Paul Eggert [Tue, 19 Nov 2002 09:33:21 +0000 (09:33 +0000)]
* data/glr.c (yygetLRActions): Replace `yyindex' with
`yytable[yyindex]' to fix typo introduced in my 2002-11-09 patch.
This fixes the regression with Sun ONE Studio 7 cc that I reported in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001892.html>.
Paul Eggert [Tue, 19 Nov 2002 07:35:29 +0000 (07:35 +0000)]
(yygetLRActions): Replace `yyindex' with
`yytable[yyindex]' to fix typo introduced in my 2002-11-09 patch.
This fixes the regression with Sun ONE Studio 7 cc that I reported in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001892.html>.
Paul Eggert [Mon, 18 Nov 2002 07:51:47 +0000 (07:51 +0000)]
(YYDPRINTF, YYDSYMPRINT, YYDSYMPRINTF) [!YYDEBUG]: Do not define to
{}, since this breaks the common use of `YYDPRINTF ((...));' if a
single statement is desired (e.g. before `else'). Work around GCC
warnings by surrounding corresponding calls with {} if needed.
(yyhasResolvedValue): Remove unused function.
(yymergeOptionSets, yyresolvStack): Use `continue;' for empty
loop body.
(yyreportSyntaxError): Renamed from yyreportParseError.
(yyrecoverSyntaxError): Renamed from yyrecoverParseError.
All uses changed.
Akim Demaille [Sat, 16 Nov 2002 12:32:27 +0000 (12:32 +0000)]
Augment the similarity between GLR and LALR traces.
* data/yacc.c (yy_stack_print, YY_STACK_PRINT, yy_reduce_print)
(YY_REDUCE_PRINT): New.
(yyparse): Use them.
* data/glr.c (yy_reduce_print): Use YYFPRINTF, no need for
YYDPRINT here.
(yyglrReduce, yyrecoverParseError, yyparse): Don't report the
state reached after the reduction/recovery, since...
(yyparse, yyprocessOneStack): Report the state we are entering in.
Akim Demaille [Sat, 16 Nov 2002 12:31:36 +0000 (12:31 +0000)]
* src/getargs.h, src/getargs.c (trace_e, trace_args, trace_types):
Add support for --trace=skeleton.
* src/scan-skel.l: %option debug.
Scan strings of non-@ or n instead of character by character.
(scan_skel): Handle trace_skeleton.
(QPUTS): New.
(@output_parser_name@, @output_header_name@): ``Restore'' their
support (used to be M4 macros).
* data/yacc.c: Quote larger chunks, a la glr.c.
* data/lalr1.cc: Likewise.
The header guards are no longer available, so use some other
string than `YYLSP_NEEDED'.
Akim Demaille [Sat, 16 Nov 2002 12:30:23 +0000 (12:30 +0000)]
Make the ``Printers and Destructors'' test more verbose, taking
`yacc.c''s behavior as (possibly wrong) reference.
* tests/actions.at (_AT_CHECK_PRINTER_AND_DESTRUCTOR): Use printf
instead of fprint on stdout.
Set and report the last_line of the symbols.
Consistently display values and locations.
Akim Demaille [Fri, 15 Nov 2002 16:42:16 +0000 (16:42 +0000)]
* src/parse-gram.y (declaration): Have %parse-param and %lex-param
take two BRACED_CODE, not two string_content.
Free the scanner's obstack when we are done.
(code_content): New.
* tests/calc.at: Adjust.
* doc/bison.texinfo: Adjust.
Also, make sure to include the `,' for these declarations.
Akim Demaille [Fri, 15 Nov 2002 09:04:37 +0000 (09:04 +0000)]
2002-11-15 Tim Van Holder <tim.van.holder@pandora.be>
* src/.cvsignore: Add src-gram.c.
* tests/.cvsignore: Add atlocal.
* config/.cvsignore, m4/.cvsignore, po/.cvsignore:
Add several files that autoreconf may bring in during
bootstrap.
Akim Demaille [Fri, 15 Nov 2002 08:56:40 +0000 (08:56 +0000)]
Always check the value returned by yyparse.
* tests/calc.at (_AT_DATA_CALC_Y): Have `main' exit with the value
returned by yyparse.
(_AT_CHECK_CALC_ERROR): Take the expected exit value as argument.
Adjust calls.
* tests/glr-regr1.at (glr-regr1.y): Have `main' exit with the value
returned by yyparse.
