the construction of symbol_type safer, using
	assertions.

For instance with:

%token ':' <std::string> ID <int> INT;

generate:

symbol_type (int token, const std::string&);
	    symbol_type (int token, const int&);
	    symbol_type (int token);

It does mean that now named token constructors (make_ID, make_INT,
	etc.) go through a useless assert, but I think we can ignore this: I
	assume any decent compiler will inline the symbol_type ctor inside the
	make_TOKEN functions, which will show that the assert is trivially
	verified, hence I expect no code will be emitted for it.  And anyway,
	that's an assert, NDEBUG controls it.

* data/c++.m4 (symbol_type): Turn into a subclass of
	basic_symbol<by_type>.
	Declare symbol constructors when variants are enabled.
	* data/variant.hh (_b4_type_constructor_declare)
	(_b4_type_constructor_define): Replace with...
	(_b4_symbol_constructor_declare, _b4_symbol_constructor_def): these.
	Generate symbol_type constructors.
	* doc/bison.texi (Complete Symbols): Document.
	* tests/types.at: Check.

2018-12-22  Akim Demaille  <akim.demaille@gmail.com>

c++: provide symbol constructors per type
	On

%token <int> FOO BAR

we currently generate make_FOO(int) and make_BAR(int).  However, in
	order to factor their scanners, some users would also like to have
	make_symbol(tok, int), where tok is FOO or BAR.  To ensure type
	safety, add assertions that do check that value type and token type
	match.  Bind this assertion to the parse.assert %define variable.

Suggested by Frank Heckenbach.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00034.html
	Should also match expectations from Аскар Сафин.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00023.html

* data/variant.hh: Use b4_token_visible_if where applicable.
	(_b4_type_constructor_declare, _b4_type_constructor_define): New.
	Use them.

2018-12-22  Akim Demaille  <akim.demaille@gmail.com>

c++: style changes
	* data/c++.m4, data/variant.hh: Improve layout of the generated code.
	Avoid casts.
	(_b4_symbol_constructor_declare, _b4_symbol_constructor_define): Rename
	as...
	(_b4_token_maker_declare, _b4_token_maker_define): these.
	* tests/types.at: Improve pair printing.

2018-12-22  Akim Demaille  <akim.demaille@gmail.com>

style: simplify tests
	* tests/types.at: Simplify C++ instantiations.

2018-12-19  Akim Demaille  <akim.demaille@gmail.com>

style: use b4_token_visible_if
	And other formatting/comment changes.

* data/variant.hh: Here.

2018-12-19  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-12-19  Akim Demaille  <akim.demaille@gmail.com>

c++: fix token constructors for types with commas
	Bitten by macros, again.
	See 680b715518795c8648360fcf05f3772c04d2eed2.

* data/variant.hh (_b4_symbol_constructor_declare)
	(_b4_symbol_constructor_define): Do not use user types, which can
	include commas as in `std::pair<int, int>`, to macros.

* tests/local.at: Adjust the lex related macros to support the
	case of token constructors.
	* tests/types.at: Also check token constructors on types with commas.

2018-12-19  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-12-17  Akim Demaille  <akim.demaille@gmail.com>

symbols: document the overhaul of symbol declarations
	* doc/bison.texi (Symbol Decls): New.

2018-12-16  Akim Demaille  <akim.demaille@gmail.com>

symbols: check more invalid declarations
	* tests/input.at (Invalid %nterm uses): Rename as...
	(Invalid symbol declarations): this.
	Extend.

2018-12-16  Akim Demaille  <akim.demaille@gmail.com>

symbols: check the previous commit
	* tests/input.at (Symbol declarations): New.

2018-12-16  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-12-16  Akim Demaille  <akim.demaille@gmail.com>

symbols: clean up their parsing
	Prompted by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html

We have four classes of directives that declare symbols: %nterm,
	%type, %token, and the family of %left etc.  Currently not all of them
	support the possibility to have several type tags (`<type>`), and not
	all of them support the fact of not having any type tag at all
	(%type).  Let's unify this.

- %type
	  POSIX Yacc specifies that %type is for nonterminals only.  However,
	  some Bison users want to use it for both tokens and nterms
	  (actually, Bison's own grammar does this in several places, e.g.,
	  CHAR).  So it should accept char/string literals.

As a consequence cannot be used to declare tokens with their alias:
	  `%type foo "foo"` would be ambiguous (are we defining foo = "foo",
	  or are these two different symbols?)

POSIX specifies that it is OK to use %type without a type tag.  I'm
	  not sure what it means, but we support it.

- %token
	  Accept token declarations with number and string literal:
	  (ID|CHAR) NUM? STRING?.

- %left, etc.
	  They cannot be the same as %token, because we accept to declare the
	  symbol with %token, and to then qualify its precedence with %left.
	  Then `%left foo "foo"` would also be ambiguous: foo="foo", or two
	  symbols.

They cannot be simply a list of identifiers, but POSIX Yacc says we
	  can declare token numbers here.  I personally think this is a bad
	  idea, precedence management is tricky in itself and should not be
	  cluttered with token declaration issues.

We used to accept declaring a token number on a string literal here
	  (e.g., `%left "token" 1`).  This is abnormal.  Either the feature is
	  useful, and then it should be supported in %token, or it's useless
	  and we should not support it in corner cases.

- %nterm
	  Obviously cannot accept tokens, nor char/string literals.  Does not
	  exist in POSIX Yacc, but since %type also works for terminals, it is
	  a nice option to have.

* src/parse-gram.y: Avoid relying on side effects.  For instance, get
	rid of current_type, rather, build the list of symbols and iterate
	over it to assign the type.
	It's not always possible/convenient.  For instance, we still use
	current_class.
	Prefer "decl" to "def", since in the rest of the implementation we
	actually "declare" symbols, we don't "define" them.
	(token_decls, token_decls_for_prec, symbol_decls, nterm_decls): New.
	Use them for %token, %left, %type and %nterm.
	* src/symlist.h, src/symlist.c (symbol_list_type_set): New.
	* tests/regression.at b/tests/regression.at
	(Token number in precedence declaration): We no longer accept
	to give a number to string literals.

2018-12-15  Akim Demaille  <akim.demaille@gmail.com>

symbols: set tag_seen when assigning a type to symbols
	* src/reader.h, src/reader.c (tag_seen): Move to...
	* src/symtab.h, src/symtab.c: here.
	(symbol_type_set): Set it to true.
	* src/parse-gram.y: Don't.

2018-12-14  Akim Demaille  <akim.demaille@gmail.com>

tests: isolate test about Yacc warnings
	* tests/input.at (Yacc warnings): New.
	(AT_CHECK_UNUSED_VALUES): Remove checks about yacc.

2018-12-14  Akim Demaille  <akim.demaille@gmail.com>

parser: warn about string literals in Yacc mode
	* src/scan-gram.l (scan_integer): Warn.
	* tests/input.at (Yacc warnings on symbols): Check.

2018-12-14  Akim Demaille  <akim.demaille@gmail.com>

parser: warn about hexadecimal token numbers in Yacc mode
	* src/scan-gram.l (scan_integer): Warn.
	* tests/input.at (Yacc warnings on symbols): Check.

2018-12-14  Akim Demaille  <akim.demaille@gmail.com>

parser: reprecate %nterm back
	After having spent quite some time on cleaning the handling of symbol
	declarations in the grammar files, I believe we should keep it.

It looks like it's a duplicate of %type, but it is not.  While POSIX
	Yacc requires %type to apply only to nonterminal symbols, it appears
	that both byacc and bison accept it for tokens too.  And some
	experienced users do actually expect this feature to group
	symbols (terminal or not) by type ("On the other hand, it is generally
	more useful IMHO to group terminals and non-terminals with the same
	type tag together",
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html).
	Even Bison's own parser does this today (see CHAR).

Basically reverts 7928c3e6fbdf47ff81184966cee937e6aa694b94.

* src/scan-gram.l (%nterm): Dedeprecate, but issue a Wyacc warning.
	* tests/input.at: Adjust expectations.
	(Yacc warnings  on symbols): New.
	* src/symtab.c (symbol_class_set): Fix error introduced in
	20b07467938cf880a1d30eb30d6e191843a21fec.

2018-12-11  Eduard Staniloiu  <edi33416@gmail.com>

CI: add dmd support
	* .travis.yml: here.

2018-12-11  Akim Demaille  <akim.demaille@gmail.com>

style: s/non-terminal/nonterminal/
	I personally prefer 'non terminal', or 'non-terminal', but
	'nonterminal' is the common spelling.

* data/glr.c, src/parse-gram.y, src/symtab.c, src/symtab.h,
	* tests/input.at, doc/refcard.tex: here.

2018-12-11  Akim Demaille  <akim.demaille@gmail.com>

style: rename error functions for clarity
	* src/symtab.c (symbol_redeclaration, semantic_type_redeclaration)
	(user_token_number_redeclaration):
	Rename as...
	(complain_symbol_redeclared, complain_semantic_type_redeclared)
	(complain_user_token_number_redeclared):
	this.

2018-12-11  Akim Demaille  <akim.demaille@gmail.com>

parser: improve the error message for symbol class redefinition
	Currently our error messages include both "symbol redeclared" and
	"symbol redefined", and they mean something different.  This is
	obscure, let's make this clearer.

I think the idea between 'definition' vs. 'declaration' is that in the
	case of the nonterminals, the actual definition is its set of rules,
	so %nterm would be about declaration.  The case of %token is less
	clear.

* src/symtab.c (complain_class_redefined): New.
	(symbol_class_set): Use it.
	Simplify the logic of this function to clearly skip its body when the
	preconditions are not met.
	* tests/input.at (Symbol class redefinition): New.

