Akim Demaille [Mon, 11 Feb 2013 08:07:25 +0000 (09:07 +0100)]
tests: enlarge the allowed duration for calc tests
Hydra "often" fails on this test:
252. calc.at:658: 252. Calculator %glr-parser api.pure
parse.error=verbose %debug %locations %defines api.prefix="calc"
%verbose %yacc %parse-param {semantic_value *result}
%parse-param {int *count} (calc.at:658): FAILED
* tests/calc.at: Give 200s instead of 100s.
Use AT_DEBUG_IF.
Akim Demaille [Sat, 9 Feb 2013 15:07:09 +0000 (16:07 +0100)]
debug: improve the display of symbol lists
* src/symtab.c (symbol_print): Remove useless quotes (the symbol already
has quotes).
Prefer fputs.
* src/symlist.c (symbol_list_syms_print): Likewise.
Fix separators.
Akim Demaille [Sat, 9 Feb 2013 17:25:15 +0000 (18:25 +0100)]
style: minor changes
* src/complain.c: Space changes.
* src/reader.c: Comment changes.
Avoid && in assertions.
* src/location.c: Move comments to...
* src/location.h: here.
* src/symlist.h, src/symlist.c: Create a pseudo section for members
that apply to the rule.
Akim Demaille [Fri, 8 Feb 2013 16:01:23 +0000 (17:01 +0100)]
news: restructure, document variants for C++
* NEWS: here.
Akim Demaille [Fri, 8 Feb 2013 14:25:18 +0000 (15:25 +0100)]
c++: api.token.constructor requires api.value.type=variant
Eventually it should also support "union".
* data/glr.cc: Move this check to...
* data/c++.m4: here, as lalr1.cc is affected too.
Akim Demaille [Tue, 5 Feb 2013 13:37:55 +0000 (14:37 +0100)]
build: restore C90 compatibility
* src/parse-gram.y, src/parse-gram.c: Don't use // comments.
Akim Demaille [Mon, 4 Feb 2013 16:55:27 +0000 (17:55 +0100)]
doc: use @group to improve page breaking
* doc/bison.texi: here.
Akim Demaille [Mon, 4 Feb 2013 17:35:51 +0000 (18:35 +0100)]
style: rename internal "stype" as "union_members" for clarity
"stype" is quite unclear, and it also collides with the former %define
variable that had the same name (replaced by api.value.type).
* src/parse-gram.y (stype): Rename as...
(union_members): this.
* data/bison.m4: Adjust.
(b4_user_stype): Rename as...
(b4_user_union_members): this.
* data/c++.m4, data/c.m4: Adjust.
* src/parse-gram.c: regen.
Akim Demaille [Mon, 4 Feb 2013 15:13:02 +0000 (16:13 +0100)]
tests: improve the language independance layer
* tests/local.at (_AT_LANG_DISPATCH): New, shamelessly stolen from
Autoconf's _AT_LANG_DISPATCH.
(AT_LANG_DISPATCH): New.
(AT_YYERROR_FORMALS, AT_YYERROR_PROTOTYPE, AT_YYERROR_DECLARE_EXTERN)
(AT_YYERROR_DECLARE, AT_YYERROR_DEFINE, AT_MAIN_DEFINE, AT_COMPILE)
(AT_FULL_COMPILE):
Use AT_LANG_DISPATCH instead of an ad hoc m4_case.
Akim Demaille [Mon, 4 Feb 2013 14:46:47 +0000 (15:46 +0100)]
regen
Akim Demaille [Mon, 4 Feb 2013 14:52:38 +0000 (15:52 +0100)]
style: space changes in the parser
* src/parse-gram.y: Fix spaces.
Akim Demaille [Mon, 4 Feb 2013 14:44:26 +0000 (15:44 +0100)]
parser: use api.pure full
* src/parse-gram.y: Use api.pure full instead of silly macro tricks.
Akim Demaille [Fri, 1 Feb 2013 16:52:01 +0000 (17:52 +0100)]
style: use a for loop instead of a while loop, and scope reduction
* src/reader.c (packgram): Improve readability.
The parser calls grammar_current_rule_end at the end of every rhs,
which adds a NULL to separate the rules. So there is no need to
check whether "p" is non-null before proceeding.
Theophile Ranquet [Fri, 1 Feb 2013 14:07:54 +0000 (15:07 +0100)]
variants: stylistic change
* data/variant.hh (tname): Respect the GNU Coding Standards for this
pointer's declaration.
Theophile Ranquet [Thu, 31 Jan 2013 23:49:59 +0000 (00:49 +0100)]
grammar: free the association tracking graph
The graph introduced by Valentin wasn't free'd after use.
* src/symtab.c (assoc_free): New, clear the array of linked lists with...
(linkedlist_free): This, new.
(print_precedence_warnings): Call assoc_free when done.