Paul Eggert [Thu, 14 Nov 2002 23:08:27 +0000 (23:08 +0000)]
* tests/headers.at (AT_TEST_CPP_GUARD_H): Don't assume that
"$CC -E foo.h" is allowed, as this doesn't work with the Portland
Group compiler. Instead, use "$CC -E bar.c". Include the .h
file twice in the grammar, as an extra check.
* tests/input.at (Torturing the Scanner): Surround the
backslash-newline tests with "#if 0", to make it less likely that
we'll run into compiler bugs. Bring back solitary \ inside
comment, but add a closing comment to work around HP C bug. Don't
test backslash-newline in C character constant. This should fix
the input.at bug reported by Nelson H. F. Beebe in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001893.html>.
Paul Eggert [Thu, 14 Nov 2002 23:08:01 +0000 (23:08 +0000)]
(AT_TEST_CPP_GUARD_H): Don't assume that "$CC -E foo.h" is allowed, as
this doesn't work with the Portland Group compiler. Instead, use "$CC
-E bar.c". Include the .h file twice in the grammar, as an extra
check.
Paul Eggert [Thu, 14 Nov 2002 22:53:47 +0000 (22:53 +0000)]
(Torturing the Scanner): Surround the backslash-newline tests with
"#if 0", to make it less likely that we'll run into compiler bugs.
Bring back solitary \ inside comment, but add a closing comment to
work around HP C bug. Don't test backslash-newline in C character
constant. This should fix the input.at bug reported by Nelson
H. F. Beebe in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001893.html>.
Akim Demaille [Thu, 14 Nov 2002 11:09:51 +0000 (11:09 +0000)]
* tests/synclines.at (AT_SYNCLINES_COMPILE): Ignore the exit
status of the compiler.
Calling `exit 1' is no longer needed.
Reported by Nelson H. F. Beebe.
Akim Demaille [Thu, 14 Nov 2002 09:58:01 +0000 (09:58 +0000)]
* tests/atlocal.in (CPPFLAGS): We have config.h.
* tests/testsuite.at (AT_DATA_GRAMMAR_PROLOGUE, AT_DATA_GRAMMAR):
New.
* tests/actions.at, tests/calc.at, tests/conflicts.at,
* tests/cxx-type.at, tests/glr-regr1.at, tests/headers.at,
* tests/regression.at, tests/torture.at: Use them for all the
grammars that are to be compiled.
* tests/cxx-type.at (_AT_TEST_GLR_CALC): Rename as...
* tests/cxx-type.at (_AT_TEST_GLR_CXXTYPES): this.
* doc/bison.texinfo (GLR Parsers): Document `inline'.
Akim Demaille [Thu, 14 Nov 2002 08:32:10 +0000 (08:32 +0000)]
* doc/bison.texinfo: Various formatting changes (alignments in
samples, additional @group/@end group, GCS in samples.
Use @deffn instead of simple @table to define the directives,
macros, variables etc.
Paul Eggert [Wed, 13 Nov 2002 07:46:19 +0000 (07:46 +0000)]
(yydoAction): Return YYRESULTTAG, not int.
(yyglrReduce): Return yyok, not 0.
This should avoid the enumerated-type warnings reported
by Nelson H. F. Beebe in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001872.html>.
Paul Eggert [Wed, 13 Nov 2002 06:48:18 +0000 (06:48 +0000)]
(inline): Remove #define. It's the user's responsibility to #define
it away, just like 'const'. This fixes one of the bugs reported by
Nelson H.F. Beebe in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001873.html>.
Paul Eggert [Wed, 13 Nov 2002 06:39:26 +0000 (06:39 +0000)]
(po-check): Scan .l and .y files instead
of the .c and the .h files that they generate. Look for N_
as well as for _. Try to avoid matching #define for N_ and _.
Akim Demaille [Tue, 12 Nov 2002 09:03:13 +0000 (09:03 +0000)]
* tests/regression.at (input.y): s/YYEOF/MYEOF/, as the skeleton
defines it.
* data/glr.c (yystos): New.
(b4_yysymprint_generate, b4_yydestruct_generate): Invoke.
(YYDSYMPRINT): New.
(yyval): Don't define it, it is handled via M4.
(yyrecoverParseError): Free verbosely the discarded symbols.
* data/yacc.c (yysymprint): Remove, rather...
(b4_yysymprint_generate): invoke.
* data/c.m4 (b4_yysymprint_generate): New.
Accept pointers as arguments, as opposed to the version from
yacc.c.