2018-12-11  Akim Demaille  <akim.demaille@gmail.com>

examples: simplify computation of yydebug
	* examples/c/lexcalc/parse.y: here.

2018-12-10  Akim Demaille  <akim.demaille@gmail.com>

C++: support variadic emplace
	Suggested by Askar Safin.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00006.html

* data/variant.hh: Implement.
	* tests/types.at: Check.
	* doc/bison.texi: Document.

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

examples: add a simple Flex+Bison example in C
	Suggested by Askar Safin.
	http://lists.gnu.org/archive/html/bug-bison/2018-12/msg00003.html

* examples/c/lexcalc/Makefile, examples/c/lexcalc/README.md,
	* examples/c/lexcalc/lexcalc.test, examples/c/lexcalc/local.mk,
	* examples/c/lexcalc/parse.y, examples/c/lexcalc/scan.l:
	New.

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

examples: sort them per language and complete them
	Convert some of the READMEs to Markdown, which is now more common, and
	nicely displayed in some git hosting services.

Add missing READMEs and Makefiles.  Generate XML, HTML and Dot files.  Be
	sure to ship the test files.  Complete CLEANFILES to remove all generated
	files.

* examples/calc++: Move into...
	* examples/c++: here.
	* examples/mfcalc, examples/rpcalc: Move into...
	* examples/c: here.

* examples/README.md, examples/c++/calc++/Makefile, examples/c/local.mk,
	* examples/c/mfcalc/Makefile, examples/c/rpcalc/Makefile,
	* examples/d/README.md, examples/java/README.md:
	New files.

* examples/test (medir): Be robust to deeper directory nesting.

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

parser: minor refactoring
	* src/parse-gram.y (symbol.prec): Reuse int.opt.

2018-12-09  Akim Demaille  <akim.demaille@gmail.com>

parser: move checks inside the called functions
	Revamping the handling of the symbols is the grammar is much more
	delicate than I anticipated.  Let's first move things around for
	clarity.

* src/symtab.c (symbol_make_alias): Don't accept to alias
	non-terminals.
	(symbol_user_token_number_set): Don't accept user token numbers
	for non-terminals.
	Don't do anything in case of redefinition, instead of trying to
	update.  The flow is eaier to follow this way.

2018-12-08  Akim Demaille  <akim.demaille@gmail.com>

d: fix double definition of YYSemanticType
	* data/lalr1.d: When moving to b4_user_union_members, it also defines
	b4_tag_seen_flag, so we had two definitions.

2018-12-08  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-12-06  Akim Demaille  <akim.demaille@gmail.com>

parser: fix incorrect condition to raise a syntax error
	* src/parse-gram.y (symbol_def): Fix test.

2018-12-06  Akim Demaille  <akim.demaille@gmail.com>

d: fix use of b4_union_members
	* data/lalr1.d: Use b4_user_union_members instead.

2018-12-06  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* data/variant.hh: here.

2018-12-06  Akim Demaille  <akim.demaille@gmail.com>

java, d: add a Makefile for the example
	* examples/java/Makefile, examples/d/Makefile: New.

2018-12-05  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in ielr.c
	* src/ielr.c: here.

2018-12-05  Akim Demaille  <akim.demaille@gmail.com>

style: scope reduction in lalr.c
	* src/lalr.c: here.

2018-12-05  Akim Demaille  <akim.demaille@gmail.com>

d, java: compute static subtractions
	* data/d.m4, data/java.m4: Use b4_subtract where appropriate.

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

d: add an example
	* examples/d/calc.test, examples/d/calc.y, examples/d/local.mk:

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

d: update the skeleton
	* data/d.m4, data/lalr1.d: Catch up with Bison.
	And actually, also catch up with D.

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

d: add experimental support for the D language
	* configure.ac (ENABLE_D): New.
	* src/getargs.c (valid_languages): Add d.

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

d: add skeleton for the D language
	Contributed by Oliver Mangold.
	https://lists.gnu.org/archive/html/help-bison/2012-01/msg00000.html

* README-D.txt, d-skel.m4, d.m4, lalr1.d: New.

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

examples: regenerate them when version.texi changes
	When we extract the examples from the documentation, %require
	"@value{VERSION}" is replaced with the current version.  If we change
	the git branch, without changing the documentation, the generated
	examples will %require a version of Bison that differs from the actual
	version.

* examples/local.mk (extracted.stamp): Depend on doc/version.texi.

2018-12-04  Akim Demaille  <akim.demaille@gmail.com>

skeletons: start some technical documentation
	* data/README: Convert to Markdown.
	Start documenting some of the macros used in all our skeletons.
	Simplify and fix the documentation of the macros in the skeletons.

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

backend: revamp the handling of symbol types
	Currently it is the front end that passes the symbol types to the
	backend.  For instance:

%token <ival> NUM
	  %type <ival> exp1 exp2
	  exp1: NUM { $$ = $1; }
	  exp2: NUM { $<ival>$ = $<ival>1; }

In both cases, $$ and $1 are passed to the backend as having type
	'ival' resulting in code like `val.ival`.  This is troublesome in the
	case of api.value.type=union, since in that the case the code this:

%define api.value.type union
	  %token <int> NUM
	  %type <int> exp1 exp2
	  exp1: NUM { $$ = $1; }
	  exp2: NUM { $<int>$ = $<int>1; }

because in this case, since the backend does not know the symbol being
	processed, it is forced to generate casts in both cases: *(int*)(&val)`.
	This is unfortunate in the first case (exp1) where there is no reason
	at all to use a cast instead of `val.NUM` and `val.exp1`.

So instead delegate the computation of the actual value type to the
	backend: pass $<ival>$ as `symbol-number, ival` and $$ as
	`symbol-number, MULL`, instead of passing `ival` before.

* src/scan-code.l (handle_action_dollar): Find the symbol the action
	is about, not just its tyye.  Pass both symbol-number, and explicit
	type tag ($<tag>n when there is one) to b4_lhs_value and b4_rhs_value.

* data/bison.m4 (b4_symbol_action): adjust to the new signature to
	b4_dollar_pushdef.

* data/c-like.m4 (_b4_dollar_dollar, b4_dollar_pushdef): Accept the
	symbol-number as new argument.

* data/c.m4 (b4_symbol_value): Accept the symbol-number as new
	argument, and use it.
	(b4_symbol_value_union): Accept the symbol-number as new
	argument, and use it to prefer ready a union member rather than
	casting the union.
	* data/yacc.c (b4_lhs_value, b4_rhs_value): Accept the new
	symbol-number argument.
	Adjust uses of b4_dollar_pushdef.
	* data/glr.c (b4_lhs_value, b4_rhs_value): Adjust.

* data/lalr1.cc (b4_symbol_value_template, b4_lhs_value): Adjust
	to the new symbol-number argument.
	* data/variant.hh (b4_symbol_value, b4_symbol_value_template): Accept
	the new symbol-number argument.

* data/java.m4 (b4_symbol_value, b4_rhs_data): New.
	(b4_rhs_value): Use them.
	* data/lalr1.java: Adjust to b4_dollar_pushdef, and use b4_rhs_data.

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

style: comment and formatting changes
	* data/bison.m4, data/c++.m4, data/glr.c, data/java.m4, data/lalr1.cc,
	* data/yacc.c, src/scan-code.l:
	Fix comments.
	Prefer POS to denote the position of a symbol in a rule, since NUM
	is also used to denote symbol numbers.

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

java: make sure the build dir exists
	* examples/java/local.mk (%D%/Calc.java): here.

2018-12-03  Akim Demaille  <akim.demaille@gmail.com>

c++: don't define variant<S>, directly define semantic_type
	Instead of defining yy::variant<S> and then alias
	yy::parser::semantic_type to variant<sizeof (union_type)>, directly
	define yy::parser::semantic_type.

This model is more appropriate if we want to sit the storage on top of
	unions in C++11.

* data/variant.hh (b4_variant_define): Specialize and inline the
	definition into...
	(b4_value_type_declare): Here.
	Define union_type here.
	* data/lalr1.cc: Adjust.

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

C++: use noexcept and constexpr
	There are probably more opportunities for them.
	So far, I observed no performance improvements.

* data/c++.m4, data/lalr1.cc, data/stack.hh: here.

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

CI: also display the examples' test suite log
	* .travis.yml: here.

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

java: add an example
	* examples/java/Calc.y: New, based on test 495: "Calculator
	parse.error=verbose %locations".
	* examples/java/Calc.test, examples/java/local.mk: New.

* configure.ac (ENABLE_JAVA): New.
	* examples/test (prog): Be ready to run Java programs.

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

style: unsigned int -> unsigned
	See
	https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00027.html

* src/output.c (muscle_insert_unsigned_int_table): Rename as...
	(muscle_insert_unsigned_table): this.

2018-12-01  Akim Demaille  <akim.demaille@gmail.com>

output: restore yyrhs and yyprhs
	This was demanded several times.  See for instance:

- David M. Warme
	  https://lists.gnu.org/archive/html/help-bison/2011-04/msg00003.html

- box12009
	  http://lists.gnu.org/archive/html/bug-bison/2016-10/msg00001.html

Basically, this reverts:

- commit 3d3bc1fe30f356cf674a979409e86ea0f88de4a0
	  Get rid of (yy)rhs and (yy)prhs

- commit d333175f63f402dbadb647175e40ad88bf1defb5
	  Avoid compiler warning.

Note that since these tables are not needed in the generated parsers,
	no skeleton requests them.  This change only brings back their
	definition to M4, making it possible to user-defined skeletons to use
	these tables.

* src/output.c (muscle_insert_item_number_table): Define.
	(prepare_rules): Generate the rhs and prhs tables.

2018-11-30  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-30  Akim Demaille  <akim.demaille@gmail.com>

parser: shorten side-effects on current_type
	* src/parse-gram.y (tag.opt): Don't change current_type.
	Rather, return its value.
	Adjust dependencies.