(print_assoc_warnings): Free used_assoc after use.
Theophile Ranquet [Thu, 31 Jan 2013 21:16:14 +0000 (22:16 +0100)]
tests: use AT_FULL_COMPILE where possible
* tests/c++.at (C++ Variant-based Symbol, Variants): Here. Rename the
generated input files to use .y instead of .yy, as a requirement for using
AT_FULL_COMPILE instead of a combination of AT_BISON_CHECK and
AT_BISON_COMPILE_CXX.
Theophile Ranquet [Tue, 29 Jan 2013 21:35:04 +0000 (22:35 +0100)]
variants: avoid type punning issue
This is based on what is recommended by both Scott Meyers, in 'Effective
C++', and Andrei Alexandrescu and Herb Sutter in 'C++ Coding Standards'.
Use a static_cast on void* rather than directly use a reinterpret_cast,
which can have nefarious effects on objects. However, even though following
this guideline is good practice in general, I am not quite sure how relevant
it is when applied to conversions from POD to objects. Actually, it might
very well be the opposite: isn't this exactly what reinterpret_cast is for?
What we really want *is* to transmit the memory map as a series of bytes,
which, if I am correct, falls into the kind of "low level" hack for which
this cast is meant.
In any case, this silences the warning, which will be greatly appreciated by
anyone using variants with a compiler supporting -fstrict-aliasing.
* data/variant.hh (as): Here.
* tests/c++.at (Exception safety, C++ Variant-based Symbols, Variants):
Don't use NO_STRICT_ALIAS_CXXFLAGS (revert commit
ddb9db15), as type punning
is no longer an issue.
* tests/atlocal.in, configure.ac (NO_STRICT_ALIAS_CXXFLAGS): Remove
definition.
* examples/local.mk (NO_STRICT_ALIAS_CXXFLAGS): Remove from AM_CXXFLAGS.
* doc/bison.texi: Don't mention type punning issues.
Theophile Ranquet [Thu, 31 Jan 2013 11:18:37 +0000 (12:18 +0100)]
todo: update
Reformulate and give more details on my thoughts concerning the graphical
visualization, and add an entry about a bug in the options processing for
warnings as errors.
* TODO: Here.
Akim Demaille [Fri, 1 Feb 2013 13:26:29 +0000 (14:26 +0100)]
regen
Akim Demaille [Fri, 1 Feb 2013 13:24:48 +0000 (14:24 +0100)]
location: pass the location first
* src/location.h, src/location.c (location_print): For consistency
with other data structures and other location_* routines, pass the
location argument first.
* src/complain.c: Adjust.
(location_caret): Likewise.
* src/parse-gram.y: Adjust.
Akim Demaille [Fri, 1 Feb 2013 13:23:49 +0000 (14:23 +0100)]
symlist: use the right stream
* src/symlist.c (symbol_list_syms_print): Use "f", not stderr.
Akim Demaille [Wed, 30 Jan 2013 20:41:41 +0000 (21:41 +0100)]
tests: put two related tests together
* tests/conflicts.at (Useless associativity warning): Move next
to "Useless precedence warning".
Akim Demaille [Wed, 30 Jan 2013 20:40:47 +0000 (21:40 +0100)]
news: name contributors
* NEWS: here.
Valentin Tolmer [Wed, 30 Jan 2013 10:30:15 +0000 (11:30 +0100)]
warnings: introduce -Wprecedence
The new warning category "precedence" flags useless precedence and
associativity. -Wprecedence can now be used, it is disabled by default.
The warnings about precedence and associativity are grouped into one, and
the testsuite was corrected accordingly.
* src/complain.h (warnings): Introduce "precedence".
* src/complain.c (warnings_print_categories): Adjust.
* src/getargs.c (warnings_args, warning_types): Likewise.
* src/symtab.h, src/symtab.c (print_associativity_warnings): Remove.
* src/symtab.h (register_assoc): Correct arguments.
* src/symtab.c (print_precedence_warnings): Print both warnings together.
* doc/bison.texi (Bison options): Document the warnings and provide an
example.
* tests/conflicts.at, tests/existing.at, tests/local.at,
* tests/regression.at: Adapt the testsuite for the new category
(-Wprecedence instead of -Wother where appropriate).
Akim Demaille [Wed, 30 Jan 2013 14:52:34 +0000 (15:52 +0100)]
build: avoid clang's colored diagnostics in the test suite
The syncline tests, which try to recognize compiler diagnostics,
are confused by escapes for colors.
* configure.ac (warn_tests): New, to factor the warnings for both
C and C++ tests.
Add -fno-color-diagnostics to it.
* tests/local.at (AT_TEST_TABLES_AND_PARSE): Do not remove glue
together compiler flags.