(b4_yydestruct_generate): Likewise.
* tests/cations.at (Printers and Destructors): Use Bison directives
instead of CPP macros.
Don't rely on internal details.
Akim Demaille [Tue, 12 Nov 2002 08:45:40 +0000 (08:45 +0000)]
* data/c.m4 (b4_yydestruct_generate, b4_symbol_actions): New.
* data/yacc.c: Rename yychar1 as yytoken, as in glr.c.
Don't work on yychar (i.e., do set it to YYEMPTY, don't match
it against YYEMPTY and so forth), work on yytoken (i.e., set
it to YYEMPTY etc.).
(yydestruct): Replace with a b4_yydestruct_generate invocation.
(b4_symbol_actions): Remove.
* data/glr.c (YYTRANSLATE): As for yacc.c, if negative, it stands
for 0, end-of-input.
Akim Demaille [Tue, 12 Nov 2002 08:30:47 +0000 (08:30 +0000)]
* src/tables.c (tables_generate): Use free for pointers that
cannot be NULL, not XFREE.
(pack_vector): Use assert, not fatal, for bound violations.
* src/state.c (state_new): Likewise.
* src/reader.c (reader): Likewise.
* src/lalr.c (set_goto_map): Likewise.
* src/location.h (LOCATION_PRINT): If first_line is 0, just issue
the file name.
Akim Demaille [Tue, 12 Nov 2002 08:26:38 +0000 (08:26 +0000)]
* src/scan-gram.l, src/reader.h (scanner_last_string_free):
Restore.
* src/scan-gram.l (last_string): Is global to the file, not to
yylex.
* src/parse-gram.y (input): Don't append the epilogue here,
(epilogue.opt): do it here, and free the scanner's obstack.
* src/reader.c (epilogue_set): Rename as...
(epilogue_augment): this.
* data/c.m4 (b4_epilogue): Defaults to empty.
Akim Demaille [Tue, 12 Nov 2002 08:05:59 +0000 (08:05 +0000)]
* src/struniq.h, src/struniq.c (struniq_t): Is const.
(STRUNIQ_EQ, struniq_assert, struniq_assert_p): New.
Use struniq for symbols.
* src/symtab.h (symbol_t): The tag member is a struniq.
(symbol_type_set): Adjust.
* src/symtab.c (symbol_new): Takes a struniq.
(symbol_free): Don't free the tag member.
(hash_compare_symbol_t, hash_symbol_t): Rename as...
(hash_compare_symbol, hash_symbol): these.
Use the fact that tags as struniqs.
(symbol_get): Use struniq_new.
* src/symlist.h, src/symlist.c (symbol_list_n_type_name_get):
Returns a strniq.
* src/reader.h (merger_list, grammar_currentmerge_set): The name
and type members are struniqs.
* src/reader.c (get_merge_function)
(grammar_current_rule_merge_set): Adjust.
(TYPE, current_type): Are struniq.
Use struniq for file names.
* src/files.h, src/files.c (infile): Split into...
(grammar_file, current_file): these.
* src/scan-gram.c (YY_USER_INIT, handle_syncline): Adjust.
* src/reduce.c (reduce_print): Likewise.
* src/getargs.c (getargs): Likewise.
* src/complain.h, src/complain.c: Likewise.
* src/main.c (main): Call struniqs_new early enough to use it for
file names.
Don't free the input file name.
Akim Demaille [Tue, 12 Nov 2002 07:55:55 +0000 (07:55 +0000)]
* src/symtab.c (symbol_free): Remove dead deactivated code:
type_name are properly removed.
Don't use XFREE to free items that cannot be NULL.
* src/struniq.h, src/struniq.c: New.
* src/main.c (main): Initialize/free struniqs.
* src/parse-gram.y (%union): Add astruniq member.
(yyprint): Adjust.
* src/scan-gram.l (<{tag}>): Return a struniq.
Free the obstack bit that used to store it.
* src/symtab.h (symbol_t): The 'type_name' member is a struniq.
Paul Eggert [Tue, 12 Nov 2002 07:33:36 +0000 (07:33 +0000)]
Revamp to fix many (but not all) of the C- and M4-related quoting
problems. Among other things, this fixes the Bison bug reported
by Jan Hubicka when processing the Bash grammar.
Paul Eggert [Tue, 12 Nov 2002 07:27:51 +0000 (07:27 +0000)]
(yylineno, yyoutname): Remove static vars, replacing
them with auto vars.