2018-11-30  Akim Demaille  <akim.demaille@gmail.com>

style: reduce scopes
	* src/symlist.c: here.

2018-11-29  Akim Demaille  <akim.demaille@gmail.com>

tests: don't name C++ files *.c
	* tests/synclines.at (syncline escapes): Here.
	Otherwise, Clang generates an error and skips the test.

2018-11-29  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-11-29  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-29  Akim Demaille  <akim.demaille@gmail.com>

parser: factor the symbol definition
	* src/parse-gram.y (int.opt, string_as_id.opt): New.
	(symbol_def): Use it.

2018-11-29  Akim Demaille  <akim.demaille@gmail.com>

parser: improve location of string alias errors
	* src/parse-gram.y (symbol_def): Pass the right location for symbol_make_alias.
	* tests/regression.at (Duplicate string): Move to...
	* tests/input.at: here.
	(Token collisions): New.

2018-11-29  Akim Demaille  <akim@lrde.epita.fr>

diagnostics: complain about Bison directives when -Wyacc
	* src/complain.h, src/complain.c (bison_directive): New.
	* src/scan-gram.l (BISON_DIRECTIVE): New.
	Use it for Bison extensions.

2018-11-28  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: here.

2018-11-27  Akim Demaille  <akim.demaille@gmail.com>

style: fix quotation in the test suite
	* tests/input.at: here.

2018-11-27  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-27  Akim Demaille  <akim.demaille@gmail.com>

%nterm: do not accept character literals
	Reported by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html

* src/complain.h: Formatting change.
	* src/parse-gram.y (id): Reject character literals used in a context
	for non-terminals.
	* tests/input.at (Invalid %nterm uses): Check that.

2018-11-27  Akim Demaille  <akim.demaille@gmail.com>

%nterm: do not accept numbers nor string alias
	Reported by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html

* src/parse-gram.y (symbol_def): Refuse string aliases and numbers
	for non-terminals.
	(prologue_declaration): Recover from errors ended with ';'.
	* tests/input.at (Invalid %nterm uses): New.

2018-11-27  Akim Demaille  <akim.demaille@gmail.com>

TODO: update

2018-11-26  Akim Demaille  <akim.demaille@gmail.com>

doc: formatting changes
	* doc/bison.texi: Here.

2018-11-26  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* tests/testsuite.at: here.

2018-11-26  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update ignores

2018-11-26  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update to use its bitsets
	Bison's bitset were moved to gnulib.

* lib/abitset.c, lib/abitset.h, lib/bbitset.h, lib/bitset.c,
	* lib/bitset.h, lib/ebitset.c, lib/ebitset.h, lib/lbitset.c,
	* lib/bitset_stats.c, lib/bitset_stats.h, lib/bitsetv-print.c,
	* lib/bitsetv-print.h, lib/bitsetv.c, lib/bitsetv.h,
	* lib/lbitset.h, lib/vbitset.c, lib/vbitset.h:
	Remove.

* gnulib: Update.
	* bootstrap.conf, lib/local.mk: Adjust.

2018-11-25  Akim Demaille  <akim.demaille@gmail.com>

gnulib: use conditional dependencies
	* bootstrap.conf: here.

2018-11-25  Akim Demaille  <akim.demaille@gmail.com>

CI: run on xenial
	Xenial (Ubuntu 16.04) is finally available on Travis.  Let's use it.

Among the changes:

- Automake 1.14.1 -> 1.15.0
	- Doxygen  1.8.6  -> 1.8.11
	- Flex     2.5.35 -> 2.6.0, with plenty of new compiler warnings
	- Gettext  0.18.3 -> 0.19.7
	- Graphviz 2.36.0 -> 2.38.0
	- Texinfo  5.2    -> 6.1

* .travis.yml: here.

2018-11-25  Akim Demaille  <akim.demaille@gmail.com>

CI: we don't need git show
	* .travis.yml: Don't run it.

2018-11-25  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-25  Akim Demaille  <akim.demaille@gmail.com>

warning: avoid warnings about unreachable code
	Reported by Uxio Prego.
	https://lists.gnu.org/archive/html/help-bison/2018-11/msg00031.html

We also need to move the unreachable 'goto' to a reachable place,
	otherwise clang complains about the code being unreachable anyway.
	See also https://bugs.llvm.org/show_bug.cgi?id=39736.

Interestingly, we don't have to apply that trick to
	`#define YYCDEBUG if (false) std::cerr`, clang does not warn when the
	code comes from macro expansion.

* configure.ac: Use -Wunreachable-code when supported.
	* data/lalr1.cc, data/yacc.c: Pacify clang's warning about `if (0)`
	by using a macro.
	Another possibility was to move this statement to a reachable place.
	* tests/actions.at, tests/c++.at: Avoid generating unreachable code.

2018-11-24  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: avoid generating dead code
	We should probably introduce some struct and functions to deal with
	stack management, rather than open coding it.  yyparse would be much
	nicer to read, and a better model for possible other skeletons.

* data/yacc.c (yyparse::yysetstate): Avoid generating code when
	neither yyoverflow nor YYSTACK_RELOCATE is defined.

2018-11-22  Akim Demaille  <akim.demaille@gmail.com>

%expect-rr: tune the number of conflicts per rule
	Currently on a grammar such as

exp : a '1' | a '2' | a '3' | b '1' | b '2' | b '3'
	    a:
	    b:

we count only one rr-conflict on the `b:` rule, i.e., we expect:

b: %expect-rr 1

although there are 3 conflicts in total.  That's because in the
	conflicted state we count only a single conflict, not three (one for
	each of the lookaheads: '1', '2', '3').

State 0

0 $accept: . exp $end
	        1 exp: . a '1'
	        2    | . a '2'
	        3    | . a '3'
	        4    | . b '1'
	        5    | . b '2'
	        6    | . b '3'
	        7 a: . %empty  ['1', '2', '3']
	        8 b: . %empty  ['1', '2', '3']

'1'       reduce using rule 7 (a)
	        '1'       [reduce using rule 8 (b)]
	        '2'       reduce using rule 7 (a)
	        '2'       [reduce using rule 8 (b)]
	        '3'       reduce using rule 7 (a)
	        '3'       [reduce using rule 8 (b)]
	        $default  reduce using rule 7 (a)

exp  go to state 1
	        a    go to state 2
	        b    go to state 3

See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00106.html.

* src/conflicts.c (rule_has_state_rr_conflicts): Rename as...
	(count_rule_state_sr_conflicts): this.
	DWIM.
	(count_rule_rr_conflicts): Adjust.
	* tests/conflicts.at (%expect-rr in grammar rules)
	(%expect-rr too much in grammar rules)
	(%expect-rr not enough in grammar rules): New.

2018-11-22  Akim Demaille  <akim.demaille@gmail.com>

%expect-rr: fix the computation of the overall number of conflicts
	On a grammar such as

exp: "num" | "num" | "num"

we currently report only one RR conflict, instead of two.

This bug is present since the origins of Bison

commit 08089d5d35ece0c7d41659cc1bc09638e2abb151
	    Author: David MacKenzie <djm@djmnet.org>
	    Date:   Tue Apr 20 05:42:52 1993 +0000

Initial revision

and was preserved in

commit 676385e29c4aedfc05d20daf1ef20cd4ccc84856
	    Author: Paul Hilfinger <Hilfinger@CS.Berkeley.EDU>
	    Date:   Fri Jun 28 02:26:44 2002 +0000

Initial check-in introducing experimental GLR parsing.  See entry in
	       ChangeLog dated 2002-06-27 from Paul Hilfinger for details.

See
	https://lists.gnu.org/archive/html/bison-patches/2018-11/msg00011.html

* src/conflicts.h, src/conflicts.c (count_state_rr_conflicts)
	(count_rr_conflicts): Use only the correct count of conflicts.
	* tests/glr-regression.at: Fix expectations.

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

tests: generate *.output files
	* tests/glr-regression.at: here.

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

%expect: tune the number of conflicts per rule
	Currently on a grammar such as

exp: "number" | exp "+" exp | exp "*" exp

we count only one sr-conflict for both binary rules, i.e., we expect:

exp: "number" | exp "+" exp  %expect 1 | exp "*" exp  %expect 1

although there are 4 conflicts in total.  That's because in the states
	in conflict, for instance that for the "+" rule:

State 6

2 exp: exp . "+" exp
	        2    | exp "+" exp .  [$end, "+", "*"]
	        3    | exp . "*" exp

"+"  shift, and go to state 4
	        "*"  shift, and go to state 5

"+"       [reduce using rule 2 (exp)]
	        "*"       [reduce using rule 2 (exp)]
	        $default  reduce using rule 2 (exp)

we count only a single conflict, although there are two (one on "+"
	and another with "*").

See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00106.html.

* src/conflicts.c (rule_has_state_sr_conflicts): Rename as...
	(count_rule_state_sr_conflicts): this.
	DWIM.
	(count_rule_sr_conflicts): Adjust.
	* tests/conflicts.at (%expect in grammar rules): New.

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-21  Akim Demaille  <akim@lrde.epita.fr>

style: reduce scopes
	* src/conflicts.c, src/reader.c: Minor style changes.

2018-11-21  Paul Hilfinger  <hilfingr@eecs.berkeley.edu>

allow %expect and %expect-rr modifiers on individual rules
	This change allows one to document (and check) which rules participate
	in shift/reduce and reduce/reduce conflicts.  This is particularly
	important GLR parsers, where conflicts are a normal occurrence.  For
	example,

%glr-parser
	    %expect 1
	    %%

...

argument_list:
	      arguments %expect 1
	    | arguments ','
	    | %empty
	    ;

arguments:
	      expression
	    | argument_list ',' expression
	    ;

...