Akim Demaille [Wed, 30 Jan 2013 14:28:08 +0000 (15:28 +0100)]
build: please Clang++ 3.2+ on Flex scanners
Clang++, with -Wall, rejects code generated by Flex (for C scanners):
CXX examples/calc++/examples_calc___calc__-calc++-scanner.o
In file included from examples/calc++/calc++-scanner.cc:1:
error: implicit conversion of NULL constant to 'bool' [-Werror,-Wnull-conversion]
if ( ! ( (yy_buffer_stack) ? (yy_buffer_stack)[(yy_buffer_stack_top)] : __null) ) {
~ ^~~~~~
false
* configure.ac (WARN_NO_NULL_CONVERSION_CXXFLAGS): Compute it.
* examples/calc++/local.mk (examples_calc___calc___CXXFLAGS): Use it.
Valentin Tolmer [Tue, 29 Jan 2013 15:27:04 +0000 (16:27 +0100)]
grammar: record used associativity and print useless ones
Record which symbol associativity is used, and display useless ones.
* src/symtab.h, src/symtab.c (register_assoc, print_assoc_warnings): New
* src/symtab.c (init_assoc, is_assoc_used): New
* src/main.c: Use print_assoc_warnings
* src/conflicts.c: Use register_assoc
* tests/conflicts.at (Useless associativity warning): New.
Due to the new warning, many tests had to be updated.
* tests/conflicts.at tests/existing.at tests/regression.at:
Add the associativity warning in the expected results.
* tests/java.at: Fix the java calculator's grammar to remove a useless
associativity.
* doc/bison.texi (mfcalc example): Fix associativity to remove
warning.
Valentin Tolmer [Tue, 29 Jan 2013 13:55:53 +0000 (14:55 +0100)]
grammar: warn about unused precedence for symbols
Symbols with precedence but no associativity, and whose precedence is
never used, can be declared with %token instead. The used precedence
relationships are recorded and a warning about useless ones is issued.
* src/conflicts.c (resolve_sr_conflict): Record precedence relation.
* src/symtab.c, src/symtab.h (prec_nodes, init_prec_nodes)
(symgraphlink_new, register_precedence_second_symbol)
(print_precedence_warnings): New.
Record relationships in a graph and warn about useless ones.
* src/main.c (main): Print precedence warnings.
* tests/conflicts.at: New.
Theophile Ranquet [Tue, 29 Jan 2013 13:53:35 +0000 (14:53 +0100)]
variants: remove the 'built' assertions
When using %define parse.assert, the variants come with additional variables
that are useful for development purposes. One is a Boolean indicating if the
variant is built (to make sure we don't read a non-built variant), and the
other is a string describing the stored type. There is no need to have both of
these, the string is enough.
* data/variant.hh (built): Remove.
Akim Demaille [Tue, 29 Jan 2013 13:23:44 +0000 (14:23 +0100)]
style: indentation fixes
* src/parse-gram.y: here.
Akim Demaille [Tue, 29 Jan 2013 13:11:46 +0000 (14:11 +0100)]
maint: be sure to neutralize out-of-tree paths from our parser
* tests/bison.in: Adjust to support fixed versions of ylwrap.
Theophile Ranquet [Mon, 28 Jan 2013 17:26:04 +0000 (18:26 +0100)]
m4: generate a basic_symbol constructor for each symbol type
Recently, there was a slightly vicious bug hidden in the make_ functions:
parser::symbol_type
parser::make_TEXT (const ::std::string& v)
{
return symbol_type (token::TOK_TEXT, v);
}
The constructor for symbol_type doesn't take an ::std::string& as
argument, but a constant variant. However, because there is a variant
constructor which takes an ::std::string&, this caused the implicit
construction of a built variant. Considering that the variant argument
for the symbol_type constructor was cv-qualified, this temporary variant
was never destroyed.
As a temporary solution, the symbol was built in two stages:
symbol_type res (token::TOK_TEXT);
res.value.build< ::std::string&> (v);
return res;
However, the solution introduced in this patch contributes to letting
the symbols handle themselves, by supplying them with constructors that
take a non-variant value and build the symbol's own variant with that
value.
* data/variant.hh (b4_symbol_constructor_define_): Use the new
constructors rather than building in a temporary symbol.
(b4_basic_symbol_constructor_declare,
b4_basic_symbol_constructor_define): New macros generating the
constructors.
* data/c++.m4 (basic_symbol): Invoke the macros here.
Theophile Ranquet [Mon, 28 Jan 2013 17:03:58 +0000 (18:03 +0100)]
c++: minor stylistic changes
* data/c++m4: Remove useless comment lines.
* data/variant.hh (self_type): Use this typedef instead of variant<S>.
(b4_symbol_constructor_define_): Remove commented-out line, and stylistic
change (avoid blank line).
Akim Demaille [Tue, 29 Jan 2013 07:58:51 +0000 (08:58 +0100)]
c++: please G++ 4.8 with -O3: type puning issue
* tests/c++.at (Exception safety): Now that this test covers
variants, pass -fno-strict-aliasing to g++.
