platform/upstream/bison.git
10 years agoversion 3.0.2 v3.0.2
Akim Demaille [Thu, 5 Dec 2013 13:58:22 +0000 (14:58 +0100)]
version 3.0.2

* NEWS: Record release date.

10 years agognulib: update
Akim Demaille [Thu, 5 Dec 2013 13:39:30 +0000 (14:39 +0100)]
gnulib: update

* gnulib: here.

10 years agooutput: do not generate source files when late errors are caught
Akim Demaille [Wed, 4 Dec 2013 09:51:47 +0000 (10:51 +0100)]
output: do not generate source files when late errors are caught

Reported by Alexandre Duret-Lutz as "second problem" in:
http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html

* bootstrap.conf: We need the "unlink" module.
* src/files.h, src/files.c (unlink_generated_sources): New.
* src/output.c: Use it.
* tests/output.at: Check the case of late errors.

10 years agooutput: record what generated files are source or report files
Akim Demaille [Wed, 4 Dec 2013 09:26:46 +0000 (10:26 +0100)]
output: record what generated files are source or report files

* src/files.h, src/files.c (output_file_name_check): Take an additional
argument to record whether a file is a source or report file.
* src/files.c (generated_file): New.
(file_names, file_names_count): Replace with...
(generated_files, generated_files_size): these.
* src/scan-skel.l: Adjust.

10 years agooutput: do not generate source files when early errors are caught
Akim Demaille [Wed, 4 Dec 2013 08:30:08 +0000 (09:30 +0100)]
output: do not generate source files when early errors are caught

Reported by Alexandre Duret-Lutz as "second problem" in:
http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html

One problem is that some errors are caught early, before the
generation of output files, while others can only be detected
afterwards (since, for instance, skeletons can raise errors
themselves).

This will be addressed in two steps: early errors do not generate
source files at all, while later errors will remove the files that
have already been generated.

* src/scan-skel.l (yyout): Open to /dev/null when there are errors.
* tests/output.at (AT_CHECK_FILES): Factored out of...
(AT_CHECK_OUTPUT): this.
Fuse the "SHELLIO" argument in the "FLAGS" one.
Use $5 to denote the expected exit status.
Add a test case for early errors.

10 years agoxml: also use "%empty" with html output
Akim Demaille [Tue, 26 Nov 2013 15:23:52 +0000 (16:23 +0100)]
xml: also use "%empty" with html output

* data/xslt/xml2xhtml.xsl: No longer issue an Epsilon, display as in
dot and text formats.

10 years agostyle: formatting changes
Akim Demaille [Tue, 26 Nov 2013 15:20:55 +0000 (16:20 +0100)]
style: formatting changes

* src/print-xml.c: here.

10 years agoxml: also display %empty for empty right-hand sides
Akim Demaille [Tue, 26 Nov 2013 15:20:52 +0000 (16:20 +0100)]
xml: also display %empty for empty right-hand sides

* data/xslt/xml2dot.xsl, data/xslt/xml2text.xsl: Display %empty where needed.

10 years agoreports: display %empty in the generated pointed-rules
Akim Demaille [Tue, 26 Nov 2013 14:24:53 +0000 (15:24 +0100)]
reports: display %empty in the generated pointed-rules

* src/print.c (print_core): Use %empty for empty rules.
* src/print_graph.c (print_core): Ditto.
* tests/conflicts.at, tests/output.at, tests/reduce.at: Adjust
expectations.

10 years agonews: YYERROR vs variants
Akim Demaille [Tue, 26 Nov 2013 14:31:33 +0000 (15:31 +0100)]
news: YYERROR vs variants

10 years agostyle: scope reduction in lalr.cc
Akim Demaille [Mon, 18 Nov 2013 09:29:23 +0000 (10:29 +0100)]
style: scope reduction in lalr.cc

* src/lalr.c: Shorten variable scopes.
(lookahead_tokens_print): Use the same variable name in two loops
iterating over the same structure.

10 years agolalr1.cc: formatting changes
Akim Demaille [Fri, 15 Nov 2013 09:10:41 +0000 (10:10 +0100)]
lalr1.cc: formatting changes

* data/lalr1.cc: Fix indentation.

10 years agolalr1.cc: fix the support of YYERROR with variants
Akim Demaille [Fri, 15 Nov 2013 09:08:31 +0000 (10:08 +0100)]
lalr1.cc: fix the support of YYERROR with variants

When variant are enabled, the yylhs variable (the left-hand side of
the rule being reduced, i.e. $$ and @$) is explicitly destroyed when
YYERROR is called.  This is because before running the user code, $$
is initialized, so that the user can properly use it.

However, when quitting yyparse, yylhs is also reclaimed by the C++
compiler: the variable goes out of scope.

Instead of trying to be too smart, let the compiler do its job: reduce
the scope of yylhs to exactly the reduction.  This way, whatever the
type of scope exit (regular, exception, return, goto...) this variable
will be properly reclaimed.

Reported by Paolo Simone Gasparello.
<http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00003.html>

* data/lalr1.cc (yyparse): Reduce the scope of yylhs.
* tests/c++.at: We now pass this test.

