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.
* 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.
"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.
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.
* 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.
* src/fixits.c (fixits_run): If erase the content of a line, also
erase the following \n.
* tests/input.at (Deprecated directives): Update expectations.
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.
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.
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.
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.
* 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.
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.
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.
* 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.
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'.
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.
Commit 90a8537e62 was right, but issued
two error messages. Commit 80ef7e7639
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.
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.
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.
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.
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.
* 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.
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
d132c2d545), 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.
Commit 112ccb5ed7 moved the skeletons
from dist_pkgdata_DATA to dist_skeletons_DATA, hence broke the dependencies.
* Makefile.am (dependencies): New.
Use it where appropriate.
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.
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.
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.
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.
* 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.
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.
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.
Bitten by macros, again.
See 680b715518.
* 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.
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.
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 7928c3e6fb.
* 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
20b0746793.
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.
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.
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.
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.
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.
* 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.
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.
* 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.
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.
* 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.
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 3d3bc1fe30
Get rid of (yy)rhs and (yy)prhs
- commit d333175f63
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.
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.
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.
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.
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)]
$default reduce using rule 7 (a)
exp go to state 1
a go to state 2
b go to state 3
See https://lists.gnu.org/archive/html/bison-patches/2013-02/msg00106.html.
* src/conflicts.c (rule_has_state_rr_conflicts): Rename as...
(count_rule_state_sr_conflicts): this.
DWIM.
(count_rule_rr_conflicts): Adjust.
* tests/conflicts.at (%expect-rr in grammar rules)
(%expect-rr too much in grammar rules)
(%expect-rr not enough in grammar rules): New.
On a grammar such as
exp: "num" | "num" | "num"
we currently report only one RR conflict, instead of two.
This bug is present since the origins of Bison
commit 08089d5d35
Author: David MacKenzie <[email protected]>
Date: Tue Apr 20 05:42:52 1993 +0000
Initial revision
and was preserved in
commit 676385e29c
Author: Paul Hilfinger <[email protected]>
Date: Fri Jun 28 02:26:44 2002 +0000
Initial check-in introducing experimental GLR parsing. See entry in
ChangeLog dated 2002-06-27 from Paul Hilfinger for details.
See
https://lists.gnu.org/archive/html/bison-patches/2018-11/msg00011.html
* src/conflicts.h, src/conflicts.c (count_state_rr_conflicts)
(count_rr_conflicts): Use only the correct count of conflicts.
* tests/glr-regression.at: Fix expectations.
Currently on a grammar such as
exp: "number" | exp "+" exp | exp "*" exp
we count only one sr-conflict for both binary rules, i.e., we expect:
exp: "number" | exp "+" exp %expect 1 | exp "*" exp %expect 1
although there are 4 conflicts in total. That's because in the states
in conflict, for instance that for the "+" rule:
State 6
2 exp: exp . "+" exp
2 | exp "+" exp . [$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.
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."
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.
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.
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.
* 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.
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.
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.
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.
* 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.
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.
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.
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.
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.
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.
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.
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++.
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).
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.
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).
* 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.
* 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.
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.
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.
* 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.
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.
* 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.
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.
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.
* 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.
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.
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.
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.
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).
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.
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.
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.
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 $$ to $1) useless. But... maybe some users depend on this, in
spite of the comments that clearly warn againt this. So let's not
turn this off just yet.
* src/reader.c (grammar_rule_check_and_complete): Rename as...
(grammar_rule_check_and_complete): this.
Install the default semantic action when applicable.
* examples/variant-11.yy, examples/variant.yy, tests/calc.at:
Exercise the default semantic action, even with variants.
Currently we use "<dir><api.location.file>" as \file argument, and as
base for the CPP guard. This is not nice when <dir> is absolute, in
which case it is expected that the user will use api.location.include
to get something nicer. If defined, use that name instead.
* data/location.cc (b4_location_path): New.
Use it.
* tests/c++.at (Shared locations): Check the guard and Doxygen doc.
Now that distcheck no longer fails (see previous commit), let's
address the shortcomings.
* Makefile.am (CLEANDIRS, clean-local): New.
* doc/local.mk, examples/calc++/local.mk, examples/local.mk,
* examples/mfcalc/local.mk, examples/rpcalc/local.mk,
* src/local.mk
(CLEANDIRS): Get rid of Apple's *.dSYM directories.
(CLEANFILES): Get rid of *.output files.
* examples/variant-11.yy, examples/variant.yy: Don't generate
any of the auxiliary files (location.hh and the like).
`make update-po` runs:
package_gnu="$(PACKAGE_GNU)"; \
test -n "$$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'.
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.
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.
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.
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.
* 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.
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.
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.
* src/getargs.c (getargs): Don't display any argv other that argv[0]
when reporting a missing argument.
* tests/bison.in: Neutralize path differences in stderr.
* tests/input.at (Invalid number of arguments): New.
* src/getargs.c (usage): Rely on setlocale (LC_MESSAGES, NULL)
trick only on glibc, as POSIX does not specify the output
of setlocale in this case, and the Gnulib localename module
source code indicates that the trick works only on glibc.
C11 no longer requires support for variable-length arrays, and
VS2015 does not have them. Redo UNIQSTR_CONCAT to use a method
that is simpler and better anyway.
* src/uniqstr.c (uniqstr_vsprintf): Remove; no longer needed.
* src/uniqstr.h (UNIQSTR_GEN_FORMAT, UNIQSTR_GEN_FORMAT_):
* src/uniqstr.c (uniqstr_concat): New function.
* src/uniqstr.h (UNIQSTR_CONCAT): Use it instead of using
uniqstr_vsprintf.
* doc/bison.texi: Minor fixes in typography.
It is no longer require to pass --defines for C++ (it was addressed
long ago).
No longer refer to the `variant` define variable, it was replaced by
`api.value.type variant`.
Prefer nullptr to 0 for the null pointer.
Use deftypeop for constructors.
(Complete Symbols): Give the expected signature of yylex.
Don't document the symbol_type constructors, as we want users to focus
on make_TOKEN.
Also show the case without locations.
Bison's test 464 (Syntax error as exception) fails on the CI.
Do not use clang with asan on Ubuntu's libc++.
https://bugs.llvm.org/show_bug.cgi?id=17379
* .travis.yml (Clang 7 libc++ and ASAN): New.
(Clang 6 -O3 and libc++): Really use libc++.
(Clang 5): Don't use libc++, nor asan (does not work either, same
reason).
* .travis.yml: Don't define CC/CXX, it does not work.
Use `[[...]]` instead of `[...]`.
Show the compiler versions.
(Clang 3.3, Clang 3.4): Specify the path to avoid using
/usr/local/clang-5.0.0/bin's clang.
The message on configure is misleading:
checking whether the linker accepts -std=c++11... yes
checking whether the linker accepts -std=c++14... yes
checking whether the linker accepts -std=c++17... no
It is the compiler that we check, not just the linker.
* m4/ax_check_link_flag.m4: Remove.
* m4/bison-check-compiler-flag.m4: New.
* m4/bison-cxx-std.m4: Use it.
This has been bugging me for while. I was hard to reproduce: it
worked only on GNU/Linux, probably because libc++ implements the small
string optimization, while libstdc++ did not and actually allocated on
the heap for this small string.
See https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00110.html.
* tests/types.at (api.value.type): Do not provide a semantic value to
EOF.
The forthcoming automove feature, to be properly checked, will require
that we can rely on the value of a moved-from string, which is not
something the C++ standard guarantees. So introduce our own wrapper.
Suggested by Frank Heckenbach.
https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00111.html
* tests/c++.at (Variants): Introduce and use a new 'string' class.
The 'Variants' tests are well suited to check support for move, and in
particular for the forthcoming automove feature. But the tests were
written to show the best practice in C++98, using swap:
list "," item { std::swap ($$, $1); $$.push_back ($3); }
This cannot work with std::move. So, make this example simpler, based
on regular assignment instead of swap, which is a regression for
C++98 (as the new traces show), but will be an improvement for modern
C++ with automove.
* tests/c++.at (Variants): Stop using swap.
We don't generate a header file, so remove the 'require' code section.
Adjust expectations.
Several types of failures. First, unable to pass the file name
properly to the linker.
./synclines.at:416: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o \"\\\"\" \"\\\"\".c $LIBS
stderr:
ld: cannot open output file "/"": No such file or directory
stdout:
Unable to save under such a file name.
./synclines.at:421: $CXX $CXXFLAGS $CPPFLAGS -c $LDFLAGS -o \"\\\"\" \"\\\"\".cc $LIBS
stderr:
error: can't open file "/"" for write
compilation aborted for "\"".cc (code 1)
Spurious output because of warning flags is failed to reject as an
error during configure:
./headers.at:343: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS c-only.o cxx-only.o -o c-and-cxx ||
exit 77
--- /dev/null 2018-09-18 21:21:37.745649000 +0000
+++ /home/travis/build/akimd/bison/tests/testsuite.dir/at-groups/222/stderr 2018-09-18 21:28:17.291919519 +0000
@@ -0,0 +1,7 @@
+icpc: command line warning #10006: ignoring unknown option '-Wcast-align'
+icpc: command line warning #10006: ignoring unknown option '-fparse-all-comments'
+icpc: command line warning #10006: ignoring unknown option '-Wdocumentation'
+icpc: command line warning #10006: ignoring unknown option '-Wnull-dereference'
+icpc: command line warning #10006: ignoring unknown option '-Wnoexcept'
+icpc: command line warning #10006: ignoring unknown option '-fno-color-diagnostics'
+icpc: command line warning #10006: ignoring unknown option '-Wno-keyword-macro'
stdout:
* tests/local.at (AT_SKIP_IF_CANNOT_LINK_C_AND_CXX): Also ignore
stderr, as with ICC we get
* tests/synclines.at (syncline escapes): Don't link the output.
The code was already using midrule only, never mid_rule. This is
simpler to remember, and matches a similar change we made from
look-ahead to lookahead.
* NEWS, doc/bison.texi, src/reader.c, src/scan-code.h, src/scan-code.l
* tests/actions.at, tests/c++.at, tests/existing.at: here.
We use both styles, let's stick to a single one. Autoconf uses the
prefix one, let's do the same.
* data/bison.m4, data/c++.m4, data/c-like.m4, data/lalr1.cc,
* data/variant.hh, data/yacc.c: Rename all the b4_*_ macros
as _b4_*.
On the CI, we had failures such as:
./c++.at:401: $PREPARSER ./list
stderr:
./list: error while loading shared libraries: libc++.so.1:
cannot open shared object file: No such file or directory
because we accepted `-std=c++ -stdlib=libc++` although libc++ is not
installed on the machine.
* m4/ax_check_compile_flag.m4 (AX_CHECK_COMPILE_FLAG): Rewrite as...
* m4/bison-check-compile-flag.m4 (BISON_CHECK_COMPILE_FLAG): this, so
that we use AC_LINK_IFELSE to check the compiler (and its std lib)
instead of AC_COMPILE_IFELSE.
The CI is littered with
# -*- compilation -*-
423. regression.at:907: testing Dancer %glr-parser ...
./regression.at:907: bison -fno-caret -o dancer.c dancer.y
./regression.at:907: $BISON_C_WORKS
stderr:
stdout:
./regression.at:907: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o dancer dancer.c $LIBS
stderr:
icc: command line warning #10006: ignoring unknown option '-Wcast-align'
icc: command line warning #10006: ignoring unknown option '-fparse-all-comments'
icc: command line warning #10006: ignoring unknown option '-Wdocumentation'
icc: command line warning #10006: ignoring unknown option '-Wnull-dereference'
icc: command line warning #10006: ignoring unknown option '-Wbad-function-cast'
icc: command line warning #10006: ignoring unknown option '-fno-color-diagnostics'
icc: command line warning #10006: ignoring unknown option '-Wno-keyword-macro'
dancer.c(755): error #1628: function declared with "noreturn" does return
}
^
dancer.c(761): error #1628: function declared with "noreturn" does return
}
^
compilation aborted for dancer.c (code 2)
ICC sees that `longjmp(buf, 1);` does not return, it sees that
`abort();` does not either, but fails to see it for
`longjmp(buf, 1); abort();`
* data/glr.c (YYLONGJMP): Be even clearer on the fact this does not
return.
Putting them in the env is useless. We don't want to pass
`CPPFLAGS="$CPPFLAGS"` to configure, as it means "set it to nothing"
when $CPPFLAGS is not set, which is not what we want.
This correctly started to use libc++, but it is not installed on the
Ubuntu. We will see later if we can use it.
* .travis.yml: Define CONFIGUREFLAGS, and pass it to configure.
* build-aux/install-icc.sh: New.
* .travis.yml (icc): New.
Use -k to get as many errors as possible from the start.
* src/complain.c (warnings_types): Use a more precise type.
On the CI, we have this spurious failure with clang 3.9 with
-std=c++17:
In file included from list.y:23:
In file included from /usr/include/c++/4.8/iostream:39:
In file included from /usr/include/c++/4.8/ostream:38:
In file included from /usr/include/c++/4.8/ios:42:
In file included from /usr/include/c++/4.8/bits/ios_base.h:41:
In file included from /usr/include/c++/4.8/bits/locale_classes.h:40:
In file included from /usr/include/c++/4.8/string:52:
In file included from /usr/include/c++/4.8/bits/basic_string.h:2815:
In file included from /usr/include/c++/4.8/ext/string_conversions.h:43:
/usr/include/c++/4.8/cstdio:120:11: error: no member named 'gets' in the global namespace
using ::gets;
~~^
This shows that our test, based on gl_WARN_ADD, is a joke. We have to
really check for at least a bit of C++.
* m4/ax_check_compile_flag.m4, m4/bison-cxx-std.m4: New.
* configure.ac: Use them to make sure the compiler actually works.
In C++ pre C++11 it is standard practice to use 0 for the null pointer.
But GCC pre 8 -std=c++98 with -Wzero-as-null-pointer-constant warns about
this.
So disable -Wzero-as-null-pointer-constant when compiling C++ pre 11.
Let's do this in AT_DATA_SOURCE_PROLOGUE (which is pasted on top of
all the test grammar files). Unfortunately, that shifts all the
locations in the expected error messages, which would be too noisy.
Instead, let's introduce testsuite.h, which can vary in length, and
include it in AT_DATA_SOURCE_PROLOGUE.
* tests/testsuite.h: New.
Disable -Wzero-as-null-pointer-constant's warning with GCC pre 8,
C++ pre 11.
* tests/local.at (AT_DATA_SOURCE_PROLOGUE): Use it.
* tests/atlocal.in (CPPFLAGS): Find it.
* tests/local.mk: Ship it.
* data/c.m4 (YY_NULLPTR): Prefer ((void*)0) to 0 in C.
For some reasons, the checkout on travis may not have any tags, so
`git describe` fails, so bootstrap fails.
* .travis.yml: If git describe fails, install some tag.
This is much of course more efficient than in the matrix of the CI (or
on our own machines), but a bit more tedious.
* configure.ac (CXX03_CXXFLAGS, CXX11_CXXFLAGS, CXX14_CXXFLAGS)
(CXX17_CXXFLAGS, CXX2A_CXXFLAGS, STDCXX_FLAGS): New.
* tests/atlocal.in: Receive them.
* tests/local.at (AT_FOR_EACH_CXX): New.
* tests/c++.at: Use AT_FOR_EACH_CXX.
Reported by Frank Heckenbach.
http://lists.gnu.org/archive/html/bug-bison/2018-03/msg00002.html
Actually the assignment operator should never be needed: the C++98
requirements for vector::push_back is CopyInsertable, which does not require
an assignment operator. However, libstdc++ shipped with GCC up to (and
including) 6 uses the assignment operator (which affects Clang on top of
libstdc++, but also ICC). So let's keep it for legacy C++.
See https://gcc.godbolt.org/z/q0XXmC.
* data/lalr1.cc (stack_symbol_type::operator=): Remove.
* data/c++.m4 (basic_symbol::operator=): Ditto.
* tests/c++.at (C++ Variant-based Symbols Unit Tests): Adjust.
Modern C++ (i.e., C++11 and later) introduced "move only" types: types such
as std::unique_ptr<T> that can never be duplicated. They must never be
copied (by assignments and constructors), they must be "moved". The
implementation of lalr1.cc used to copy symbols (including their semantic
values). This commit ensures that values are only moved in modern C++, yet
remain compatible with C++98/C++03.
Suggested by Frank Heckenbach, who provided a full implementation on
top of C++17's std::variant.
See http://lists.gnu.org/archive/html/bug-bison/2018-03/msg00002.html,
and https://lists.gnu.org/archive/html/bison-patches/2018-04/msg00002.html.
Symbols (terminal/non terminal) are handled by several functions that used
to take const-refs, which resulted eventually in a copy pushed on the stack.
With modern C++ (C++11 and later) the callers must use std::move, and the
callees must take their arguments as rvalue refs (foo&&). In order to avoid
duplicating these functions to support both legacy C++ and modern C++, let's
introduce macros (YY_MOVE, YY_RVREF, etc.) that rely on copy-semantics for
C++98/03, and move-semantics for modern C++.
That's easy for inner types, when the parser's functions pass arguments to
each other. Functions facing the user (make_NUMBER, make_STRING, etc.)
should support both rvalue-refs (for instance to support move-only types:
make_INT (std::make_unique<int> (1))), and lvalue-refs (so that we can pass
a variable: make_INT (my_int)). To avoid the multiplication of the
signatures (there is also the location), let's take the argument by value.
See:
https://lists.gnu.org/archive/html/bison-patches/2018-09/msg00024.html.
* data/c++.m4 (b4_cxx_portability): New.
(basic_symbol): In C++11, replace copy-ctors with move-ctors.
In C++11, replace copies with moves.
* data/lalr1.cc (stack_symbol_type, yypush_): Likewise.
Use YY_MOVE to avoid useless copies.
* data/variant.hh (variant): Support move-semantics.
(make_SYMBOL): In C++11, in order to support both read-only lvalues,
and rvalues, take the argument as a copy.
* data/stack.hh (yypush_): Use rvalue-refs in C++11.
* tests/c++.at: Use move semantics.
* tests/headers.at: Adjust to the new macros (YY_MOVE, etc.).
* configure.ac (CXX98_CXXFLAGS, CXX11_CXXFLAGS, CXX14_CXXFLAGS)
(CXX17_CXXFLAGS, ENABLE_CXX11): New.
* tests/atlocal.in: Receive them.
* examples/variant.yy: Don't define things in std.
* examples/variant-11.test, examples/variant-11.yy: New.
Check the support of move-only types.
* examples/README, examples/local.mk: Adjust.
This generates less code, which is nicer to read, but also takes less
chances with compilers such as G++ 4.8 that are too strict and check
"dead code" (templated code that is not instantiated).
* data/c++.m4 (b4_symbol_type_declare, b4_symbol_type_define): When
variants are used, don't generate code meant for non variants.
See ea0db44fed. We also need to disable
the warning in the examples (but don't want to clutter the
documentation with such details).
* doc/bison.texi (scanner.ll): Disable Clang's -Wdocumentation.
While at it, hide the other kludges.
That's the version on Ubuntu Precise.
See also 1dac131ec4.
* src/flex-scanner.h: Disable -Wdocumentation.
* doc/bison.texi: Turn off a warning triggered by Flex 2.6.4.
We have failures on Flex output, which are probably related to an old
release. Let's check.
In file included from src/scan-code-c.c:3:
src/scan-code.c:2198:21: error: empty paragraph passed to '@param' command
[-Werror,-Wdocumentation]
* @param line_number
~~~~~~~~~~~~~~~~~^
* .travis.yml: here.
Running all these tests might be overkill: it is very long, and don't
need full portability checks. Besides, some tests under Valgrind are
too slow and get killed by the CI (timeout of 10min without output).
* .travis.yml: here.
We have some maintainer-check-foo and some maintainer-foo-check. Keep
only the former.
* tests/local.mk (maintainer-push-check, maintainer-xml-check)
(maintainer-release-check): Rename as...
(maintainer-check-push, maintainer-check-xml)
(maintainer-check-release): these.
Most of our variables for C++ flags are named FOO_CXXFLAGS, not
CXXFLAGS_FOO.
* configure.ac, tests/atlocal.in, tests/calc.at
(NO_EXCEPTIONS_CXXFLAGS): Rename as...
(CXXFLAGS_NO_EXCEPTIONS): this.
Clang++ issues warnings when it's used to compile C. This make target
is precisely checking whether we can do that.
* configure.ac (NO_DEPRECATED_CXXFLAGS): New.
* tests/atlocal.in: Use it.
This completes 2d7e743802.
Some shells don't grok "local var=`cmd`" very well: they need the rhs
to be quoted.
./examples/test: 72: local: you.,: bad variable name
FAIL examples/variant.test (exit status: 2)
Reported by Étienne Renault.
* examples/test (run): Quote the values in 'local' assignments.
With GCC7 we have warnings (false positive):
x8.c: In function 'x8_parse':
x8.c:1233:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized]
yylval = *yypushed_val;
~~~~~~~^~~~~~~~~~~~~~~
x8.c: In function 'x8_pull_parse':
x8.c:1233:16: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized]
yylval = *yypushed_val;
~~~~~~~^~~~~~~~~~~~~~~
See also 9645a2b20e.
* tests/local.at (AT_PUSH_IF): New.
(AT_BISON_OPTION_POPDEFS): Pop it, and pop AT_PURE_IF.
* tests/headers.at (Several parsers, Several parsers): Disable these
warnings when in push parser.
Currently, in glr.cc, we emit the definitions of basic_symbol and
symbol_type, although there are not used.
* data/c++.m4 (b4_public_types_declare): Extract these definitions from
here, and move them...
(b4_symbol_type_declare): here.
(b4_public_types_declare): Also remove the definition of the symbol
constructors.
* data/lalr1.cc (b4_shared_declarations): Adjust: call
b4_symbol_type_declare and b4_symbol_constructor_declare.
Some shells don't grok `local var=$val` very well: they need the rhs
to be quoted.
./examples/test: 66: local: you.,: bad variable name
FAIL examples/variant.test (exit status: 2)
Reported by Étienne Renault.
* examples/test (run): Quote the values in 'local' assignments.
The header generated for variants with assertions but without
locations, is not self-contained. Prepare a check for this.
* tests/headers.at (Sane headers): New, extracted from...
(Several parsers): here.
Benchmarks show that it is more efficient to keep this copy
constructor, rather than forcing the use of the default constructor
and then assignment.
This reverts commit 7ab25ad020.
Currently, in bison's C++ parser template (`lalr.cc`), the `variant<>`
struct's `build()` method uses placement-new in the form `new (...) T`
to initialize a variant type. However, for POD variant types, this
will leave the memory space uninitialized. If we subsequently tries
to `::move` into a variant object in such state, the call can trigger
clang's undefined behavior sanitizer due to accessing the
uninitialized memory.
https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00098.html
* data/variant.hh (build): Always initialize the stored value.
Signed-off-by: Akim Demaille <[email protected]>
* doc/bison.texi (A Complete C++ Example): Prefer throw exceptions
from the scanner.
Show the invalid characters.
Since the scanner sends exceptions, it no longer needs to report
errors, so we can get rid of the driver's routine to report error,
do it in yyerror.
Use @group/@end group to improve rendering.
This example uses the calcxx_ prefix for each class. That's uselessly
heavy.
* doc/bison.texi (A Complete C++ Example): Simplify the class names.
Since now 'driver' denotes the class, use 'drv' for the values.
Formatting changes.
On these tests, at -O2 and above, GCC 8 complains that yylval may be
uninitialized. But it seems wrong: it is initialized. Rather than
turning off the warning in the skeleton (hence possibility hiding
relevant warnings of user parsers), let's turn it off in the tests
only.
163: parse.error=verbose and consistent errors: FAILED (conflicts.at:625)
165: parse.error=verbose and consistent errors: lr.default-reduction=consistent FAILED (conflicts.at:635)
166: parse.error=verbose and consistent errors: lr.default-reduction=accepting FAILED (conflicts.at:641)
167: parse.error=verbose and consistent errors: lr.type=canonical-lr FAILED (conflicts.at:645)
168: parse.error=verbose and consistent errors: parse.lac=full FAILED (conflicts.at:650)
169: parse.error=verbose and consistent errors: parse.lac=full lr.default-reduction=accepting FAILED (conflicts.at:655)
We get:
input.c: In function 'yyparse':
input.c:980:9: error: 'yylval' may be used uninitialized in this function [-Werror=maybe-uninitialized]
YYSTYPE yylval YY_INITIAL_VALUE (= yyval_default);
^~~~~~
cc1: all warnings being treated as errors
See https://lists.gnu.org/archive/html/bison-patches/2018-08/msg00063.html.
* tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Disable
-Wmaybe-uninitialized.
Suggested by Rici Lake.
https://lists.gnu.org/archive/html/bug-bison/2018-08/msg00033.html
* doc/bison.texi: Complete description of the first node in the main
@menu.
(Push Decl): Remove the 'experimental' warnings about push parser.
Clarify that the push parser object can be reused in several parses.
Reported by Brooks Moses <[email protected]>
http://lists.gnu.org/archive/html/bison-patches/2018-02/msg00000.html
* data/lalr1.cc (YY_EXCEPTIONS): New.
Use it to disable try/catch clauses.
* doc/bison.texi (C++ Parser Interface): Document it.
* configure.ac (CXXFLAGS_NO_EXCEPTIONS): New.
* tests/atlocal.in: Receive it.
* tests/local.at (AT_FULL_COMPILE, AT_LANG_COMPILE):
Accept a new argument, extra compiler flags.
* tests/calc.at: Run the C++ calculator with exception support disabled.