Akim Demaille [Tue, 29 Jan 2013 07:52:57 +0000 (08:52 +0100)]
c++: please G++ 4.8 with -O3: array bounds
* data/c++.m4, data/lalr1.cc (by_state, by_type): Do not use -1 to
denote the absence of value, as GCC then fears that this -1 might
be used to dereference arrays (such as yytname).
Use 0, which corresponds to $accept, which is valueless (the needed
property: the symbol destructor must not try to reclaim the memory
associated with the symbol).
Akim Demaille [Tue, 29 Jan 2013 07:16:15 +0000 (08:16 +0100)]
c++: use more explicit types than int
* data/c++.m4 (b4_public_types_declare): Declare token_number_type soon.
Introduce symbol_number_type (wider than token_number_type).
Clarify the requirement that kind_type from by_state and by_type
denote the _input_ type (required by the constructor), not the stored type.
Use symbol_number_type and token_number_type where appropriate, instead
of int.
* data/lalr1.cc: Adjust to these changes.
Propagate "symbol_number_type".
Invoke "type_get ()" instead of read "type" directly.
Akim Demaille [Tue, 29 Jan 2013 06:07:45 +0000 (07:07 +0100)]
c++: value_type -> kind_type
* data/c++.m4, data/lalr1.cc (by_type, by_state): Rename 'value_type'
as 'kind_type', as it is clearer.
Akim Demaille [Tue, 29 Jan 2013 07:14:15 +0000 (08:14 +0100)]
c++: improve the signature of yysyntax_error_
* data/lalr1.cc: This function is const.
It takes a symbol_number_type.
Akim Demaille [Tue, 29 Jan 2013 07:15:37 +0000 (08:15 +0100)]
c++: style changes
* data/lalr1.cc: Formatting changes.
And name changes.
Akim Demaille [Mon, 28 Jan 2013 17:27:15 +0000 (18:27 +0100)]
doxygen: upgrade Doxyfile, and complete it
* doc/Doxyfile.in: Let doxygen upgrade it.
(INCLUDE_PATH): Point to lib too.
(PROJECT_BRIEF): New.
(EXCLUDE): Update to reflect the current file hierarchy.
Akim Demaille [Mon, 28 Jan 2013 16:17:12 +0000 (17:17 +0100)]
maint: fix syntax-check issues
* cfg.mk: Ignore strcmp in local.at.
* tests/conflicts.at: Use AT_PARSER_CHECK.
* tests/regression.at: Preserve the exit status of the generated parsers.
* tests/local.mk ($(TESTSUITE)): Map @tb@ to a tabulation.
* tests/c++.at, tests/input.at, tests/regression.at: Use @tb@.
* cfg.mk: (space-tab): There are no longer exceptions.
Akim Demaille [Mon, 28 Jan 2013 16:10:30 +0000 (17:10 +0100)]
tests: please C90 compilers
* tests/actions.at, tests/conflicts.at: Use /* ... */ comments.
Let "main" return a value.
Akim Demaille [Mon, 28 Jan 2013 16:08:45 +0000 (17:08 +0100)]
maint: update todo
* TODO: Remove fixed items.
Akim Demaille [Mon, 28 Jan 2013 16:02:55 +0000 (17:02 +0100)]
news: minor improvements
* NEWS: Name some more contributors.
Restructure slightly.
Akim Demaille [Mon, 28 Jan 2013 15:50:50 +0000 (16:50 +0100)]
tests: please clang and use ".cc", not ".c", for C++ input
When fed with foo.c, clang++ 3.2 answers:
clang: error: treating 'c' input as 'c++' when in C++ mode,
this behavior is deprecated
* tests/output.at (AT_CHECK_OUTPUT_FILE_NAME): Use *.cc and *.hh
for C++.
Akim Demaille [Mon, 28 Jan 2013 15:44:00 +0000 (16:44 +0100)]
tests: formatting changes
* tests/local.at: Restore proper indentation.
Theophile Ranquet [Mon, 28 Jan 2013 16:41:31 +0000 (17:41 +0100)]
c++: better inline expansion
Many 'inline' keywords were in the declarations. They rather belong in
definitions, so move them.
* data/c++.m4 (basic_symbol, by_type): Many inlines here.
* data/lalr1.cc (yytranslate_, yy_destroy_, by_state, yypush_, yypop_): Inline
these as well.
(move): Move the definition outside the struct, where it belongs.
Akim Demaille [Mon, 28 Jan 2013 15:05:09 +0000 (16:05 +0100)]
tests: check that using variants is exception safe
* tests/local.at: (Slightly) improve the regexp by escaping '.'
when it denotes a point.
(AT_VARIANT_IF): New.
* tests/c++.at (Exception Safety): Run it for variants too.