10 years agotests: check $$'s destruction with variant, YYERROR, and no error recovery
Akim Demaille [Fri, 15 Nov 2013 08:43:01 +0000 (09:43 +0100)]
tests: check $$'s destruction with variant, YYERROR, and no error recovery

When variant are enabled, the yylhs variable (the left-hand side of
the rule being reduced, i.e. $$ and @$) is explicitly destroyed when
YYERROR is called.  This is because before running the user code, $$
is initialized, so that the user can properly use it.

However, when quitting yyparse, yylhs is also reclaimed by the C++
compiler: the variable goes out of scope.

This was not detected by the test suite because (i) the Object tracker
was too weak, and (ii) the problem does not show when there is error
recovery.

Reported by Paolo Simone Gasparello.
<http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00003.html>

* tests/c++.at (Exception safety): Improve the objects logger to make
sure that we never destroy twice an object.
Also track copy-constructors.
Use a set instead of a list.
Display the logs before running the function body, this is more
useful in case of failure.
Generalize to track with and without error recovery.

10 years agotests: simplify useless obfuscation
Akim Demaille [Thu, 14 Nov 2013 15:58:42 +0000 (16:58 +0100)]
tests: simplify useless obfuscation

* tests/c++.at: $$ is not special for M4, there is no need to "escape"
it.

10 years agoskeletons: use better names when computing a "goto"
Akim Demaille [Thu, 14 Nov 2013 15:14:39 +0000 (16:14 +0100)]
skeletons: use better names when computing a "goto"

* data/glr.c (yyLRgotoState): Name the symbol argument yysym, instead
of yylhs.
* data/lalr1.cc (yy_lr_goto_state_): Likewise.
* data/lalr1.java (yy_lr_goto_state_): New, modeled after the previous
two routines.
Use it.

10 years agomaint: post-release administrivia
Akim Demaille [Tue, 12 Nov 2013 17:37:14 +0000 (18:37 +0100)]
maint: post-release administrivia

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

10 years agoversion 3.0.1 v3.0.1
Akim Demaille [Tue, 12 Nov 2013 14:55:33 +0000 (15:55 +0100)]
version 3.0.1

* NEWS: Record release date.

10 years agoaver: it is no longer "protected against NDEBUG"
Akim Demaille [Tue, 12 Nov 2013 14:46:27 +0000 (15:46 +0100)]
aver: it is no longer "protected against NDEBUG"

Apply the same rules for aver as for assert: no side effects,
especially not important ones.

* src/AnnotationList.c, src/muscle-tab.c: Adjust aver uses to resist
to -DNDEBUG.

10 years agoparsers: rename YY_NULL as YY_NULLPTR to avoid conflicts with Flex
Akim Demaille [Fri, 8 Nov 2013 09:52:15 +0000 (10:52 +0100)]
parsers: rename YY_NULL as YY_NULLPTR to avoid conflicts with Flex

Flex also defines YY_NULL (to 0).  Avoid gratuitous conflicts.

* data/c.m4 (b4_null_define): Rename YY_NULL as YY_NULLPTR.

* data/glr.c, data/lalr1.cc, data/location.cc, data/variant.hh,
* data/yacc.c, src/parse-gram.c, tests/actions.at, tests/c++.at,
* tests/cxx-type.at, tests/glr-regression.at, tests/headers.at,
* tests/push.at, tests/regression.at:
Adjust.

10 years agobuild: use Automake 1.14's non-recursive Makefile features
Akim Demaille [Tue, 5 Nov 2013 13:32:20 +0000 (14:32 +0100)]
build: use Automake 1.14's non-recursive Makefile features

* configure.ac: Require Automake 1.14.
* examples/calc++/local.mk, examples/local.mk, examples/mfcalc/local.mk,
* examples/rpcalc/local.mk, tests/local.mk: Use %D% and %C%.

10 years agobuild: restore maintainer-push-check
Akim Demaille [Tue, 5 Nov 2013 13:15:15 +0000 (14:15 +0100)]
build: restore maintainer-push-check

* tests/local.mk: here.

10 years agoc++: use __attribute__((__pure__)) to avoid warnings
Akim Demaille [Wed, 23 Oct 2013 08:57:51 +0000 (10:57 +0200)]
c++: use __attribute__((__pure__)) to avoid warnings

Building C++ parsers with -Wsuggest-attribute=const and
-Wsuggest-attribute=noreturn triggers warning in generated code.

* data/lalr1.cc: Call b4_attribute_define.
(debug_stream, debug_level): Flag as pure.
* tests/headers.at (Several parsers): There are now more YY macros
that "leak".

10 years agoskeletons: update the handling of compiler attributes
Akim Demaille [Wed, 23 Oct 2013 08:57:09 +0000 (10:57 +0200)]
skeletons: update the handling of compiler attributes

* data/c.m4 (b4_attribute_define): Instead of defining __attribute__,
define YY_ATTRIBUTE conditionally.
(YY_ATTRIBUTE_PURE, YY_ATTRIBUTE_UNUSED, _Noreturn): New.
Use them.
* data/glr.c: Use them.