Looking at the output from -v, one can see that the shift-reduce
	conflict here is due to the fact that the parser does not know whether
	to reduce arguments to argument_list until it sees the token AFTER the
	following ','.  By marking the rule with %expect 1 (because there is a
	conflict in one state), we document the source of the 1 overall shift-
	reduce conflict.

In GLR parsers, we can use %expect-rr in a rule for reduce/reduce
	conflicts.  In this case, we mark each of the conflicting rules.  For
	example,

%glr-parser
	    %expect-rr 1

stmt:
	      target_list '=' expr ';'
	    | expr_list ';'
	    ;

target_list:
	      target
	    | target ',' target_list
	    ;

target:
	      ID %expect-rr 1
	    ;

expr_list:
	      expr
	    | expr ',' expr_list
	    ;

expr:
	      ID %expect-rr 1
	    | ...
	    ;

In a statement such as

x, y = 3, 4;

the parser must reduce x to a target or an expr, but does not know
	which until it sees the '='.  So we notate the two possible reductions
	to indicate that each conflicts in one rule.

See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00105.html.

* doc/bison.texi (Suppressing Conflict Warnings): Document %expect,
	%expect-rr in grammar rules.
	* src/conflicts.c (count_state_rr_conflicts): Adjust comment.
	(rule_has_state_sr_conflicts): New static function.
	(count_rule_sr_conflicts): New static function.
	(rule_nast_state_rr_conflicts): New static function.
	(count_rule_rr_conflicts): New static function.
	(rule_conflicts_print): New static function.
	(conflicts_print): Also use rule_conflicts_print to report on individual
	rules.
	* src/gram.h (struct rule): Add new fields expected_sr_conflicts,
	expected_rr_conflicts.
	* src/reader.c (grammar_midrule_action): Transfer expected_sr_conflicts,
	expected_rr_conflicts to new rule, and turn off in current_rule.
	(grammar_current_rule_expect_sr): New function.
	(grammar_current_rule_expect_rr): New function.
	(packgram): Transfer expected_sr_conflicts, expected_rr_conflicts
	to new rule.
	* src/reader.h (grammar_current_rule_expect_sr): New function.
	(grammar_current_rule_expect_rr): New function.
	* src/symlist.c (symbol_list_sym_new): Initialize expected_sr_conflicts,
	expected_rr_conflicts.
	* src/symlist.h (struct symbol_list): Add new fields expected_sr_conflicts,
	expected_rr_conflicts.
	* tests/conflicts.at: Add tests "%expect in grammar rule not enough",
	"%expect in grammar rule right.", "%expect in grammar rule too much."

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-11-21  Akim Demaille  <akim.demaille@gmail.com>

remove ancient lint directives
	* data/c++.m4, data/yacc.c: Remove surprising remains of lint
	directives.

2018-11-21  Jannick  <thirdedition@gmx.net>

doc: calc++: remove ancient fix for flex
	* doc/bison.texi (Calc++ Scanner): Remove fix for Flex 2.5.31-2.5.33.

2018-11-21  Jannick  <thirdedition@gmx.net>

doc: calc++: ignore \r in the scaner
	* doc/bison.texi (Calc++ Scanner): Ignore \r.

2018-11-20  Akim Demaille  <akim.demaille@gmail.com>

style: harmonize the labels of yyparse
	* data/glr.c, data/lalr1.cc, data/yacc.c: Fix indentation and
	other formatting issues.

2018-11-20  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-20  Akim Demaille  <akim.demaille@gmail.com>

style: avoid lengthy actions
	We also lack a consistent naming for directive implementations.
	`directive_skeleton` is too long, `percent_skeleton` is not very nice
	looking, `process_skeleton` looks ambiguous, `do_skeleton` is somewhat
	ambiguous too, but seems a better track.

* src/parse-gram.y (version_check): Rename as...
	(do_require): this.
	(do_skeleton): New.
	Use it.

2018-11-20  Akim Demaille  <akim.demaille@gmail.com>

c++: using macros around user types breaks when they include comma
	We may generate code such as

basic_symbol (typename Base::kind_type t, YY_RVREF (std::pair<int,int>) v);

which, of course, breaks, because YY_RVREF sees two arguments.  Let's
	not play tricks with _VA_ARGS__, I'm unsure about it portability.
	Anyway, I plan to change more things in this area.

Reported by Sébastien Villemot.
	http://lists.gnu.org/archive/html/bug-bison/2018-11/msg00014.html

* data/variant.hh (b4_basic_symbol_constructor_declare)
	(b4_basic_symbol_constructor_define): Don't use macro on user types.
	* tests/types.at: Check that we support pairs.

2018-11-16  Akim Demaille  <akim.demaille@gmail.com>

README: update

2018-11-16  Akim Demaille  <akim.demaille@gmail.com>

tests: remove duplicate definition
	Probably imported by 6d58c632025cb6928a90e4176577982bfb9c3981, a merge
	commit.

* tests/atlocal.in (POSIXLY_CORRECT_IS_EXPORTED): Define it once.

2018-11-16  Akim Demaille  <akim.demaille@gmail.com>

glr.c: fix use of _Noreturn
	In C++, [[noreturn]] must not be between "static" and the rest of the
	function signature, it must precede it.  C's _Noreturn does not seem
	to have such a constraint, but it is therefore compatible with the C++
	constraint.  Since we #define _Noreturn as [[noreturn]] is modern C++,
	be sure to push the _Noreturn first.

Unfortunately this was not caught by the test suite, because it always
	loads config.h first, and config.h contains another definition of
	_Noreturn that does not use [[noreturn]], and hides ours.  That's
	probably a sign we should avoid always loading config.h.

* data/glr.c (yyFail, yyMemoryExhausted): here.

2018-11-16  Akim Demaille  <akim.demaille@gmail.com>

tests: run the api.value.type tests for all C++ standards
	* tests/local.at (AT_LANG_FOR_EACH_STD): New.
	(AT_REQUIRE_CXX_VERSION): Rename as...
	(AT_REQUIRE_CXX_STD): this.
	Accept an argument for what to do when the requirement is not met.
	* tests/types.at (api.value.type): Check all the C++ stds.

2018-11-16  Akim Demaille  <akim.demaille@gmail.com>

CI: split the ASAN job in two
	The following commit introduce even more compilations/runs than
	before, and with ASAN on, we go beyond to 50min credit from Travis.

* .travis.yml (Clang 7 libc++ and ASAN): Split in two.

2018-11-14  Akim Demaille  <akim.demaille@gmail.com>

c++: use YY_CPLUSPLUS
	* data/c++.m4: here.

2018-11-13  Akim Demaille  <akim.demaille@gmail.com>

c++: factor the handling of __cplusplus into YY_CPLUSPLUS
	* data/c++.m4 (b4_cxx_portability): Define it.
	Use it.
	* data/lalr1.cc, data/variant.hh: Use it.

2018-11-13  Akim Demaille  <akim.demaille@gmail.com>

style: formatting changes
	* src/scan-gram.l: here.

2018-11-13  Akim Demaille  <akim.demaille@gmail.com>

tests: clarify some magic constant
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): here.

2018-11-13  Akim Demaille  <akim.demaille@gmail.com>

examples: remove useless includes
	* examples/c++/variant-11.yy, examples/c++/variant.yy: here.
	Fix warning when storing a long into an int.

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

tests: compile the C++ examples with warnings
	* examples/c++/local.mk: Pass $(WARN_CXXFLAGS_TEST).

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

parser: deprecate %error-verbose
	It is unfortunate that %error_verbose was properly diagnosed as
	obsoleted by "%define parse.error verbose", but %error-verbose was
	not.

* src/parse-gram.y (%error-verbose): Remove support.
	* src/scan-gram.l: Do it here instead, with a warning.
	* tests/input.at (Deprecated directives): Check it.

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

tests: migrate from %error-verbose to %define parse.error verbose
	* tests/actions.at, tests/c++.at, tests/conflicts.at,
	* tests/cxx-type.at, tests/existing.at, tests/glr-regression.at,
	* tests/headers.at, tests/input.at, tests/java.at, tests/javapush.at,
	* tests/local.at, tests/regression.at, tests/skeletons.at,
	* tests/torture.at:
	Here.

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

parser: deprecate %nterm
	It has several weaknesses.
	Reported by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html

* src/scan-gram.l: here.

2018-11-12  Akim Demaille  <akim.demaille@gmail.com>

tests: fix syncline tests
	These tests are skipped with GCC:

"\"".c:1:5: error: function declaration isn't a prototype [-Werror=strict-prototypes]
	     int main() { return 0; }
	         ^~~~

* tests/synclines.at: Stop writing C++ in C.
	* tests/local.at: Formatting changes.

2018-11-11  Akim Demaille  <akim.demaille@gmail.com>

NEWS: expected features of Bison 3.3

2018-11-11  Akim Demaille  <akim.demaille@gmail.com>

yacc: reduce scope in push mode
	* data/yacc.c (yypull_parse): Here.

2018-11-11  Akim Demaille  <akim.demaille@gmail.com>

style: comment changes
	* data/c++.m4, data/glr.c, data/lalr1.java, data/yacc.c
	(yytranslate, YYTRANSLATE): Harmonize comments.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

c++: simplify a default construction
	* data/lalr1.cc (yytnamerr_): here.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

reader: no longer accept %define variable names in quotes
	It was never documented.

* src/parse-gram.y (variable): Here.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

dogfooding: use api.value.type union
	* src/parse-gram.y (api.value.type): Set to union.
	Replace occurrences of %union with explicit %types.
	* src/scan-gram.l: Adjust yylval's field names.
	(RETURN_VALUE): No longer needs the Field argument.
	Use it more.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

djgpp: remove
	Support for DJGPP was announced to be removed in the NEWS of Bison
	3.1 (2018-08-27) unless someone expressed interest.  There was no answer.