Akim Demaille [Mon, 28 Jan 2013 14:37:49 +0000 (15:37 +0100)]
tests: remove useless %defines
Many tests were using %defines because C++ skeletons used to require
it.
* tests/actions.at, tests/c++.at, tests/input.at, tests/regression.at:
Remove useless %defines.
Akim Demaille [Mon, 28 Jan 2013 13:56:16 +0000 (14:56 +0100)]
c++: remove now-useless operators
Now that symbols behaves properly, we can eliminate special routines
that are no longer needed.
* data/c++.m4, data/glr.cc, data/lalr1.cc, data/variant.hh:
Remove useless assignment operators and copy constructors.
As a consequence, remove useless includes for "abort".
Akim Demaille [Mon, 28 Jan 2013 13:55:14 +0000 (14:55 +0100)]
tests: enable support for --debug
* tests/c++.at (Variants): Here.
And remove useless clutter when api.token.constructor is enabled.
Akim Demaille [Mon, 28 Jan 2013 13:29:43 +0000 (14:29 +0100)]
c++: revamp the support for variants
The current approach was too adhoc: the symbols were not sufficiently
self-contained, in particular wrt memory management. The "new"
guideline is the one that should have been followed from the start:
let the symbols handle themslves, instead of leaving their users to
it. It was justified by the will to avoid gratuitious moves and
copies, but the current approach does not seem to be slower, yet it
will probably be simpler to adjust to support move semantics from
C++11.
The documentation says that the %parse-param are available from the
%destructor. In retrospect, that was a silly design decision, which
we can break for variants, as its a new feature. It should be phased
out for non-variants too.
* data/variant.hh: A variant never knows if it stores something or
not, it is up to its users to store this information.
Yet, in parse.assert mode, make sure the empty/filled variants
are properly used.
(b4_symbol_constructor_define_): Don't call directly the symbol
constructor, to save a useless temporary.
* data/stack.hh (push): Steal the pushed value instead of duplicating
it.
This will simplify the callers of push, who handled this "move"
approach themselves.
* data/c++.m4 (basic_symbol): Let -1, as kind, denote the fact that
a symbol is empty.
This is needed for instance when shifting the lookahead: yyla
is given as argument to "push", and its value is then moved on
the stack. But then yyla must be declared "empty" so that its
destructor won't be called.
(basic_symbol::move): New.
Move the responsibility of calling the destructor from yy_destroy
to ~basic_symbol in the case of variants.
* data/lalr1.cc (stack_symbol_type): Now a derived class from its
previous value, so that we can add a constructor from a symbol_type.
(by_state): State -1 means empty.
(yypush_): Factor, by calling one overload from the other one, and
using the new semantics of stack::push.
No longer reclaim by hand the memory from rhs symbols, since now
that we store objects with proper destructors, they will be reclaimed
automatically.
Conversely, be sure to delete yylhs.
* tests/c++.at (C++ Variant-based Symbols): New "unit" test for
symbols.
Akim Demaille [Mon, 28 Jan 2013 13:22:51 +0000 (14:22 +0100)]
c++: formatting and comment changes
* data/c++.m4, data/lalr1.cc, data/stack.hh, data/variant.hh:
Fix indentation.
Fix some comments.
Valentin Tolmer [Fri, 25 Jan 2013 10:12:49 +0000 (11:12 +0100)]
tests: add token declaration order test
* tests/conflicts.at: New test.
Akim Demaille [Sun, 27 Jan 2013 09:41:16 +0000 (10:41 +0100)]
regen
Valentin Tolmer [Fri, 25 Jan 2013 10:12:47 +0000 (11:12 +0100)]
grammar: preserve token declaration order
In a declaration %token A B, the token A is declared before B, but in %left
A B (or with %precedence or %nonassoc or %right), the token B was declared
before A (tokens were declared in reverse order).
* src/symlist.h, src/symlist.c (symbol_list_append): New.
* src/parse-gram.y: Use it instead of symbol_list_prepend.
* tests/input.at: Adjust expectations.
Akim Demaille [Fri, 25 Jan 2013 10:06:32 +0000 (11:06 +0100)]
tests: improve test group titles
* tests/local.at (AT_SETUP_STRIP): AT_SETUP does not behave properly
with new-lines in its argument.
Remove them.
Fix the handling of %define with quotes.
Akim Demaille [Fri, 25 Jan 2013 12:51:33 +0000 (13:51 +0100)]
c: no longer require stdio.h when locations are enabled
Recent changes (in 2.7) introduced a dependency on both FILE and
fprintf, which are "available" only in %debug mode. This was to
define yy_location_print_, which is used only in %debug mode by the
parser, but massively used by the test suite to output the locations
in yyerror.
Break this dependency: the test suite should define its own routines
to display the locations. Eventually Bison will provide the user with
a means to display locations, but not yet.
* data/c.m4 (b4_yy_location_print_define): Use YYFPRINTF instead of
fprintf directly.