Currently the examples are too dense, let's put empty lines where
'#line' would be issued. And also remove some spurious empty
lines (remains from @group, @end group, etc.).
* examples/extexi: Do that.
* examples/local.mk (extexiFLAGS): Rename as...
(EXTEXIFLAGS): this.
* examples/mfcalc/local.mk, examples/rpcalc/local.mk: Define the
programs in a more natural order, source, preproc, then linker.
* examples/test: Be ready to work on programs that are not in
a subdir.
* examples/variant.test: New.
* examples/local.mk: Use it.
* examples/variant.yy: Don't use 0 for nullptr.
Use a more natural output for a list of string.
Visual Studio issues a C4146 warning on '-static_cast<unsigned>(rhs)'.
The code is weird, probably to cope with INT_MIN. Let's go back to
using std::max (whose header is still included in position.hh...) like
originally, but with the needed casts.
Reported by 長田偉伸, and with help from Rici Lake.
See also
http://lists.gnu.org/archive/html/bug-bison/2013-02/msg00000.html
and commit 75ae829984.
* data/location.cc (position::add_): Take min as an int.
Use std::max.
While here, get rid of a couple of useless inlines.
* doc/bison.texi (Mid-Rule Actions): Restructure to insert...
(Typed Mid-Rule Actions): this new section.
Move the manual translation of mid-rule actions into regular actions
to...
(Mid-Rule Action Translation): here.
Reported by Jannick.
http://lists.gnu.org/archive/html/bug-bison/2017-05/msg00001.html
"Amusingly" enough, we have the same problem with %defines when the
parser file name has backslashes or quotes: we generate #includes with
an incorrect C string.
* src/output.c (prepare_symbol_definitions): Escape properly the file
names before passing them to M4.
* data/bison.m4, data/lalr1.cc: Don't simply put the file name between
two quotes (that should have been strong enough a smell...), expect
the string to be properly quoted.
* tests/synclines.at: New tests to check this.
Suggested by Paul Eggert.
* src/reader.c (find_start_symbol): Don't check 'res', we know it is
not null. That suffices to avoid the GCC warnings.
* bootstrap.conf: We don't need 'assume', which doesn't exist anyway.
In 0931d14728 I removed too many
initializations from some ctors: some were not about base ctors, but
about member variables. In fact, more of them were missing to please
GCC 8.
While at it, generate more natural code for C++ without variant:
instead of
template <typename Base>
parser::basic_symbol<Base>::basic_symbol (const basic_symbol& other)
: Base (other)
, value ()
{
value = other.value
}
generate
template <typename Base>
parser::basic_symbol<Base>::basic_symbol (const basic_symbol& other)
: Base (other)
, value (other.value)
{}
* data/c++.m4 (basic_symbol::basic_symbol): Always initialize 'value',
it might be a POD without a ctor.
* data/lalr1.cc (stack_symbol_type::stack_symbol_type): Likewise.
* data/variant.hh (variant::variant): Default initialize the buffer too.
GCC 8 issues warnings whose root cause was a bit hard to find.
calc.cc: In member function 'virtual int yy::parser::parse()':
calc.cc:810:18: warning: '*((void*)&<anonymous> +8)' may be used uninitialized in this function [-Wmaybe-uninitialized]
, location (l)
^
calc.cc: In member function 'void yy::parser::yypush_(const char*, yy::parser::stack_symbol_type&)':
calc.cc:810:18: warning: '*((void*)&<anonymous> +8)' may be used uninitialized in this function [-Wmaybe-uninitialized]
, location (l)
^
calc.cc: In member function 'void yy::parser::yypush_(const char*, yy::parser::state_type, yy::parser::symbol_type&)':
calc.cc:810:18: warning: '*((void*)&<anonymous> +8)' may be used uninitialized in this function [-Wmaybe-uninitialized]
, location (l)
^
The problem is with locations that don't have a constructor, such as
Span (in calc.cc) which is POD. It is POD on purpose: so that we can
use that structure to test glr.cc which cannot use non POD in its
(C) stacks.
* data/c++.m4 (basic_symbol): Also ensure that 'location' is
initialized.
With GCC 7.3.0 and Flex 2.6.4, we get warnings on all the generated
scanners:
examples/calc++/calc++-scanner.cc: In function 'void yyrestart(FILE*)':
examples/calc++/calc++-scanner.cc:1611:20: error: potential null pointer dereference [-Werror=null-dereference]
/* %endif */
~~~~~~~~~~~ ^
examples/calc++/calc++-scanner.cc:1607:19: error: potential null pointer dereference [-Werror=null-dereference]
/* %if-c-only */
~~~~~~~~~~~~~~~ ^
examples/calc++/calc++-scanner.cc:1611:20: error: potential null pointer dereference [-Werror=null-dereference]
/* %endif */
~~~~~~~~~~~ ^
examples/calc++/calc++-scanner.cc:1607:19: error: potential null pointer dereference [-Werror=null-dereference]
/* %if-c-only */
~~~~~~~~~~~~~~~ ^
cc1plus: all warnings being treated as errors
Obviously the lines are incorrect, and the warnings are emitted twice.
Still, let's get rid of these warnings.
* doc/bison.texi, src/flex-scanner.h: Disable these warnings in code
generated by Flex.
Commit 3df32101e7 introduced invalid C
code. Caught by GCC 7.3.0.
* bootstrap.conf (gnulib_modules): We need assume.
* src/reader.c (find_start_symbol): Fix the signature (too much C++,
sorry...).
Prefer 'assume' to 'assert', so that we don't have these warnings even
when NDEBUG is defined.
This file was meant to be shown as an example. Install it.
* README, data/README: Put Emacs metadata in the final section.
* examples/README: New.
* examples/variant.yy: Use %empty.
* examples/local.mk: Install both these files.
We currently generate copy constructors such as the following
one (taken from examples/variant.yy):
parser::stack_symbol_type::stack_symbol_type (const stack_symbol_type& that)
: super_type (that.state, that.location)
{
switch (that.type_get ())
{
case 3: // TEXT
case 8: // item
value.copy< ::std::string > (that.value);
break;
case 7: // list
value.copy< ::std::vector<std::string> > (that.value);
break;
case 4: // NUMBER
value.copy< int > (that.value);
break;
default:
break;
}
}
they are actually useless: we never need it.
* data/lalr1.cc: Don't generate the stack_symbol_type copy ctor.
Fix a typo so that instead of
basic_symbol::basic_symbol (typename Base::kind_type t, const int v)
we now generate
basic_symbol::basic_symbol (typename Base::kind_type t, const int& v)
* data/variant.hh (b4_basic_symbol_constructor_declare)
(b4_basic_symbol_constructor_define): Add missing reference.
Several features were flagged 'experimental' and waiting for user
feedback to 'stabilize', but i. AFAIK, no user ever reported anything
about them, ii. they'be been here long enough to prove they don't do
harm.
* doc/bison.texi: No longer experimental: default %printer and
%destructor (typed: <*> and untyped: <>), %define api.value.type union
and variant, Java parsers, XML output, LR family (lr, ielr, lalr),
semantic predicates (%?).
* tests/actions.at: Fix some log messages.
Prefer #error to fprintf: it fixes the invalid use of yyoutput in
%destructor, and it is an even stronger check: that the code is not
even emitted. The portability of #error is not really a problem here,
since the point is anyway to have the compilation fail.
Make sure that we cannot apply a type to the (main) action of a rule.
* src/reader.c (grammar_rule_check): Issue the warning.
* tests/input.at (Cannot type action): Check the warning.
* tests/input.at (_AT_UNUSED_VALUES_DECLARATIONS): Check
typed mid-rule actions.
* tests/report.at (Reports): Check that types of typed mid-rule
actions are reported.
* tests/actions.at (Typed mid-rule actions): Check that
the values of typed mid-rule actions are correct.
Prompted on Piotr Marcińczyk's message:
http://lists.gnu.org/archive/html/bug-bison/2017-06/msg00000.html.
See also http://lists.gnu.org/archive/html/bug-bison/2018-06/msg00001.html.
Because their type is unknown to Bison, the values of midrule actions are
not treated like the others: they don't have %printer and %destructor
support. In addition, in C++, (Bison) variants cannot work properly.
Typed midrule actions address these issues. Instead of:
exp: { $<ival>$ = 1; } { $<ival>$ = 2; } { $$ = $<ival>1 + $<ival>2; }
write:
exp: <ival>{ $$ = 1; } <ival>{ $$ = 2; } { $$ = $1 + $2; }
* src/scan-code.h, src/scan-code.l (code_props): Add a `type` field to
record the declared type of an action.
(code_props_rule_action_init): Add a type argument.
* src/parse-gram.y: Accept an optional type tag for actions.
* src/reader.h, src/reader.c (grammar_current_rule_action_append): Add
a type argument.
(grammar_midrule_action): When a mid-rule is typed, pass its type to
the defined dummy non terminal symbol.
grammar_current_rule_action_append was used in two different places:
for actual action (`{...}`), and for predicates (`%?{...}`). Let's
split this in two different functions.
* src/reader.h, src/reader.c (grammar_current_rule_predicate_append): New.
Extracted from...
(grammar_current_rule_action_append): here.
Remove arguments that don't apply.
Adjust dependencies.
Predicates with GLR are issued with synclines in the middle of C code:
case 2:
if (! (#line 6 "sempred.y" /* glr.c:816 */
new_syntax)) YYERROR;
#line 793 "sempred.tab.c" /* glr.c:816 */
break;
Reported by Rici Lake.
http://lists.gnu.org/archive/html/bug-bison/2018-05/msg00033.html
* data/c.m4 (b4_predicate_case): Be sure to start on column 0.
It would be nicer if b4_syncline could ensure this by itself
(that would avoid ugly code when synclines are disabled), but that's
way more work.
* tests/glr-regression.at (Predicates): Be a real end-to-end test.
This would have caught this error years ago...
For instance on test 99:
In file included from @@.cc:56:
@@.hh:409:26: error: definition of implicit copy constructor for
'stack_symbol_type' is deprecated because it
has a user-declared copy assignment operator
[-Werror,-Wdeprecated]
stack_symbol_type& operator= (const stack_symbol_type& that);
^
Reported by Derek Clegg.
https://lists.gnu.org/archive/html/bison-patches/2018-05/msg00036.html
* configure.ac (warn_tests): Add -Wdeprecated.
* data/lalr1.cc (stack_symbol_type): Add an explicit copy ctor.
We cannot rely on the explicit default implementation (`= default`)
as we support C++ 98.
With recent compilers:
input.yy:49:5: error: definition of implicit copy assignment
operator for 'Object' is deprecated because
it has a user-declared destructor
[-Werror,-Wdeprecated]
~Object ()
^
input.yy:130:35: note: in implicit copy assignment operator for
'Object' first required here
{ yylhs.value.as< Object > () = yystack_[0].value.as< Object > (); }
* tests/c++.at (Object): Add missing assignment operator.
sc_two_space_separator_in_usage complains about bootstrap:
two_space_separator_in_usage
/Users/akim/src/gnu/bison/bootstrap:905: --aux-dir $build_aux\
/Users/akim/src/gnu/bison/bootstrap:906: --doc-base $doc_base\
/Users/akim/src/gnu/bison/bootstrap:907: --lib $gnulib_name\
/Users/akim/src/gnu/bison/bootstrap:908: --m4-base $m4_base/\
/Users/akim/src/gnu/bison/bootstrap:909: --source-base $source_base/\
/Users/akim/src/gnu/bison/bootstrap:910: --tests-base $tests_base\
/Users/akim/src/gnu/bison/bootstrap:911: --local-dir $local_gl_dir\
maint.mk: help2man requires at least two spaces between an option and its description
* cfg.mk: Exclude bootstrap from this check.
At least GCC 7.3, with -O1 or -O2 (but not -O0 or -O3) generates
warnings with -Wnull-dereference when using yyformat: it fails to see
yyformat cannot be null.
Reported by Frank Heckenbach, https://savannah.gnu.org/patch/?9620.
* configure.ac: Use -Wnull-dereference if supported.
* data/glr.c, data/lalr1.cc, data/yacc.c: Define yyformat in such
a way that GCC cannot not see that yyformat is defined.
Using `default: abort();` also addresses the issue, but forces
the inclusion of `stdlib.h`, which we avoid.
Sometimes `inline` would be used in *.cc files on symbols that are not
exported (useless but harmless), and sometimes on exported symbols
such as the constructor of syntax_error (harmful: linking fails).
Reported several times, including:
- by Dennis T
http://lists.gnu.org/archive/html/bug-bison/2016-03/msg00002.html
- by Frank Heckenbach
https://savannah.gnu.org/patch/?9616
* data/c++.m4 (b4_inline): New: expands to `inline` or nothing.
Use it where appropriate.
* data/lalr1.cc: Use it where appropriate.
* tests/c++.at (Syntax error as exception): Put the scanner in another
compilation unit to exercise the constructor of syntax_error.
* data/lalr1.cc, doc/bison.texi, etc/bench.pl.in, examples/variant.yy,
* tests/actions.at, tests/atlocal.in, tests/c++.at, tests/headers.at,
* tests/local.at, tests/types.at:
Don't use std::endl, it flushes uselessly, and is considered bad
style.
* README-hacking: Commit before bootstrapping.
* bootstrap.conf: gnulib_mk is no longer defined by bootstrap.
* bootstrap, gnulib, lib/.gitignore, m4/.gitignore: Update/regen.
With Clang 6.0:
CC src/bison-getargs.o
bison/src/getargs.c:67:12: error: parameter 'option' not found in the
function declaration [-Werror,-Wdocumentation]
* \param option option being decoded.
^~~~~~
bison/src/getargs.c:67:12: note: did you mean 'rpl_option'?
* src/getargs.c: Don't use `option` as a documentation argument.
Currently lalr1.cc makes an out-of-bound access when trying to read @1
in rules with an empty rhs (i.e., when there is no @1) that raises an
error (YYERROR).
glr.c already gracefully handles this by using @$ as initial location
for the errors. Let's do that in yacc.c and lalr1.cc.
* data/lalr1.cc, data/yacc.c: Use @$ to initialize the error location.
* tests/actions.at: Check that case.
This style appears to be more traditional, at least in C++.
For instance in the standard, [facets.examples].
There are occurrences using "unsigned int" too though.
* data/lalr1.cc, data/location.cc, data/stack.hh: here.
We use this trick to write some test about internal details. But
since we use -Werror, clang++ 3.6 dies issueing a warning about it.
* configure.ac (warn_tests): Disable this warning.
Because musl supports UTF-8 with LC_ALL=C, gcc produces:
input.y: In function ‘yyparse’:
instead of:
input.y: In function 'yyparse':
Reported by Ferdinand Thiessen.
http://lists.gnu.org/archive/html/bug-bison/2015-02/msg00001.html
* tests/synclines.at (AT_SYNCLINES_COMPILE): Skip syncline tests when
we can't trust error messages issued about a function body.
Sun C 5.13 SunOS_sparc 2014/10/20 reports errors on tests 430-432.
Reported by Dennis Clarke.
<http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00087.html>
* tests/c++.at (Variants): Be sure to emit operator<< before using it:
use "%code top" rather than "%code".
Prefer std::vector to std::list.
Do not define anything in std::, to avoid undefined behavior.
* origin/maint:
maint: post-release administrivia
version 3.0.4
gnulib: update
build: re-enable compiler warnings, and fix them
tests: c++: fix a C++03 conformance issue
tests: fix a title
c++: reserve 200 slots in the parser's stack
tests: be more robust to unrecognized synclines, and try to recognize xlc
tests: fix C++ conformance
build: fix some warnings
build: avoid infinite recursions on include_next
There are warnings (-Wextra) in generated C++ code:
ltlparse.cc: In member function 'ltlyy::parser::symbol_number_type
ltlyy::parser::by_state::type_get() const':
ltlparse.cc:452:33: warning: enumeral and non-enumeral type in
conditional expression
return state == empty_state ? empty_symbol : yystos_[state];
Reported by Alexandre Duret-Lutz.
It turns out that -Wall and -Wextra were disabled because of a stupid
typo.
* configure.ac: Fix the stupid typo.
* data/lalr1.cc, src/AnnotationList.c, src/InadequacyList.c,
* src/ielr.c, src/print.c, src/scan-code.l, src/symlist.c,
* src/symlist.h, src/symtab.c, src/tables.c, tests/actions.at,
* tests/calc.at, tests/cxx-type.at, tests/glr-regression.at,
* tests/named-refs.at, tests/torture.at:
Fix warnings, mostly issues about variables used only with assertions,
which are disabled with -DNDEBUG.
This fixes test 241 on xLC:
"input.y", line 42.11: 1540-0274 (S) The name lookup for "report" did not find a declaration.
"input.y", line 42.11: 1540-1292 (I) Static declarations are not considered for a function call if the function is not qualified.
where report is:
static void
report (std::ostream& yyo, int ival, float fval)
{
yyo << "ival: " << ival << ", fval: " << fval;
}
and line 42 is:
%printer { report (yyo, $$, $<fval>$); } <ival>;
It turns out that indeed this function must not be declared static,
<http://stackoverflow.com/a/17662745/1353549>. Let's put it into an
anonymous namespace.
Reported by Thomas Jahns.
http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html
* tests/actions.at (Qualified $$ in actions): Don't use "static",
prefer anonymous namespace.
This is consistent with what is done with yacc.c and glr.c. Because
it also avoids that the stack needs to be resized very soon, it should
help keeping tests about destructors more reliable.
Indeed, if the stack is created too small, very soon the C++ library
needs to enlarge it, which means creating a new one, copying the
elements from the initial one onto it, and then destroy the elements
of the initial stack: that would be a spurious call to a destructor.
Reported by Thomas Jahns.
http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html
* data/stack.hh (stack::stack): Reserve 200 slots.
* tests/c++.at: Remove traces of stack expansions.
Reported by Thomas Jahns.
http://lists.gnu.org/archive/html/bug-bison/2015-01/msg00059.html
* tests/synclines.at (AT_SYNCLINES_COMPILE): Rename as...
(_AT_SYNCLINES_COMPILE): this.
Try to recognize xlc locations.
(AT_SYNCLINES_COMPILE): New. Skips the test if we can't read the
synclines.
* origin/maint:
doc: minor fixes
gnulib: strtoul is considered obsolete and now useless
c++: avoid warnings when destructors don't use $$
maint: post-release administrivia
version 3.0.3
gnulib: update
On MacOS X 10.5 PPC with Apple's GCC 4.0.1:
% uname -a
Darwin aria.cielonegro.org 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul 15 16:57:0
1 PDT 2009; root:xnu-1228.15.4~1/RELEASE_PPC Power Macintosh
% gcc --version
powerpc-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5493)
Copyright (C) 2005 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
building in place enters into an infinite recursion on "#include_next":
% gmake V=1
[snip]
depbase=`echo lib/math.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
gcc -std=gnu99 -I. -Ilib -I. -I./lib -g -O2 -MT lib/math.o -MD -MP -MF $depbase.Tpo -c -o lib/math.o lib/math.c &&\
mv -f $depbase.Tpo $depbase.Po
In file included from lib/math.h:27,
from lib/math.h:27,
from lib/math.h:27,
from lib/math.h:27,
[snip]
from lib/math.h:27,
from lib/math.h:27,
from lib/math.c:3:
lib/math.h:27:23: error: #include nested too deeply
Makefile:3414: recipe for target 'lib/math.o' failed
gmake[2]: *** [lib/math.o] Error 1
Using -I./lib instead of -Ilib fixes the problem.
Reported by Pho.
<https://lists.gnu.org/archive/html/bison-patches/2014-01/msg00000.html>
* Makefile.am (AM_CPPFLAGS): Use -I./lib instead of -Ilib.
Currently on the following grammar:
%type <foo> foo
%%
start: foo | bar | "baz"
foo: foo
bar: bar
bison reports:
warning: 2 nonterminals useless in grammar [-Wother]
warning: 4 rules useless in grammar [-Wother]
1.13-15: warning: nonterminal useless in grammar: foo [-Wother]
%type <foo> foo
^^^
3.14-16: warning: nonterminal useless in grammar: bar [-Wother]
start: foo | bar | "baz"
^^^
[...]
i.e., the location of the first occurrence of a symbol is taken as its
definition point. In the case of nonterminals, the first occurrence
as a left-hand side of a rule makes more sense:
warning: 2 nonterminals useless in grammar [-Wother]
warning: 4 rules useless in grammar [-Wother]
4.1-3: warning: nonterminal useless in grammar: foo [-Wother]
foo: foo
^^^
5.1-3: warning: nonterminal useless in grammar: bar [-Wother]
bar: bar
^^^
[...]
* src/symtab.h, src/symtab.c (symbol::location_of_lhs): New.
(symbol_location_as_lhs_set): New.
* src/parse-gram.y (current_lhs): Use it.
* tests/reduce.at: Update locations.
In the following grammar, the 'exp' nonterminal is trivially useless.
So, of course, its rules are useless too.
%%
input: '0' | exp
exp: exp '+' exp | exp '-' exp | '(' exp ')'
Previously all the useless rules were reported, including those whose
left-hand side is the 'exp' nonterminal:
warning: 1 nonterminal useless in grammar [-Wother]
warning: 4 rules useless in grammar [-Wother]
2.14-16: warning: nonterminal useless in grammar: exp [-Wother]
input: '0' | exp
^^^
2.14-16: warning: rule useless in grammar [-Wother]
input: '0' | exp
^^^
! 3.6-16: warning: rule useless in grammar [-Wother]
! exp: exp '+' exp | exp '-' exp | '(' exp ')'
! ^^^^^^^^^^^
! 3.20-30: warning: rule useless in grammar [-Wother]
! exp: exp '+' exp | exp '-' exp | '(' exp ')'
! ^^^^^^^^^^^
! 3.34-44: warning: rule useless in grammar [-Wother]
! exp: exp '+' exp | exp '-' exp | '(' exp ')'
! ^^^^^^^^^^^
The interest of being so verbose is dubious. I suspect most of the
time nonterminals are not expected to be useless, so the user wants to
fix the nonterminal, not remove its rules. And even if the user
wanted to get rid of its rules, the position of these rules probably
does not help more that just having the name of the nonterminal.
This commit discard these messages, marked with '!', and keep the
others. In particular, we still report:
2.14-16: warning: rule useless in grammar [-Wother]
input: '0' | exp
^^^
All the useless rules (including the '!' ones) are still reported in
the reports (xml, text, etc.); only the diagnostics on stderr change.
* src/gram.c (grammar_rules_useless_report): Don't complain about
useless rules whose lhs is useless.
* src/reduce.h, src/reduce.c (reduce_nonterminal_useless_in_grammar):
Take a sym_content as argument.
Adjust callers.
* tests/reduce.at (Useless Rules, Underivable Rules, Reduced Automaton):
Adjust.
* origin/maint:
tests: split a large test case into several smaller ones
package: a bit of trouble shooting indications
doc: liby's main arms the internationalization
bison: avoid warnings from static code analysis
c++: fix the use of destructors when variants are enabled
style: tests: simplify the handling of some C++ tests
c++: symbols can be empty, so use it
c++: variants: don't leak the lookahead in error recovery
c++: provide a means to clear symbols
c++: clean up the handling of empty symbols
c++: comment and style changes
c++: variants: comparing addresses of typeid.name() is undefined
c++: locations: complete the API and fix comments
build: do not clean figure sources in make clean
* tests/conflicts.at (AT_CONSISTENT_ERRORS_CHECK): Move AT_SETUP/AT_CLEANUP
into it, so that we don't skip non Java tests following a test case in Java.
A static analysis tool reports that some callers of symbol_list_n_get
might get NULL and not handle it properly. This is not the case, yet
we can suppress this pattern.
Reported by Mike Sullivan.
<https://lists.gnu.org/archive/html/bug-bison/2013-12/msg00027.html>
* src/symlist.c (symbol_list_n_get): Actually it is never called
to return 0. Enforce this postcondition via aver.
(symbol_list_n_type_name_get): Simplify accordingly. In particular,
discards a (translated) useless error message.
* src/symlist.h: Adjust documentation.
* src/scan-code.l: Style change.
When using variants, destructors generate invalid code.
<http://lists.gnu.org/archive/html/bug-bison/2014-09/msg00005.html>
Reported by Michael Catanzaro.
* data/c++.m4 (~basic_symbol): b4_symbol_foreach works on yysym:
define it.
* tests/c++.at (Variants): Check it.
The previous patches ensure that symbols (symbol_type and
stack_symbol_type) can be empty, cleared, and their emptiness can be
checked. Therefore, yyempty, which codes whether yyla is empty or
not, is now useless.
In C skeletons (e.g., yacc.c), the fact that the lookahead is empty is
coded by "yychar = YYEMPTY", which is exactly what this patch
restores, since yychar/yytoken corresponds to yyla.type.
* data/lalr1.cc (yyempty): Remove.
Rather, depend on yyla.empty ().
During error recovery, when discarding the lookeahead, we don't
destroy it, which is caught by parse.assert assertions.
Reported by Antonio Silva Correia.
With an analysis and suggested patch from Michel d'Hooge.
<http://savannah.gnu.org/support/?108481>
* tests/c++.at (Variants): Strengthen the test to try syntax errors
with discarded lookahead.
The symbol destructor is currently the only means to clear a symbol.
Unfortunately during error recovery we might have to clear the
lookahead, which is a local variable (yyla) that has not yet reached
its end of scope.
Rather that duplicating the code to destroy a symbol, or rather than
destroying and recreating yyla, let's provide a means to clear a
symbol.
Reported by Antonio Silva Correia, with an analysis from Michel d'Hooge.
<http://savannah.gnu.org/support/?108481>
* data/c++.m4, data/lalr1.cc (basis_symbol::clear, by_state::clear)
(by_type::clear): New.