10 years agognulib: update
Akim Demaille [Wed, 23 Oct 2013 08:25:47 +0000 (10:25 +0200)]
gnulib: update

10 years agostyle: use /* ... */ comments
Akim Demaille [Wed, 23 Oct 2013 13:03:45 +0000 (15:03 +0200)]
style: use /* ... */ comments

* src/complain.c: Here.

10 years agotests: skip C++ tests that are too demanding for some compilers
Akim Demaille [Wed, 23 Oct 2013 13:04:15 +0000 (15:04 +0200)]
tests: skip C++ tests that are too demanding for some compilers

Some tests now fail when compiled with G++ 4.3 or 4.4 on MacPorts.

* tests/local.at (AT_SKIP_IF_EXCEPTION_SUPPORT_IS_POOR): New.
* tests/c++.at (Exception safety): Use it.

10 years agoinstall: do not install yacc.1 when --disable-yacc
Akim Demaille [Tue, 22 Oct 2013 15:32:49 +0000 (17:32 +0200)]
install: do not install yacc.1 when --disable-yacc

* configure.ac (ENABLE_YACC): New conditional.
(YACC_SCRIPT, YACC_LIBRARY): Remove.
* lib/local.mk, src/local.mk: Use the former instead of the latter.
* doc/local.mk: Use ENABLE_YACC to avoid installing yacc.1.

10 years agostyle: avoid tabs
Akim Demaille [Tue, 22 Oct 2013 14:43:17 +0000 (16:43 +0200)]
style: avoid tabs

* src/scan-code.l: here.

10 years agoc++: fix generated doxygen comments
Akim Demaille [Tue, 22 Oct 2013 14:30:12 +0000 (16:30 +0200)]
c++: fix generated doxygen comments

* configure.ac: Enable -Wdocumentation if supported.
* data/lalr1.cc: Fix comments.

10 years agofix: uniqstr are already pointers
Akim Demaille [Tue, 22 Oct 2013 14:29:22 +0000 (16:29 +0200)]
fix: uniqstr are already pointers

* src/uniqstr.c (uniqstr_assert): Remove incorrect double indirection,
and now useless cast.

10 years agobison: pacify Sun C 5.12
Paul Eggert [Mon, 21 Oct 2013 22:21:32 +0000 (15:21 -0700)]
bison: pacify Sun C 5.12

* src/scan-code.l (show_sub_message):
Redo initializations to work around a bogus Sun C 5.12 warning.
(parse_ref): Remove unreachable code that Sun C 5.12 complains about.
* src/uniqstr.h (uniqstr_vsprintf): Use
_GL_ATTRIBUTE_FORMAT_PRINTF (...)  instead of __attribute__
((__format__ (__printf__, ...))).  Otherwise, Sun C 5.12
complains about an unknown attribute.

10 years agomaint: git now ignores rpcalc
Paul Eggert [Mon, 21 Oct 2013 22:17:16 +0000 (15:17 -0700)]
maint: git now ignores rpcalc

* examples/rpcalc/.gitignore: Ignore rpcalc.

10 years agobuild: examples/calc++/calc++ requires flex
Paul Eggert [Mon, 21 Oct 2013 22:16:43 +0000 (15:16 -0700)]
build: examples/calc++/calc++ requires flex

* configure.ac (FLEX_CXX_WORKS): New AM_CONDITIONAL.
* examples/calc++/local.mk (examples/calc++/calc++):
Build if FLEX_CXX_WORKS, not BISON_CXX_WORKS.

10 years agomaint: mention help2man, texinfo, apt-get
Paul Eggert [Mon, 21 Oct 2013 22:13:52 +0000 (15:13 -0700)]
maint: mention help2man, texinfo, apt-get

* README-hacking: Add help2man, texinfo.
Describe how to add packages if you're using Debian.

10 years agomaint: git now ignores .log and .trs files
Paul Eggert [Mon, 21 Oct 2013 22:12:13 +0000 (15:12 -0700)]
maint: git now ignores .log and .trs files

* .gitignore: Add *.log, *.trs.

10 years agotests: fix incorrect object construction
Akim Demaille [Fri, 18 Oct 2013 14:22:23 +0000 (16:22 +0200)]
tests: fix incorrect object construction

Reported by Ken Moffat.
http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00009.html

* tests/c++.at (Exception safety): Here.

10 years agoglr: allow spaces between "%?" and "{" in predicates
Akim Demaille [Wed, 16 Oct 2013 13:19:44 +0000 (15:19 +0200)]
glr: allow spaces between "%?" and "{" in predicates

Reported by Rici Lake.
http://lists.gnu.org/archive/html/bug-bison/2013-10/msg00004.html
http://stackoverflow.com/questions/19330171/

* src/scan-gram.l: Do not try to be too smart when diagnosing invalid
directives.
* tests/glr-regression.at (Predicates): New test.