* djgpp: Remove.
	* NEWS, Makefile.am, cfg.mk, po/POTFILES.in: Adjust.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

scanner: simplify use of gettext
	* src/scan-gram.l (unexpected_end): Leave the actual call to gettext
	to the caller.

2018-11-10  Akim Demaille  <akim.demaille@gmail.com>

style: clean up the scanner and parser
	* src/scan-gram.l: Formatting changes.
	Add "missing" assertion for symmetry.
	* src/parse-gram.y: Formatting changes.

2018-11-09  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2018-11-09  Akim Demaille  <akim.demaille@gmail.com>

version 3.2.1
	* NEWS: Record release date.

2018-11-08  Akim Demaille  <akim.demaille@gmail.com>

build: fix issues in the generated tarball
	Reported by Andre da Costa Barros.
	https://savannah.gnu.org/patch/?9716

* examples/calc++/local.mk: We no longer generate position.hh and
	stack.hh.  Leaving them here triggers their concurrent generation,
	which fails.
	(%C%_calc___CPPFLAGS): Fix the extracted headers in the source tree.
	* examples/mfcalc/local.mk (%C%_mfcalc_CPPFLAGS): Ditto.

2018-11-07  Akim Demaille  <akim.demaille@gmail.com>

build: fix typo
	Reported by Horst Von Brand.
	https://savannah.gnu.org/support/?109580

* examples/local.mk (.PHOMY): Rename as...
	(.PHONY): this.

2018-11-06  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-11-06  Akim Demaille  <akim.demaille@gmail.com>

examples: ship them
	Currently, the examples are extracted on the user's side.
	Unfortunately, that requires that the user has Perl, which is
	otherwise not needed for Bison.  Let's ship the examples instead.

The examples were handled this way so that we could depend on
	configure flags: if --enable-gcc-warnings is passed, it is understood
	as "I'm a maintainer", so the examples are generated with `#line`s.
	Regular users should not see them, so they are now unconditionally
	removed when rolling a tarball.

Reported by Mike Frysinger.
	https://lists.gnu.org/archive/html/bison-patches/2015-04/msg00000.html

* examples/local.mk: Ship all the extracted files.
	(examples-unline): New.
	Make sure that the generated tarballs do not contain the #lines.

2018-11-06  Akim Demaille  <akim.demaille@gmail.com>

build: minor fixes in doc/
	* doc/local.mk: Consistently use *.tmp for temporary files.
	Fix comments.

2018-11-05  Akim Demaille  <akim.demaille@gmail.com>

CI: maximize chances of errors sooner
	* .travis.yml: Try clang and icc soon, so that we don't have to wait
	for the end of the run to know that they fail.

2018-11-04  Akim Demaille  <akim.demaille@gmail.com>

c++: improve the generated documentation
	* data/lalr1.cc, data/location.cc: Improve documenting comments.
	* tests/c++.at (Doxygen Documentation): Fix AT_BISON_OPTION_PUSHDEFS,
	so that the generated yyerror is correct.
	* tests/c++.at, tests/headers.at: Prefer %empty.

2018-11-04  Akim Demaille  <akim.demaille@gmail.com>

tests: don't fail if the C++ compiler does not work
	Also, make sure that `make dist` generates a correct tarball even if
	the C++ compiler does not work.

Reported by Nelson H. F. Beebe.

* m4/cxx.m4 (BISON_CXX_WORKS): Define to true/false instead of
	true/exit 77.  The latter is too dangerous to use (it directly quits).
	(ENABLE_CXX): New name for the Automake conditional, for consistency
	with ENABLE_CXX11 etc.
	* tests/local.at (AT_COMPILE, AT_COMPILE_CXX): Adjust to the new
	semantics of BISON_CXX_WORKS.
	* examples/c++/local.mk: Skip the variant test if C++ does not work.
	* examples/calc++/local.mk: Likewise.

2018-11-04  Akim Demaille  <akim.demaille@gmail.com>

tests: don't disable C++ warnings in C files
	This triggers error with GCC.
	See eff6739124c61bb5660d78453210d1d6a17d30e7.

* tests/testsuite.h: Disable -Wzero-as-null-pointer-constant only for
	C++.

2018-11-04  Akim Demaille  <akim.demaille@gmail.com>

c++: workaround portability issue
	On some systems (x86_64-pc-solaris2.11), with Developer Studio 12.5's
	CC, we get:

".../include/CC/Cstd/vector.cc", line 127: Error: Cannot assign const yy::parser::stack_symbol_type to yy::parser::stack_symbol_type without "yy::parser::stack_symbol_type::operator=(const yy::parser::stack_symbol_type&)";.
	    ".../include/CC/Cstd/vector", line 475:     Where: While instantiating "std::vector<yy::parser::stack_symbol_type>::__insert_aux(yy::parser::stack_symbol_type*, const yy::parser::stack_symbol_type&)".
	    ".../include/CC/Cstd/vector", line 475:     Where: Instantiated from non-template code.
	    1 Error(s) detected.

Don't expect __cplusplus to be always defined.  If it's not, consider
	this is C++98.

Reported by Nelson H. F. Beebe.

* data/c++.m4, data/lalr1.cc, examples/c++/variant.yy, tests/local.at,
	* tests/testsuite.h:
	An undefined __cplusplus means pre C++11.

2018-11-03  Akim Demaille  <akim.demaille@gmail.com>

tests: work around getopt portability issues
	On some systems, we don't use our getopt.  As a consequence the error
	messages vary:

$ bison --skeleton
	    bison: option requires an argument -- skeleton
	    Try 'bison --help' for more information.

instead of

bison: option '--skeleton' requires an argument
	    Try 'bison --help' for more information.

Reported by Jannick and Nelson H. F. Beebe.
	https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00140.html

* tests/input.at (Invalid number of arguments): work around getopt
	portability issues.

2018-11-03  Akim Demaille  <akim.demaille@gmail.com>

doc: -Wzero-as-null-pointer-constant was added to GCC 4.7
	It is not supported by previous versions.
	See https://www.gnu.org/software/gcc/gcc-4.7/changes.html
	Reported by Nelson H. F. Beebe.

* doc/bison.texi (Calc++ Scanner): here.

2018-11-02  Adam Sampson  <ats@offog.org>

examples: #include <cstring> in calc++
	strerror is defined by <string.h>, and recent versions of GNU libstdc++
	no longer include this automatically from <string>.

2018-10-31  Akim Demaille  <akim.demaille@gmail.com>

c: provide a definition of _Noreturn that works for C++
	On Solaris, GCC 7.3 defines:

-std=c++14  -std=c++17
	    __cplusplus       201402L       201703L
	    __STDC_VERSION__  199901L       201112L

So the current #definition of _Noreturn sees that 201112 <=
	__STDC_VERSION__, i.e., that C11 is supported, so it expects _Noreturn
	to be supported.  Apparently it is not.

This is only for C++, the test suite works for C.  However, the test
	suite does not try several C standards, maybe we should...

http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00064.html

* data/c.m4 (b4_attribute_define): Define _Noreturn as [[noreturn]] in
	modern C++.

2018-10-30  Akim Demaille  <akim.demaille@gmail.com>

c: update the definition of _Noreturn
	Does not work on Solaris 11.3 x86/64:

479. c++.at:1293: testing C++ GLR parser identifier shadowing ...
	    ======== Testing with C++ standard flags: '-std=c++17'
	    ./c++.at:1332: $BISON_CXX_WORKS
	    stderr:
	    stdout:
	    ./c++.at:1332: $CXX $CXXFLAGS $CPPFLAGS  $LDFLAGS -o input input.cc $LIBS
	    stderr:
	    input.cc:837:8: error: '_Noreturn' does not name a type
	     static _Noreturn void
	            ^~~~~~~~~
	    input.cc:845:8: error: '_Noreturn' does not name a type
	     static _Noreturn void
	            ^~~~~~~~~

Reported by Kiyoshi Kanazawa.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00051.html

* data/c.m4 (b4_attribute_define): Use the snippet which is currently
	in gnulib's m4/gnulib-common.m4 (which seems a little more advanced
	than lib/_Noreturn.h).

2018-10-30  Akim Demaille  <akim.demaille@gmail.com>

tests: don't expect the shell to support 'local'
	It doesn't work on Solaris 11.3 x86/64.
	Reported by Kiyoshi Kanazawa.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00051.html

* examples/test: Don't use 'local'.

2018-10-30  Akim Demaille  <akim.demaille@gmail.com>

bitset: fix warning
	Reported by Hans Åberg.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00047.html

* lib/bitset.c (bitset_count_): here.

2018-10-30  Akim Demaille  <akim.demaille@gmail.com>

build: fix use of gnulib Make variables
	Reported by Kiyoshi Kanazawa.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00048.html

* lib/local.mk (lib_libbison_a_LIBADD): Merge into...
	* src/local.mk (src_bison_LDADD): here.

2018-10-29  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2018-10-29  Akim Demaille  <akim.demaille@gmail.com>

version 3.2
	* NEWS: Record release date.

2018-10-29  Akim Demaille  <akim.demaille@gmail.com>

build: don't depend on the libc to generate bison.help
	The "Report translation bugs to..." part of --help is issued only on
	glibc systems.  So if the tarball is not wrapped on such a system, and
	used on such a system (or the converse), then bison.help will differ
	on the user's system, and help2man will be called to update bison.1.

But help2man should not be a requirement.

Reported by Alexandre Duret-Lutz.

* doc/local.mk (doc/bison.help): Remove the possible doc about
	translation bugs.
	Pass LC_ALL=C, as reported in src/getargs.c's usage().
	(doc/cross-options.texi): Use bison.help instead of calling bison
	--help.