Switch to new escape scheme, where @ is the escape character uniformly.
Abort if a stray escape character is found. Avoid unbounded input
buffer when parsing non-escaped text.
Paul Eggert [Tue, 12 Nov 2002 07:27:04 +0000 (07:27 +0000)]
(<SC_CHARACTER>): Don't worry about any backslash
escapes other than \\ and \'; this simplifies the code.
(<SC_STRING>): Likewise, for \\ and \".
(<SC_COMMENT,SC_LINE_COMMENT,SC_STRING,SC_CHARACTER,SC_BRACED_CODE,
SC_PROLOGUE,SC_EPILOGUE>): Escape $ and @, too.
Use new escapes @{ and @} for [ and ].
Paul Eggert [Tue, 12 Nov 2002 07:25:36 +0000 (07:25 +0000)]
(escaped_file_name_output): New function.
(prepare_symbols): Quote tokens for M4.
(prepare): Don't insert output_infix, output_prefix,
output_parser_name, output_header_name; this is now down by scan-skel.
Insert skeleton as a C string.
(user_actions_output, symbol_destructors_output,
symbol_printers_output): Quote filenames for C and M4.
Paul Eggert [Tue, 12 Nov 2002 07:22:01 +0000 (07:22 +0000)]
Use new @ escapes consistently.
(b4_input_suffix, b4_output_parser_suffix, b4_output_parser_name,
b4_output_header_suffix, b4_output_header_name, b4_header_guard):
Remove, since they couldn't handle arbitrary characters in file
names.
Remove use of "#ifdef b4_header_guard", since it
mishandled funny characters in file names, and anyway it isn't
needed any more.
(YYSTYPE_IS_TRIVIAL): Define when the .h file would.
Paul Eggert [Tue, 12 Nov 2002 07:21:01 +0000 (07:21 +0000)]
Use new @ escapes consistently.
(b4_input_suffix, b4_output_parser_suffix, b4_output_parser_name,
b4_output_header_suffix, b4_output_header_name, b4_header_guard):
Remove, since they couldn't handle arbitrary characters in file
names.
Use YYSLP_NEEDED instead of b4_header_guard.
Paul Eggert [Tue, 12 Nov 2002 07:15:59 +0000 (07:15 +0000)]
(b4_copyright): Omit file name from comment, since
the file name could contain "*/".
(b4_synclines_flag): Don't quote the 2nd argument; it should already
be quoted. All uses changed.
Paul Eggert [Sun, 10 Nov 2002 07:10:08 +0000 (07:10 +0000)]
Fix the test failure due to GCC warnings described in
<http://mail.gnu.org/pipermail/bug-bison/2002-November/001815.html>.
* data/glr.c (yyis_pact_ninf, yyis_table_ninf): New macros, which
evaluate to 0 if it's impossible for NINF to be in the respective
table.
(yygetLRActions, yyrecoverParseError): Use them.
* src/scan-gram.l (unexpected_end_of_file): Fix bug: columns were
counted in the token inserted at end of file. Now takes
location_t *, not location_t, so that the location can be
adjusted. All uses changed.
* tests/regression.at (Invalid inputs): Adjust wording in
diagnostic to match the new behavior.
* tests/torture.at (AT_DATA_TRIANGULAR_GRAMMAR,
AT_DATA_HORIZONTAL_GRAMMAR, AT_DATA_LOOKAHEADS_GRAMMAR,
AT_DATA_STACK_TORTURE): Replace `assert (x);' with `if (! (x))
abort ();'. This reduces the runtime of the "Many lookaheads"
test from 27.6 to 2.7 minutes on a 440 MHz Ultrasparc III running
GCC 3.2.
Paul Eggert [Sun, 10 Nov 2002 07:04:07 +0000 (07:04 +0000)]
(AT_DATA_TRIANGULAR_GRAMMAR, AT_DATA_HORIZONTAL_GRAMMAR,
AT_DATA_LOOKAHEADS_GRAMMAR, AT_DATA_STACK_TORTURE): Replace `assert
(x);' with `if (! (x)) abort ();'. This reduces the runtime of the
"Many lookaheads" test from 27.6 to 2.7 minutes on a 440 MHz
Ultrasparc III.
Paul Eggert [Sun, 10 Nov 2002 05:17:56 +0000 (05:17 +0000)]
(unexpected_end_of_file): Fix bug: columns were counted in the token
inserted at end of file. Now takes location_t *, not location_t, so
that the location can be adjusted. All uses changed.