10 years agodiagnostics: "-Werror -Wno-error=foo" must not emit errors
Akim Demaille [Wed, 16 Oct 2013 08:55:28 +0000 (10:55 +0200)]
diagnostics: "-Werror -Wno-error=foo" must not emit errors

Currently "-Werror -Wno-error=foo" still turns "foo" warnings into errors.
Reported by Alexandre Duret-Lutz.
See http://lists.gnu.org/archive/html/bug-bison/2013-09/msg00015.html.

* src/complain.c (errority, errority_flag): New.
(complain_init): Initialize the latter.
(warning_argmatch): Extract the loop iterating on the flag's bits.
Set and unset errority_flag here.
(warnings_argmatch): -Wno-error is not the same as -Wno-error=everything:
we must remember if category foo was explicitly turned in an error/warning
via -W(no-)error=foo.
(warning_severity): Use errority_flag.

* tests/input.at (Symbols): Just check --yacc, not -Wyacc, that's the
job of tests on -W.
(-Werror is not affected by -Wnone and -Wall): Rename as...
(-Werror combinations): this.
Tests more combinations of -W, -W(no-)error, and -W(no-)error=foo.
* tests/local.at (AT_BISON_CHECK_WARNINGS): Don't expect -Werror
to turn runs that issue warnings into runs with errors, as the
warnings might be enforced as warnings by -Wno-error=foo, in which
case -Werror does not change anything.

* doc/bison.texi (Bison Options): Try to be clearer about how
-W(no-)error and -W(no-)error=foo interact.

10 years agocomment changes
Akim Demaille [Tue, 8 Oct 2013 11:01:24 +0000 (13:01 +0200)]
comment changes

* src/complain.h, src/complain.c: More documentation, more comments.

10 years agolocation: fix EOF check
Andreas Schwab [Fri, 27 Sep 2013 22:00:34 +0000 (00:00 +0200)]
location: fix EOF check

* location.c (location_caret): Use int, not char, for values from
getc.

10 years agostyle: variant: remove empty line
Akim Demaille [Thu, 19 Sep 2013 14:45:09 +0000 (16:45 +0200)]
style: variant: remove empty line

* data/variant.hh (b4_symbol_constructor_define_): Remove
stray eol.

10 years agoglr: more assertions
Akim Demaille [Thu, 19 Sep 2013 13:27:19 +0000 (15:27 +0200)]
glr: more assertions

* data/glr.c (yyaddDeferredAction, yyglrShiftDefer, yypdumpstack):
More assertions.

10 years agoglr: shorten scopes
Akim Demaille [Thu, 19 Sep 2013 13:01:24 +0000 (15:01 +0200)]
glr: shorten scopes

* data/glr.c (yyglrReduce): Define yyflag with its value.

10 years agoglr: formatting changes
Akim Demaille [Thu, 19 Sep 2013 13:00:32 +0000 (15:00 +0200)]
glr: formatting changes

* data/glr.c: here.

10 years agoglr: better use of tracing macros
Akim Demaille [Thu, 19 Sep 2013 12:58:57 +0000 (14:58 +0200)]
glr: better use of tracing macros

* data/glr.c (yydestroyGLRState): Use YY_SYMBOL_PRINT instead of
yy_symbol_print.

10 years agoexamples: improve the output of the "variant" example
Akim Demaille [Thu, 19 Sep 2013 13:20:22 +0000 (15:20 +0200)]
examples: improve the output of the "variant" example

* examples/variant.yy: Improve the printing of lists.

10 years agovariant: remove useless assertion
Akim Demaille [Thu, 19 Sep 2013 13:03:01 +0000 (15:03 +0200)]
variant: remove useless assertion

* data/variant.hh (move): Remove precondition assertion which is
ensured by the first call of the body (this precondition is also one
of "build").

10 years agotests: remove stray debugging traces
Akim Demaille [Thu, 19 Sep 2013 13:51:32 +0000 (15:51 +0200)]
tests: remove stray debugging traces

* tests/atlocal.in: Remove traces.
Be ready to remove conftest.dSYM generated on OS X.

11 years agotests: do not use grep -q
Akim Demaille [Wed, 4 Sep 2013 15:23:37 +0000 (17:23 +0200)]
tests: do not use grep -q

Reported by Daniel Galloway.
http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00020.html

* tests/java.at: Ignore grep's output instead.

11 years agobuild: don't require flex for ordinary builds
Paul Eggert [Sun, 25 Aug 2013 01:22:22 +0000 (18:22 -0700)]
build: don't require flex for ordinary builds

* configure.ac (LEX): Don't fail if this is lex, as flex is not
required for ordinary builds.  Instead, issue a warning and
substitute a no-op LEX.  Reported by Michael Felt in
<http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00009.html>.

11 years agomaint: update .gitignore
Paul Eggert [Sun, 25 Aug 2013 01:00:25 +0000 (18:00 -0700)]
maint: update .gitignore

* .gitignore: Add *.eps, *.o, *.pdf, *.png, *.stamp, *~,
.deps, .dirstamp.  Needed to suppress unwanted chatter from
'git status' after a bootstrap build.