2018-10-29  Akim Demaille  <akim.demaille@gmail.com>

c++: always issue the "generated by" message
	Some users rely on this sentence to know that the file can be ignored.
	Reported by Alexandre Duret-Lutz.

* data/bison.m4 (b4_generated_by): New.
	(b4_copyright): Use it.
	* data/location.cc, data/stack.hh: Use it too, for the stub files
	(position.hh and stack.hh).

2018-10-28  Akim Demaille  <akim.demaille@gmail.com>

cfg.mk: remove exceptions for timevar
	They appear to be no longer needed.

* cfg.mk: here.

2018-10-28  Akim Demaille  <akim.demaille@gmail.com>

style: clean up src/AnnotationList.c
	* src/AnnotationList.c: Reduce scopes.

2018-10-28  Akim Demaille  <akim.demaille@gmail.com>

style: clean up print.c
	* src/print.c: Reduce scopes.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up bbitset.h
	* lib/libiberty.h: Inline in...
	* lib/bbitset.h: here.
	* lib/local.mk: Adjust.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up bitset.h
	* lib/bitset.h: Fix include order.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up vbitset.c
	* lib/vbitset.c: Reduce scopes, etc.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up lbitset.c
	* lib/lbitset.c: Reduce scopes, etc.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up ebitset.c
	* lib/ebitset.c: Reduce scopes, etc.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up bitset_stats.c
	* lib/bitset_stats.c: Reduce scopes, etc.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up bitset.c
	* lib/bitset.c: Reduce scopes, etc.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

bitset: clean up abitset.c
	* lib/abitset.c: Reduce scopes, etc.

2018-10-27  Jannick  <thirdedition@gmx.net>

xml2dot.xsl: fix typos in comments

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

doc: fix distcheck
	The extracted example, simple.yy, does not use %require "3.2", so it
	generates a stack.hh, which breaks distcheck.

* doc/bison.texi: Fix it.

2018-10-27  Akim Demaille  <akim.demaille@gmail.com>

NEWS: prepare for 3.2

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

tests: beware of Windows file name constraints
	Don't expect to be able to build a file named '"\"".y' (6 characters)
	on Windows.

Reported by Jannick.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00042.html

* tests/synclines.at (syncline escapes): Skip if we failed to create
	the file.

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

tests: fix invocation of m4_map
	* tests/actions.at, tests/synclines.at: m4_map takes a list of
	arguments in $2, m4_map_args takes arguments in $2, $3, etc.

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check
	* README, doc/bison.texi, examples/README, examples/c++/README: here.

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

examples: add a Makefile for C++ short examples
	* examples/c++/Makefile: New.
	* examples/c++/local.mk, examples/c++/README: Adjust.

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

doc: some improvements
	* doc/bison.texi (Calc++ Scanner): Show how exception can be thrown
	from auxiliary functions.
	Clarify the meaning of the various flex %options we use.
	Get rid of a warning.
	(Calc++ Parsing Driver): Use the parser as a functor.

2018-10-26  Akim Demaille  <akim.demaille@gmail.com>

examples: check the errors
	* examples/test (run): Check stderr, unless -noerr is passed.
	* examples/calc++/calc++.test, examples/mfcalc/mfcalc.test: Check
	errors.

2018-10-25  Akim Demaille  <akim.demaille@gmail.com>

doc: minor fixes
	* doc/bison.texi: Simplify wording.
	Fix Texinfo error.
	(Complete Symbols): Handle the token EOF.
	(Calc++ Parser): In the modern C++ world, prefer assignment to swap.
	(Strings are Destroyed): Prefer an explicit 'continue' to a comment.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

configure: quit on trying to get ICC and Flex be friends
	The CI is using Flex 2.5.35.  And ICC is too picky for it.  Let's stop
	making these warnings errors.  I wish I could disable them in the
	source files using the ICC version and the Flex version, but ICC's
	pragma support is unclear, and I'm tired of fighting it.

* configure.ac (FLEX_SCANNER_CXXFLAGS): Make warnings warnings.
	* examples/c++/local.mk: Comment changes.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

doc: mention earlier how to disable the generation of location.hh
	Suggested by Victor Khomenko.

* doc/bison.texi (C++ Bison Interface): Here.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

c++: std::to_string is available in C++11
	Reported by Victor Khomenko.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00033.html

* doc/bison.texi, examples/c++/variant-11.yy: Use std::to_string
	instead of ours.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

examples: move the variant examples into the C++ directory
	* examples/variant-11.test examples/variant-11.yy,
	* examples/variant.test examples/variant.yy:
	Move into examples/c++/.
	* examples/c++/README: New.
	* examples/README, examples/c++/local.mk, examples/local.mk:
	Adjust.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

doc: an introductory example for C++
	Suggested by Victor Khomenko.
	http://lists.gnu.org/archive/html/bug-bison/2018-08/msg00037.html

* doc/bison.texi (A Simple C++ Example): New.
	* examples/c++/local.mk, examples/c++/simple.test: New.
	Extract, check, and install this new example.
	* examples/local.mk: Adjust.
	* examples/test: Adjust to the case where the dirname differs
	from the test name.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

build: remove a few copies of the Copyright from the generated Makefile
	* build-aux/local.mk, cfg.mk, examples/calc++/local.mk,
	* examples/local.mk, examples/mfcalc/local.mk,
	* examples/rpcalc/local.mk, lib/local.mk, src/local.mk,
	* tests/local.mk:
	Use Automake comments so that we don't get a copy of each in the
	generated Makefile.

2018-10-24  Akim Demaille  <akim.demaille@gmail.com>

c++: make operator() an alias to the parse function
	* data/glr.cc, data/lalr1.cc (operator()): New.
	* doc/bison.texi: Update.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

build: enable more warnings during tests
	Prompted by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00018.html

* configure.ac: here.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: work around strange typing issues
	On the CI, both GCC and Clang report:

src/parse-gram.c: In function 'yy_lac':
	    src/parse-gram.c:1479:29: error: format '%hd' expects argument of type 'int',
	       but argument 3 has type 'yytype_int16 {aka long int}' [-Werror=format=]
	             YYDPRINTF ((stderr, " G%hd", yystate));
	                                 ^

Although yytype_int16 is supposed to be a short int, not a long int.
	This must be explored.

* data/yacc.c (yy_lac): Work around typing issue.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: don't define _Noreturn uselessly
	Clang warns:

aux/x.h:97:11: error: macro name is a reserved identifier
	           [-Werror,-Wreserved-id-macro]
	     #  define _Noreturn YY_ATTRIBUTE ((__noreturn__))

Reported by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00024.html

* data/c.m4 (b4_attribute_define): Don't define _Noreturn unconditionally.
	* data/glr.c: Ask for _Noreturn.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

pacify ICC 16.0.3 20160415
	Found on the CI.

yacc.c:

error #2259: non-pointer conversion from "int" to "yytype_int16={short}" may lose significant bits
	        yystate = (0 <= yyi && yyi <= YYLAST && yycheck[yyi] == *yyesp
	                ^

glr.c:

error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits
	      yybool yynormal YY_ATTRIBUTE_UNUSED = (yystackp->yysplitPoint == YY_NULLPTR);
	                                          ^

error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits
	      return yypact_value_is_default (yypact[yystate]);
	             ^

error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits
	      return 0 < yyaction;
	             ^

error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits
	      return yyaction == 0;
	             ^

error #2259: non-pointer conversion from "int" to "yybool={unsigned char}" may lose significant bits
	      yystackp->yytops.yylookaheadNeeds[yys] = yychar != YYEMPTY;
	                                                     ^

* data/glr.c, data/yacc.c: Avoid these warnings.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

flex: work around more warnings
	* doc/bison.texi: here.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

examples: clang 5 defines nullptr as a macro
	* examples/variant.yy: So don't do it.
	* examples/variant-11.yy: Fix comment.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

glr.c: be strict about types
	* data/glr.c: Don't use `foo |= bar` with foo and bar being yybool:
	the result appears to be an int, not a yybool.
	Use yybool where appropriate.
	Add casts where needed.

2018-10-23  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: fix warnings about integral types
	Reported by Derek Clegg.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00018.html

Rather than adding casts, we should be more careful with types.  For
	instance yystate should be a yytype_int16.  But currently we can't: it
	is also used sometimes for storing other things that state numbers.

* data/yacc.c (yyparse): Add missing casts.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

yacc.c: clarify the computation of yystate
	The yacc.c skeleton is old, and was using many tricks to save
	registers.  Today's register allocators can do this themselves.  Let's
	keep the code simpler to read and let compilers do their job.

* data/yacc.c: Avoid using yystate for different types of content.
	An inline function would be better, but doing this portably will be
	a problem.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

printf returns a signed int
	* tests/local.at: Adjust location_print's signature.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

tests: be strict about types
	* tests/actions.at, tests/c++.at, tests/cxx-type.at,
	* tests/glr-regression.at, tests/local.at, tests/torture.at,
	* tests/types.at:
	Pay stricter attention to types to avoid warnings.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

c++: fix signedness issues
	* data/lalr1.cc, data/stack.hh: The callers of stack use int, while
	stack is based on size_type.  Add overloads to avoid warnings.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

c++: minor changes
	* data/lalr1.cc: Fix oldish comment.
	* data/stack.hh: Prefer typename for type names.
	Use size() instead of duplicating it.
	* examples/variant-11.yy, examples/variant.yy (yylex): Use int,
	as this is the type of the semantic value.

2018-10-22  Akim Demaille  <akim.demaille@gmail.com>

all: display a clear warning about private macros
	* data/bison.m4 (b4_disclaimer): New.
	* data/glr.c, data/glr.cc, data/lalr1.cc, data/yacc.c: Use it.

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

c++: minor simplification
	* data/stack.hh: Prefer a default argument value to two constructors.