* data/yacc.c (b4_yy_location_print_define): Invoke it only in %debug
mode, so that stdio.h is included (needed for FILE*), and YYFPRINTF
is defined.
* tests/local.at (AT_YYERROR_DECLARE, AT_YYERROR_DEFINE): Declare
and define location_print and LOCATION_PRINT.
* tests/actions.at, tests/existing.at, tests/glr-regression.at,
* tests/input.at, tests/named-refs.at, tests/regression.at: Adjust
to use them.
Fix the expected line numbers (as the prologue's length has changed).
Akim Demaille [Fri, 25 Jan 2013 13:57:42 +0000 (14:57 +0100)]
c: minor simplification in the debug code
* data/c.m4 (yy_symbol_print): Minor factoring.
Akim Demaille [Fri, 25 Jan 2013 13:14:01 +0000 (14:14 +0100)]
c++: display locations as C does
See commit
3804aa260b956dd012adde3894767254422a5fcf.
* data/location.cc (operator<<): Display location exactly as is
done in C skeletons.
* tests/local.at (AT_LOC_PUSHDEF, AT_LOC_POPDEF): Also define
AT_FIRST_LINE, AT_LAST_LINE, AT_FIRST_COLUMN, AT_LAST_COLUMN.
* tests/actions.at (Location Print): Also check C++ skeletons.
Akim Demaille [Fri, 25 Jan 2013 11:01:10 +0000 (12:01 +0100)]
tests: highlight empty right-hand sides
* tests/actions.at, tests/c++.at, tests/headers.at,
* tests/input.at: here.
Akim Demaille [Fri, 25 Jan 2013 15:30:56 +0000 (16:30 +0100)]
news: prepare for 2.8
* NEWS: Restructure.
Name contributors.
Akim Demaille [Mon, 21 Jan 2013 15:01:53 +0000 (16:01 +0100)]
tests: generalize default main for api.namespace
* tests/local.at (AT_NAME_PREFIX): Also match api.namespace.
(AT_MAIN_DEFINE): Take it into account.
* tests/c++.at, tests/headers.at: Use AT_NAME_PREFIX.
(AT_CHECK_NAMESPACE): Rename as...
(AT_TEST): this.
Akim Demaille [Mon, 21 Jan 2013 14:38:49 +0000 (15:38 +0100)]
tests: improve factoring of the main function
* tests/local.at (AT_MAIN_DEFINE): If %debug is used, check if
-d/--debug is passed to the generated parser, and enable the traces.
Return exactly the result of yyparse, so that we can check exit code
2 too.
* tests/actions.at, tests/glr-regression.at, tests/regression.at:
Use AT_MAIN_DEFINE, helping AT_BISON_OPTION_PUSHDEFS where needed,
preferably to option -t.
Akim Demaille [Tue, 15 Jan 2013 17:29:45 +0000 (18:29 +0100)]
tests: factor the definition of main
With Théophile Ranquet.
* tests/local.at (AT_MAIN_DEFINE): New.
(AT_YYERROR_DEFINE): Improve formatting.
* tests/actions.at, tests/c++.at, tests/conflicts.at,
* tests/glr-regression.at, tests/input.at, tests/regression.at,
* tests/skeletons.at, tests/torture.at: Adjust.
* tests/c++.at: Add missing %skeleton for a PUSHDEFS, and a missing
PUSH/POPDEFS for another test.
Akim Demaille [Wed, 16 Jan 2013 07:25:50 +0000 (08:25 +0100)]
tests: minor refactoring
* tests/named-refs.at: Use AT_FULL_COMPILE where applicable.
Akim Demaille [Tue, 15 Jan 2013 16:02:03 +0000 (17:02 +0100)]
diagnostics: avoid useless caret stuttering
* src/scan-code.l: When reporting a missing ending ';', don't display
the guilty action twice.
Theophile Ranquet [Tue, 15 Jan 2013 16:51:45 +0000 (17:51 +0100)]
examples: please clang
* doc/bison.texi (calc++-scanner.ll): Don't output useless yyinput function.
Theophile Ranquet [Tue, 15 Jan 2013 17:03:39 +0000 (18:03 +0100)]
tests: better silencing of unused argument warnings
input.yy:35:44: error: unused parameter 'msg' [-Werror,-Wunused-parameter]
void yy::parser::error (std::string const& msg)
^
* tests/c++.at (C++ GLR parser identifier shadowing): Don't name unused
argument, use YYUSE instead of a direct cast to void.
Theophile Ranquet [Tue, 15 Jan 2013 16:54:44 +0000 (17:54 +0100)]
bench: compatibility for Bison <= 2.7
There used to be a bug in some skeletons, which caused the expansion of
'yylval' and 'yylloc', generating these errors:
input.cc:547:16: error: expected ',' or '...' before '(' token
#define yylval (yystackp->yyval)
^
input.yy:29:39: note: in expansion of macro 'yylval'
int yylex (yy::parser::semantic_type *yylval)
^
This bug is fixed by 'skel: better aliasing of identifiers', but a workaround
is useful when benchmarking against older versions of Bison, which are still
affected by the bug.