11 years agobuild: port to pre-5.8.7 perl
Paul Eggert [Sat, 24 Aug 2013 14:52:27 +0000 (07:52 -0700)]
build: port to pre-5.8.7 perl

* examples/local.mk (extract): Omit -f from perl options.
This doesn't work with perl versions before 5.8.7
that are configured without USE_SITECUSTOMIZE.
Reported by Michael Felt in
<http://lists.gnu.org/archive/html/bug-bison/2013-08/msg00006.html>.

11 years agotests: minor change to make it easier to test other skeletons
Akim Demaille [Thu, 1 Aug 2013 14:11:27 +0000 (16:11 +0200)]
tests: minor change to make it easier to test other skeletons

* tests/c++.at (Variants): Pass the skeleton as argument.

11 years agouniqstr: fix assertion
Valentin Tolmer [Thu, 1 Aug 2013 12:53:54 +0000 (14:53 +0200)]
uniqstr: fix assertion

* src/uniqstr.c (uniqstr_assert): Really make sure str is a uniqstr,
not just whether some uniqstr with the same content was registered.

11 years agobuild: ship the ASCII art figures
Akim Demaille [Thu, 1 Aug 2013 09:41:49 +0000 (11:41 +0200)]
build: ship the ASCII art figures

We don't ship the *.txt files that are used to build the info
file.
Reported by Colin Daley.

* doc/figs/example.txt: New.
* doc/local.mk (bison.info): Depend on the txt files.
And ship them.

11 years agodoc: prefer the ".gv" extension to ".dot"
Akim Demaille [Thu, 1 Aug 2013 09:19:05 +0000 (11:19 +0200)]
doc: prefer the ".gv" extension to ".dot"

See http://marc.info/?l=graphviz-devel&m=129418103126092 for the
motivation (basically, some word processor now uses *.dot).

* doc/figs/example-reduce.dot: Rename as...
* doc/figs/example-reduce.gv: this.
* doc/figs/example-shift.dot: Rename as...
* doc/figs/example-shift.gv: this.
* doc/figs/example.dot: Rename as...
* doc/figs/example.gv: this.
* doc/local.mk: Adjust.

11 years agomaint: post-release administrivia
Akim Demaille [Thu, 25 Jul 2013 16:13:53 +0000 (18:13 +0200)]
maint: post-release administrivia

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

11 years agoversion 3.0 v3.0
Akim Demaille [Thu, 25 Jul 2013 15:55:58 +0000 (17:55 +0200)]
version 3.0

* NEWS: Record release date.

11 years agoregen
Akim Demaille [Thu, 25 Jul 2013 15:52:15 +0000 (17:52 +0200)]
regen

11 years agonews: prepare 3.0
Akim Demaille [Thu, 25 Jul 2013 15:50:11 +0000 (17:50 +0200)]
news: prepare 3.0

* NEWS (3.0): Reorder.

11 years agotests: fix invalid assignment when using variants in C++11
Akim Demaille [Thu, 25 Jul 2013 15:45:19 +0000 (17:45 +0200)]
tests: fix invalid assignment when using variants in C++11

* tests/c++.at (Exception safety): In variant mode $$ is an instance
of Object.  Assigning YY_NULL in C++98 is incorrect, but behaves ok,
as it assigns YY_NULL=0 using Object::operator= (char v).  It is wrong
in C++11 as there is operator for "$$ = nullptr".

11 years agoyacc: beware of "uninitialized uses" warnings
Akim Demaille [Thu, 25 Jul 2013 14:04:26 +0000 (16:04 +0200)]
yacc: beware of "uninitialized uses" warnings

Again some issues with the fact that yylval is reported by GCC as
possibly not initialized in some cases.  Here, the case at hand is the
%destructor.

I am still not convinced that it is worth going all the trouble of
using pragmas to disable temporarily some warnings, instead of just
initializing the looking symbol once for all, but that's what Paul
voted for, see
<http://lists.gnu.org/archive/html/bison-patches/2012-10/msg00050.html>.

* data/c.m4 (b4_attribute_define): Define
YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN, YY_IGNORE_MAYBE_UNINITIALIZED_END,
YY_INITIAL_VALUE here, as we will need them in the generation of the
destructor function, which is defined in yacc.c before yyparse, which
was in charge of defining these macros.
* data/yacc.c (b4_declare_scanner_communication_variables): Simplify:
trying to factor the definitions of the case pure and impure is
too complex.
Actually, it is not even clear that this macro should really exist,
as even the calls are complex.
Be careful not to issue a lone ";", as this is a statement, and C90
forbids declarations after statements ; so write
"YY_INITIAL_VALUE(Decl;)", not "YY_INITIAL_VALUE(Decl);".

11 years agognulib: update
Akim Demaille [Thu, 25 Jul 2013 11:33:28 +0000 (13:33 +0200)]
gnulib: update

11 years agotests: skip C++ tests if we can't compile a simple program
Akim Demaille [Wed, 3 Jul 2013 15:16:04 +0000 (17:16 +0200)]
tests: skip C++ tests if we can't compile a simple program

There are possible conflicts between gnulib replacement functions (in
<stdio.h>) and their C++ wrappers (in <stream>).  Trying to address
these in configure seems too hard, and I don't know how to fix the issue
in gnulib.  Cowardly avoid the problem by skipping C++ tests when this
happens.
Reported by Stefano Lattarini.
http://lists.gnu.org/archive/html/bug-bison/2013-06/msg00001.html

* tests/atlocal.in (BISON_CXX_WORKS): Also set it to "skip" if we can't
compile a simple program using <stream>.
* tests/local.at: Comment changes.