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

all: avoid useless comments and #lines
	Currently we emit useless code for places where we might issue user
	content, but there is none.  This commit avoids this.  Besides, some
	of the comments looked like implementation details ("Copy the first
	part of user declarations"), rather than made for the reader of the
	result ("First part of user prologue").

On Bison's parse-gram.c we get:

@@ -76,10 +76,6 @@
	     #define yynerrs         gram_nerrs

-/* Copy the first part of user declarations.  */
	    -
	    -#line 82 "src/parse-gram.c" /* yacc.c:339  */
	    -

* data/bison.m4 (b4_define_user_code): Accept a comment to document
	the section.
	Do not emit any code if the content is empty.
	Adjust callers to not emit the comment.
	Do not
	* data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, data/yacc.c:
	Adjust.

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

tests: refactor
	* tests/actions.at, tests/synclines.at: Prefer iteration to
	copy-paste.

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

tests: rename AT_SKEL_CC_IF/AT_SKEL_JAVA_IF as AT_CXX_IF/AT_JAVA_IF
	The previous name is too obscure, and the other macros for C++ use
	CXX, not CC.

* tests/local.at (AT_SKEL_CC_IF, AT_SKEL_JAVA_IF): Rename as...
	(AT_CXX_IF, AT_JAVA_IF): these.
	Adjust callers.

2018-10-21  Akim Demaille  <akim.demaille@gmail.com>

c++: check that emplace for rvalues works
	See the previous commit.

* tests/local.at (AT_REQUIRE_CXX_VERSION): New.
	* tests/types.at (api.value.type): Check emplace in C++14.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

c++: prefer a perfect forwarding version of variant's emplace
	* data/variant.hh (emplace): In modern C++, use only a perfect
	forwarding version.
	And use it.
	* doc/bison.texi: Document it.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

c++: prefer 'emplace' to 'build'
	When we introduced variants in Bison, C++ did not have the 'emplace'
	functions, and we chose 'build'.  Let's align with modern C++ and
	promote 'emplace' rather than 'build'.

* data/lalr1.cc, data/variant.hh (emplace): New.
	(build): Deprecate in favor of emplace.
	* doc/bison.texi: Adjust.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

%printer: promote yyo rather than yyoutput
	* doc/bison.texi: Promote yyo rather than yyoutput.

* data/c.m4, data/glr.cc, tests/types.at, tests/calc.at,
	tests/regression.at: Adjust.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

doc: improve the C++ section
	* doc/bison.texi (C++ Parser): file.hh and location.hh are no longer
	mandatory.
	Various minor fixes.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

doc: reorder C++ sections
	* doc/bison.texi (C++ Parser Interface): Document before semantic_type
	and location_type.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

git: don't ignore auxiliary Texinfo files
	As a matter of fact, I think it is wrong to gitignore generated files
	that belong to the build tree.  There should be the strict minimum,
	and it's up to people that build in place to adjust their own
	~/.gitignore.

* doc/.gitignore: here.
	Remove files we no longer produce (thanks to texi2dvi).

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

c++: do not exhibit private macros
	* examples/variant-11.yy: here.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

c++: don't obfuscate std::move when not needed
	* data/lalr1.cc, data/variant.hh: Avoid macros that depend on the
	version of C++ when not needed.

2018-10-20  Akim Demaille  <akim.demaille@gmail.com>

build: add missing gnulib libs
	Reported by Denis Excoffier.

* lib/local.mk, src/local.mk: here.

2018-10-18  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2018-10-18  Akim Demaille  <akim.demaille@gmail.com>

version 3.1.91
	* NEWS: Record release date.

2018-10-18  Akim Demaille  <akim.demaille@gmail.com>

NEWS: update

2018-10-18  Akim Demaille  <akim.demaille@gmail.com>

examples: force stack resizing with unique_ptr
	In the previous commit we fixed a problem when the C++ stack was
	resized.  The test was using ints.  Let's add a test with someone
	quite touchy: unique_ptr

* examples/variant-11.yy: Accept an argument, which is the number of
	numbers to send to the parser.
	* examples/variant-11.test: Check with many numbers.

2018-10-18  Akim Demaille  <akim.demaille@gmail.com>

lalr1.cc: fix stack symbol move
	In some casing, once we moved a stack symbol, we forget to mark the
	source stack symbol as emptied.  As a consequence, it may be destroyed
	a second time.

This happens when the stack has to be resized.

* data/lalr1.cc (stack_symbol_type::stack_symbol_type): Record that
	the source was emptied.
	(stack_symbol_type::operator=): Likewise.
	* tests/c++.at (C++ Variant-based Symbols Unit Tests): Force the stack
	to be resized.  Check its content.

2018-10-17  Akim Demaille  <akim.demaille@gmail.com>

doc: improve the doc of the examples
	* examples/README: here.

2018-10-17  Akim Demaille  <akim.demaille@gmail.com>

reader: recognize C++ even when it's not lalr1.cc or glr.cc
	* src/reader.c (grammar_rule_check_and_complete): If a user uses her
	own skeleton but sets the language to C++, recognize it as C++.

2018-10-17  Akim Demaille  <akim.demaille@gmail.com>

maint: post-release administrivia
	* NEWS: Add header line for next release.
	* .prev-version: Record previous version.
	* cfg.mk (old_NEWS_hash): Auto-update.

2018-10-17  Akim Demaille  <akim.demaille@gmail.com>

version 3.1.90
	* NEWS: Record release date.

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

examples: don't generate the position/stack files
	* examples/variant-11.yy, examples/variant.yy: Require 3.2.

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

pacify syntax-checks
	* lib/lbitset.c, tests/c++.at: here.
	* cfg.mk: Add exceptions.

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

generate the default action only for C++
	This commit adds restrictions to what was done in
	01898726e27c8cf64f8fcea7f26f8ce62f3f5cf2 [1].

Rici Lake [2] has shown that it's risky to disable the pre-action, at
	least now.  Also, generating the default $$ = $1 action can have bad
	effects in some cases [3].

The original change [1] was prompted for C++.  Let's try it there
	only, for a start.  We could restrict it further to lalr1.cc with
	variants, but we need to see in the wild how this change behaves.  And
	it is not unreasonable to expect grammar files in C++ to behave better
	wrt types.

See
	[1] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00050.html
	[2] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00061.html
	[3] https://lists.gnu.org/archive/html/bison-patches/2018-10/msg00066.html

* src/getargs.c: Style changes.
	* src/reader.c (grammar_rule_check_and_complete): Complete only for
	C++.

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

regen

2018-10-16  Akim Demaille  <akim.demaille@gmail.com>

C++: let %require "3.2" disable the generation of obsolete files
	The files stack.hh and position.hh are deprecated.  Rather than
	devoting specify %define variables to discard them (api.position.file
	and api.stack.file), and rather than having to use special rules when
	api.location.file is used, let's simply decide that from %require
	"3.2" onwards, these files will not be generated.

The only noticeable thing here is that, in order to be able to check
	the behavior of %require "3.2", to have this version (which is still
	3.1-*) to accept %require "3.2".

* src/gram.h, src/gram.c (required_version): New.
	* src/parse-gram.y (version_check): Set it.
	* src/output.c (prepare): Pass it m4.
	* data/bison.m4 (b4_required_version_if): Receive it and use it.
	* data/location.cc, data/stack.hh: Replace the api.*.file with only
	required version comparison.
	* tests/input.at: No longer check api.stack.file and api.position.file.
	* NEWS, doc/bison.texi: Don't mention them.
	Document the %require 3.2 behavior.
	* tests/output.at: Use %require 3.2 instead.

2018-10-15  Akim Demaille  <akim.demaille@gmail.com>

java: bump to Java SE 7
	macOS 10.14 no longer supports versions of Java earlier than 5.
	And Java 6 will be deprecated by the end of this year.  So let's move
	our requirement to Java 7.
	Reported by Yu Yijun.
	https://lists.gnu.org/archive/html/bug-bison/2018-09/msg00060.html
	Suggested by Paul Eggert and Bruno Haible.
	http://lists.gnu.org/archive/html/bug-gnulib/2018-10/msg00094.html

* configure.ac: Require Java 7, both compiler and runtime.

2018-10-15  Akim Demaille  <akim.demaille@gmail.com>

doc: do not advertise %nterm
	Reported by Rici Lake.
	http://lists.gnu.org/archive/html/bug-bison/2018-10/msg00000.html

* NEWS, doc/bison.texi: here.

2018-10-14  Akim Demaille  <akim.demaille@gmail.com>

generate the default semantic action
	Currently, in C, the default semantic action is implemented by being
	always run before running the actual user semantic action.  As a
	consequence, when the user action is run, $$ is already set as $1.

In C++ with variants, we don't do that, since we cannot manipulate the
	semantic value without knowing its exact type.  When variants are
	enabled, the only guarantee is that $$ is default contructed and ready
	to the used.

Some users still would like the default action to be run with
	variants.  Frank Heckenbach's parser in
	C++17 (http://lists.gnu.org/archive/html/bug-bison/2018-04/msg00011.html)
	provides this feature, but relying on std::variant's dynamic typing,
	which we forbid in lalr1.cc.

The simplest seems to be actually generating the default semantic
	action (in all languages/skeletons).  This makes the pre-action (that
	sets $$ to $1) useless.  But...  maybe some users depend on this, in
	spite of the comments that clearly warn againt this.  So let's not
	turn this off just yet.

* src/reader.c (grammar_rule_check_and_complete): Rename as...
	(grammar_rule_check_and_complete): this.
	Install the default semantic action when applicable.
	* examples/variant-11.yy, examples/variant.yy, tests/calc.at:
	Exercise the default semantic action, even with variants.