(basic_symbol::~basic_symbol): Use clear.
Instead of storing and comparing pointers to names of types, store
pointers to the typeids, and compares the typeids.
Reported by Thomas Jahns.
<http://lists.gnu.org/archive/html/bug-bison/2014-03/msg00001.html>
* data/variant.hh (yytname_): Replace with...
(yytypeid_): this.
There are no support for += between locations, and some comments are wrong.
Reported by Alexandre Duret-Lutz.
* data/location.cc: Fix.
* doc/bison.texi: Document.
* tests/c++.at: Check.
* origin/maint:
build: don't try to generate docs when cross-compiling
package: fix a reporter's name
%union: fix the support for named %union
package: bump to 2015
flex: don't trust YY_USER_INIT
yacc.c: fix broken union when api.value.type=union and %defines are used
doc: fix missing xref
gnulib: update
location: remove some ugly debugging code traces
build: use abort to pacify compiler errors
package: bump to 2014
doc: specify documentation encoding
Bison supports a union tag, for obscure reasons. But it does a poor
job at it, especially since Bison 3.0.
Reported by Stephen Cameron and Tobias Frost.
It did not ensure that the name was not given several times. An easy
way to do this is to make the %union tag be handled as a %define
variable, as they cannot be defined several times.
Since Bison 3.0, the synclines were wrongly placed, resulting in
invalid code. Addressing this issue, because of the way the union tag
was stored (as a code muscle), would have been tedious. Unless we
rather define the %union tag as a %percent variable, whose synclines
are easier to manipulate.
So replace the b4_union_name muscle by the api.value.union.name
%define variable, document, and check.
* data/bison.m4: Make sure that api.value.union.name has a keyword value.
* data/c++.m4: Make sure that api.value.union.name is not defined.
* data/c.m4 (b4_union_name): No longer use it, use api.value.union.name.
* doc/bison.texi (%define Summary): Document it.
* src/parse-gram.y (union_name): No longer define b4_uion_name, but
api.value.union.name.
* tests/input.at (Redefined %union name): New.
* tests/synclines.at (%union name syncline): New.
* tests/types.at: Check named %unions.
Reported by Rich Wilson.
* data/c.m4 (b4_symbol_type_register): Append to b4_union_members,
not b4_user_union_members.
The latter invokes the former, but it is the former which is reinitialized
to empty by b4_value_type_setup_union.
* tests/types.at: Check it.
This reveals another bug, this time in the case of glr.c parsers.
* data/glr.c: Generate the header file before the implementation file,
to be sure that the setup is run before what depends on it.
clang, with -DNDEBUG and -Werror fails on some functions that might
lack a return. This is because aver is just another assert, discarded
with -DNDEBUG. So use abort.
* src/muscle-tab.c, src/scan-skel.l: here.
* doc/bison.texi: Add '@documentencoding UTF-8'; needed since the
manual contains UTF-8 characters. This will cause the .info files
to contain UTF-8 quotes and the like, which should be OK nowadays.
Add @documentlanguage while we're at it.
This completes and fixes a728075710.
Reported by Valentin Tolmer.
Before it Bison used to put the properties of the symbols
(associativity, printer, etc.) in the 'symbol' structure. An
identifier-named token (FOO) and its string-named alias ("foo")
duplicated these properties, and symbol_check_alias_consistency()
checked that both had compatible properties and fused them, at the end
of the parsing of the grammar.
The commit a728075710 introduces a
sym_content structure that keeps all these properties, and ensures
that both aliases point to the same sym_content (instead of
duplicating). However, it removed symbol_check_alias_consistency,
which resulted in the non-fusion of *existing* properties:
%token FOO "foo"
%left FOO %left "foo"
was properly diagnosed as a redeclaration, but
%left FOO %left "foo"
%token FOO "foo"
was not, as the properties of FOO and "foo" were not checked before
fusion. It certainly also means that
%left "foo"
%token FOO "foo"
did not transfer properly the associativity to FOO.
The fix is simple: reintroduce symbol_check_alias_consistency (under a
better name, symbol_merge_properties) and call it where appropriate.
Also, that commit made USER_NUMBER_HAS_STRING_ALIAS useless, but left
it.
* src/symtab.h (USER_NUMBER_HAS_STRING_ALIAS): Remove, unused.
Adjust dependencies.
* src/symtab.c (symbol_merge_properties): New, based on the former
symbol_check_alias_consistency.
* tests/input.at: Re-enable tests that we now pass.
* origin/maint:
package: install the examples
package: install README and the like in docdir
diagnostics: fix the order of multiple declarations reports
symbol: provide an easy means to compare them in source order
Conflicts:
src/symtab.c
tests/input.at
* tests/input.at: Comment out a test that master currently does not
pass (because of a728075710).
Currently, we do not install the various examples extracted from the
documentation. Let's do it, as they are useful starting points.
* configure.ac: When --enable-gcc-warnings is set, enable ENABLE_GCC_WARNINGS.
* examples/extexi: No longer issue synclines by default.
* examples/local.mk: Except if ENABLE_GCC_WARNINGS.
* examples/calc++/local.mk, examples/mfcalc/local.mk,
* examples/rpcalc/local.mk: Install the example files.
* origin/maint: (43 commits)
maint: post-release administrivia
version 3.0.2
gnulib: update
output: do not generate source files when late errors are caught
output: record what generated files are source or report files
output: do not generate source files when early errors are caught
xml: also use "%empty" with html output
style: formatting changes
xml: also display %empty for empty right-hand sides
reports: display %empty in the generated pointed-rules
news: YYERROR vs variants
style: scope reduction in lalr.cc
lalr1.cc: formatting changes
lalr1.cc: fix the support of YYERROR with variants
tests: check $$'s destruction with variant, YYERROR, and no error recovery
tests: simplify useless obfuscation
skeletons: use better names when computing a "goto"
maint: post-release administrivia
version 3.0.1
aver: it is no longer "protected against NDEBUG"
...
Conflicts:
data/glr.c
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.
* 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.
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.
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.
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.
* 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.
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.
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".
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.
* 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.
* 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.
* configure.ac (FLEX_CXX_WORKS): New AM_CONDITIONAL.
* examples/calc++/local.mk (examples/calc++/calc++):
Build if FLEX_CXX_WORKS, not BISON_CXX_WORKS.
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.
* origin/maint:
glr: more assertions
glr: shorten scopes
glr: formatting changes
glr: better use of tracing macros
examples: improve the output of the "variant" example
variant: remove useless assertion
tests: remove stray debugging traces
tests: do not use grep -q
build: don't require flex for ordinary builds
maint: update .gitignore
build: port to pre-5.8.7 perl
tests: minor change to make it easier to test other skeletons
uniqstr: fix assertion
The commit which introduces yyresolveLocations (commit
8710fc41aa) saves and restores the
look-ahead (type, value and location) for no clear reason. This
appears to be useless.
* data/glr.c (yyresolveLocations): Don't save/restore the current
look-ahead to call YYLLOC_DEFAULT.
Minor style changes.
Rather than having duplicate info in the symbol and the alias that has
to be resolved later on, both the symbol and the alias have a common
pointer to a separate structure containing this info.
* src/symtab.h (sym_content): New structure.
* src/symtab.c (sym_content_new, sym_content_free, symbol_free): New
* src/AnnotationList.c, src/conflicts.c, src/gram.c, src/gram.h,
* src/graphviz.c, src/ielr.c, src/output.c, src/parse-gram.y, src/print.c
* src/print-xml.c, src/print_graph.c, src/reader.c, src/reduce.c,
* src/state.h, src/symlist.c, src/symtab.c, src/symtab.h, src/tables.c:
Adjust.
* tests/input.at: Fix expectations (order changes).
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.
* 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".
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);".
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.
* 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.
* 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 <[email protected]>
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.
* --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 <[email protected]>
Signed-off-by: Akim Demaille <[email protected]>
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.
* 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.
-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.
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.
* src/parse-gram.y (version_check, gram_error, char_name, lloc_default):
Move their prototypes from %{...%} to %code.
(YYLLOC_DEFAULT, YY_LOCATION_PRINT): Move from %{...%} to %code.
(current_lhs): Move its implementation to the epilogue.
* src/muscle-tab.c (muscle_percent_define_get_kind)
(muscle_percent_define_check_kind): New.
(muscle_percent_define_default): Variables with a default value are
of "keyword" kind.
(muscle_percent_define_flag_if, muscle_percent_define_check_values):
Check that the variable is of keyword kind.
* data/bison.m4: Likewise, but in M4. That is to say...
(b4_percent_define_default): Define the kind when the variable is undefined.
(b4_percent_define_check_kind): Use a better error message.
(_b4_percent_define_check_values, _b4_percent_define_check_values):
Former "enum" variables should be defined using the keyword syntax.
* doc/bison.texi: Update.
A couple of fixes.
* tests/input.at (%define keyword variables): New.
* src/muscle-tab.c (muscle_percent_define_ensure): Discover the virtues
of || to factor conditionals.
* NEWS: As api.pure is no longer flagged as "used" by accident,
we now have warnings for useless definitions.
* tests/calc.at: So remove api.pure settings when running C++ tests,
since C++ skeletons use a pure interface.
* src/muscle-tab.c (muscle_percent_define_get_raw): New.
Use it where appropriate.
(location_decode): No longer fetch the value from the table,
take the value as argument.
* data/bison.m4 (b4_percent_define_check_kind): New.
Use it to check api.token.prefix.
* data/c++.m4: Check the kind of api.namespace.
* doc/bison.texi: Update a reference to former 'namespace' variable.
* tests/input.at ("%define" code variables): Check api.namespace.
Reported by Daniel Frużyński.
http://lists.gnu.org/archive/html/bug-bison/2013-02/msg00000.html
* data/location.cc (position::columns, position::lines): Check for
underflow.
Fix some weird function signatures.
(location): Accept signed integers as arguments where appropriate.
Add operator- and operator+=.
* doc/bison.texi (C++ position, C++ location): Various fixes
and completion.
* tests/c++.at (C++ Locations): New tests.
Recently "braceless" in the parser was changed so that an eol was no
longer added to the value. This is not correct when a %code is used
multiple times, because the syncline of the next snippet might be
appended to the last (and not ended) line of the previous snippet.
* src/muscle-tab.h (muscle_grow): Make it private.
* src/muscle-tab.c (muscle_grow): Accept a fourth argument: a required
terminator.
Adjust callers.
* tests/input.at (Multiple %code): New.
* origin/maint:
glr.cc: fix a clang warning
maint: update copyright years
build: fix VPATH issue
build: avoid clang's colored diagnostics in the test suite
tests: please clang and use ".cc", not ".c", for C++ input
gnulib: update
skeletons: avoid empty switch constructs
lalr1.cc: fix compiler warnings
yacc.c: do not use __attribute__ unprotected
tests: style changes
Suggested by Joel E. Denny.
http://lists.gnu.org/archive/html/bison-patches/2013-03/msg00016.html
* data/bison.m4 (b4_percent_define_get_kind): New.
(b4_variant_flag): Check that api.value.type is defined as the 'variant'
keyword value.
* data/c.m4 (_b4_value_type_setup_keyword): New.
(b4_value_type_setup): Use it to simplify reading.
Use b4_define_silent.
Decode api.value.type, including its type.
(b4_value_type_define): Likewise.
* data/c++.m4 (b4_value_type_declare): Adjust the decoding of api.value.type,
taking its kind into account.
* doc/bison.texi: Adjust all the examples to the new syntax.
* NEWS: Ditto.
* tests/types.at: Adjust
Instead of generating invalid C code, generate an error when --yacc and
'%define api.value.type union' are used together.
* data/bison.m4: Issue an error in this case.
* tests/types.at (%yacc vs. %define api.value.type union): New, check this
error.
* doc/bison.texi (Type Generation): Document it.
* tests/output.at: Check that '-o y.tab.c' and '-y' behave equally
wrt generated file names.
* NEWS (Use of YACC='bison -y'): New.
Promote the use of 'bison -o y.tab.c'.
Provide a means to dereference pointers when defining tags. One
example could be:
%code requires
{
typedef struct ListElementType
{
union value
{
int intVal;
float floatVal;
char* charptrVal;
} value;
struct ListElementType* next;
} ListElementType;
}
%union
{
ListElementType* list;
}
%token <list->value.charptrVal> STRING
%token <list->value.intVal> INTEGER
%token <list->value.floatVal> REAL
%type <list> ElementList LiteralType
* src/scan-code.l, src/scan-gram.l: Accept "->" in tags.
* tests/types.at: Add more test cases to cover this case.
* src/scan-gram.l: Remove the rule for simple tags: the "complex" case
subsumes it. It was more efficient, but duplicated the code for a
negligible benefit.
* data/c.m4 (b4_symbol_type_register, b4_type_define_tag)
(b4_symbol_value_union, b4_value_type_setup_union)
(b4_value_type_setup_variant, b4_value_type_setup):
New.
(b4_value_type_define): Use it to set up properly the type.
Handle the various possible values of api.value.type.
* data/c++.m4 (b4_value_type_declare): Likewise.
* data/lalr1.cc (b4_value_type_setup_variant): Redefine.
* tests/types.at: New.
Exercise all the C/C++ skeletons with different types of
api.value.type values.
* tests/local.mk, tests/testsuite.at: Use it.
* doc/bison.texi (%define Summary): Document api.value.type.
* NEWS: Advertise it, together with api.token.constructor.
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.
Conflicts:
configure.ac
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++.
Reported by Rob Conde.
http://lists.gnu.org/archive/html/bug-bison/2013-03/msg00003.html
* data/c.m4 (b4_symbol_actions): Rename as...
(_b4_symbol_actions): this.
(b4_symbol_actions): New wrapper.
Do not emit empty switches.
Adjust all b4_symbol_actions callers.
Reported by Rob Conde.
http://lists.gnu.org/archive/html/bug-bison/2013-03/msg00003.html
* data/stack.hh (operator=, stack(const stack&)): Make this class
uncopyable, i.e., "undefine" these operators: make them private and
don't implement them.
(clear): New.
* data/lalr1.cc: Use it instead of an assignment.
(parser): Make this class uncopyable.
* tests/c++.at, tests/input.at: Use "%define api.namespace {foo}" instead
of using quotes.
* tests/local.at (AT_SETUP_STRIP, AT_NAME_PREFIX): Recognize uses of
braces instead of quotes.
* doc/bison.texi: Use braces for api.namespace's values.
Now that we use "braceless" (which is {...} blocks of code with
initial and final braces stripped) to denote "short" values (such as
api.namespaces), the added end-of-line is a nuisance. As a matter of
fact, this extra-safety was useless, as every expansion of "braceless"
(aka, "user code") is followed by an end of line.
* src/parse-gram.y, src/parse-gram.c (braceless): Instead of replacing
the final brace by \n, just delete the brace.
Provide a means to tell the difference between "keyword" values (e.g.,
%define api.pull both), "string" values (e.g., %define file.name
"foo"), and "code" values (e.g., %define api.namespace {calc}).
Suggested by Joel E. Denny.
http://lists.gnu.org/archive/html/bison-patches/2013-03/msg00016.html
* src/muscle-tab.h, src/muscle-tab.c (muscle_kind, muscle_kind_new)
(muscle_kind_string): New.
(muscle_percent_define_insert): Take the kind as new argument.
Insert it in the muscle table.
Adjust callers.
* src/getargs.c: Adjust callers.
* src/parse-gram.y: Ditto.
(content.opt): Remove, replaced by...
(value): this new non-terminal, whose semantics value is stored
in the new "value" union member.
Provide a printer.
Support values in braces in additions to keyword and string values.
fuse me
* src/parse-gram.y (PARAM_TYPE): Remove useless typedef guard.
There's a header guard.
Use 'yyo' with %printer.
Use a consistent style for %union one-liners.
* src/parse-gram.y (INT): Fuse the %type and %token declaration.
Move its %union right before its introduction.
(%union): Split in several %unions, right before their use.
In a precedence declaration, when tokens are declared with a litteral
character (e.g., 'a') or with a identifier (e.g., B), Bison behaved
differently: the litteral tokens would be numbered first, and then the
other ones, leading to the following grammar:
%right A B 'c' 'd'
being numbered as such: 'c' 'd' A B.
* src/parse-gram.y (symbol.prec): Set the symbol number when reading the
symbols.
* tests/conflicts.at (Token declaration order: literals vs. identifiers):
New.
Signed-off-by: Akim Demaille <[email protected]>
It would be a pity to warn the users against Bison features...
http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00107.html
* src/complain.h, src/complain.c (Wall): Disable Wyacc.
(Weverything): New (hidden so far) category which really denotes all
the categories (what used to be Wall).
(warnings_args, warnings_types): Adjust.
(warning_argmatch): Now !none = Weverything and conversely, no longer Wall.
* NEWS, doc/bison.texi, src/getargs.c: Adjust the documentation.
* tests/input.at (-Werror is not affected by -Wnone and -Wall): Adjust
by not using a -Wyacc type of warning.
Bison 3.0 is already breaking backward compatibility with other
features. It is an appropriate time to drop this feature. Note that
it was disabled when --yacc is passed. See
http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00102.html
Basically, revert e8cd1ad655.
* src/scan-code.l (braces_level, need_semicolon, in_cpp): Remove.
Remove every rule needed to detect and add missing semicolon.
* tests/actions.at (Fix user actions without a trailing semicolon):
Remove.
* NEWS: Adjust.
This is to match the names used in C and api.value.type, even if the
parser actually defines semantic_type.
* data/c++.m4 (b4_semantic_type_declare): Rename as...
(b4_value_type_declare): this.
* data/variant.hh: Likewise.
* tests/c++.at (Object): Somehow instances of Object were assigned
YY_NULL, which is 0 most of the time (that case passes), but is
nullptr in C++11, and there is nothing in Object to support such an
assignment (failure). Use 0 as value, and provide the needed
assignment operator.
Also, use a more natural order within the class definition.
* configure.ac (WARN_NO_NULL_CONVERSION_CXXFLAGS): Rename as...
(FLEX_SCANNER_CXXFLAGS): this.
Pass -Wno-zero-as-null-pointer-constant to G++ if it supports it.
* examples/calc++/local.mk: Adjust.
The changes by Théophile Ranquet about type punning issues need
to be extend to in-place new to please G++ 4.4.7.
* data/variant.hh (variant::as_): New, factors the casts that avoid
compiler warnings.
(as, build): Use them.
When reporting a duplicate directive on a rule, point to its first
occurrence:
one.y:11.10-15: error: only one %empty allowed per rule
%empty {} %empty
^^^^^^
one.y:11.3-8: previous declaration
%empty {} %empty
^^^^^^
And consistently discard the second one.
* src/complain.h, src/complain.c (duplicate_directive): New.
* src/reader.c: Use it where appropriate.
* src/symlist.h, src/symlist.c (symbol_list): Add a dprec_location member.
* tests/actions.at: Adjust expected output.
* doc/bison.texi (Grammar Rules): Make it a @section which
contains...
(Rules Syntax): this new subsection (with the previous contents of
"Grammar Rules".
(Empty Rules): New subsection, extracted from the former
"Grammar Rules".
Document %empty.
(Recursion): New a subsection of "Grammar Rules".
Complete a few index entries.
(Bison Options): Document -Wempty-rule.
* src/complain.h, src/complain.c (warning_is_unset): New.
* src/reader.c (grammar_current_rule_empty_set): If enabled -Wempty-rule,
if not disabled.
* tests/actions.at (Implicitly empty rule): Check this feature.
Also check that -Wno-empty-rule does disable this warning.
Provide a means to explicitly denote empty right-hand sides of rules:
instead of
exp: { ... }
allow
exp: %empty { ... }
Make sure that %empty is properly used.
With help from Joel E. Denny and Gabriel Rassoul.
http://lists.gnu.org/archive/html/bison-patches/2013-01/msg00142.html
* src/reader.h, src/reader.c (grammar_current_rule_empty_set): New.
* src/parse-gram.y (%empty): New token.
Use it.
* src/scan-gram.l (%empty): Scan it.
* src/reader.c (grammar_rule_check): Check that %empty is properly used.
* tests/actions.at (Invalid uses of %empty, Valid uses of %empty): New.
This allows to reduce the number of public interfaces.
* src/getargs.c (--yacc): Use warning_argmatch instead of tweaking
directly warnings_flag (which will be private).
(warning_argmatch, warnings_argmatch): Move to...
* src/complain.h, src/complain.c: here.
* src/getargs.h, src/getargs.c (warnings_args, warnings_types): Move to...
* src/complain.c: here, now private.
* src/complain.h (severity, warnings_flag): Move to...
* src/complain.c: here, now private.
Recent discussions with Joel E. Denny
(http://lists.gnu.org/archive/html/bison-patches/2013-02/msg00026.html)
show that it is desirable to tell the difference between an option
that was explicitly disabled with -Wno-foo, as opposed to be left
unset. The current framework does not allow this.
Instead of having a first int to store which options are enabled, and
another to store which are turned into errors, use an array that for
each warning category tells its status: disabled, unset, warning,
error.
* src/complain.h, src/complain.c (warning_bit): New enum.
(warnings): Use it.
(severity): New enum.
(warnings_flag): Now an array of severity.
(errors_flag): Remove, now done by warnings_flag.
(complain_init): New function, to initialie warnings_flag.
(warnings_are_errors): New Boolean, for -Werror.
* src/complain.c (warning_severity): New.
(warnings_print_categories, complains): Use it.
* src/getargs.c (warning_argmatch): Adjust to use warnings_flag.
(warnings_argmatch): Ditto.
Handle -Werror and -Wno-error here.
(getargs): Adjust.
* src/main.c (main): Call complain_init.
* tests/input.at (Invalid options): Add more corner cases.
* src/getargs.c (warnings_argmatch, warning_argmatch): Simplify by
replacing function arguments with their actual values.
(WARNING_ARGMATCH): Remove, useless.
Adjust callers.
* src/getargs.c (warning_argmatch, warnings_argmatch, WARNINGS_ARGMATCH):
New.
Use them for -W/--warning.
They are copied from...
(flag_argmatch, flags_argmatch, FLAGS_ARGMATCH): these.
Simplify by removing the support for "error".
* tests/input.at (Invalid options): New.
* TODO (Laxism in Bison invocation arguments): Remove.
* src/getargs.h, src/getargs.c (warnings_args, warnings_types): Make
them public.
* src/complain.h, src/complain.c (warnings_print_categories): Its
only use outside complain.c was removed in a recent commit, so
make it static.
Simplify its implementation.
Use warnings_args and warnings_types.
* src/muscle-tab.c (muscle_percent_define_check_values): Make it
silent.
* src/gram.c (grammar_rules_useless_report): Let -fcaret handle the
pretty-printing of the guilty rules.
(rule_print): Inline in its only use.
* tests/conflicts.at, tests/existing.at, tests/reduce.at,
* tests/regression.at: Adjust.
* NEWS: Document.
The argmatch functions accept prefixes of the alternatives (like
getopt does for long options). Bison uses this to document the
warning categories. This is troublesome: it duplicates the --help
documentation, it is not gettextized, it is displayed with ugly quotes
(because argmatch uses it to display the list of possible answers),
and it prevents straighforward uses of the tables of valid warning
categories (for instance so that warning diagnostics end with the name
of the warning).
The "hidden" option --trace uses the same trick, but it does not need
to be translated, nor to be described in --help.
* src/getargs.c (warnings_args): Remove pseudo documentation.
Comment 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.
"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.
* 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.
* 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.
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.
* 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.
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.
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.
* 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.
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).
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.
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.
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.
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.
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.
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.
* 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 $accept, which is valueless (the needed
property: the symbol destructor must not try to reclaim the memory
associated with the symbol).
* data/c++.m4 (b4_public_types_declare): Declare token_number_type soon.
Introduce symbol_number_type (wider than token_number_type).
Clarify the requirement that kind_type from by_state and by_type
denote the _input_ type (required by the constructor), not the stored type.
Use symbol_number_type and token_number_type where appropriate, instead
of int.
* data/lalr1.cc: Adjust to these changes.
Propagate "symbol_number_type".
Invoke "type_get ()" instead of read "type" directly.
* doc/Doxyfile.in: Let doxygen upgrade it.
(INCLUDE_PATH): Point to lib too.
(PROJECT_BRIEF): New.
(EXCLUDE): Update to reflect the current file hierarchy.
* cfg.mk: Ignore strcmp in local.at.
* tests/conflicts.at: Use AT_PARSER_CHECK.
* tests/regression.at: Preserve the exit status of the generated parsers.
* tests/local.mk ($(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.
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++.
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.
* 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.
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.
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".
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.
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.
* 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.
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).
See commit 3804aa260b.
* 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.
* 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.
* 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.
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'.
* 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.
* 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.
* 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.
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.
* 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.
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.
So far we were issuing two lines for each syncline change:
/* Line 356 of yacc.c */
#line 1 "src/parse-gram.y"
This is a lot of clutter, especially when reading diffs, as these
lines change often. Fuse them into a single, shorter, line:
#line 1 "src/parse-gram.y" /* yacc.c:356 */
* data/bison.m4 (b4_syncline): Issue a single line.
Comment improvements.
(b4_sync_start, b4_sync_end): Issue a shorter comment.
* data/c++.m4 (b4_semantic_type_declare): b4_user_code must be
on its own line as it might start with a "#line" directive.
* src/gram.c (print_lhs): Use %*s to indent.
* src/print_graph.c (print_lhs): Use obstack_printf.
Became simple enough to be inlined in...
(print_core): here.
Use a "rule*" instead of an index in "rules[]".
* src/location.c (location_caret): finish the line with one whether or not it
is present in input. Rewrite code without getline.
(cleanup_caret): Reset the caret_info global.