11 years agotests: fix 'find' portability issues
Akim Demaille [Fri, 28 Jun 2013 06:33:33 +0000 (08:33 +0200)]
tests: fix 'find' portability issues

Reported by Stefano Lattarini.
http://lists.gnu.org/archive/html/bug-bison/2013-06/msg00000.html

* tests/output.at (AT_CHECK_OUTPUT): Use Perl instead.

11 years agomaint: post-release administrivia
Akim Demaille [Mon, 24 Jun 2013 08:29:36 +0000 (10:29 +0200)]
maint: post-release administrivia

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

11 years agoversion 2.7.91 v2.7.91
Akim Demaille [Mon, 24 Jun 2013 08:21:33 +0000 (10:21 +0200)]
version 2.7.91

* NEWS: Record release date.

11 years agoNEWS: prepare for 2.7.91
Akim Demaille [Mon, 24 Jun 2013 08:20:46 +0000 (10:20 +0200)]
NEWS: prepare for 2.7.91

* NEWS (2.7.91): Java push parsers.

11 years agojava: rename YYMORE as YYPUSH_MORE for consistency with C
Akim Demaille [Mon, 24 Jun 2013 07:32:22 +0000 (09:32 +0200)]
java: rename YYMORE as YYPUSH_MORE for consistency with C

http://lists.gnu.org/archive/html/bison-patches/2013-06/msg00008.html

* data/lalr1.java, doc/bison.texi, tests/javapush.at:
s/YYMORE/YYPUSH_MORE.

11 years agotests: fix Java push failure when running with BISON_USE_PUSH_FOR_PULL
Akim Demaille [Fri, 21 Jun 2013 09:39:02 +0000 (11:39 +0200)]
tests: fix Java push failure when running with BISON_USE_PUSH_FOR_PULL

* tests/javapush.at (Trivial Push Parser with api.push-pull verification):
When push for pull is enabled, there is one such function generated.

11 years agostyle: minor changes in the Java tests
Akim Demaille [Fri, 21 Jun 2013 09:28:01 +0000 (11:28 +0200)]
style: minor changes in the Java tests

* tests/java.at (AT_CHECK_JAVA_GREP): Ignore the exit status.
* tests/javapush.at (AT_CHECK_JAVA_GREP): Be more alike the previous
one.
Formating changes.
Remove stray debugging "jj" file.

11 years agojava: push: do not reset the error counter
Akim Demaille [Fri, 21 Jun 2013 09:17:05 +0000 (11:17 +0200)]
java: push: do not reset the error counter

* data/lalr1.java (parse): here, when in push-pull is in "both" mode.
This breaks the test suite, for instance
make check TESTSUITEFLAGS='-d 388 BISON_USE_PUSH_FOR_PULL=1'.
More generally make maintainer-push-check.

11 years agobuild: add Valgrind suppression file for GNU/Linux
Akim Demaille [Thu, 13 Jun 2013 14:08:33 +0000 (16:08 +0200)]
build: add Valgrind suppression file for GNU/Linux

* build-aux/linux-gnu.valgrind: New.
* build-aux/local.mk: Ship it.
* configure.ac: Use it.

11 years agojava: add push-parser support
Dennis Heimbigner [Thu, 13 Jun 2013 08:08:19 +0000 (10:08 +0200)]
java: add push-parser support

* data/lalr1.java: Capture the declarations as m4 macros to avoid
duplication.  When push parsing, the declarations occur at the class
instance level rather than within the parse() function.

Change the way that the parser state is initialized.  For
push-parsing, the parse state declarations are moved to
"push_parse_initialize()", which is called on the first invocation of
"push_parse()". The %initial-action code is also inserted after the
invocation of "push_parse_initialize()".

The body of the parse loop is modified to return values at appropriate
points when doing push parsing.  In order to make push parsing work,
it is necessary to divide YYNEWSTATE into two states: YYNEWSTATE and
YYGETTOKEN. On the first call to push_parse(), the state is
YYNEWSTATE. On all later entries, the state is set to YYGETTOKEN. The
YYNEWSTATE switch arm falls through into YYGETTOKEN. YYGETTOKEN
indicates that a new token is potentially needed.  Normally, with a
pull parser, this new token would be obtained by calling "yylex()". In
the push parser, the value YYMORE is returned to the caller. On the
next call to push_parse(), the parser will return to the YYGETTOKEN
state and continue operation.

* tests/javapush.at: New test file for java push parsing.
* tests/testsuite.at: Use it.
* tests/local.mk: Adjust.
* doc/bison.texi (Java Push Parser Interface): New.