* etc/bench.pl.in: Rename yylval to yylvalp and yylloc to yyllocp in base
grammar 'list'.
Theophile Ranquet [Tue, 15 Jan 2013 12:05:21 +0000 (13:05 +0100)]
c++: remove useless inlines
* data/c++.m4 (basic_symbol): Keep 'inline' in the prototypes, but don't
duplicate it in the implementation.
* data/variant.hh (variant): 'inline' is not needed when the implementation is
provided in the class definition.
Theophile Ranquet [Tue, 15 Jan 2013 11:44:37 +0000 (12:44 +0100)]
c++: m4 stylistic change
* data/c++.m4 (syntax_error): Fix the indentation of 'inline'.
Theophile Ranquet [Mon, 14 Jan 2013 18:25:35 +0000 (19:25 +0100)]
c++: silence warnings
* data/c++.m4 (basic_symbol<Base>::operator=): Unused parameter.
* tests/c++.at (C++ GLR parser identifier shadowing): Here too.
-
Theophile Ranquet [Mon, 14 Jan 2013 10:02:12 +0000 (11:02 +0100)]
news: typos
* NEWS: Fix a typo, use YYSTYPE rather than semantic_type.
Akim Demaille [Sat, 12 Jan 2013 15:22:13 +0000 (16:22 +0100)]
regen
Akim Demaille [Sat, 12 Jan 2013 15:14:04 +0000 (16:14 +0100)]
maint: update copyright years
Suggested by Stefano Lattarini.
Run "make update-copyright".
Akim Demaille [Sat, 12 Jan 2013 15:03:15 +0000 (16:03 +0100)]
build: fix VPATH issue
* Makefile.am (update-b4-copyright, update-package-copyright-year): Fix
path to build-aux.
Theophile Ranquet [Fri, 11 Jan 2013 12:47:57 +0000 (13:47 +0100)]
carets: document default activation
* NEWS: Announce it.
* doc/bison.texi: Adjust.
Theophile Ranquet [Fri, 28 Dec 2012 12:33:04 +0000 (13:33 +0100)]
carets: show them in more tests
* tests/input.at, tests/named-refs.at: Here.
Theophile Ranquet [Fri, 28 Dec 2012 12:32:14 +0000 (13:32 +0100)]
carets: activate by default
* src/getargs.c (feature_flag): Here.
* tests/local.at (AT_BISON_CHECK_, AT_BISON_CHECK_NO_XML): Deactivate carets
for the testsuite, by default.
* tests/input.at: Adjust the locations for command line definitions.
Theophile Ranquet [Fri, 11 Jan 2013 11:38:35 +0000 (12:38 +0100)]
variants: document move and swap
* data/variant.hh (swap): Doc.
(build): Rename as...
(move): This, more coherent naming with clearer meaning.
* data/c++.m4 (move): Adjust.
Theophile Ranquet [Fri, 11 Jan 2013 10:41:07 +0000 (11:41 +0100)]
c++: privatize variant blind copies
* data/variant.hh (variant, operator=): Make private.
* data/c++.m4 (operator=): New, to avoid needing a definition of that operator
for each class member (such as a possible variant).
* data/glr.cc, data/lalr.cc: Add the necessary include for the abort.
Akim Demaille [Fri, 4 Jan 2013 15:01:07 +0000 (16:01 +0100)]
glr.c: fix an unused argument issue
* data/glr.c (yyuserAction): "Use" yyrhslen, as in variant mode, we might
not use it.
Akim Demaille [Fri, 4 Jan 2013 15:00:59 +0000 (16:00 +0100)]
glr.c: style changes
* data/glr.c (yyuserAction): Use a size_t for sizes.
Akim Demaille [Fri, 4 Jan 2013 14:58:50 +0000 (15:58 +0100)]
c.m4: style fix
* data/c.m4 (b4_parse_param_use): Add missing space before paren.
Theophile Ranquet [Fri, 4 Jan 2013 11:30:01 +0000 (12:30 +0100)]
skel: better aliasing of identifiers
* data/glr.c, data/yacc.c: Avoid emitting useless defines.
* data/glr.cc: Restore prefixes for epilogue.
Theophile Ranquet [Fri, 4 Jan 2013 17:52:21 +0000 (18:52 +0100)]
glr.cc: fatal if using api.token.ctor without variants
* data/glr.cc: Here.
Theophile Ranquet [Wed, 9 Jan 2013 11:46:55 +0000 (12:46 +0100)]
skel: correctly indent switch cases
* data/bison.m4 (b4_type_action_): Here.