* bootstrap.conf: No longer require getline.
Unput was no longer used since a POSIX-compatiblity issue with Flex 2.5.31,
which has been adressed in newer versions of Flex. See this discussion:
<http://lists.gnu.org/archive/html/bug-bison/2003-04/msg00029.html>
This partially reverts commit aa4180418f.
* src/scan-gram.l (unexpected_end): Here.
* tests/input.at: Adjust for new order of error reports.
The commit 38de4e570f underquoted the
content of the comments, which resulted in losing square brackets in
the comments. Besides, some other invocations were underquoting the
effective arguments.
* data/c.m4 (b4_comment_): Properly quote the comment.
(b4_comment_, b4_comment): Move to...
* data/c-like.m4: here, so that...
* data/java.m4: can use it instead of its own copy.
* data/bison.m4 (b4_integral_parser_tables_map): Fix some comments.
* data/lalr1.cc, data/lalr1.java, data/yacc.c: Comment fixes.
* data/lalr1.cc: Reorder a bit to factor some CPP directives.
* src/tables.c (save_column, pack_vector): Reduce the scope to
emphasize the structure of the code.
Rename the returned value "res" to make understanding easier.
The YYLEX existed only to support YYLEX_PARAM, which is now removed.
This macro was a nuisance, since incorrect yylex calls where pointed
the macro _use_, instead of its definition.
* data/c.m4 (b4_lex_formals, b4_lex): New.
* data/glr.c, data/yacc.c: Use it.
* data/lalr1.cc (b4_lex): New.
Use it.
squash! skeletons: no longer call yylex via a CPP macro
* data/c.m4 (b4_token_enum): Improve comments.
(b4_value_type_define, b4_location_type_define): New, extracted
from...
(b4_declare_yylstype): here.
Separate the typedefs from the union/struct definitions.
* tests/input.at: Rename some AT_SETUP to avoid that
AT_SETUP_STRIP thinks they contain %define directives.
("%define" backward compatibility): Merge tests together
to speed up the test suite, and to make maintenance easier
(multiple AT_CHECK means multiple runs of the test suite to
be sure to have updated all the error messages).
Check the "%define variant" is properly obsoleted.
* src/muscle-tab.c (define_directive): Be robust to "assignment"
containing '='.
(muscle_percent_variable_update): Upgrade "variant" to "api.value.type".
Support such upgrade patterns.
Adjust callers.
* data/bison.m4: Use api.value.type for variants.
* tests/c++.at: Adjust tests.
Instead of
warning: deprecated %define variable name: 'namespace', use 'api.namespace' [-Wdeprecated]
display (in -fno-caret mode):
warning: deprecated directive: '%define namespace foo', use '%define api.namespace foo' [-Wdeprecated]
and (in -fcaret mode):
warning: deprecated directive, use '%define api.namespace toto' [-Wdeprecated]
%define namespace toto
^^^^^^^^^
This is in preparation of cases where not only the variable is
renamed, but the values are too:
warning: deprecated directive: '%define variant', use '%define api.value.type variant' [-Wdeprecated]
* src/muscle-tab.c (define_directive): New.
(muscle_percent_variable_update): Take the value as argument, and use it
in the diagnostics.
Loop with a pointer instead of an index.
* tests/input.at (%define backward compatibility): Adjust.
Improve the output by ensuring a well defined order for type switches.
* src/uniqstr.h: Style changes for macro arguments.
(UNIQSTR_CMP): Replace by...
(uniqstr_cmp): this.
* src/uniqstr.c (uniqstr_cmp): New.
Produce well defined results.
* src/output.c: Use it.
* tests/local.at (AT_SETUP_STRIP): New.
(AT_SETUP): Use it to shorten the test titles: remove %defines, %language
and %skeleton whose arguments suffice.
* tests/synclines.at: Use more precise AT_SETUP.
This is mostly used for the license header, the synclines, and the
generated tables:
- /* STOS_[STATE-NUM] -- The (internal number of the) accessing
- symbol of state STATE-NUM. */
+ // STOS_[STATE-NUM] -- The (internal number of the) accessing
+ // symbol of state STATE-NUM.
static const unsigned char yystos_[];
* data/c.m4: Comment changes.
(b4_comment_): Expand the text argument.
Before this change, we were actually formatting M4 code as a
C comment, and then expand it.
(b4_comment): Fix the closing of comments: there is no reason to
add the (line) prefix before the closing "*/".
* data/c++.m4 (b4_comment): New.
The "variant" structure provides a means to store, in a typeless way,
C++ objects. Manipulating it without provide the type of the stored
content is doomed to failure. So provide a means to copy in a type
safe way, and prohibit typeless assignments.
* data/c++.m4 (symbol_type::move): New.
* data/lalr1.cc: Use it.
* data/variant.hh (b4_variant_define): Provide variant::copy.
Let variant::operator= abort.
We cannot undefine it, yet, as it is still uses by the implicit
assigment in symbols, which must also be disabled.
Equip variants with more checking code. Provide a means to request
includes.
* data/variant.hh (b4_variant_includes): New.
* data/lalr1.cc: Use it.
* data/variant.hh (variant::built): Define at the end, as a private member.
(variant::tname): New.
Somewhat makes "built" useless, but let's keep both for a start, in
case using "typeinfo" is considered unacceptable in some environments.
Fix some formatting issues.
The fact that glr.cc uses glr.c makes the handling of parse params
more complex, as the parser object of glr.cc must be passed to the
parse function of glr.c. Yet not all the functions need access to
the parser object.
* data/glr.cc (b4_parse_param_wrap): New.
Use them.
* src/symlist.c (symbol_list_free): Deep free it.
* src/symtab.c (symbols_free, semantic_types_sorted): Free it too.
(symbols_do, sorted): Call by address.
* data/glr.c, src/graphviz.h, src/ielr.c, src/scan-gram.l,
* src/system.h, tests/actions.at, tests/glr-regression.at: Do not
use // comments.
Do not introduce variables after statements.
Provide "main" with a return value.
* origin/maint:
news: prepare for forthcoming release
doc: explain how mid-rule actions are translated
error: use better locations for unused midrule values
doc: various minor improvements and fixes
tests: ignore more useless compiler warnings
tests: be robust to C being compiled with a C++11 compiler
build: beware of Clang++ not supporting POSIXLY_CORRECT
maint: post-release administrivia
version 2.6.90
build: fix syntax-check error.
cpp: simplify the Flex version checking macro
news: improve the carets example and fix a typo
cpp: improve the Flex version checking macro
carets: improve the code
maint: update news
build: keep -Wmissing-declarations and -Wmissing-prototypes for modern GCCs
build: drop -Wcast-qual
gnulib: update
Conflicts:
NEWS
doc/Makefile.am
doc/bison.texi
gnulib
src/reader.c
tests/actions.at
tests/atlocal.in
tests/input.at
* doc/bison.texi (Actions in Mid-Rule): Mention and use named references.
Split into three subsections, among which...
(Mid-Rule Action Translation): this new section.
On
%%
exp: {;} {$$;} { $$ = $1; }
instead of reporting (with -fcaret -Wmidrule-value)
midrule.y:2.6-8: warning: unset value: $$ [-Wmidrule-values]
exp: {;} {$$;} { $$ = $1; }
^^^
midrule.y:2.6-27: warning: unused value: $2 [-Wmidrule-values]
exp: {;} {$$;} { $$ = $1; }
^^^^^^^^^^^^^^^^^^^^^^
report
midrule.y:2.6-8: warning: unset value: $$
exp: {;} {$$;} { $$ = $1; }
^^^
midrule.y:2.10-14: warning: unused value: $2
exp: {;} {$$;} { $$ = $1; }
^^^^^
* src/reader.c (grammar_rule_check): When warning about the value of a
midrule action, use the location of the midrule action instead of the
location of the rule.
the location of the part of the rule.
* tests/actions.at (Default %printer and %destructor for mid-rule values):
Adjust expectations
* tests/input.at (Unused values with default %destructor): Ditto.
(AT_CHECK_UNUSED_VALUES): Ditto.
And use -fcaret.
* doc/figs/example.dot, doc/figs/example.y: New.
* doc/bison.texi: Prefer "token" to TOKEN.
Use @group where appropriate.
Adjust with style changes in the output (State 0, not state 0).
Fix some @ref that were missing the third argument.
Fix some incorrect line numbers.
Use "nonterminal", not "non-terminal".
Fix overfull and underfull TeX hboxes.
Put the comments in the index.
Remove duplicate index entries.
Fuse glossary entries where appropriate.
(Understanding): Improve the continuity between sections.
Use example.dot to show the whole graph.
* doc/Makefile.am: Adjust.
* m4/c-working.m4 (BISON_LANG_COMPILER_POSIXLY_CORRECT): New.
(BISON_C_COMPILER_POSIXLY_CORRECT): Use it.
For consistency with C++, also define BISON_C_WORKS.
* m4/cxx.m4 (BISON_CXX_COMPILER_POSIXLY_CORRECT): New.
* configure.ac: Use it.
* tests/atlocal.in: Get its result.
Propagate properly CXX values when used to compile C.
When POSIXLY_CORRECT, adjust BISON_C_WORKS and BISON_CXX_WORKS.
* tests/local.at (AT_COMPILE): Use BISON_C_WORKS.
* cfg.mk: Exclude names-refs, it includes a "double" if (end of first
line, first of second line below).
test.y:43.12-44.59: symbol not found in production: if
if-stmt-a: IF expr[cond] THEN stmt.list[then] ELSE stmt.list[else] FI
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Fixes a -Werror failure of xalloc.h used in src.
From Eric Blake.
http://lists.gnu.org/archive/html/bug-gnulib/2012-12/msg00006.html
* configure.ac: Check whether GCC pragma diagnostic push/pop works.
Enable these warnings for bison if it does.
Enable these warnings for the test suite anyway.
* origin/maint:
misc: pacify the Tiny C Compiler
cpp: make the check of Flex version portable
misc: require getline
c++: support wide strings for file names
doc: document carets
tests: enhance existing tests with carets
errors: show carets
getargs: add support for --flags/-f
Conflicts:
doc/bison.texi
m4/.gitignore
src/complain.c
src/flex-scanner.h
src/getargs.c
src/getargs.h
src/gram.c
src/main.c
tests/headers.at
* tests/actions.at: Unset value.
* tests/conflicts.at: Rule useless due to conflicts.
* tests/input.at: Missing terminator, unexpected end of file, command line
redefinition of variable.
* tests/named-refs.at: Many errors.
* tests/reduce.at: Useless nonterminals and rules.
* tests/regression.at: Large token.
* origin/maint:
parser: accept #line NUM
m4: use a safer pattern to enable/disable output
tests: beware of gnulib's need for config.h
gnulib: update
yacc.c, glr.c: check and fix the display of locations
formatting changes
glr.c: remove stray macro
Conflicts:
data/c.m4
data/glr.cc
data/lalr1.cc
data/lalr1.java
data/location.cc
data/stack.hh
data/yacc.c
src/scan-gram.l
Work on some other areas of Bison revealed that some macros expanded
to be expanded only once were actually expanded several times. This
was due to the fact that changecom was not properly restored each
time, and macro names appearing in comments were then expanded.
Introduce begin/end macros which are easier to match that
changecom()/changecom(#).
* data/bison.m4 (b4_output_begin, b4_output_end): New.
* data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java,
* data/location.cc, data/stack.hh, data/yacc.c:
Use them.
In some case, negative column number could be displayed.
Make YY_LOCATION_PRINT similar to bison's own implementation of
locations. Since the macro is getting fat, make it a static
function.
Reported by Jonathan Fabrizio.
* data/c.m4 (yy_location_print_define): Improve the implementation,
and generate the yy_location_print_ function.
Adjust YY_LOCATION_PRINT.
* tests/actions.at (Location Print): New tests.
* data/c.m4: Fix comments, put macros in a more natural order.
Space changes (from M-x whitespace-cleanup).
* src/location.c: Fix spaces.
* tests/actions.at: Space changes.
* NEWS: Add entry.
* doc/bison.texi (%define Summary): Show the old Yacc behaviour.
(Parser Function): Move parse-param examples here.
(Pure Calling): Remove parse-param examples.
(Error Reporting): Don't show the old behavior, stick to 'full'.
This makes the interface for yyerror() pure without the need for a spurious
parse_param.
* data/yacc.c (b4_pure_if, b4_pure_flag): New definition, accept three states.
(b4_yacc_pure_if): Rename as...
(b4_yyerror_arg_loc_if): This, and use b4_pure_flag.
* tests/actions.at (%define api.pure): Modernize.
* test/calc.at (Simple LALR Calculator): Modernize.
* tests/local.at (AT_YYERROR_ARG_LOC_IF): Adjust.
* tests/local.at (AT_YYERROR_FORMALS): Make llocp const.
(AT_PURE_AND_LOC_IF, AT_GLR_OR_PARAM_IF): Remove, expand...
(AT_YYERROR_ARG_LOC_IF): Here, and use m4_join for readability.
* origin/maint:
yacc.c: always initialize yylloc
scanner: issue a single error for groups of invalid characters
tests: formatting changes
doc: one of the fixes for an ambiguous grammar was ambiguous too
doc: fix the dangling else with precedence directives
doc: prefer "token" to TOKEN
doc: formatting changes
scanner: use explicit "ignore" statements
Conflicts:
src/scan-gram.l
* origin/branch-2.6:
yacc.c: always initialize yylloc
doc: one of the fixes for an ambiguous grammar was ambiguous too
doc: fix the dangling else with precedence directives
doc: prefer "token" to TOKEN
doc: formatting changes
Conflicts:
NEWS
doc/bison.texi
The initial location might be used if the parser starts by an empty
reduction, so really ensure proper initialization of the initial
location. The previous approach fails for PostgreSQL, which uses
Reported by Peter Eisentraut.
http://lists.gnu.org/archive/html/bug-bison/2012-11/msg00023.html
With help from Théophile Ranquet.
* data/yacc.c (b4_declare_scanner_communication_variables): Be sure
to initialize yylloc, even when its structure is unknown.
(yyparse): Simplify the call to b4_dollar_pushdef.
* tests/actions.at (Initial location): Check of similar pattern
as in the case of PostgreSQL.
* origin/maint:
tests: close files in glr-regression
xml: match DOT output and xml2dot.xsl processing
xml: factor xslt space template
graph: fix a memory leak
xml: documentation
output: capitalize State
Make the DOT produced by XSLT processing equivalent to the one made with the
--graph option.
* data/xslt/xml2dot.xsl: Stylistic changes, and add support for reductions.
* doc/bison.texi (Xml): Update.
* src/graphviz.c (conclude_red): Minor stylistic changes to DOT internals.
(output_red): Swap enabled and disabled reductions output, for coherence
with XSLT output.
* src/print_graph.c (print_core): Minor stylistic change to States' output.
(print_actions): Swap order of output for reductions and transitions.
* tests/local.at (AT_BISON_CHECK_XML): Ignore differences in order.
* tests/output.at: Adjust to changes in DOT internals.
The XML output combined with the XSL Transformations provided in data/ are
incredibly useful, they should be documented.
* doc/bison.texi (Xml): New node.
* origin/maint:
maint: address syntax-check errors.
tests: use valgrind where appropriate
tests: use valgrind where appropriate
tests: don't expect $EGREP to support -w
tests: more possible error compiler messages for "#error"
Conflicts:
cfg.mk
tests/headers.at
* cfg.mk: Ignore the "error" call in tests/c++.at, it is not to be
translated.
* doc/bison.texi: Fix incorrect @pxref use.
* po/POTFILES.in: Add missing file.
* src/print_graph.c: Remove useless include.
Reported by Théophile Ranquet.
* tests/glr-regression.at: Rewrite some test cases so that AT_PARSER_CHECK,
which runs valgrind, is exposed with the parser, not with "echo".
* tests/local.at, tests/regression.at, tests/headers.at:
Use AT_PARSER_CHECK for generated parsers.
Does not work on Solaris 10. Reported by Dennis Clarke.
http://lists.gnu.org/archive/html/bug-bison/2012-11/msg00009.html
* tests/headers.at (Several parsers): Use Perl instead.
While at it, run it only once, on all the generated headers.
Adjust to YY_NULL be defined in position.hh.
* origin/maint:
regen
maint: post-release administrivia
version 2.6.5
regen
tests: syntax-check
tests: beware of compilers that do not support POSIXLY_CORRECT
gnulib: update
Conflicts:
src/parse-gram.c
src/parse-gram.h
tests/atlocal.in
* origin/branch-2.6:
maint: post-release administrivia
version 2.6.5
regen
tests: syntax-check
tests: beware of compilers that do not support POSIXLY_CORRECT
gnulib: update
Conflicts:
NEWS
src/parse-gram.c
src/parse-gram.h
Running "maintainer-release-check" on OS X with Clang 2.9 fails,
because "clang-mp-2.9 -o test -g test.c" launches "/usr/bin/dsymutil
test -o test.dSYM" which fails with "error: unable to open executable
'-o'".
* m4/c-working.m4 (BISON_CHECK_WITH_POSIXLY_CORRECT)
(BISON_C_COMPILER_POSIXLY_CORRECT): New.
* configure.ac: Use the latter.
* tests/atlocal.in (POSIXLY_CORRECT_IS_EXPORTED): New.
* tests/local.at (AT_BISON_CHECK_WARNINGS_): Use it instead of computing its
value each time.
(AT_QUELL_VALGRIND): Skip tests that cannot work because of compilers
that do not support POSIXLY_CORRECT.
* tests/calc.at: Don't initialize the location, let the parser
do it.
Use a $printer.
Change some testing input to be easier to distinguish (instead of always
"0 0" for instance).
There are several issues to address here. One is that yylloc should
be initialized when possible. Another is that the push parser needs
to update yypushed_loc when the user modified it. And if the parser
starts by a reduction of an empty, it uses the first location on the
stack, which, therefore, must also be initialized to this initial
location.
This is getting complex, especially since because initializing a
global (impure interface) is different from initializing a local
variable. To simplify, the local yylloc is not initialized during its
definition.
* data/c.m4 (b4_yyloc_default_define): Replace by...
(b4_yyloc_default): this.
Adjust dependencies.
* data/glr.cc: Initialize yylloc.
* data/yacc.c (b4_declare_scanner_communication_variables):
Initialize yylloc during its definition.
Don't define yyloc_default.
(yypush_parse): The location formal is not const, as we might
initialize it.
(yyparse): Define yyloc_default.
Use it before running the user initial action.
Possibly update the first location on the stack, and the pushed
location after the user initial action.
* tests/actions.at (Initial locations): Check that the initial
location is correct.
* data/c.m4 (b4_yy_location_print_define): New.
Now issues "short" locations, e.g., "1.1" instead of "1.1-1.1".
Was initially a function, but then we face "static but unused"
warnings.
Simpler as a macro.
* tests/local.at, data/glr.c, data/yacc.c: Use it instead of duplicating.
* tests/actions.at: Adjust expectations.
* data/yacc.c: Fuse the initializations of yyssp, yyss and the like.
Remove an obsolete comment: we do initialize these initial stack
members (in some cases).
* tests/local.at (AT_YYERROR_DEFINE): Don't display the end of the location
if it is not after its beginning.
* tests/actions.at, tests/cxx-type.at: Adjust the expected output.
* tests/local.at (AT_PARSE_PARAMS): New.
(AT_YYERROR_FORMALS, AT_YYERROR_DEFINE): Use it to add the parse-param
to yyerror.
* tests/calc.at, tests/regression.at: Use AT_YYERROR_DEFINE and
AT_YYERROR_DECLARE, now that they handle properly the parse-params.
Be sure to let AT_BISON_OPTION_PUSHDEFS now what parse-params are used.
* data/bison.m4 (b4_lex_symbol_if): Rename as...
(b4_token_ctor_if): this.
Depend upon api.token.constructor.
* data/c++.m4, data/lalr1.cc: Adjust.
* doc/bison.texi: Fix all the occurrences of lex_symbol.
* etc/bench.pl.in: Adjust.
* examples/variant.yy: Likewise.
* tests/local.at (AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS):
Handle AT_TOKEN_CTOR_IF.
* tests/c++.at: Adjust to using api.token.constructor and AT_TOKEN_CTOR_IF.
Simplify the test of both build call styles.
(AT_CHECK_VARIANTS): Rename as...
(AT_TEST): this.
And undef when done.
When generating a pure push parser, the initialization of yylval and
yylloc may not be visible to the compiler. With warnings enabled, GCC
4.3.6, 4.4.7, 4.5.4, and 4.6.3 report uninitialized uses of
yylval/yylloc. Using local pragmas to disable these warnings is not
supported before 4.6, and 4.6 does not support it properly. So
initialize yylval and yylloc at their definition. Reported by Peter
Simons. See
http://lists.gnu.org/archive/html/bison-patches/2012-10/msg00133.html
* data/c.m4 (b4_yyloc_default_define): New.
* data/yacc.c: Use it when locations are requested.
(YYLVAL_INITIALIZE): Replace by...
(YY_INITIAL_VALUE): this.
(yyparse): Initialize yylloc and yylval.
Therefore, remove the initialization of yylloc's field.
* data/glr.c: Likewise.
Treating warnings as errors caused Bison to exit earlier than needed, making it
hide warnings that would have been printed had -Werror not been set.
Also, fix a bug that caused some context information of errors to not be
shown.
* src/complain.c (complaint_issued): Rename as...
(complaint_status): This, and change its type from boolean to
* src/complain.h (err_status): This, new enumeration.
* src/main.c (main): Adjust (only finish early if an actual complaint was
risen, not a mere warning treated an error).
* src/reader.c: Adjust.
* origin/maint: (46 commits)
doc: minor style change
maint: use gendocs's new -I option
regen
yacc.c: do not define location support when not using locations
maint: be compilable with GCC 4.0
tests: address a warning from GCC 4.4
tests: don't use options that Clang does not support
tests: restore the tests on -Werror
regen
parse-gram: update the Bison interface
fix comment
maint: post-release administrivia
version 2.6.4
regen
2.6.4: botched 2.6.3
maint: post-release administrivia
version 2.6.3
gnulib: update
tests: check %no-lines
NEWS: warnings with clang
...
Conflicts:
NEWS
TODO
data/c.m4
data/java.m4
doc/Makefile.am
src/getargs.c
src/getargs.h
src/output.c
src/parse-gram.c
src/parse-gram.h
src/parse-gram.y
src/reader.h
* origin/branch-2.6:
regen
yacc.c: do not define location support when not using locations
maint: be compilable with GCC 4.0
tests: address a warning from GCC 4.4
tests: don't use options that Clang does not support
tests: restore the tests on -Werror
regen
parse-gram: update the Bison interface
fix comment
The "shadows a global declaration" warning in GCC 4.0 was a bit
annoying. It does not like that a type name be used in a prototype of
a function (not the implementation, just the declaration):
In file included from src/LR0.c:38:
src/reader.h:56: warning: declaration of 'named_ref' shadows a
global declaration
src/named-ref.h:35: warning: shadowed declaration is here
It does not like either when a global variable name is used in a
prototype. Flex 2.5.37 generates this prototype:
void gram_set_debug (int debug_flag );
* src/getargs.h, src/getargs.c (debug_flag): Rename as...
(debug): this.
Adjust dependencies.
* src/reader.h: Don't use "named_ref" as a formal argument name.
236. torture.at:465: testing Exploding the Stack Size with Alloca ...
../../../tests/torture.at:474: bison -o input.c input.y
../../../tests/torture.at:474: $CC $CFLAGS $CPPFLAGS $LDFLAGS -o input input.c $LIBS
stderr:
cc1: warnings being treated as errors
input.y: In function 'main':
input.y:60: error: 'status' may be used uninitialized in this function
* tests/torture.at (AT_DATA_STACK_TORTURE): Initial status to avoid
the previous error.
When run as /bin/sh, Bash sets the shell variable POSIXLY_CORRECT to
y. The test suite checks for the envvar POSIXLY_CORRECT to turn of
some tests not supported in POSIX mode. Restore these tests.
Reported by the Hydra build farm, from Rob Vermaas.
* tests/local.at (AT_BISON_CHECK_WARNINGS_): Check the envvar
POSIXLY_CORRECT, not the shell variable.
* 2.6.3: (22 commits)
maint: post-release administrivia
version 2.6.3
gnulib: update
tests: check %no-lines
NEWS: warnings with clang
warnings: avoid warnings from clang
tests: no longer disable -O compiler options
yacc.c: initialize yylval in pure-parser mode
skeletons: style changes
tests: minor improvements
tests: use $PERL instead of perl
build: look for Perl in configure.
tests: fix sed portability issues
tests: diff -u is not portable
maint: word changes
lalr1.cc: fix test suite portability
maint: fix an erroneous include
tests: check that headers are self contained
doc: add missing documentation for --report
headers: move CPP guards into YY_*_INCLUDED to avoid collisions
...
Conflicts:
NEWS
doc/bison.texi
tests/headers.at
Note that 'make web-manual' fails.
* NEWS: Document these changes.
* doc/Makefile.am: Adjust to generate example files.
* doc/bison.texi: Add a Graphviz section after "Understanding::", the section
describing the .output file, because these are similar.
* doc/figs/example-reduce.dot, doc/figs/example-reduce.txt,
doc/figs/example-shift.dot, doc/figs/example-shift.txt: New, minimal
examples to illustrate the documentation.
Signed-off-by: Akim Demaille <[email protected]>
Use something similar to the report file.
* src/print_graph.c (print_lhs): New, obstack equivalent of rule_lhs_print.
(print_core): Use here.
Signed-off-by: Akim Demaille <[email protected]>
* src/graphviz.c (start_graph): Use courier font.