Signed-off-by: Akim Demaille <akim@lrde.epita.fr>
11 years agobuild: ship all the files, even if the C++ compiler is broken
Akim Demaille [Tue, 11 Jun 2013 13:26:48 +0000 (15:26 +0200)]
build: ship all the files, even if the C++ compiler is broken

* examples/calc++/local.mk: Be sure to ship calc++.test even if
the current C++ compiler is not sufficient to run the tests.

11 years agostyle: comment changes in Java skeleton
Dennis Heimbigner [Wed, 5 Jun 2013 08:03:16 +0000 (10:03 +0200)]
style: comment changes in Java skeleton

* data/lalr1.java: Here.

11 years agotests: fix a G++ warning
Akim Demaille [Mon, 3 Jun 2013 14:40:53 +0000 (16:40 +0200)]
tests: fix a G++ warning

* tests/c++.at: Use YY_NULL instead of 0 for the null pointer.
And formatting changes.

11 years agobuild: fix a warning from clang
Akim Demaille [Mon, 3 Jun 2013 14:40:30 +0000 (16:40 +0200)]
build: fix a warning from clang

* src/muscle-tab.c: Declare local functions static.

11 years agomaint: post-release administrivia
Akim Demaille [Thu, 30 May 2013 13:28:44 +0000 (15:28 +0200)]
maint: post-release administrivia

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

11 years agoversion 2.7.90 v2.7.90
Akim Demaille [Thu, 30 May 2013 13:02:52 +0000 (15:02 +0200)]
version 2.7.90

* NEWS: Record release date.

11 years agostyle: syntax-check fixes
Akim Demaille [Thu, 30 May 2013 12:08:41 +0000 (14:08 +0200)]
style: syntax-check fixes

* data/yacc.c, src/Sbitset.c, src/Sbitset.h, src/muscle-tab.h,
* src/output.c, src/parse-gram.y, src/reader.c, src/symtab.c,
* src/uniqstr.c, src/uniqstr.h: Fix space before parens.
* cfg.mk (_space_before_paren_exempt): Add needed exceptions.

11 years agoxml: use %empty in the text output
Akim Demaille [Wed, 29 May 2013 15:07:32 +0000 (17:07 +0200)]
xml: use %empty in the text output

* data/xslt/xml2text.xsl: here.

11 years agobuild: locally disable new GCC warnings that fail on Flex generated code
Akim Demaille [Wed, 29 May 2013 14:27:59 +0000 (16:27 +0200)]
build: locally disable new GCC warnings that fail on Flex generated code

* configure.ac: here.

11 years agofix a memory leak
Akim Demaille [Wed, 29 May 2013 14:26:53 +0000 (16:26 +0200)]
fix a memory leak

* src/print-xml.c (num_escape_bufs): New.
(print_xml): Be sure to release all the escape_bufs.

11 years agoregen
Akim Demaille [Wed, 29 May 2013 07:27:59 +0000 (09:27 +0200)]
regen

11 years agobuild: be sure to include config.h first in the generated parser
Akim Demaille [Wed, 29 May 2013 07:17:48 +0000 (09:17 +0200)]
build: be sure to include config.h first in the generated parser

Using %code for config.h is wrong, as some headers will already have
been included by Bison.  In some cases, e.g., glibc's string.h, this
results in some declaration not being made for lack of definition of
_GNU_SOURCE, which is performed by config.h.

* src/parse-gram.y: here.

11 years agoMerge remote-tracking branch 'origin/maint'
Akim Demaille [Wed, 29 May 2013 07:05:12 +0000 (09:05 +0200)]
Merge remote-tracking branch 'origin/maint'

* origin/maint:
  maint: post-release administrivia
  version 2.7.1
  regen

11 years agodrop unused options --raw, -n, -e, --include and -I
Petr Machata [Fri, 17 May 2013 15:23:14 +0000 (17:23 +0200)]
drop unused options --raw, -n, -e, --include and -I

* --raw appears to be ignored.  It was marked as obsolete in the
  commit ec3bc39, and documented as no longer supported as of 1.29
  (2001-09-07).  Support for %raw appears to have been dropped in
  e9955c83 on 2002-06-11, but --raw was kept around.  Maybe it's time
  to drop it as well?

* Commit e9955c83 dropped support for %no-parser as well, and
  converted it to option.  --no-parser was later dropped in 728c4be2
  on 2007-08-12, but -n was kept around, probably as an omission.  All
  three are documented as removed since 2.3b (2008-05-27).

* -e existed for a single day in 2001.  It was introduced in eeeb962b
  on 2001-11-27.  The handling was removed in c7925b99 on 2001-11-28,
  but "e" was kept in the list of short options.  Probably an
  omission.

* --include appears to be dead code.  The option sets a variable, but
  that variable is not used anywhere.  It was added in f6bd5427 on
  2001-11-26 as a %-directive, with comments that it's not yet
  implemented.  It was converted to a command-line option later, but
  doesn't seem to ever have been actually implemented.