2018-10-14  Jannick  <thirdedition@gmx.net>

xml2xhtml.xsl: add UTF-8 encoding
	To make arrows appear nicely in the browser. Currently it is shown as
	some garbled something in mine (Firefox).

* data/xslt/xml2xhtml.xsl: here.

2018-10-14  Akim Demaille  <akim.demaille@gmail.com>

reader: reorder some calls to separate checks from assignments
	* src/reader.c (packgram): Move assignments to rules[ruleno] after the
	checks on the rule.

2018-10-14  Akim Demaille  <akim.demaille@gmail.com>

C++: style: add missing space before parens
	* data/c++.m4, data/lalr1.cc, data/stack.hh, data/variant.hh,
	* examples/variant-11.yy: here.

2018-10-14  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update timevar
	See
	https://lists.gnu.org/archive/html/bug-gnulib/2018-10/msg00005.html.

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/vbitset.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/vbitset.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/lbitset.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/lbitset.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/ebitset.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/ebitset.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitsetv.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitsetv.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitsetv-print.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitsetv-print.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitset_stats.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitset_stats.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitset.c

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bitset.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/bbitset.h

2018-10-11  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/abitset.c

2018-10-10  Akim Demaille  <akim.demaille@gmail.com>

style: modernize lib/abitset.h

2018-10-09  Akim Demaille  <akim.demaille@gmail.com>

C++: issue a better CPP guard and Doxygen file command
	Currently we use "<dir><api.location.file>" as \file argument, and as
	base for the CPP guard.  This is not nice when <dir> is absolute, in
	which case it is expected that the user will use api.location.include
	to get something nicer.  If defined, use that name instead.

* data/location.cc (b4_location_path): New.
	Use it.
	* tests/c++.at (Shared locations): Check the guard and Doxygen doc.

2018-10-09  Akim Demaille  <akim.demaille@gmail.com>

C++: remove stray empty line
	* data/stack.hh: here.

2018-10-09  Akim Demaille  <akim.demaille@gmail.com>

doc: spell check

2018-10-09  Akim Demaille  <akim.demaille@gmail.com>

doc: document api.*.file and the like
	* doc/bison.texi (Exposing the Location Classes): New.
	(%define Summary): Document api.location.file, api.location.include,
	api.stack.file and api.position.file.
	(C++ Bison Interface): stack.hh and position.hh are deprecated.

2018-10-07  Akim Demaille  <akim.demaille@gmail.com>

build: add missing gnulib libs
	* src/local.mk (LDADD): Here.

2018-10-07  Akim Demaille  <akim.demaille@gmail.com>

build: let timevar be dealt with by gnulib
	* lib/local.mk (lib_libbison_a_SOURCES): Remove timevar.

2018-10-07  Akim Demaille  <akim.demaille@gmail.com>

build: fix distcheck
	Now that distcheck no longer fails (see previous commit), let's
	address the shortcomings.

* Makefile.am (CLEANDIRS, clean-local): New.
	* doc/local.mk, examples/calc++/local.mk, examples/local.mk,
	* examples/mfcalc/local.mk, examples/rpcalc/local.mk,
	* src/local.mk
	(CLEANDIRS): Get rid of Apple's *.dSYM directories.
	(CLEANFILES): Get rid of *.output files.
	* examples/variant-11.yy, examples/variant.yy: Don't generate
	any of the auxiliary files (location.hh and the like).

2018-10-07  Akim Demaille  <akim.demaille@gmail.com>

README: work around a nasty behavior of gettext
	`make update-po` runs:

package_gnu="$(PACKAGE_GNU)"; \
	    test -n "$$package_gnu" || { \
	      if { if (LC_ALL=C find --version) 2>/dev/null | grep GNU >/dev/null; then \
	             LC_ALL=C find -L $(top_srcdir) -maxdepth 1 -type f \
	                           -size -10000000c -exec grep 'GNU @PACKAGE@' \
	                           /dev/null '{}' ';' 2>/dev/null; \
	           else \
	             LC_ALL=C grep 'GNU @PACKAGE@' $(top_srcdir)/* 2>/dev/null; \
	           fi; \
	         } | grep -v 'libtool:' >/dev/null; then \
	         package_gnu=yes; \
	       else \
	         package_gnu=no; \
	       fi; \
	    }; \

and based on the result, put GNU or not in the following line from
	bison.pot:

# This file is distributed under the same license as the GNU bison package.

It turns out that in my environment some log files had the 'GNU bison'
	string (note the lower case), but in distcheck, these files are no
	longer visible, so the generate bison.pot was different, and distcheck
	failed because we try to update bison.pot, which is read only in
	distcheck.

The heuristics should look accept 'GNU Bison', not just 'GNU bison'.
	But let's please it to make sure we have our 'GNU'.

* README: Mention 'GNU bison'.

2018-10-07  Akim Demaille  <akim.demaille@gmail.com>

NEWS: document api.*.file changes

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: provide a means to control how location.hh is included
	Users may want to generate the location file elsewhere, say
	$top_srcdir/include/ast/location.hh.  Yet, we should not generate
	`#include "$top_srcdir/include/ast/location.hh"` but probably
	something like `#include <ast/location.hh>`, or `#include
	"ast/location.hh", or `#include <location.hh>`.  It entirely depends
	on the compiler flags (-I/-isystem) that are used.  Bison cannot guess
	what is expected, so let's give the user a means to tell how the
	location file should be included.

* data/location.cc (b4_location_file): New.
	* data/glr.cc, data/lalr1.cc: Use it.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: support absolute api.location.file names
	In the case a user wants to create location.hh elsewhere, it can be
	helpful to define api.location.file to some possibly absolute path
	such as -Dapi.location.file='"$(top_srcdir)/include/ast/location.hh"'.
	Currently this does not work with `-o foo/parser.cc`, as we join foo/
	and $(top_srcdir) together, the latter starting with slash.

We should not try to do that in m4, manipulating file names is quite
	complex when you through Windows file name in.  Let m4 delegate this
	to gnulib.

* src/scan-skel.l (at_output): Accept up to two arguments.
	* data/bison.m4 (b4_output): Adjust.
	* tests/skeletons.at (Fatal errors but M4 continues producing output):
	Adjust to keep the error.

* data/location.cc, data/stack.hh: Leave the concatenation to @output.
	* tests/output.at: Exercise api.location.file with an absolute path.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: when api.location.file is defined, don't generate stack.hh
	Make it easier to have fewer files.

* data/stack.hh: Don't generate stack.hh when api.location.file is
	specified.
	* tests/calc++.at, tests/output.at: Adjust tests.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: make position.hh completely useless
	Let's put the definition of position into location.hh, there's no real
	value in keeping them separate: they are small, and share the same
	requirements.

To help users transition to this new model, still generate position.hh
	by default, but as a simple include to location.hh.

* data/location.cc (api.position.file): Accept only 'none' as possible
	value.
	(position.hh): Make it a stub.
	(location.hh): Adjust.
	(b4_position_define): Merge into...
	(b4_location_define): this.
	* data/glr.cc, data/lalr1.cc, tests/input.at, tests/output.at: Adjust.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: make stack.hh completely useless
	Let's completely deprecate stack.hh.  Don't provide a means to give it
	a new name, allow only its removal.

See https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00151.html
	and https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00182.html.

* data/stack.hh: Reduce stack.hh to a simple comment explaining how to
	get rid of it.
	* data/lalr1.cc: Adjust
	* tests/input.at (%define file variables): Adjust.
	* tests/output.at: Remove cases where stack.hh was removed.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: add support for api.position.file and api.location.file
	* data/location.cc: Sort includes.
	(b4_position_file, b4_location_file): New.
	When there's a file for locations but not for positions, include the
	definition of position in the location file.
	* data/lalr1.cc (b4_shared_declarations): Include the
	position/location file when it exists.
	Otherwise, define the class.
	* data/glr.cc: Likewise.
	* tests/input.at (%define file variables): Check them.
	* tests/output.at (C++ output): Check various cases with
	api.position.file and api.location.file.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: provide control over the stack.hh file name
	It was not a good idea to generate the file stack.hh.  It never was.
	But now we have to deal with backward compatibility: if we stop
	generating it, the build system of some build system will probably
	break.

So offer the user a means to (i) decide what the name of the output
	file should be, and (ii) not generate this file at all (its content
	will be inline where the parser is defined).

* data/lalr1.cc (b4_percent_define_check_file_complain)
	(b4_percent_define_check_file): New.
	* data/stack.hh: Generate the file only if api.stack.file is not
	empty.
	In that case, use it as file name.
	* data/lalr1.cc: Adjust to include the right file, or to include
	the definition of stack.
	* tests/calc.at, tests/output.at: Exercise api.stack.file.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

tests: c++: don't fuse prefix and namespace
	They are not the same concept.  It appears that we still consider that
	api.prefix is the default for api.namespace.  We should stop that.

* tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS):
	Separate namespace and prefix.
	Adjust dependencies.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

style: tests: factor file extension computations
	* tests/local.at (AT_LANG_HDR): New.
	* tests/calc.at, tests/headers.at, tests/synclines.at: Use it, and
	AT_LANG_EXT.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

c++: style changes
	* data/lalr1.cc: Formatting changes.
	Remove duplicate definition of YY_NULLPTR.
	Add quotes to help Emacs track balanced parens.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

lib: introduce xpath_join
	* lib/path-join.h, lib/path-join.c: New.
	* lib/local.mk: Adjust.
	* src/output.c: Use it.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

CI: travis finally knows about llvm-toolchain-trusty-7
	See https://github.com/travis-ci/apt-source-safelist/pull/392.

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

gnulib: update

2018-10-06  Akim Demaille  <akim.demaille@gmail.com>

doc: restore spello made on purpose

2018-10-05  Akim Demaille  <akim.demaille@gmail.com>

THANKS: add Josh Soref