(conclude_red): Use commas to separate attributes. Show the acceptation
as a special reduction, with a blue color and an "Acc" label. Show the
lookahead tokens between square brackets.
(output_red): No longer label default reductions.
* src/print_graph.c (print_core): Refactor spacing, and print an
additional space between a rule's rhs and its lookahead tokens. Also,
capitalize "State".
(print_actions): Style, move a declaration.
Signed-off-by: Akim Demaille <[email protected]>
All disabled reductions should now be shown as such.
* src/graphviz.c (output_red): Here.
(conclude_red): New.
Signed-off-by: Akim Demaille <[email protected]>
See
http://lists.gnu.org/archive/html/bison-patches/2012-02/msg00045.html
Note that api.tokens.prefix has not been released, yet.
* NEWS, data/bison.m4, doc/bison.texi, tests/c++.at,
* tests/calc.at, tests/java.at, tests/local.at: Do it.
* src/muscle-tab.c (muscle_percent_variable_update): Ensure
backward compatibility.
Because argv[0] is never used, shift it out from the argument list.
* src/complain.c (complain_args): Here.
* src/scan-skel.l (at_complain): Adjust argv and argc.
Signed-off-by: Akim Demaille <[email protected]>
* origin/maint:
tests: check %no-lines
tests: minor simplification
graphs: stylistic changes.
graphs: minor style changes
graphs: show reductions
graphs: style: prefix state number with "state"
graphs: style: use left justification for states
graphs: style: prefix rules and change shapes
obstack: import obstack_finish0 from master
c++: api.location.type
muscles: a function for backward compatibility
maint: more macros
Conflicts:
data/glr.cc
data/java.m4
data/lalr1.cc
doc/bison.texi
src/muscle-tab.c
src/system.h
tests/calc.at
* src/graphviz.c (output_red): New, show reductions on the graph.
(no_reduce_bitset_init): New, initialize a bitset.
(print_token): New, print a lookahead token.
(escape): New, print "foo" as \"foo\" because Dot doesn't like quotes within
a label.
* src/graphviz.h : Adjust.
* src/print_graph.c (print_actions): Call output_red here.
Signed-off-by: Akim Demaille <[email protected]>
The label text of nodes is centered "by default" (by the use of '\n' as
a line feed). This gives bad readability to the grammar rules shown in
state nodes, a left justification is much nicer. This is done by using '\l'
as the line feed.
In order to allow \l in the DOT file, changes to the quoting system seem
necessary.
* src/print_graph.c (print_core): Escape tokens here, instead of...
* src/graphviz.c (output_node): Here...
(escape): Using this, new.
Signed-off-by: Akim Demaille <[email protected]>
* src/graphviz.c (start_graph): Use box rather than ellipsis.
* src/print_graph.c (print_core): Prefix rules with their number.
Signed-off-by: Akim Demaille <[email protected]>
This feature was introduced in 95a2de5695
(which is part of 2.5), but not documented. Give it a proper name, and
make it public.
* data/c++.m4, data/lalr1.cc, data/glr.cc, data/java.m4: Use
api.location.type instead of location_type.
* src/muscle-tab.c (muscle_percent_variable_update): Map the latter to
the former.
* tests/local.at: Adjust.
* tests/calc.at: Use api.location.type.
Leave tests/java.at with location_type, at least for the time being,
to cover both names.
* doc/bison.texi: Document api.location.type.
(User Defined Location Type): New.
* NEWS: Update.
Based on commit 171ad99d64 from master.
* src/muscle-tab.c (muscle_percent_variable_update): New.
(muscle_percent_define_insert): Use it.
Define the variables with their initial value.
Fix the following warning
parse-gram.c:2078:14: error: equality comparison with extraneous parentheses
[-Werror,-Wparentheses-equality]
if (((yyn) == (-91)))
~~~~~~^~~~~~~~
parse-gram.c:2078:14: note: remove extraneous parentheses around the
comparison to silence this warning
if (((yyn) == (-91)))
~ ^ ~
parse-gram.c:2078:14: note: use '=' to turn this equality comparison into
an assignment
if (((yyn) == (-91)))
^~
=
1 error generated.
and the following one:
input.cc:740:1: error: function declared 'noreturn' should not return
[-Werror,-Winvalid-noreturn]
static void yyMemoryExhausted (yyGLRStack* yystackp)
__attribute__ ((__noreturn__));
static void
yyMemoryExhausted (yyGLRStack* yystackp)
{
YYLONGJMP (yystackp->yyexception_buffer, 2);
}
^
1 warning and 1 error generated.
This is Apple clang version 3.1 (tags/Apple/clang-318.0.61).
* data/c.m4 (b4_table_value_equals): Use (!!(A == B)) instead of (A == B)
to avoid this warning.
Any reasonable compiler should generate the same code.
* src/uniqstr.h (UNIQSTR_EQ): Likewise.
* data/glr.c (LONGJMP): abort after longjmp to pacify clang.
Tests are running without -O since
f377f69fec because some warnings (about
yylval not being initialized) show only when GCC is given -O2. The
previous patch fixes the warnings. Run the test suite with compiler
options unmodified.
* tests/atlocal.in (O0CFLAGS, O0CXXFLAGS): Remove, use CFLAGS and
CXXFLAGS.
See http://lists.gnu.org/archive/html/bison-patches/2012-08/msg00024.html
(spreading over September and October).
* data/yacc.c (YY_IGNORE_MAYBE_UNINITIALIZED_BEGIN)
(YY_IGNORE_MAYBE_UNINITIALIZED_END, YYLVAL_INITIALIZE):
New macros. Use them to suppress an unwanted GCC diagnostic.
* tests/c++.at (Exception safety): Don't use swap here, it
is useless.
Cover more test cases: yyerror, YYERROR, YYABORT, and
error recovery.
(Object): Instead of just keeping a counter of instances, keep
a list of them.
* tests/c++.at (Exception safety): Let the parser support the --debug
option.
On 'p', throw an exception from the %printer.
* data/lalr1.cc (yyparse): Do not display the values we discard, as it
uses %printer, which might have thrown the exception.
* data/lalr1.cc: Check size > 1, rather than size != 1, when cleaning
the stack, as at the beginning, size is 0.
* tests/c++.at (Exception safety): Check exception safety in
%initial-action.
lalr1.cc does not reclaim its memory when ended by an exception.
Reported by Oleksii Taran:
http://lists.gnu.org/archive/html/help-bison/2012-09/msg00000.html
* data/lalr1.cc (yyparse): Protect the whole yyparse by a try-catch
block that cleans the stack and the lookahead.
* data/lalr1.cc (YY_SYMBOL_PRINT, YY_REDUCE_PRINT, YY_STACK_PRINT):
Define to something so that, for instance, "if (foo) YY_SYMBOL_PRINT"
is valid even when !YYDEBUG.
Bison uses "/usr/bin/perl" or "perl" in several places, and it does
not appear to be a problem. But, at least to make it simpler to
change PERL on the make command line, check for perl in configure.
* configure.ac (PERL): New.
* doc/Doxyfile.in, doc/Makefile.am, tests/bison.in: Use it.
* origin/maint:
maint: word changes
lalr1.cc: fix test suite portability
maint: fix an erroneous include
tests: check that headers are self contained
doc: add missing documentation for --report
Conflicts:
NEWS
data/location.cc
* src/scan-skel.l: Use a simpler and more consistent pattern escaping
scheme.
Catch all the invalid directives here by just removing the previous
catch-all-but-alphabetical rule.
* src/scan-skel.l (at_directive, at_init): New.
(at_ptr): New, function pointer used to call the right at_directive
function (at_basename, etc.).
(outname): Rename as...
(out_name): this, for consistency with out_lineno.
Signed-off-by: Akim Demaille <[email protected]>
* TODO: Address the issue, so remove it.
* data/bison.m4: Use b4_error with [[note]] rather than a complain_at
for context information.
* src/complain.c (complain_args): Take an additional argument, an
indentation pointer, to allow the dispatching of context information.
* src/complain.h (complain_args): Adjust prototype.
* src/scan-skel.l (at_directive_perform): Recognize the new @note mark.
* tests/input.at: Adjust.
Signed-off-by: Akim Demaille <[email protected]>
Instead of @complain, @warn, and @fatal, use a unique @complain
directive. This directive's first argument is "complain", "warn", etc.
* data/bison.m4 (m4_error): Here.
* src/scan-skel.l (at_directive_perform): Adjust.
(flag): Replace the switch by safer and more explicit if branches.
Signed-off-by: Akim Demaille <[email protected]>
Reported by Rob Vermaas' Hydra build farm on x86_64-darwin 10.2.0 with
G++ 4.6.3.
* tests/headers.at (Several parsers): Include AT_DATA_SOURCE_PROLOGUE
in the files to compile.
* data/location.cc: Do not include twice string and iostream (once
by position.hh, and then by location.hh).
* README-hacking (Typical errors): Some hints for other maintainers.
This fixes test 130 (Several parsers).
* data/location.cc: Include <iostream> rather than <iosfwd> since
we really need << on strings for instance.
* NEWS: Document this.
Signed-off-by: Akim Demaille <[email protected]>
Suggested by Jim Meyering.
* etc/prefix-gnulib-mk: Remove, as it is now provided by...
* bootstrap.conf (modules): the non-recursive-gnulib-prefix-hack module.
* build-aux/.gitignore, configure.ac, m4/.gitignore: Adjust.
This function is now a mere iterator that calls flag_argmatch,
a new function, that matches a single option parameter.
* src/getargs.c (flag_argmatch): New, taken from...
(flags_argmatch): Here.
Signed-off-by: Akim Demaille <[email protected]>
The code here was too confusing, this seems more natural.
* src/complain.c (error_message): Move the indentation check and the category
output to complains. Also, no longer take a 'warnings' argument.
(complains): Factor calls to error_message.
Signed-off-by: Akim Demaille <[email protected]>
Move the dispatch of variadic complains to complain.c, rather than do
it in a scanner.
* src/complain.h, src/complain.c (complain_args): New.
* src/scan-skel.l (at_directive_perform): Use it.
Signed-off-by: Akim Demaille <[email protected]>
Rectify a bug that introduced suffixes out of place.
* src/complainc.c (complains): Handle all three special warning bits.
* src/scan-code.l (show_sub_message): Remove useless argument.
* tests/named-refs.at: Adjust.
Signed-off-by: Akim Demaille <[email protected]>
This is the continuation of the work on the readability of errors
context.
* src/symtab.c (user_token_number_redeclaration): Use
complain_at_indent to output with increased indentation level.
* tests/input:at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
This line doesn't add any meaningful information anymore, the appended
[-Werror=CATEGORY] is enough. It is actually more insightful, as it
allows to distinguish warnings treated as errors from those that
aren't. This line is also removed by gcc 4.8.
* src/complain.c (set_warnings_issued): The only action left was
checking if the error bit corresponding to the warning issued was set,
and that function was only called once. Therefore, remove it, and do
its job directly in the caller...
(complains): here.
* src/complains.h: Adjust.
* tests/input.at: Adjust.
* NEWS: Document this change.
Signed-off-by: Akim Demaille <[email protected]>
The prefix of warnings treated as errors is now "error: ". Also, their
suffix now reflects the changes in the Werror option format.
An output for -Werror=other used to be:
bison: warnings being treated as errors
input.y:1.1: warning: stray ',' treated as white space [-Wother]
It is now:
bison: warnings being treated as errors
input.y:1.1: error: stray ',' treated as white space [-Werror=other]
The line "warnings being treated as errors" no longer adds any info,
it will be removed in a forthcoming change.
* NEWS: Add entry "Enhancement of the -Werror"
* doc/bison.texi: Move the warnings-as-error to a new bullet.
* src/complain.c (complains): Refactor, change the prefix of warnings
that are treated as errors.
(warnings_print_categories): Support for [-Werror=CATEGORY] display
* src/getargc.c (getargs): -y implies -Werror=yacc
* tests/input.at: Update expected --yacc output for coherence.
Signed-off-by: Akim Demaille <[email protected]>
This new option is a lot more flexible than the previous one. Its
details will be discussed in the NEWS and info file, in a forthcoming
change.
If no category is specified (ie: used as simply "-Werror"), the
functionality is the same as before.
* src/complain.c (errors_flag): New variable.
(set_warning_issued): Accept warning categories as an argument.
* src/complain.h (Wall): Better definition.
* src/getargs.c (flags_argmatch): Support for the new format.
(usage): Update -Werror to -Werror[=CATEGORY] format.
* src/complain.c (errors_flag): New variable.
(set_warning_issued): Accept warning categories as an argument.
* src/complain.h (Wall): Better definition.
* src/getargs.c (flags_argmatch): Support for the new format.
(usage): Update -Werror to -Werror=[CATEGORY] format.
Signed-off-by: Akim Demaille <[email protected]>
The deprecated warning, introduced some time ago, was not displayed in
the usage message. This patch addresses the issue.
* src/getargs.c (usage): Insert here.
Signed-off-by: Akim Demaille <[email protected]>
This improves readability. This is also what gcc does.
* NEWS: Document this change.
* src/complain.c (complain_at): Prefix all errors with "error: ".
(complain_at_indent, warn_at_indent): Do not prefix the context
information of errors, which are basically just indented errors.
* tests/conflicts.at, tests/glr-regression.at, tests/input.at,
tests/named-refs.at, tests/output.at, tests/push.at,
tests/regression.at, tests/skeletons.at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
This is the continuation of the work on the readability of errors
context.
For example, what used to be:
input.y:1.9-29: invalid value for %define variable 'foo' : 'bar'
input.y:1.9-29: accepted value: 'most'
is now:
input.y:1.9-29: invalid value for %define variable 'foo' : 'bar'
input.y:1.9-29: accepted value: 'most'
* src/muscle-tab.c (muscle_percent_define_check_values): Use
complain_at_indent to output with increased indentation level.
* tests/input:at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
This is the continuation of the work on the readability of errors
context.
For example, what used to be:
input.y:2.9-11: %define variable 'var' redefined
input.y:1.9-11: previous definition
is now:
input.y:2.9-11: %define variable 'var' redefined
input.y:1.9-11: previous definition
* src/muscle-tab.c (muscle_percent_define_insert): Use
complain_at_indent to output with increased indentation level.
* tests/input.at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
This is the continuation of the work on the readability of errors
context.
For example, what used to be:
input.y:5.10-24: %printer redeclaration for <field2>
input.y:3.11-25: previous declaration
is now:
input.y:5.10-24: %printer redeclaration for <field2>
input.y:3.11-25: previous declaration
* NEWS: Document this change.
* src/symtab.c (symbol_redeclaration, semantic_type_redeclaration,
user_token_number_redeclaration, default_tagged_destructor_set,
default_tagless_destructor_set, default_tagged_printer_set,
default_tagless_printer_set): Use complain_at_indent to
output with increased indentation level.
* tests/input.at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
This used to be the format of the error report:
input.y:6.5-10: result type clash on merge function 'merge': [...]
input.y:2.4-9: previous declaration
In order to distinguish the actual error from the context provided, we
rather this new output:
input.y:6.5-10: result type clash on merge function 'merge': [...]
input.y:2.4-9: previous declaration
Another patch will introduce an "error: " prefix to all non-indented
lines, giving yet better readability to the reports.
* src/complain.h (SUB_INDENT): Move to here.
* src/reader.c (record_merge_function_type): Use complain_at_indent to
output with increased indentation level.
* src/scan-code.l (SUB_INDENT): Remove from here.
* tests/glr-regression.at: Apply this change.
Signed-off-by: Akim Demaille <[email protected]>
The current routines used to display s/r and r/r conflicts are both
inconvenient from the programmer point of view (they do not use the
warning infrastructure) and for the user (the messages are rather
terse, not necessarily pleasant to read, and because they don't use
the same routines, they look different).
It was due to the belief (dating back to the initial checked-in
version of Bison) that, at some point, POSIX Yacc mandated the format
for these messages. Today, the Open Group's manual page for Yacc,
<http://pubs.opengroup.org/onlinepubs/009695399/utilities/yacc.html>,
explicitly states that the format of these messages is unspecified.
See commit be7280480c and
<http://lists.gnu.org/archive/html/bison-patches/2002-12/msg00027.html>.
For a discussion on the chosen warning format, see
http://lists.gnu.org/archive/html/bison-patches/2012-09/msg00039.html
In an effort to factor the handling of errors and warnings, use the
Bison warning routines to report these messages.
* src/conflicts.c (conflicts_print): Rewrite with clearer sections
about S/R and then R/R conflicts.
(conflict_report): Remove, inlined in its sole
caller...
(conflicts_output): here.
* tests/conflicts.at, tests/existing.at, tests/glr-regression.at,
* tests/reduce.at, tests/regression.at: Adjust the expected results.
* NEWS: Update.
* src/conflicts.c (conflicts_print): Complain about %expect-rr if not
in GLR mode, regardless of the number of reduce/reduce conflicts.
* tests/conflicts.at (%expect-rr non GLR): New test.
* NEWS: Update.
* src/complain.h : errors_flag variable
* src/complain.c : actual stuff happens here
* src/conflits.c : differentiated SR and RR conflicts
* src/getargs.c : flags_argmatch recognizes the new -Werror format
* origin/maint:
maint: remove useless file
update files to ignore
remove useless include
use locale-indep. c_is* functions for parsing, not isspace, isprint etc
gnulib: update
--help: include a place to report translation issues
tests: style changes
tests: fix push-pull test
yacc.c: style changes
Conflicts:
src/system.h
GCC seems to be using "deprecated" consistently over "obsoleted", so
use -Wdeprecated rather than -Wobsolete.
* src/complain.h (warnings): Add Wdeprecated.
* src/complain.c (warnings_print_categories): Adjust.
* src/getargs.c: Likewise.
* doc/bison.texi: Document it.
* src/scan-code.l: Use this category for the trailing ';' support.
* tests/actions.at: Adjust expected output.
* src/symtab.c (symbol_check_defined): Undeclared symbols are only
a warning.
* tests/input.at (Undeclared symbols used for a printer or destructor):
Rename as...
(Undefined symbols): this, and check this case.
* NEWS: Doc it.
* data/glr.cc: No longer require %defines.
When it is not given, define the position and location classes instead
of including their headers.
(b4_shared_declarations): Use the original parse-params.
* data/glr.c (b4_shared_declarations): Define only if undefined.
* tests/actions.at, tests/calc.at: No longer force the use of %defines
for glr.cc.
* NEWS: Doc it.
* src/scan-skel.l (at_directive_argc, at_directive_argv)
(AT_DIRECTIVE_ARGC_MAX): Rename as...
(argc, argv, ARGC_MAX): these, as there is no possible confusion.
(flags): New.
(QPUTS): Remove, inline its only use.
* src/system.h (obstack_finish0): New.
Use it to simplify several uses.
* src/muscle-tab.h (MUSCLE_INSERTF): New.
* src/muscle-tab.c: Use obstack_printf where simpler.
* src/scan-code.l (show_sub_message, show_sub_messages): Instead of a
Boolean, take a "warnings" argument.
Avoid storing printf-like format strings in a variable, so that GCC
can check them.
* origin/maint:
maint: post-release administrivia
version 2.6.2
NEWS: update.
yacc: remove trailing end of line at end of file
thanks: fix a contributor name
gnulib: update
tests: synch line -> syncline, for consistency
tests: synclines: style changes
tests: synclines: fix perl invocation
regen
c++: trailing end-of-lines in %parse-param
tests: simplify
Conflicts:
cfg.mk
data/glr.cc
data/yacc.c
src/parse-gram.c
src/parse-gram.h
src/parse-gram.y
Basically revert commit 12ce2df60d.
http://lists.gnu.org/archive/html/bison-patches/2012-08/msg00004.html
* data/c.m4, data/glr.c, data/yacc.c (YYID): Remove.
No longer use ARGSUSED.
* src/getargs.c: Restore simpler inclusion of getopt.h (anyway, since
then we now use gnulib which certainly protects us from such issues).
* data/c.m4 (b4_yydestruct_generate, b4_yy_symbol_print_generate):
They no longer need an argument, it has a single possible value.
* data/glr.c, data/yacc.c: Adjust.
* maint:
use obstack_printf
scanner: restore a missing start condition
gnulib: update
maint: post-release administrivia
version 2.6.1
gnulib: update
maint: fix some syntax-check issues
tests: do not depend on __cplusplus to decide for C++ or C output
Conflicts:
NEWS
bootstrap.conf
cfg.mk
lib/.gitignore
* tests/output.at (AT_CHECK_OUTPUT): It used to check that the
expected files are indeed generated, but it did not check that
there are no additional ones.
Do that, and adjust expectations (in particular alphabetical order).
* data/location.cc (b4_position_define, b4_location_define): New.
(location.hh, position.hh): Generate only if %defines.
* data/lalr1.cc: therefore, define these classes when locations are
needed, but headers are not generated.
* tests/output.at: Check that these files are not generated.
* NEWS: Document.
* data/glr.cc: Use b4_locations_if where appropriate.
* data/lalr1.cc: M4 quotation changes to highlight code duplication
with glr.cc.
* tests/calc.at: Check glr.cc with and without %location.
While at it, fuse multiple %parse-params into one.
* tests/actions.at: Simplify.
* NEWS: Doc this.
Some other wording changes.
* origin/maint: (29 commits)
regen
synclines: remove spurious empty line
also support $<foo>$ in the %initial-action
skeletons: b4_dollar_pushdef and popdef to simpify complex definitions
regen
printer/destructor: translate only once
factor the handling of m4 escaping
news: schedule the removal of the ";" hack
style changes in the scanners
regen
support $<tag>$ in printers and destructors
scan-code: factor the handling of the type in $<TYPE>$
muscles: fix another occurrence of unescaped type name
glr.cc: fix the handling of yydebug
gnulib: update
formatting changes
tests: fix an assertion
tests: adjust to GCC 4.8, which displays caret errors
be sure to properly escape type names
obstack_quote: escape and quote for M4
muscles: shuffle responsabilities
muscles: make private functions static
muscles: rename private functions/macros
obstack_escape: escape M4 characters
remove dead macro
maint: style changes
doc: avoid problems with case insensitive file systems
configure: fix botched quoting
news: fix typo.
Conflicts:
NEWS
data/c.m4
data/glr.cc
data/lalr1.cc
examples/rpcalc/local.mk
src/muscle-tab.h
src/output.c
src/parse-gram.c
src/parse-gram.h
src/parse-gram.y
src/scan-code.l
src/symlist.c
src/symlist.h
src/symtab.h
tests/calc.at
Currently they are treated in separated variables, contrary to other
<TYPE> code_props. This duplicates code (and messages for translators)
uselessly, as demonstrated by the fact that thanks to this patch, now
useless <*> and <> code_props are reported like the others.
* src/parse-gram.y (generic_symlist_item): Treat "<*>" and "<>" as regular
type tags.
* src/symlist.h, src/symlist.c (symbol_list_default_tagged_new)
(symbol_list_default_tagless_new,SYMLIST_DEFAULT_TAGGED)
(SYMLIST_DEFAULT_TAGLESS): Remove.
* src/symtab.h, src/symtab.c (default_tagged_code_props)
(default_tagless_code_props, default_tagged_code_props_set)
(default_tagless_code_props_set): Remove.
(symbol_code_props_get): Default to <*> or <>'s code_props.
* tests/actions.at: Complete expected errors: there are new warnings.
* tests/input.at: Likewise.
(Useless printers or destructors): Extend.
The logic to compute the %printer or %destructor to used (i.e., a
code_props) is implemented twice: one, of course, in
symbol_code_props_get, and another time in symbol_check_defined to
record the fact that a code_props is used (so that we can reported
unused ones). Let the former use the latter.
I would probably use "mutable" in C++ and keep these guys const,
but this is C. And casting away constness triggers warnings.
* src/scan-code.h, src/scan-code.l (code_props_none): Is not const.
* src/symtab.h, src/symtab.c (symbol_code_props_get): The symbol
is not const.
(symbol_check_defined): Use it.
Because I'm using a VPATH build with an absolute srcdir, I have
GRAM__USERS_AKIM_SRC_GNU_BISON_SRC_PARSE_GRAM_H. Before, I was using
a relative srcdir, and had GRAM_______SRC_PARSE_GRAM_H (coming from
../../). Let it be GRAM_SRC_PARSE_GRAM_H.
* tests/bison.in: Do not depend on the value of $top_srcdir for
Bison itself.
If we were to use relative paths from .c to .y, we would not have
this problem.
* data/glr.c, data/glr.cc, data/lalr1.cc, data/lalr1.java, data/yacc.c:
Since the previous commit, b4_tokens_define and the like no longer need
b4_tokens.
* src/output.c (token_definitions_output): Remove.
There are currently two systems used to pass information about tokens to
m4: the original one, and another, which is used for instance for
printers and destructors, variants etc. Move to using only the latter.
* data/bison.m4 (b4_symbol_map, b4_token_visible_if)
(b4_token_has_definition, b4_any_token_visible_if, b4_token_format): New.
* data/c++.m4, data/c.m4, data/glr.c, data/java.m4: Adjust to use them.
* origin/maint:
maint: update gnu-web-doc-update.
maint: post-release administrivia
version 2.6
maint: prepare for release 2.6
maint: post-release administrivia
version 2.5.91
maint: prepare NEWS.
maint: fix spaces.
tests: adjust to case where the C compiler is actually a C++ compiler
tests: fix dependencies
doc: fix Texinfo command
maint: Valgrind on OS X.
tests: be sure that backups are safe.
maint: dead comment.
tests: refactor for legibility.
tests: refactor the bison invocations.
maint: fix syntax-check ignore patterns.
gnulib: update
gnulib: update.
gnulib: update
Conflicts:
build-aux/Makefile.am
cfg.mk
tests/Makefile.am
list.yy: In function 'yy::parser::symbol_type yylex()':
list.yy:107:29: error: typedef 'token' locally defined but not used
[-Werror=unused-local-typedefs]
typedef yy::parser::token token;
^
* tests/c++.at (AT_CHECK_VARIANTS): here.