Theophile Ranquet [Fri, 21 Dec 2012 15:48:54 +0000 (16:48 +0100)]
variants: assert changes
* data/variant.hh (swap): More asserts can't hurt. Don't perform useless swaps.
(build): Deactivate problematic asserts, pending further investigation.
(variant): Prohibit copy construction.
Theophile Ranquet [Fri, 21 Dec 2012 15:46:05 +0000 (16:46 +0100)]
lalr1.cc: use a vector for the symbol stack
* data/lalr1.cc: Adjust includes.
* data/stack.hh (push, pop): Use push_back and pop_back.
(operator []): Access vector from the end.
Theophile Ranquet [Fri, 21 Dec 2012 15:49:48 +0000 (16:49 +0100)]
lalr1.cc: change symbols implementation
A "symbol" groups together the symbol type (INT, PLUS, etc.), its
possible semantic value, and its optional location. The type is
needed to access the value, as it is stored as a variant/union.
There are two kinds of symbols. "symbol_type" are "external symbols":
they have type, value and location, and are returned by yylex.
"stack_symbol_type" are "internal symbols", they group state number,
value and location, and are stored in the parser stack. The type of
the symbol is computed from the state number.
The class template symbol_base_type<Exact> factors the code common to
stack_symbol_type and symbol_type. It uses the Curiously Recurring
Template pattern so that we can always (static_) downcast to the exact
type. symbol_base_type features value and location, and delegates the
handling of the type to its parameter.
When trying to generalize the support for variant, a significant issue
was revealed: because stack_symbol_type and symbol_type _derive_ from
symbol_base_type, the type/state member is defined _after_ the value
and location. In C++ the order of the definition of the members
defines the order in which they are initialized, things go backward:
the value is initialized _before_ the type. This is wrong, since the
type is needed to access the value.
Therefore, we need another means to factor the common code, one that
ensures the order of the members.
The idea is simple: define two (base) classes that code the symbol
type ("by_type" codes it by its type, and "by_state" by the state
number). Define basic_symbol<Base> as the class template that
provides value and location support. Make it _derive_ from its
parameter, by_type or by_state. Then define stack_symbol_type and
symbol_type as basic_symbol<by_state>, basic_symbol<by_type>. The
name basic_symbol was chosen by similarity with basic_string and
basic_ostream.
* data/c++.m4 (symbol_base_type<Exact>): Remove, replace by...
(basic_symbol<Base>): which derives from its parameter, one of...
(by_state, by_type): which provide means to retrieve the actual type of
symbol.
(symbol_type): Is now basic_symbol<by_type>.
(stack_symbol_type): Is now basic_symbol<by_state>.
* data/lalr1.cc: Many adjustments.
Theophile Ranquet [Wed, 26 Dec 2012 15:26:17 +0000 (16:26 +0100)]
bench: add %b directive to use a specific Bison
For example,
$ bench.pl -v '%s lalr1.cc & %d variant & ( %b ~/old-bison/bin/bison
| %b ~/new-bison/bin/bison )' -g list -i 10000
* etc/bench.pl.in: Here.
Akim Demaille [Fri, 4 Jan 2013 15:30:16 +0000 (16:30 +0100)]
regen
Akim Demaille [Fri, 28 Dec 2012 10:49:19 +0000 (11:49 +0100)]
gnulib: update
Akim Demaille [Mon, 31 Dec 2012 15:21:34 +0000 (16:21 +0100)]
doc: use deffn to declare the list of %define variables
* doc/bison.texi (%define Summary): Use @deffn instead of @table, it
spares a lot of width, especially in PDF, and looks nicer in the other
formats too.
It is also more consistent with the rest of the document.
Akim Demaille [Mon, 31 Dec 2012 10:18:48 +0000 (11:18 +0100)]
doc: minor completion and fixes
* doc/bison.texi (%define Summary): Provide more history to some
variables.
Akim Demaille [Mon, 31 Dec 2012 10:13:50 +0000 (11:13 +0100)]
java: stype is obsoleted by api.value.type
This is consistent with the other %define variable names.
* data/java.m4: Use api.value.type instead of stype.
* doc/bison.texi, NEWS: Document that change.
* src/muscle-tab.c (muscle_percent_variable_update): Provide backward
compatibility.
* tests/java.at: Adjust.
Akim Demaille [Mon, 31 Dec 2012 15:31:14 +0000 (16:31 +0100)]
doc: fix html build
* doc/local.mk (bison.html): Fix dependencies.
Akim Demaille [Sun, 30 Dec 2012 08:28:42 +0000 (09:28 +0100)]
todo: remove erroneous task
* tests/input.at: Check that there are no warnings about stray $ and @
in the epilogue.
* TODO: Remove the correponding task.
Akim Demaille [Fri, 28 Dec 2012 10:49:19 +0000 (11:49 +0100)]
gnulib: update