Signed-off-by: Petr Machata <pmachata@redhat.com>
Signed-off-by: Akim Demaille <akim@lrde.epita.fr>
11 years agognulib: update
Akim Demaille [Tue, 28 May 2013 14:38:26 +0000 (16:38 +0200)]
gnulib: update

11 years agodiagnostics: always point to the first directive
Akim Demaille [Mon, 22 Apr 2013 11:55:00 +0000 (13:55 +0200)]
diagnostics: always point to the first directive

Some directives cannot be used several times (e.g., a given symbol may
only have a single printer).  In case of repeated definitions, an
error is issued for the second definition, yet it is not discarded,
and becomes the definition used for the rest of the file.

This is not consistent with the idea that multiple definitions are not
allowed: discard any repeated directive.

* src/symtab.c (symbol_type_set, symbol_code_props_set)
(semantic_type_code_props_set, symbol_class_set, symbol_translation):
Discard repeated directives.
* tests/input.at (Default %printer and %destructor redeclared)
(Per-type %printer and %destructor redeclared): Update expectations.

11 years agotests: factor test for printer/desctructor redefined
Akim Demaille [Mon, 22 Apr 2013 11:47:50 +0000 (13:47 +0200)]
tests: factor test for printer/desctructor redefined

* tests/input.at (Default %printer and %destructor redeclared):
Introduce AT_TEST to factor.

11 years agodiagnostics: use appropriate location for useless precedence/associativity
Akim Demaille [Mon, 22 Apr 2013 09:26:28 +0000 (11:26 +0200)]
diagnostics: use appropriate location for useless precedence/associativity

* src/symtab.c (symbol_precedence_set): Use prec_location, not
location (which is the first occurrence of the symbol, possibly just
%token).
Also, as redefinitions are not allowed, keep the first values, not
the subsequent ones.
* tests/conflicts.at, tests/existing.at, tests/regression.at: Adjust.

11 years agotests: factor duplicate expected warnings
Akim Demaille [Mon, 22 Apr 2013 09:13:09 +0000 (11:13 +0200)]
tests: factor duplicate expected warnings

* tests/existing.at: Instead of "t ? abc : aBc", write "a(t?b:B)c".

11 years agotests: enable -Wsign-compare and fix corresponding warnings
Akim Demaille [Fri, 19 Apr 2013 15:13:07 +0000 (17:13 +0200)]
tests: enable -Wsign-compare and fix corresponding warnings

-Wsign-compare was disabled for bison's own code, following gnulib's
approach.  However, the generated parsers should not trigger such
warnings.

Reported by Efi Fogel.
http://lists.gnu.org/archive/html/help-bison/2013-04/msg00018.html

See also http://stackoverflow.com/questions/16101062 for the weird
"-(unsigned)i" piece of code.

* configure.ac (warn_tests): Enable -Wsign-compare.
* data/location.cc (position::add_): New.
(position::lines, position::columns): Use it.
* tests/actions.at (AT_CHECK_PRINTER_AND_DESTRUCTOR): Fix signedness issues.

11 years agomuscle: check more cases of %define variables with code values
Akim Demaille [Thu, 18 Apr 2013 14:52:55 +0000 (16:52 +0200)]
muscle: check more cases of %define variables with code values

* data/bison.m4 (b4_percent_define_check_kind): Fix overquotation.
(api.location.type, api.position.type): Check they have code values here.
* data/c++.m4 (api.location.type): No longer checked here.
(parser_class_name): Check it here.
* data/java.m4 (api.value.type, init_throws, lex_throws, parser_class_name)
(throws, annotations, extends, implements): Check they have code values.
* doc/bison.texi: Fix every incorrect occurrence of %define.
Document the additional syntax for %define: code values.
Document the additional syntax for -D/-F: string and code values.
* tests/calc.at, tests/headers.at, tests/input.at, tests/java.at,
* tests/local.at: Fix dependencies.

11 years agoregen
Akim Demaille [Thu, 18 Apr 2013 14:47:41 +0000 (16:47 +0200)]
regen

11 years agoparser: do not convert $ and @ in code values of %define variables
Akim Demaille [Thu, 18 Apr 2013 14:37:23 +0000 (16:37 +0200)]
parser: do not convert $ and @ in code values of %define variables

* src/parse-gram.y (value: "{...}"): Just strip the braces, but pass
the value as is.

11 years agoparser: no longer use the "braceless" non-terminal
Akim Demaille [Thu, 18 Apr 2013 14:36:54 +0000 (16:36 +0200)]
parser: no longer use the "braceless" non-terminal

The purpose of this symbol was only to factor function calls.  As a
result the actions were indeed simpler, but the grammar was somewhat
uselessly obfuscated.  Get rid of this symbol, but introduce functions
to simplify dependencies.

There is no (intended) changes of behavior here.

* src/parse-gram.y (strip_braces, translate_code(
(translate_code_braceless): New.
(braceless): Remove, use "{...}" instead, and one of the previous
functions depending on the context.
(STRING, "%{...%}", EPILOGUE): Declare as <code>, instead of <chars>,
the difference between both is useless (well, I couldn't make sense of
it, even after having read the initial commit that introduced them).
(%union): Remove the now useless "chars" type.
Adjust the printers.
* src/scan-gram.l: Adjust.

11 years agoregen
Akim Demaille [Thu, 18 Apr 2013 13:12:24 +0000 (15:12 +0200)]
regen