* origin/maint:
tests: headers.at: strengthen.
glr.cc: do not override C++ definitions by C macros.
YYLLOC_DEFAULT: factor, and don't export it in headers.
api.prefix: do not use #define to handle YYSTYPE_IS_TRIVIAL etc.
tests: portability fixes.
c++: fewer #includes in the headers.
glr.cc: formatting changes.
tests: more logs.
api.prefix: also rename YYDEBUG.
Conflicts:
data/c.m4
data/glr.c
data/glr.cc
data/lalr1.cc
data/yacc.c
A function to print warnings categories, like -Wyacc, -Wother, etc.
* src/complain.h, src/complain.c (print_warning_categories): New function.
* src/output.c (ARRAY_CARDINALITY): Move it to file 'src/system.h'.
* src/complain.h (enum warnings): New value, 'silent', "complain"
must not display the warning type.
Forthcoming changes will use the warning categories much more often,
so shortening them will improve readability.
* src/complain.c, src/complain.h, src/conflicts.c,
* src/getargs.c, src/getargs.h, src/gram.c (enum warnings):
s/warnings_/W/g.
* origin/maint:
tests: use the generalized default yylex.
tests: AT_YYERROR_DEFINE: prepare for list of ints.
skeletons: no longer define YYLSP_NEEDED.
c++: do not export YYTOKEN_TABLE and YYERROR_VERBOSE.
Conflicts:
data/c.m4
data/glr.cc
data/lalr1.cc
doc/bison.texi
tests/regression.at
* origin/maint:
tests: do not output m4 set up.
tests: use the generic yyerror function.
tests: use assert instead of plain abort.
tests: improve the generic yylex implementation.
tests: generalize the compilation macros.
tests: fix confusion between api.prefix and name-prefix.
maint: gitignores.
yacc: work around the ylwrap limitation.
Conflicts:
NEWS
tests/local.at
* src/scan-code.h (code_props): Add field 'is_used'.
(CODE_PROPS_NONE_INIT): Adjust.
* src/scan-code.l (code_props_plain_init, code_props_symbol_action_init)
(code_props_rule_action_init): Instead of implementing several
times the initialization of the code_props structures,
use code_props_none_init.
* src/symtab.c (symbol_check_defined): If a symbol does not have a
destructor (resp. printer) but has a type which has a destructor (resp.
printer), then set field 'is_used' to true.
(semantic_type_check_defined): If a type has a destructor (resp. printer)
but all symbols of this type have already a destructor (resp. printer),
then raise a warning.
* tests/input.at (Useless printers or destructors): New.
Signed-off-by: Akim Demaille <[email protected]>
* src/symtab.h (symbol_list): Represent semantic types as structure
'semantic_type'.
* src/symlist.c (symbol_list_type_new): Allocate this structure.
(symbol_list_code_props_set): Set this semantic type's status to used if it
was not declared.
* src/symtab.c (semantic_types_sorted): New.
(semantic_type_new): Set the new semantic type's location appropriately.
(symbol_check_defined): If a symbol has a type, then set this type's status
to "declared".
(semantic_type_check_defined, semantic_type_check_defined_processor): Same
as symbol_check_defined and symbol_check_defined_processor, but for semantic
types.
(symbol_check_defined): Check semantic types usefulness.
* src/symtab.h (semantic_type): New fields 'location' and 'status'.
* src/symtab.h, src/symtab.c (semantic_type_new)
(semantic_type_from_uniqstr, semantic_type_get): Accept a location as a
supplementary argument.
* tests/input.at (Unassociated types used for printer of destructor): New.
* tests/c++.at (AT_CHECK_VARIANTS): Fix an error caught by this commit.
The previous "code_props: factor more" patch sends has_%printer
etc. to m4, instead of has_printer.
* src/output.c (prepare_symbol_definitions): Fix value of pname.
* origin/maint:
tests: more uniformity.
tests: handle locations in a more generic way.
tests: handle locations in the generic yyerror functions.
tests: fix AT_CHECK_CALC.
tests: improve infrastructure
tests: factor.
skeletons: minor style changes
tests: AT_LANG.
c skeletons: factor the declaration of yylloc and yylval.
news: condemn YYPARSE_PARAM and YYLEX_PARAM.
maint: regen.
Conflicts:
tests/calc.at
tests/local.at
tests/regression.at
* src/symtab.h, src/symtab.c (code_props_type_string): No longer static.
* src/output.c (CODE_PROPS): Remove, we can now iterate on both the
destructor and the printer.
(SET_KEY2): New.
There is too much code duplication between %printer and %destructor.
We used to have two functions for each action: the first one for
destructors, the second one for printers. Factor using a
'code_props_type', and an array of code_props instead of two
members.
* src/symlist.h, src/symlist.c (symbol_list_destructor_set)
(symbol_list_printer_set): Fuse into...
(symbol_list_code_props_set): this.
* src/symtab.h, src/symtab.c (default_tagged_destructor)
(default_tagged_printer): Fuse into...
(default_tagged_code_props): this.
(default_tagless_destructor, default_tagless_printer)
(default_tagless_code_props): Likewise.
(code_props_type_string): new.
(symbol_destructor_set, symbol_destructor_get, semantic_type_destructor_set)
(default_tagged_destructor_set, default_tagless_destructor_set)
(symbol_printer_set, symbol_printer_get, semantic_type_printer_set)
(default_tagged_printer_set, default_tagless_printer_set): Replace by...
(symbol_code_props_set, symbol_code_props_get, semantic_type_code_props_set)
(default_tagged_code_props_set, default_tagless_code_props_set): these.
* src/parse-gram.y (grammar_declaration): Adjust.
* src/output.c (CODE_PROP, grammar_declaration): Ditto.
* src/reader.c (symbol_should_be_used): Ditto.
The previous commit, which turns into a warning what used to be an
error:
%printer {} foo;
%%
exp: '0';
has two shortcomings: the warning is way too long (foo is reported
to be useless later), and besides, it also turns into a warning much
more serious errors:
%printer {} foo;
%%
exp: foo;
Reduce the amount to warnings in the first case, restore the error in
the second.
* src/symtab.h (status): Add a new inital state: undeclared.
* src/symtab.c (symbol_new): Initialize to undeclared.
(symbol_class_set): Simplify the logic of the code that neutralize
the "redeclared" warning after the "redefined" one.
(symbol_check_defined): "undeclared" is also an error.
* src/reader.c (grammar_current_rule_symbol_append): Symbols appearing
in a rule are "needed".
* src/symlist.c (symbol_list_destructor_set, symbol_list_printer_set):
An unknown symbol appearing in a %printer/%destructor is "used".
* src/reduce.c (nonterminals_reduce): Do not report as "useless" symbols
that are not used (e.g., those that for instance appeared only in a
%printer).
* tests/input.at (Undeclared symbols used for a printer or destructor):
Improve the cover the cases described above.
We used to raise an error if a symbol appears only in a %printer or
%destructor. Make it a warning.
* src/symtab.h (status): New enum.
(symbol): Replace the binary "declared" with the three-state "status".
Adjust dependencies.
* src/symtab.c (symbol_check_defined): Needed symbols are an error,
whereas "used" are simply warnings.
* src/symlist.c (symbol_list_destructor_set, symbol_list_printer): Set
symbol status to 'used' when associated to destructors or printers.
* input.at (Undeclared symbols used for a printer or destructor): New.
* origin/maint:
maint: formatting changes.
tests: support api.prefix.
tests: pacify font-lock-mode.
tests: remove test covered elsewhere.
tests: factor the declaration/definition of yyerror and yylex.
regen.
tests: portability issues.
tests: call the parser from another compilation unit.
glr.c, yacc.c: declare yydebug in the header.
skeletons: use header guards.
tests: improve AT_FULL_COMPILE.
tests: reorder.
tests: strengthen the test on generated headers inclusion
yacc.c: instead of duplicating y.tab.h inside y.tac.c, include it.
yacc.c: factor.
Conflicts:
NEWS
data/glr.c
data/yacc.c
src/parse-gram.c
src/parse-gram.h
tests/conflicts.at
tests/regression.at
Basically, revert ba60c39547a445dee3e07920931b4d7a81843868's move to
mbs* functions, which was prompted by -DGNULIB_POSIXCHECK. See
<http://lists.gnu.org/archive/html/bison-patches/2012-05/msg00052.html>
and following.
* bootstrap.conf: No longer ask for them.
* src/files.c, src/getargs.c, src/location.c,
* src/parse-gram.c, src/parse-gram.y, src/scan-gram.l,
* src/symtab.c: s/mbs(r?chr)/str$1/g.
* bootstrap.conf (gnulib_modules): Request it.
* src/output.h, src/output.c (compute_pkgdatadir): Rename as...
(pkgdatadir): this.
Adjust dependencies.
* src/output.c (output_skeleton): Reduce the scope of "in".
Use xconcatenated_filename to simplify the construction of the
qualified paths to m4sugar.m4, bison.m4, and the selected skeleton.
There are a few minor differences: the new code uses strchr instead of
mbschr (but this was not really justified), and the new code does not
garantee a single slash even if $BISON_PKGDATADIR ends with several
(which was considered more accurate). See the discussion at
<http://lists.gnu.org/archive/html/bison-patches/2012-05/msg00052.html>.
lalr1.cc used to support yyo, but not yyoutput. Support both,
but document only yyoutput (at least until there is some consensus
on this).
* data/c.m4 (yy_symbol_value_print): Also support yyo.
* data/glr.cc (yy_symbol_value_print_): Support both yyo and yyoutput.
* data/lalr1.cc: Also support yyoutput.
* doc/bison.texinfo: Explicitly use yyoutput in the examples.
* examples/mfcalc/mfcalc.test: Test the -p option.
* doc/bison.texinfo (Printer Decl): New.
Number mfcalc.y snippets so that they are output in
the proper order.
(The mfcalc Main): Use yydebug.
(Debugging): Simplify the text.
(Enabling Traces, Mfcalc Traces, The YYPRINT Macro): New.
(Table of Symbols): Document YYPRINT and YYFPRINTF.
* tests/input.at: Use "print" in %printer instead of "destroy".
It is unused, so we don't care, yet it is less surprising.
* tests/actions.at: Comment changes.
Bison uses "/usr/bin/perl" or "perl" in several places, and it does
not appear to be a problem. But, at least to make it simpler to
change PERL on the make command line, check for perl in configure.
* configure.ac (PERL): New.
* doc/Doxyfile.in, doc/local.mk, examples/local.mk,
* tests/bison.in: Use it.
For some reason it seems that texi2dvi -o no longer forces --clean
mode, so we have stray TeX compilation files staying in top_builddir
since TeX is run from there.
While at it, upgrade the generation of the (completely obsolete)
reference card. Target PDF.
* doc/local.mk (TEXI2DVI): Pass --build-dir.
(CLEANDIRS): More accurate.
(doc/refcard.dvi): Replace with...
(doc/refcard.pdf): this.
Adjust dependencies.
Caught by maintainer-check-g++.
* data/glr.c, data/lalr1.cc, data/yacc.c, tests/cxx-type.at,
* tests/glr-regression.at, tests/push.at:
When simple to do, avoid expliciting the null ptr.
Otherwise use YY_NULL.
* origin/maint:
bump to 2012 in skeletons.
build: remove ancient Autoconf tests.
doc: c++: complete the location documentation.
c++: locations: provide convenience constructors.
c++: locations: remove useless "inline".
glr: do not use locations when they are not requested
c++: use nullptr for C++11.
build: simplify and improve the compiler warnings for tests.
gnulib: update.
maint: formatting changes.
NEWS: update.
Java: Fix syntax error handling without error token.
tests: beware of -pedantic on large #line numbers.
tests: when using the C++ compiler, use its flags too.
Conflicts:
data/glr.c
data/glr.cc
data/lalr1.cc
data/location.cc
data/yacc.c
tests/Makefile.am
* tests/local.mk: Prefer passing variable assignment by
the command line, instead of the environment, so that it
is reported in the logs.
Prefer single quotes for shell literal strings.
* tests/local.at: Go for colors.
(--compile-c-with-cxx): New option.
We used to pass "CC=$CXX" as command line argument,
but it was not possible to adjust CFLAGS accordingly
in atlocal, since it is loaded before assignments on
the command line are honored (so that the command line
takes precedence).
* tests/atlocal.in: Implement it.
* tests/local.mk: Use it.
* tests/local.mk (check_SCRIPTS): Add atlocal and atconfig so
that they are properly updated before running tests.
(RUN_TESTSUITE_deps): New.
Use it to factor the dependencies of "*-check" targets,
especially those that don't bounce to the regular
"check-local" target, since then they don't benefit from the
proper dependencies (such as atlocal).
* doc/bison.texinfo (Language and Grammar): Use the same
layout for an example in all the versions, i.e., keep
as general case what used to be used only for Info.
* doc/bison.texinfo: Avoid line width issues with TeX.
Upgrade ancient messages.
Move some comments to better looking places.
Add more @group.
(Mfcalc Symbol Table): Reduce variable scopes.
Prefer size_t for sizes.
Prefer declarations with an initial value.
Fix a @group environment.
* data/c.m4: Use m4_ifvaln instead of m4_ifval where
applicable.
(b4_parse_param_use): Switch order between two nested
"if"s to avoid useless empty lines.
Adjust callers to avoid useless lines.
YYPRINT uses yytoknum which glr does not define. Since YYPRINT
is considered obsolete, and did not work, don't fix its support,
remove it from glr.c.
* data/c.m4 (yy_symbol_value_print): Use YYPRINT only for yacc.c.
* TODO: Done.
Before this change, bison would accept either .tab and _tab equivalently,
whatever the current platform. Besides, it was not obeying everywhere
to the possible definition of TAB_EXT to something else than .tab.
For consistency, handle only TAB_EXT (".tab" on non DJGPP platforms).
Support for "_tab" is neither documented, nor tested.
* src/system.h (STRNCMP_LIT): New.
From Jim Meyering.
(STRPREFIX_LIT): New.
* src/files.c, src/getargs.c: Use it.
* Makefile.am (dist_TESTS): New.
(TESTS, EXTRA_DIST): Run and ship them.
* examples/calc++/local.mk: examples/calc++/calc++.stamp no longer
exists, don't try to ship it.
(.yy.stamp): New recipe.
Use it.
* examples/calc++/local.mk, examples/mfcalc/local.mk,
* examples/rpcalc/local.mk:
Don't ship the sources.
Adjust the CPPFLAGS: there is nothing left in srcdir.
(MAINTAINERCLEANFILES): Remove, now we are in builddir.
(TESTS): Rename as...
(dist_TESTS): this.
calc++: don't rely on Automake to compile a C++ parser.
Basically, revert commit 609b3d8096,
Automake 1.11.3 is not safe enough for C++ parser.
* examples/calc++/calc++-parser.hh: Remove.
* examples/calc++/local.mk (examples/calc++/calc++-parser.stamp):
New.
examples: factor the extractions into a single step
extexi had to be run in the extraction directory. Now, it can be
given the files with expected output directory. This allows to
use $(*_extracted) variables (before we had to list again their
members, but limited to their base names). In turn, this also
allows fusing the extraction recipes into a single one.
Also, it is currently too hard (or requires too much duplication,
since Make wants all the occurrences of the files to be prefixed with
$(srcdir)/, which is something Automake cannot support for *_SOURCES)
to work in the source tree. So extract, and compile scanners and parsers
in the build tree.
* examples/extexi (basename): New.
(BEGIN): Now "file_wanted" maps base name to extracted file name.
* examples/calc++/local.mk, examples/mfcalc/local.mk,
* examples/rpcalc/local.mk: Fuse extraction rules into...
* examples/local.mk: Here.
(extract, extracted): New.
* doc/bison.texinfo: Space change.
* src/system.h (STREQ, STRNEQ): New.
* src/files.c, src/ielr.c, src/lalr.c, src/muscle-tab.c,
* src/output.c, src/print.c, src/print_graph.c,
* src/reader.c, src/scan-skel.l, src/tables.c,
* src/uniqstr.c:
Use them.
* src/scan-gram.l: Do not use streq.h, use system.h's STREQ.
* cfg.mk: The documentation is an exception.
* doc/bison.texinfo: Do not use @verbatim, in particular when
we use @group inside.
Use @quotation instead of @display for frequently asked questions,
it looks much nicer.
Exit status 63 is documented for version-mismatch.
* bootstrap.conf (gnulib_modules): Remove sysexits.
* src/system.h (EX_MISMATCH): Define.
* src/parse-gram.y (version_check): Use it instead of EX_CONFIG.
Missing includes.
* tests/calc.at, tests/named-refs.at: Include assert.h.
* cfg.mk: No longer skip it, except where EXIT_SUCCESS is used
as a witness for stdlib.h.
Skip this test when appropriate.
* data/yacc.c: Drop a note about why EXIT_SUCCESS is defined here.
* bootstrap.conf (gnulib_modules): Require progname.
* src/complain.c, src/getargs.c, src/getargs.h, src/main.c: Use it.
* cfg.mk (exclude): New.
Use it.
Skip lib/main.c for bindtextdomain and set_program_name.
* cfg.mk (local-checks-to-skip): No longer skip it.
* bootstrap.conf (gnulib_modules): Add sysexits.
* doc/bison.texinfo, etc/bench.pl.in, src/parse-gram.y,
* src/system.h, tests/calc.at, tests/named-refs.at: Use assert
where appropriate instead of "if (...) exit".
Use symbolic exit status elsewhere.
* examples/mfcalc/test, examples/calc++/test: Extract the
common bits into...
* examples/test: here.
(cwd): New.
Use it to avoid a race on the temporary directory.
Reported by Jim Meyering.
* examples/mfcalc/test, examples/calc++/test: Rename into...
* examples/mfcalc/mfcalc.test, examples/calc++/calc++.test: these.
* examples/calc++/local.mk, examples/mfcalc/local.mk,
* examples/local.mk: Adjust.
* examples/calc++/test, examples/mfcalc/test (me): Be more
meaningfull: include the example name.
(prog): Factor.
(run): Avoid printf, use echo.
Add missing parens.
(cleanup): New.
Call it on trap.
Remove the previous "rm" that did the cleanup.
Move into a private directory to avoid concurrency issues.
Reported by Jim Meyering.
* Makefile.am (TESTS, check_PROGRAMS): Initialize here.
* examples/local.mk (doc, extexi): Define here.
* examples/calc++/local.mk: Adjust accordingly.
* configure.ac: Ask for parallel-tests (for the way the logs
are handled).
* examples/calc++/test: As a consequence, always be verbose.
($prog): New.
(run): Use it.
Sort the tests in a more natural order (simplest first).
* doc/bison.texinfo (Multi-function Calc): Add missing includes.
Fix the rendering of the result: use @result and remove the
initial tabulation in the actual code.
Fix stylistic issues: avoid the , operator.
Add extexi mark-up.
* examples/extexi: Also support @smallexample.
* tests/c++.at (Syntax error as exception): Avoid spurious failure
at least when compiling with g++-4.7.x due to lack of declaration
of getenv. Include <stdlib.h>.
Rely on $(YACC) being the bison being built, and let Automake do the
rest. It turned out to be much more difficult than anticipated, for
various reasons, including some bad behavior from Automake 1.11.2
which (i) generates calc++-parser.h instead of calc++-parser.hh, and
(ii) leaves an #include "y.tab.h" in the generated parser instead
of #include "calc++-parser.h".
The authors of Automake appear to be aware of the problem,
http://lists.gnu.org/archive/html/automake/2011-05/msg00008.html
so a simple work around will suffice for the time being.
* examples/calc++/y.tab.h, examples/calc++/calc++-parser.hh: New.
To work around Automake 1.11.2 issues.
* examples/calc++/local.mk: Remove all the rules for compilation
with bison, leave them to Automake.
So provide it with "calc++-parse.yy" as a source file.
(calc_sources_generated, calc_sources_extracted): Rename as.
(calc_generated, calc_extracted): these.
(calc_sources): New.
Fix them.
* src/local.mk: Put yacc related variables together.
(AUTOMAKE_OPTIONS): Move to...
* Makefile.am: here.
Remove an old Emacs mode request which disables Automake support.
* src/local.mk (YACC, AM_YFLAGS): Move to...
* Makefile.am: here, as they will be used by other local.mks.
The directory was still using a local Makefile.am because it provides
"scoped" Make variables: these examples are not meant to use the same
CPPFLAGS etc. If we were to use the same -I set, we'd pick up
gnulib's stdio.h for instance, which we do not want for these simple
examples.
Yet, as a result, the dependencies are less accurate, there is code
duplication, etc. This is especially perceptible when trying to
extract more examples from the documentation, as will be done in
forthcoming changes.
In order to make the tuning of CPPFLAGS easier, discard the predefined
-I from Automake.
* examples/calc++/Makefile.am: Rename as...
* examples/calc++/local.mk: this.
Adjust the paths which are now rooted in top_srcdir/top_builddir.
Handle BISON_CXX_WORKS here, instead of the too crude previous
approach that completely discarded the whole directory.
($(BISON)): Remove now useless bouncing recipe.
(calc___CPPFLAGS): New.
Stay away from -Ilib.
* Makefile.am, configure.ac, examples/local.mk,
* examples/calc++/test: Adjust.
* configure.ac: Pass nostdinc to Automake.
* src/local.mk, lib/local.mk (AM_CPPFLAGS): Move to...
* Makefile.am: here.
* src/local.mk, examples/calc++/Makefile.am (BISON, BISON_IN): Factor
to...
* Makefile.am: here.
* tests/local.mk: Use it.
* data/lalr1.cc (parse): Catch syntax_error around yylex and
forward them to errlab1.
* tests/c++.at (Syntax error as exception): Check support for
syntax exceptions raised by the scanner.
* NEWS, doc/bison.texinfo: Document it.
* tests/local.mk (TESTSUITE_AT): Include plackage.m4.
Adjust dependencies.
Make testsuite.at its first argument.
(package.m4): Be silent.
(testsuite): Be silent.
Use $<.
* data/c.m4 (b4_parse_param_use): Also accept optional arguments
to "use".
Simplify callers.
* data/glr.c (yyuserAction): Simplify use of b4_parse_param_use.
(yy_reduce_print): Don't use b4_parse_param_use, as all the arguments
_are_ used.
* data/lalr1.cc (YY_SYMBOL_PRINT): Even when disabled, "use" the
symbol argument.
This neutralizes a warning in yypush_ when there are no symbols
with a semantic values.
(yy_destroy_): Remove useless "use" of yymsg.
See <http://lists.gnu.org/archive/html/bug-bison/2012-01/msg00120.html>.
Use quotearg as often as possible instead of leaving the choice of
the quotes to the translators. Use shorter messages. Factor similar
messages to a single format, to make localization easier.
* src/files.c, src/getargs.c, src/muscle-tab.c, src/reader.c
* src/scan-code.l, src/scan-gram.l, src/symtab.c:
Use quote() or quotearg_colon() on printf arguments instead of
quotes in the format string.
* data/bison.m4: Keep sync with the changes in muscle-tab.c.
* tests/skeletons.at, tests/input.at, tests/regression.at: Adjust
expected messages.
Standard output was never explicitly closed, so we could not
detect failure. Thus, bison would ignore the errors of writing
to a full file system and getting an I/O error on write, but only
for standard output, e.g., for --print-localedir, --print-datadir,
--help and some verbose output.
Now, "bison --print-datadir > /dev/full" reports the write failure:
bison: write error: No space left on device
Before, it would exit 0 with no diagnostic, implying success.
This is not an issue for "--output=-" or the other FILE-accepting
command-line options, because unlike most other GNU programs,
an output file argument of "-" is treated as the literal "./-",
rather than standard output.
* bootstrap.conf (gnulib_modules): Add closeout.
* src/main.c: Include "closeout.h".
Use atexit to ensure we close stdout.
* .gitignore: Ignore new files pulled in via gnulib-tool.
* lib/yyerror.c (yyerror): Use fputs and fputc rather than fprintf
with a mere "%s\n" format. Always return 0 now, on the assumption
that the return value was never used anyway.
Don't include <config.h> after all. This avoids a problem
reported by Thiru Ramakrishnan in
http://lists.gnu.org/archive/html/help-bison/2011-11/msg00000.html
* cfg.mk: Exempt lib/yyerror.c from the sc_require_config_h_first test.
* THANKS: Update.
This script hard-coded "libbison" and lib/gnulib.mk.
Adjust the script to require a --lib-name=$gnulib_name option
and a FILE argument like lib/$gnulib_mk.
Also add support for --help and --version.
* etc/prefix-gnulib-mk: Generalize.
* bootstrap.conf (bootstrap_post_import_hook): Update its invocation.
Updating to gnulib pulled in new quote and quotarg modules,
by which quoting is now done like 'this' rather than `this'.
That change induces many "make check" test failures. This change
adapts code and tests so that "make check" passes once again.
* src/scan-code.l: Quote like 'this', not like `this'.
* src/scan-gram.l: Likewise.
* src/symtab.c: Likewise.
* tests/actions.at: Adjust tests to match.
* tests/input.at: Likewise.
* tests/named-refs.at: Likewise.
* tests/output.at: Likewise.
* tests/regression.at: Likewise.
* lib/.gitignore: Regenerate.
* m4/.gitignore: Likewise.
Updating to the latest bootstrap from gnulib involves more of a
change than usual, and updating to the latest gnulib would involve
its own set of challenges with the upcoming quoting changes, so
we update bootstrap manually and separately.
* bootstrap: Update from gnulib.
* Makefile.am: Initialize more variables to empty, so that gnulib.mk
can append to them with "+=".
* bootstrap.conf (gnulib_mk_hook): Remove. No longer honored.
(bootstrap_post_import_hook): Instead, run the same command,
etc/prefix-gnulib-mk lib/$gnulib_mk, via slightly different API.
Temporarily disable "bootstrap_sync=true".
* etc/prefix-gnulib-mk: Don't prepend "lib/" to tokens like -I$(...
or "\".
* cfg.mk (exclude_file_name_regexp--sc_space_tab): Exempt
tests/input.at and tests/c++.at, since they appear to use
SP-TAB sequences deliberately.
* OChangeLog: Remove space-before-TAB.
* bootstrap.conf (gnulib_modules): Add gpl-3.0.
* doc/gpl-3.0.texi: Remove from version control, now that
we get it via gnulib.
* doc/.gitignore: Ignore it.
* lib/lbitset.c (lbitset_elt_find): Remove unnecessary test of "elt",
at a point where we know it is non-NULL, due to prior dereference.
Copyright (C) 1987-1988, 1991-2012 Free Software Foundation,
* lib/bitset_stats.c (bitset_stats_list): Remove declaration
and set of unused local, "type".
* lib/ebitset.c (ebitset_init): Likewise, for "size".
* lib/lbitset.c (lbitset_not): Likewise, for "elt".
This should help to avoid some portability problems. For example,
it would have revealed the empty unions fixed by the last patch
* configure.ac (WARN_CFLAGS_TESTS, WARN_CXXFLAGS_TEST): Implement.
* tests/synclines.at (AT_TEST_SYNCLINE): Avoid -pedantic warning
about an empty translation unit.
(cherry picked from commit 5422d56a71)
Empty unions are not accepted by the ISO C99 grammar or by at
least some versions of Sun Studio. Reported by Wolfgang S. Kechel
at
<http://lists.gnu.org/archive/html/bug-bison/2011-08/msg00003.html>.
* NEWS (2.5.1): Document fix.
* THANKS (Wolfgang S. Kechel): Add.
* tests/existing.at (GNU Cim Grammar)
(GNU pic (Groff 1.18.1) Grammar.): Remove empty %union.
(cherry picked from commit 7451638148)
This change was made by applying emacs' untabify function to
nearly all files in Bison's repository. Required tabs in make
files, ChangeLog, regexps, and test code were manually skipped.
Other notable exceptions and changes are listed below.
* bootstrap: Skip because we sync this with gnulib.
* data/m4sugar/foreach.m4
* data/m4sugar/m4sugar.m4: Skip because we sync these with
Autoconf.
* djgpp: Skip because I don't know how to test djgpp properly, and
this code appears to be unmaintained anyway.
* README-hacking (Hacking): Specify that tabs should be avoided
where not required.
The typo shows up in the `configure --help' output. Reported by
W.C.A. Wijngaards and Paul Eggert at
<http://lists.gnu.org/archive/html/bug-bison/2011-07/msg00000.html>.
* README-hacking (Release Procedure): Remind about updating
maintainer tools before a release.
* THANKS (W.C.A. Wijngaards): Add.
* configure.ac (AC_PREREQ): Set to 2.68, which fixes the YACC
typo.
(cherry picked from commit 845346b3bb)
* doc/bison.texinfo (Locations Overview): Rename node to...
(Locations): ... this, which is its section name. Update menus to
match that and not to use the title `Tracking Locations', which is
a different section.
(cherry picked from commit 8348436580)
* doc/bison.texinfo (Locations): Rename node to...
(Tracking Locations): ... this, which is its section name, and
update all cross references. The trouble is that there is another
node about locations. Its section name is `Locations', which was
easily confused with the node name of this node. Moreover, its
node name is `Locations Overview', which was mistakenly used to
label some cross references to this node.
(cherry picked from commit 7404cdf3a1)
Conflicts:
doc/bison.texinfo
* doc/bison.texinfo (Named References): Rename section to match
the node name, and update all cross references. Don't use the
acronyms RHS and LHS given that we don't expand them anywhere in
the manual. Try to use @samp and @code properly. Mention that
named references are experimental as we've already stated in NEWS.
(Destructor Decl): Expand the only remaining use of RHS or LHS.
(cherry picked from commit ce24f7f59b)
Reported by Hans Aberg at
<http://lists.gnu.org/archive/html/bug-bison/2011-05/msg00008.html>.
* NEWS (2.5.1): Document.
* doc/bison.texinfo (Named References): Because it discusses
locations in addition to semantic values, move this subsection out
of the section `Defining Language Semantics', where locations have
not yet been introduced, to be a new section after the following
section, `Tracking Locations'.
(cherry picked from commit 908c8647de)
It creates unnecessary differences between the sources that Bison
maintainers build and test (given that maintainers normally
configure with --enable-gcc-warnings) and the sources that Bison
users build. Instead, use PACIFY_CC, which doesn't. This change
fixes compiler warnings reported by Tys Lefering at
<http://lists.gnu.org/archive/html/bison-patches/2011-05/msg00004.html>.
* configure.ac: Don't AC_DEFINE lint regardless of the configure
options. This change affects imported gnulib sources, where
IF_LINT still appears and depends on lint.
* src/scan-gram.l, src/scan-skel.l: Replace uses of IF_LINT with
PACIFY_CC.
* src/system.h (IF_LINT): Remove cpp macro.
(PACIFY_CC): New cpp macro.
(cherry picked from commit 77bb73e7af)
Conflicts:
src/scan-gram.l
Since Bison 2.3b, which restored the ability of precedence
directives to assign user token numbers, doing so for user token
number 0 has produced an assertion failure.
* NEWS (2.5): Document fix.
* src/symtab.c (symbol_user_token_number_set): In the case of the
end token, don't decrement ntokens if it was never incremented.
* tests/regression.at (Token number in precedence declaration):
Extend.
(cherry picked from commit 9d6af15318)
* bootstrap.conf (gnulib_modules): Add all modules suggested by
-DGNULIB_POSIXCHECK.
* src/files.c (file_name_split)
* src/getargs.c (getargs)
* src/location.c (boundary_set_from_string)
* src/output.c (output_skeleton)
* src/parse-gram.y (prologue_declaration)
* src/scan-gram.l (handle_syncline)
* src/symtab.c (symbol_new): Use mbschr and mbsrchr instead of
strchr and strrchr. In the cases of command-line options, file
names, and thus locations, functionality may be improved. In the
case of symbol names, there should be no functional difference as
all characters are ASCII, so the intended benefit is just warning
suppression.
(cherry picked from commit d143e9c33f)
Conflicts:
src/output.c
src/parse-gram.c
src/parse-gram.h
* NEWS (2.5): For multi-character tokens in example grammar rules,
use "token" not 'token'. In English, use a consistent quotation
style; we might as well follow the precedent seen in info pages:
`quote' not "quote".
(cherry picked from commit ac3297d501)
This prevents gcc 4.6.0's -Wunused-but-set-variable from warning
about yyresolveLocations's local yyrhsloc. When locations are
enabled, there's no such warning because YYLLOC_DEFAULT then uses
yyrhsloc. Reported by Jim Meyering at
<http://lists.gnu.org/archive/html/bug-bison/2011-04/msg00002.html>.
* data/glr.c (yyresolveLocations): Omit definition when locations
are disabled.
(yyresolveValue): Omit yyresolveLocations invocation when
locations are disabled.
(cherry picked from commit 55dd356336)
* README-hacking (Updating a submodule): Give advice on how to
determine the versions of gnulib and autoconf to which we should
update.
(Release Procedure): Note that submodules should be updated.
* bootstrap.conf (gnulib_modules): Rename pipe to spawn-pipe as
suggested in updated gnulib NEWS.
* gnulib: Choose a stable snapshot according to advice in Bison's
README-hacking.
* po/POTFILES.in (lib/pipe.c): Rename to...
(lib/spawn-pipe.c): ... this.
* src/output.c: Update to include spawn-pipe.h.
* submodules/autoconf: Update to latest for improvement in m4.m4
that excludes M4 with buggy strstr. The only other changes to
files that we use are copyright updates.
(cherry picked from commit a898435b25)
Conflicts:
build-aux/.gitignore
Thus, conflict reports are now affected by -Werror and -Wnone
(unless %expect or %expect-rr is specified). Reported by George
Neuner at
<http://lists.gnu.org/archive/html/bug-bison/2010-08/msg00002.html>.
* NEWS (2.5): Document.
* doc/bison.texinfo (Bison Options): Document.
* src/complain.c, src/complain.h (set_warning_issued): Export
function.
* src/conflicts.c (conflicts_print): Suppress conflict report
based on -Wno-conflicts-sr and -Wno-conflicts-rr, and treat
conflicts as errors if -Werror.
* src/getargs.c (warnings_flag): Initialize with
warnings_conflicts_sr and warnings_conflicts_rr as well.
(warnings_args, warnings_types): Add entries for
warnings_conflicts_sr and warnings_conflicts_rr.
(usage): Update.
* src/getargs.h (enum warnings): Add entries for
warnings_conflicts_sr and warnings_conflicts_rr.
* tests/conflicts.at (-W versus %expect and %expect-rr): New test
group.
* tests/local.at (AT_BISON_CHECK_NO_XML): Update now that the
conflict report can produce a "warnings being treated as errors"
message. Also, check that stderr is now fully scrubbed by -Wnone
when the exit status is 0.
(cherry picked from commit 6f8bdce25d)
Adding command-line options after the grammar file name is not
permitted, so disable checks that do that when
maintainer-check-posix is running.
* tests/local.at (AT_BISON_CHECK_NO_XML): Don't run the
problematic checks when POSIXLY_CORRECT=1. Also, for readability,
remove an unnecessary m4_if.
(cherry picked from commit fc7ce9970f)
Reported by George Neuner at
<http://lists.gnu.org/archive/html/bug-bison/2010-08/msg00002.html>.
* NEWS (2.5): Document.
* THANKS (George Neuner): Add.
* doc/bison.texinfo (Bison Options): Document.
* src/complain.c, src/complain.h
(warn_at, warn_at_indent, warn): Suppress warning if -Wno-other.
(midrule_value_at): New warning function, similar to yacc_at in
that it's controlled by its own warning category.
* src/getargs.c (warnings_flag): Initialize to warnings_other.
(warnings_args, warnings_types): Add entry for warnings_other.
(usage): Update.
* src/getargs.h (enum warnings): Add entry for warnings_other.
* src/gram.c (grammar_rules_useless_report): If -Wno-other, then
don't print useless rules.
* src/reader.c (symbol_should_be_used): Rather than adjusting the
return value based on whether midrule value warnings are enabled,
accept a new parameter for telling the caller whether true is
being returned for a potential midrule warning.
(grammar_rule_check): Use midrule_value_at for midrule value
warnings, and continue to use warn_at for all other warnings. Let
them check whether the warnings are enabled.
* tests/local.at (AT_BISON_CHECK): Update documentation.
(AT_BISON_CHECK_NO_XML): Check that -Wnone and --warnings=none
disable all warnings exercised in the test suite.
(cherry picked from commit 8ffd7912e3)
Discussed at
<http://lists.gnu.org/archive/html/bison-patches/2010-08/msg00009.html>.
* NEWS (2.5): Document.
* src/getargs.c (flags_argmatch): Accept a new argument that
specifies what flags "all" and thus "none" affect.
(FLAGS_ARGMATCH): Update flags_argmatch invocation.
* tests/input.at (-Werror is not affected by -Wnone and -Wall):
New test group.
(cherry picked from commit dab9663283)
In `rhs[name]: "a" | "b"', do not free "name" twice.
Reported by Tys Lefering.
<http://lists.gnu.org/archive/html/bug-bison/2010-06/msg00002.html>
* src/named-ref.h, src/named-ref.c (named_ref_copy): New.
* src/parse-gram.y (current_lhs): Rename as...
(current_lhs_symbol): this.
(current_lhs): New function. Use it to free the current lhs
named reference.
* src/reader.c: Bind lhs to a copy of the current named reference.
* src/symlist.c: Rely on free (0) being valid.
* tests/named-refs.at: Test this.
(cherry picked from commit 8f462efe92)
Conflicts:
src/parse-gram.y
* NEWS (2.5): Document.
* data/lalr1.java (YYParser::YYStack::pop): Fix off-by-one error
in clearing the location stack. Also fix pop function that
accepts no arguments.
(cherry picked from commit 4c2a6e42ba)
Conflicts:
data/lalr1.java
Reported at
<http://lists.gnu.org/archive/html/bug-bison/2011-02/msg00005.html>.
* THANKS (Angelo Borsotti): Add.
* data/lalr1.java (YYParser::YYStack::pop): Fix off-by-one error
in clearing the value stack. Previously, the top element of the
stack wasn't cleared and so the value was not garbage collected.
(cherry picked from commit 6f75992be5)
Conflicts:
data/lalr1.java
* doc/bison.texinfo (Mystery Conflicts): Rename node to...
(Mysterious Conflicts): ... this, which is already the section
title and the name used in the index. Update all cross-references
to this node. Also, don't imply that R/R conflicts are the only
kind of mysterious conflict.
(cherry picked from commit 5da0355aff)
States that shift the error token do not have default reductions,
and GLR disables some default reductions, so "all" was a misnomer.
* doc/bison.texinfo (%define Summary): Update.
(Default Reductions): Update.
* src/print.c (print_reductions): Update.
* src/reader.c (prepare_percent_define_front_end_variables):
Update.
* src/tables.c (action_row): Update.
* tests/input.at (%define enum variables): Update.
* tests/reduce.at (%define lr.default-reductions): Update.
(cherry picked from commit d815ec4a62)
And clean up all other documentation of the features described
there.
* NEWS (2.5): Tweak wording of lr.type and parse.lac entries a
bit, update the cross-references to the manual, and point out that
LAC has caveats. Don't be so adamant that IELR+LAC=canonical LR.
That is, as the referenced section in the manual documents, LAC
does not fix infinite parsing loops on syntax errors.
* doc/bison.texinfo: Consistently drop the "(1)" suffix from LALR,
IELR, and LR in @cindex.
(%define Summary): Condense the entries for lr.default-reductions,
lr.keep-unreachable-states, lr.type, and parse.lac into brief
summaries, and cross-reference the appropriate subsections of
Tuning LR. For parse.lac, mention that it's only implemented for
deterministic parsers in C. In parse.error entry, mention LAC,
and add cross-reference to the LAC section.
(Error Reporting): When mentioning parse.error, mention LAC, and
add cross-reference to the LAC section.
(Tuning LR): New section with an extended version of the
documentation removed from %define Summary. Change all
cross-references in the manual to point here instead of there.
(Calc++ Parser): When mentioning parse.error, mention LAC, and add
cross-reference to the LAC section.
(Table of Symbols): In %error-verbose entry, add cross-reference
to Error Reporting.
(Glossary): Capitalize entry titles consistently. Add definitions
for "defaulted state" and "unreachable state". Expand IELR
acronym in IELR's entry.
(cherry picked from commit 6f04ee6c78)
Conflicts:
doc/bison.texinfo
* doc/bison.texinfo (Mystery Conflicts): Cross-reference
bibliography instead of citing publications directly.
(Generalized LR Parsing): Likewise.
(Bibliography): New section. Not all entries are cross-referenced
yet, but that will come in future patches.
(cherry picked from commit 71caec0661)
* NEWS (2.5): Document it.
* tests/java.at (_AT_DATA_JAVA_CALC_Y): To one of the yyerror
invocations, pass a location that spans multiple tokens. Change
yyerror to report all of a location rather than just the begin
position. Extend yylex and Position to count tokens on a line.
Remove getHashCode as it's unused. Update all expected output.
(cherry picked from commit 7776816565)
* doc/bison.texinfo (%define Summary): Fix mislabeled entry for
lex_symbol.
(%code Summary): For consistency with the variable list in the
%define Summary, enclose the list of %code qualifiers in a table
instead of an itemize.
* doc/bison.texinfo (%define Summary): Reword so it reads well as
a separate section. For example, add an intro, and move most of
the text outside of the @deffn so it is not indented so far.
(%code Summary): Likewise.
(Table of Symbols): Reword %code entry to match the %code entry in
Decl Summary.
(cherry picked from commit 406dec8286)
Conflicts:
doc/bison.texinfo
Suggested by Akim Demaille at
<http://lists.gnu.org/archive/html/bison-patches/2009-05/msg00013.html>.
* doc/bison.texinfo (Decl Summary): Extract most of the %define
entry into...
(%define Summary): ... this new subsection, and update all
cross-references. For readability of the patches, rewriting of
the text so it makes sense as a separate subsection will come in a
later patch. Moreover, the majority of the text describing the
various new LR features should likely move to another new section
somewhere.
(cherry picked from commit 2f4518a126)
Conflicts:
doc/bison.texinfo
Discussed in thread starting at
<http://lists.gnu.org/archive/html/bison-patches/2009-05/msg00013.html>.
* doc/bison.texinfo (Decl Summary): Extract most of the %code
entry into...
(%code Summary): ... this new subsection, and update all
cross-references. For readability of the patches, rewriting of
the text so it makes sense as a separate subsection will come in a
later patch.
(cherry picked from commit 8e6f226630)
Conflicts:
doc/bison.texinfo
The Bison manual's names for various files associated with a Bison
parser has devolved into inconsistency. This patch makes the
naming consistent for the most important files. First, it chooses
"grammar file" over "input file". The former appears to be more
traditional in the Bison manual, and Bison has other input
files (skeletons). Second, it chooses "parser implementation
file" over names like "parser file", "parser source file", "parser
source code file", and "parser output file". The new name makes
it clearer where Bison generates the main parser implementation,
and it is easily distinguishable from "parser header file".
* doc/bison.texinfo: Implement throughout.
(cherry picked from commit 9913d6e45a)
Conflicts:
doc/bison.texinfo
* doc/bison.texinfo (Introduction): Don't imply that only Robert
Corbett, Richard Stallman, and Wilfred Hansen have contributed to
Bison. However, I don't have time to write a full history, so
just point readers to THANKS and ChangeLog.
(cherry picked from commit 840341d67d)
* doc/bison.texinfo (Introduction): Say that IELR(1) and canonical
LR(1) are experimental. Mention Java. Normally experimental
features probably shouldn't be mentioned in the introduction.
However, if Bison's limitations to LALR(1), C, and C++ are so
important that they should be mentioned here, then it's important
to point out that Bison is beginning to escape those limitations.
Moreover, these particular experimental features have very little
chance of being removed.
* src/getargs.c (usage): Say that IELR(1) and canonical LR(1) are
experimental.
(cherry picked from commit d89e48b3c9)
This cleans up our previous fixes for a bug whereby Bison
discarded `.field' in `$-1.field'. The previous fixes were less
restrictive about where a dash could appear in an identifier, but
the restrictions were hard to explain. That bug was reported and
this final fix was originally suggested by Paul Hilfinger. This
also fixes a remaining bug reported by Paul Eggert whereby Bison
parses `%token ID -123' as `%token ID - 123' and handles `-' as an
identifier. Now, `-' cannot be an identifier. Discussed in
threads beginning at
<http://lists.gnu.org/archive/html/bug-bison/2011-01/msg00000.html>,
<http://lists.gnu.org/archive/html/bug-bison/2011-01/msg00004.html>.
* NEWS (2.5): Update entry describing the dash extension to
grammar symbol names. Also, move that entry before the named
references entry because the latter mentions the former.
* doc/bison.texinfo (Symbol): Update documentation for symbol
names. As suggested by Paul Eggert, mention the effect of periods
and dashes on named references.
(Decl Summary): Update documentation for unquoted %define values,
which, as a side effect, can no longer start with dashes either.
* src/scan-code.l (id): Implement.
* src/scan-gram.l (id): Implement.
* tests/actions.at (Exotic Dollars): Extend test group to exercise
bug reported by Paul Hilfinger.
* tests/input.at (Symbols): Update test group, and extend to
exercise bug reported by Paul Eggert.
* tests/named-refs.at (Stray symbols in brackets): Update test
group.
($ or @ followed by . or -): Likewise.
* tests/regression.at (Invalid inputs): Likewise.
* doc/bison.texinfo (Mid-Rule Actions): Mention that periods and
dashes make symbol names less convenient for named references.
* src/scan-code.l:
(handle_action_dollar): New arg textlen. All callers changed.
(handle_action_at): Likewise. Also, args are pointers to const.
(ref_tail_fields): Remove; no longer used.
(letter): Now includes '-' and '.', since this is for Bison
identifiers.
(id): Now the simpler traditional defn, since letters now include
'-' and '.'.
(c_letter, c_id): New defns.
(ref): Use c_id for unbracketed IDs.
(<SC_RULE_ACTION>): Simplify, now that the distinction between
Bison and unbracketed IDs are now in the regular expressions.
(VARIANT_BAD_BRACKETING): Remove.
(VARIANT_NOT_VISIBLE_FROM_MIDRULE): Renumber.
(find_prefix_end): Remove, replacing with ....
(identifier_matches): New function.
(variant_add): Use it. Omit EXPLICIT_BRACKETING arg; no longer
needed. CP arg is pointer to constant. All callers changed.
(show_sub_messages): Remove args CP, EXPLICIT_BRACKETING, DOLLAR_OR_AT.
New arg TEXT. All callers changed. Do not worry about showing
trailing context.
(parse_ref): Args CP, RULE, TEXT are now pointers to const. New
arg TEXTLEN. Remove arg DOLLAR_OR_AT. All callers changed.
Simplify code now that the regular expressions capture the
restrictions.
* src/scan-gram.l (letter, id): Adjust to match scan-code.l.
* src/symlist.c (symbol_list_null): Arg is now pointer to const.
* src/symlist.h: Likewise.
* tests/named-refs.at (Misleading references): These are now caught
by the C compiler, not by Bison; that's good enough. Adjust test
to reflect this.
(Many kinds of errors, Unresolved references): Adjust expected
diagnostics to match new behavior. The same errors are caught,
though the diagnostics are not quite as fancy.
($ or @ followed by . or -): Likewise. Also, Make the grammar
unambiguous, so that diagnostics are not complicated by ambiguity
warnings.
* data/glr.c, data/yacc.c: Do not use apostrophes in '#' comments,
as they confuse xgettext, which tries to parse them as C character
constants in a preprocessor directive.
* data/yacc.c (yy_lac): Don't use printf %d format on *yyesp, as
that expression might not promote to int on some platforms.
* src/parse-gram.c, src/parse-gram.h: Regenerate.
Previously, for this special case of an invalid reference, the
usual "symbol not found in production:" was printed. However,
because the symbol name was parsed as the empty string, that
message was followed immediately by a newline instead of a symbol
name. In reality, this is a syntax error, so the reference is
invalid regardless of the symbols actually appearing in the
production. Discussed at
<http://lists.gnu.org/archive/html/bison-patches/2011-01/msg00012.html>.
* src/scan-code.l (parse_ref): Report the above case as a syntax
error. Other than that, continue to handle this case like any
other invalid reference that Bison manages to parse because
"possibly meant" messages can still be helpful to the user.
* tests/named-refs.at ($ or @ followed by . or -): New test group.
Reported by Paul Hilfinger as a side effect of named references
support at
<http://lists.gnu.org/archive/html/bug-bison/2011-01/msg00000.html>.
Suggested by Paul Eggert.
* src/scan-code.l ({letter}, {id}): Adjust lexical definitions.
* src/scan-gram.l ({letter}, {id}): Likewise.
* .x-update-copyright: Add entry for bootstrap.
* bootstrap: Remove 2011 from copyright years. The bootstrap
version we're currently using comes from an older version of
gnulib.
* bootstrap.conf (bootstrap_sync): Add comments explaining this
issue.
* README (Copyright statements): New section explaining the range
notation. The GNU coding standards require this explanation. I
copied ours from coreutils.
* build-aux/update-b4-copyright: Revert 2010-06-17 changes that
disabled Bison's automated use of ranges.
* cfg.mk (update-copyright-env): Likewise.
* NEWS (2.5): Make some minor improvements to wording, and format
entries more consistently.
* doc/bison.texinfo (Language and Grammar): Point out that IELR
and canonical LR are experimental features.
(Decl Summary): In list of %define variables, make wording more
consistent. Improve discussion of using LALR for GLR.
* NEWS (2.5): Add entry for LAC, and mention LAC in entry for
other corrections to verbose syntax error messages.
* doc/bison.texinfo (Decl Summary): Rewrite entries for
lr.default-reductions and lr.type to be clearer, to mention
%nonassoc's effect on canonical LR, and to mention LAC. Add entry
for parse.lac.
(Glossary): Add entry for LAC.
* data/yacc.c: Rename %define variable parse.lac.es-capacity to
parse.lac.es-capacity-initial. Accept parse.lac.memory-trace
with values of "failures" (default) or "full".
(b4_declare_parser_state_variables): Add yyesa, yyes, and
yyes_capacity variables.
(YYSTACK_USE_ALLOCA): Ignore it if LAC requested.
(YYSTACK_ALLOC, YYSTACK_FREE, YYSTACK_ALLOC_MAXIMUM): Define if
LAC requested.
(YYCOPY_NEEDED): New cpp macro.
(YYCOPY): Define if LAC requested.
(yy_lac_stack_realloc): New function implementing stack
reallocations. Use YYMAXDEPTH for maximum stack size given that
the stack should never need to grow larger than the main state
stack needs to grow without LAC.
(YY_LAC_ESTABLISH): Update yy_lac invocation.
(yy_lac): Add arguments for exploratory stack memory data
recorded in the main parser. Invoke yy_lac_stack_realloc when
reallocation is necessary.
(yysyntax_error): Add the same new arguments and pass them to
yy_lac.
(yypstate_delete): Free yyes if necessary.
(yyesa, yyes, yyes_capacity): #define these to yypstate members
in the case of push parsing.
(yyparse, yypush_parse): Initialize yyes and yyes_capacity.
Update yysyntax_error invocations. At yyreturn, free yyes if
necessary.
* src/parse-gram.y: %define parse.lac full.
* tests/input.at (LAC: errors for %define): Extend for
parse.lac-memory-trace.
* tests/regression.at (LAC: Exploratory stack): Extend to check
that stack reallocs happen when expected.
(LAC: Memory exhaustion): Update to use YYMAXDEPTH and
parse.lac.es-capacity-initial.
LAC = lookahead correction. See discussion at
<http://lists.gnu.org/archive/html/bison-patches/2009-09/msg00034.html>.
However, one point there must be corrected: because of %nonassoc,
LAC is *not* always redundant for lr.type=canonical-lr.
* data/yacc.c: Accept values of "none" (default) or "full" for
parse.lac. Accept %define parse.lac.es-capacity to specify
capacity of LAC's temporary exploratory stack. It defaults to 20
and, for now, will not grow dynamically.
(b4_lac_flag, b4_lac_if): New m4 macros. Evaluate as true for
parse.lac!=none.
(YYBACKUP): Invoke YY_LAC_DISCARD.
(YY_LAC_ESTABLISH, YY_LAC_DISCARD): New cpp macros that invoke
yy_lac and track when it needs to be invoked
(yy_lac): New function that, given the current stack, determines
whether a token can eventually be shifted. Return status mimics
yyparse return status.
(yysyntax_error): Change yystate argument to yyssp so stack top
can be passed to yy_lac. If LAC is requested, build expected
token list by invoking yy_lac for every token instead of just
checking the current state for lookaheads. Return 2 if yy_lac
exhausts memory.
(yyparse, yypush_parse): Use local variable yy_lac_established and
cpp macros YY_LAC_ESTABLISH and YY_LAC_DISCARD to implement LAC.
Update yysyntax_error invocation. Add yyexhaustedlab code if LAC
is requested.
* tests/conflicts.at (%nonassoc and eof): Extend to check the
effect of each of -Dlr.type=canonical-lr and -Dparse.lac=full.
(parse.error=verbose and consistent errors): Likewise.
(LAC: %nonassoc requires splitting canonical LR states): New test
group demonstrating how LAC can fix canonical LR.
* tests/input.at (LAC: Errors for %define): New test group.
* tests/regression.at (LAC: Exploratory stack): New test group.
(LAC: Memory exhaustion): New test group.
Now, whenever we update bison's copy of gnulib, bootstrap will
update itself the next time it's run.
* bootstrap: Copy from latest gnulib for bootstrap_sync support.
* bootstrap.conf (bootstrap_sync): Set to true.
* gnulib: Update to latest so bootstrap is in sync now.
* NEWS (2.5): Document.
* data/yacc.c (yysyntax_error): In a verbose syntax error
message while in a consistent state with a default action (which
must be an error action given that yysyntax_error is being
invoked), continue to drop the expected token list, but don't
drop the unexpected token unless there actually is no lookahead.
Moreover, handle that internally instead of returning 1 to tell
the caller to do it. With that meaning of 1 gone, renumber
return codes more usefully.
(yyparse, yypush_parse): Update yysyntax_error usage. Most
importantly, set yytoken to YYEMPTY when there's no lookahead.
* data/glr.c (yyreportSyntaxError): As in yacc.c, don't drop the
unexpected token unless there actually is no lookahead.
* data/lalr1.cc (yy::parser::parse): If there's no lookahead,
pass yyempty_ not yyla.type to yysyntax_error_.
(yy::parser::yysyntax_error_): Again, don't drop the unexpected
token unless there actually is no lookahead.
* data/lalr1.java (YYParser::parse): If there's no lookahead,
set yytoken to yyempty_ before invoking yysyntax_error.
(YYParser::yysyntax_error): Again, don't drop the unexpected
token unless there actually is no lookahead.
* tests/conflicts.at (parse.error=verbose and consistent
errors): Extend test group to further reveal how the previous
use of the simple "syntax error" message was too general. Test
yacc.c, glr.c, lalr1.cc, and lalr1.java. No longer an expected
failure.
* tests/java.at (AT_JAVA_COMPILE, AT_JAVA_PARSER_CHECK): Move
to...
* tests/local.at: ... here.
(_AT_BISON_OPTION_PUSHDEFS): Push AT_SKEL_JAVA_IF definition.
(AT_BISON_OPTION_POPDEFS): Pop it.
(AT_FULL_COMPILE): Extend to handle Java.
These are purely whitespace changes that result in ugly code
but that make the next couple of patches much easier to read.
* data/glr.c (yyreportSyntaxError): Reindent.
* data/lalr1.cc (yy::parser::yysyntax_error_): Reindent.
* data/lalr1.java (YYParser::yysyntax_error): Reindent.
* data/yacc.c (yysyntax_error): Reindent.
* data/yacc.c (yyparse, yypush_parse): For yysyntax_error
invocation, get rid of the while loop, which is misleading
because there are really at most two iterations.
* bootstrap: Copy from gnulib's latest for the fix to
automatically anchor entries it constructs.
* gnulib: Update to latest just so it has the same bootstrap.
* .gitignore, build-aux/.gitignore, doc/.gitignore:
* lib/.gitignore, m4/.gitignore, po/.gitignore:
* runtime-po/.gitignore: Re-anchor all entries.
Jim Meyering reported this in
<http://lists.gnu.org/archive/html/bug-bison/2010-10/msg00007.html>.
* data/yacc.c: Use EXIT_SUCCESS as a witness for stdlib.h,
not _STDLIB_H. EXIT_SUCCESS has been defined by the standard
for quite some time.
* src/parse-gram.c, src/parse-gram.h: Regenerate.
* tests/regression.at: Tamper with the renamed witness.
* .cvsignore, build-aux/.cvsignore, doc/.cvsignore, etc/.cvsignore:
* examples/calc++/.cvsignore, lib/.cvsignore, m4/.cvsignore:
* po/.cvsignore, runtime-po/.cvsignore, src/.cvsignore:
* tests/.cvsignore: Remove; I don't use CVS to maintain Bison
anymore and don't know of anybody else who does. If someone needs
these files, they can resurrect them.
* .gitignore, build-aux/.gitignore, doc/.gitignore, lib/.gitignore:
* m4/.gitignore, po/.gitignore, runtime-po/.gitignore:
Omit leading '/', since bootstrap omits it.
Adjust file names to match current contents better.
* bootstrap: Sync from gnulib: this contains the new gnulib_mk_hook
installed just for us.
* bootstrap.conf (excluded_files): Don't exclude codeset.m4,
glibc21.m4, inttypes_h.m4, size_max.m4, xsize.m4, as they are now
needed somehow. Don't have time to look into why.
(gnulib_modules): Change malloc to malloc-gnu. Do we really assume
the GNU malloc behavior, where malloc (0) != NULL unless we're
out of storage? If not, we can omit malloc-gnu; but for now I left
it in to be safe.
(vc_ignore): Remove.
(gnulib_mk_hook): New function.
* README-hacking: Renamed from HACKING, since gnulib bootstrap now
uses that convention.
* data/glr.c (yy_reduce_print): Change yyrhsVals to yyvsp to remove
compiler errors when using %debug.
Add declaration of yylow when locations in use to avoid compilation
error.
(yyglrReduce): Conditionalize message "Parse ... rejected by rule..."
on whether we are in split mode, for consistency with behavior of
non-GLR parsing.
* NEWS (2.4.3): Mention that there are documentation fixes.
* doc/bison.texinfo (Expect Decl): Make it clear that %expect
turns conflicts into errors not warnings.
(Shift/Reduce): Likewise.
(Bison Options): Don't mention -Wsyntax. It doesn't exist.
(cherry picked from commit 3ffdd5f125)
* NEWS (2.4.3): Document fix.
* src/complain.c (error_message): Extend to handle incomplete
error messages so warn and warn_at can be used in more cases.
* src/gram.c (grammar_rules_useless_report): Use warn_at so that
-Werror is always obeyed.
* src/reduce.c (reduce_print): Use warn so that the "warnings
being treated as errors" message is printed consistently before
the first warning message. This makes testing easier.
* tests/local.at (AT_BISON_WERROR_MSG): New macro.
(AT_BISON_CHECK_NO_XML): Extend to check -Werror and
--warnings=error when warnings appear in bison's stderr.
(cherry picked from commit 954474bfa1)
* data/glr.c: Remove tests for __cplusplus related to definition of
YYSTACKEXPANDABLE.
* doc/bison.texinfo: Rewrite paragraph saying C++ stacks are not
expandable to instead indicate conditions that prevent their expansion
in C++. Reorganize section on GLR semantic values a bit. Remove
discussion of YYLLOC_DEFAULT. Mention restrictions to POD data.
* configure.ac (AM_INIT_AUTOMAKE): Underscore or dash in a
version string should disable gnits. Explain in comments.
(cherry picked from commit 122ff44429)
Conflicts:
configure.ac
* configure.ac: For prepending lib/ to the values of config
variables, fix detection of empty values. Also, due to recent
gnulib changes, add LIBUNISTRING_UNITYPES_H and
LIBUNISTRING_UNIWIDTH_H to the list of those variables.
* tests/output.at (AT_CHECK_OUTPUT): Extend to accept pre-tests.
(Output files: -dv >&-): Skip test group if running
maintainer-check-valgrind.
(cherry picked from commit c027cccb35)
Conflicts:
tests/output.at
These changes allow users to prefix an action with %? to indicate that it
is a semantic predicate---an expression that is evaluated immediately (not
deferred, even in GLR nondeterministic mode) and causes a syntax error if
false. In GLR parsers, this has the effect of killing one of a set of
split-off parses, just as would an ordinary syntax error.
Changelog:
* NEWS: Describe new semantic-predicate feature.
* data/c.m4 (b4_predicate_case): New definition.
* data/java.m4 (b4_predicate_case): New definition.
* data/glr.c (yyimmediate): Add definition.
(yydoAction): Remove comment, now obsolete.
Do YY_REDUCE_PRINT here.
(yyglrReduce): Alter comment to indicate that semantic values
need not be deferred.
Remove YY_REDUCE_PRINT from here; done in yydoAction.
(yyprocessOneStack): Pass immediate flag.
Delete stacks rejected by predicates in newly split-off parsers.
Change handling of yyerr so that only current stack gets deleted
when semantic predicate fails.
(yyfillin): Don't crash if a semantic value is unresolved (as may
happen in predicate rules).
Copy lr state as well in debugging mode.
Update comment on setting of yysval to include yyloc as well.
(yy_reduce_print): Add yynormal argument. Perform fillin properly.
Report unresolved RHS values.
(yyimmediate): New table.
* src/gram.h (struct rule): Add is_predicate field.
* src/output.c (user_actions_output): Use b4_predicate_case for
predicates.
(prepare_symbols): Output yyimmediate.
* src/scan-gram.l: Add %? token, SC_PREDICATE state.
* src/scan-code.l (code_props_rule_action_init): Add is_predicate
argument.
* src/scan-code.h (struct code_props): Add is_predicate field.
(code_props_rule_action_init): New interface.
* src/parse-gram.y (%?{...}): New token.
(rhs): Add %?{...} rule.
* src/parse-gram.c: Regenerate.
* src/parse-gram.h: Regenerate.
* src/reader.c (grammar_current_rule_action_append): Add
immediate argument.
(grammar_midrule_action): Use new interface for
code_props_rule_action_init.
(grammar_current_rule_action_append): Ditto.
(packgram): Transfer is_predicate value.
* src/reader.h (grammar_current_rule_action_append): New interface.
* doc/bison.texinfo: Document semantic predicates (%?).
* data/glr.c (yylhsNonterm, yyisDefaultedState,yyDefaultAction)
(yygetLRActions,yynewGLRStackItem,yyaddDeferredAction,yyinitStateSet)
(yyinitGLRStack,yyexpandGLRStack,yyupdateSplit,yymarkStackDeleted)
(yyundeleteLastStack,yyglrShift,yyglrShiftDefer,yydoAction,yyglrReduce)
(yyidenticalOptions,yymergeOptionSets,yyresolveStates,yyresolveAction)
(yyresolveLocations,yyresolveValue,yyreducePrint): Update parameter
names in comments and mention all parameters.
(struct yyGLRState): Fix description of yyposn field.
(yyresolveLocations): Correct comment so as not to imply action when
yyn1==0.
* data/lalr1.cc (YYLLOC_DEFAULT): Move its definition from the
header file to the implementation file, after the user %code
sections.
* NEWS: Document it.
* doc/bison.texinfo (Local Variables): Move this after the
LocalWords, since the latter are looked for in the whole document,
while the former are looked for only at its end.
Require american spell checking.
* data/lalr1.cc (YYRHSLOC): New.
(YYLLOC_DEFAULT): Use it.
* data/glr.cc: If location_type was user defined, do not include
location.hh, and do not produce location.hh and position.hh.
* tests/calc.at (YYLLOC_DEFAULT): Use YYRHSLOC.
Check that glr.cc supports user defined location_type.
* NEWS: Document this.
* tests/synclines.at (AT_SYNCLINES_COMPILE): More distcc patterns.
(AT_TEST_SYNCLINE): Remove GCC 4.5 protection which is already
taken care of in AT_SYNCLINES_COMPILE.
* data/bison.m4 (b4_percent_define_get): Accept a default value.
* data/c++.m4: Do not provide a default value for the %define
variable location_type, rather, use b4_percent_define_get with a
default argument where its value is needed.
* data/lalr1.cc: Do not load location.cc (which outputs both
location.hh and position.hh) if the user defined location_type.
Do not include location.hh either.
Currently, there is no check that %printer ... <foo> (nor
%destructor) is about an existing <foo> type. This C++ test had
it wrong (<::std::string> vs. <std::string>).
* tests/c++.at (AT_CHECK_VARIANTS): In list.yy, redefine the
pretty-printing of lists into something better for parser traces.
Update the expected output.
Fix correspondance between %type/%token and %printer.
* tests/calc.at (Span): Instead of begin/end, as in the built-in
location class, use first and last.
Define YYLLOC_DEFAULT to adjust to these changes.
* tests/local.at (_AT_BISON_OPTION_PUSHDEFS): Adjust to the
location_type changes.
* data/xslt/xml2dot.xsl (xsl:template match="bison-xml-report"):
Update output to include comments now produced by --graph.
(xsl:template match="automaton"): As for --graph, name the
digraph after the grammar file.
* src/print-xml.c (escape_bufs): Enlarge array.
(print_xml): Add bug-report and url attributes to
bison-xml-report element.
This handles the possibility that a "*/" might appear in
variable portions of those comments at some point in the future.
* src/graphviz.c (start_graph): Implement.
That is, keep them as warnings because that should be sufficient
to satisfy POSIX without creating backward compatibility issues.
Suggested by Richard Stallman at
<http://lists.gnu.org/archive/html/bison-patches/2010-03/msg00033.html>.
* NEWS (2.5): Remove mention of complaint.
* src/reader.c (grammar_rule_check): Convert warning back to
complaint.
* tests/input.at (%prec's token must be defined): Update.
* tests/local.at (AT_LOC_PUSHDEF, AT_LOC_POPDEF): New.
(_AT_BISON_OPTION_PUSHDEFS, AT_BISON_OPTION_POPDEFS): Use them to
define the appropriate AT_LOC accessors.
* tests/calc.at: Use AT_LOC accessors.
* tests/local.at (_AT_BISON_OPTION_PUSHDEFS):
Define AT_LOCATION_TYPE_IF.
(_AT_BISON_OPTION_POPDEFS): Undefine AT_LOCATION_TYPE_IF.
* tests/calc.at (_AT_DATA_CALC_Y): When %define location_type is
used, provide a user location type and use it.
(Simple LALR1 C++ Calculator): Add a test case for location_type.
* tests/calc.at: Use yy::parser::location_type rather than
yy::location, since the former is always right, and might point to
another type than the latter.
* data/lalr1.cc: Include location.hh before stack.hh.
Remove the useless forward declarations of position and location.
Reported by Chris Morley.
* data/glr.cc: Likewise.
Reported by Dagobert Michelsen at
<http://lists.gnu.org/archive/html/bug-bison/2010-03/msg00023.html>.
* THANKS (Dagobert Michelsen): Add.
* configure.ac (AC_PREREQ): Set to 2.64 so we get the latest
Autoconf macro for handling the restrict keyword.
* gnulib: Update to latest, which no longer overrides that macro
from Autoconf.
(cherry picked from commit 4333ba11c6)
Conflicts:
configure.ac
Reported by Tys Lefering at
<http://lists.gnu.org/archive/html/bug-bison/2010-03/msg00035.html>.
This fix is already implemented in glr.c and does not apply to
lalr1.java.
* data/lalr1.cc (yy::parser::parse): Increase size of
yyerror_range and adjust subscripting so you don't have to
subtract one from the beginning of the array.
* data/yacc.c (b4_declare_parser_state_variables,
yyparse, yypush_parse): Likewise.
(cherry picked from commit 48f4100a82)
Conflicts:
data/lalr1.cc
data/yacc.c
src/parse-gram.c
src/parse-gram.h
Reported by Johan van Selst at
<http://lists.gnu.org/archive/html/bug-bison/2010-03/msg00014.html>.
* NEWS (2.4.3): New.
* THANKS (Johan van Selst): Add.
* etc/prefix-gnulib-mk (prefix): Adjust regex for makefile
targets so that gnulib's new c++defs.h is matched.
* gnulib: Update to latest.
(cherry picked from commit 311b2e56ea)
Conflicts:
NEWS
* NEWS (2.4.2): Set version and date. For the recent test suite
portability fixes, don't be so optimistic about their success
given the lack of feedback on the affected platforms.
(cherry picked from commit 08090fe64c)
Conflicts:
NEWS
* tests/output.at (AT_CHECK_CONFLICTING_OUTPUT): Use
AT_BISON_CHECK_NO_XML rather than AT_BISON_CHECK because an
output file whose name conflicts with a previous output file
is now never generated.
(cherry picked from commit a3d760ef5d)
M4's output pipe was not being drained upon fatal errors during
scan_skel. As a result, broken-pipe messages from M4 were seen
on at least AIX, HP-UX, Solaris, and RHEL4, and this caused a
failure in the test suite. The problem was that, on platforms
where the default disposition for SIGPIPE is ignore instead of
terminate, M4 sometimes saw fwrite fail with errno=EPIPE and
then reported it. However, there's some sort of race condition,
because the new test group occasionally succeeded.
Reported by Albert Chin at
<http://lists.gnu.org/archive/html/bug-bison/2010-02/msg00004.html>.
There were also problems with the test suite livelocking on
Tru64 5.1b. Reported by Didier Godefroy at
<http://lists.gnu.org/archive/html/bug-bison/2009-05/msg00005.html>.
Switching to create_pipe_bidi suggested by Akim Demaille.
To attempt to solve both of these problems, switch to gnulib's
create_pipe_bidi and register M4 process as a slave. Along the
way, clean up file name conflict handling, which was affected by
the broken-pipe problem before the switch.
* NEWS (2.4.2): Document.
* THANKS (Didier Godefroy): Add.
* bootstrap.conf (gnulib_modules): Add pipe.
* gnulib: Update to latest to make sure we have all the latest
fixes.
* lib/local.mk (lib_libbison_a_SOURCES): Remove subpipe.h and
subpipe.c.
* po/POTFILES.in (lib/subpipe.c): Remove.
* src/files.c (compute_output_file_names): Update invocations
of output_file_name_check.
(output_file_name_check): In the case that the grammar file
would be overwritten, use complain instead of fatal, but replace
the output file name with /dev/null. Use the /dev/null solution
for the case of two conflicting output files as well because it
seems safer in case Bison one day tries to open both files at
the same time.
* src/files.h (output_file_name_check): Update prototype.
* src/output.c (output_skeleton): Use create_pipe_bidi and
wait_subprocess. Assert that scan_skel completely drains the
pipe.
* src/scan-skel.l (at_directive_perform): Update
output_file_name_check invocation.
* tests/output.at (AT_CHECK_CONFLICTING_OUTPUT): Check that the
grammar file actually isn't overwritten.
(Conflicting output files: -o foo.y): Update expected output.
* tests/skeletons.at (Fatal errors but M4 continues producing
output): New test group.
(cherry picked from commit 22cc8d813e)
Conflicts:
NEWS
bootstrap.conf
lib/.cvsignore
lib/.gitignore
lib/Makefile.am
m4/.cvsignore
m4/.gitignore
src/output.c
Reported by Tys Lefering.
* HACKING (Release checks): Add note about trying a recent GCC.
* tests/regression.at (_AT_DATA_DANCER_Y): For C++, include
cstdlib for abort.
(_AT_DATA_EXPECT2_Y): Likewise.
* HACKING (Release Procedure): Update notes on copyright years.
* Makefile.am (update-package-copyright-year): New target rule.
* build-aux/update-package-copyright-year: New file.
* cfg.mk (update-copyright): Add update-package-copyright-year
as a dependency.
(cherry picked from commit af8a609a90)
Stop equating LR(0) with nondeterminism and LALR(1) with
determinism. That is, if all states are consistent, then LR(0)
tables are deterministic. On the other hand, LALR(1) tables
might be nondeterministic before conflict resolution, and GLR
permits LALR(1) tables to remain nondeterministic.
* src/LR0.c, src/LR0.h: Here.
* src/lalr.c, src/lalr.h: Here.
* src/main.c (main): Here.
* src/state.c, src/state.h: Here.
* src/ielr.h (ielr): In preconditions, expect LR(0) not LALR(1)
parser tables.
(cherry picked from commit 1c4ad777cb)
Reported by Florian Krohm at
<http://lists.gnu.org/archive/html/bug-bison/2009-12/msg00005.html>.
* NEWS (2.4.2): Document.
* src/reader.c (grammar_rule_check): Implement.
(grammar_current_rule_prec_set): Add comments explaining that we
here assume a %prec identifier is a token, but we still manage
to support POSIX.
* tests/input.at (%prec's token must be defined): New test
group.
(cherry picked from commit f1ceed0774)
Specifically, don't sort objects by their memory addresses when
they're not allocated in the same array or other object. Though
I haven't found a test case where that fails on my platform, C
says the behavior is undefined.
* src/AnnotationList.c (AnnotationList__insertInto): Remove
FIXME. Use new id field of InadequacyList nodes rather than
their memory addresses when sorting.
(AnnotationList__compute_from_inadequacies): Add
inadequacy_list_node_count argument to pass to
InadequacyList__new_conflict.
* src/AnnotationList.h
(AnnotationList__compute_from_inadequacies): Update prototype
and documentation for new argument.
* src/InadequacyList.c (InadequacyList__new_conflict): Add
node_count argument and use it to assign a unique ID.
* src/InadequacyList.h (InadequacyListNodeCount): New typedef.
(InadequacyList): Add id field.
(InadequacyList__new_conflict): Update prototype and
documentation for new argument.
* src/ielr.c (ielr_compute_annotation_lists): Update
AnnotationList__compute_from_inadequacies invocation.
(cherry picked from commit 2728ac7ecd)
The problem was that yacc.c didn't always update the yychar
translation afterwards. However, other skeletons appear to be
fine. glr.c appears to already translate yychar before every
use. lalr1.cc does not define yychar and does not document its
replacement, yyla, for users. It does provide yyclearin, but
that does not manipulate yyla and thus requires no translation
update. In lalr1.java, yychar is out of scope during semantic
actions.
* NEWS (2.5): Document.
* data/yacc.c (YYBACKUP): Don't bother translating yychar into
yytoken here.
(yyparse, yypush_parse): Instead, translate before every use of
yytoken, and add comments explaining this approach.
* tests/actions.at (Destroying lookahead assigned by semantic
action): New test group checking that translation happens before
lookahead destructor calls at parser return. Previously,
incorrect destructors were called.
* tests/conflicts.at (parse.error=verbose and consistent
errors): New test group checking that translation happens at
syntax error detection before the associated verbose error
message and the associated lookahead destructor calls. While
the destructor call is fixed by this patch, the verbose error
message is currently incorrect due to another bug (see
comments in test group), so this is an expected failure for now.
* NEWS (2.5): Document.
* data/lalr1.java (parser::YYStack::YYFAIL): Rename to YYERRLAB,
and make it private. Update all uses.
* src/scan-code.l (SC_RULE_ACTION): Implement warning.
* NEWS (2.4.2): Update description of -Wundef fix.
* configure.ac (WARN_CXXFLAGS_TEST): New substitution.
(WARN_CFLAGS_TEST): New substitution.
* data/glr.c: Avoid warning about __STRICT_ANSI__.
* tests/atlocal.in (CFLAGS): Use WARN_CFLAGS_TEST instead of
WARN_CFLAGS.
(NO_WERROR_CFLAGS): Likewise.
(CXXFLAGS): Use WARN_CXXFLAGS_TEST instead of WARN_CXXFLAGS.
* data/yacc.c: Do not check YYLTYPE_IS_TRIVIAL if location
tracking is not enabled. Instead, unconditionally define
YY_LOCATION_PRINT as a no-op for backward compatibility.
Reported by Eric Blake. See discussions at
<http://lists.gnu.org/archive/html/bug-bison/2009-09/msg00008.html>
and
<http://lists.gnu.org/archive/html/bug-gnulib/2009-10/msg00108.html>.
* HACKING (Release checks): Suggest -DGNULIB_POSIXCHECK.
* bootstrap.conf (gnulib_modules): Add all the printf modules
suggested by -DGNULIB_POSIXCHECK. Add realloc-posix as
suggested by -DGNULIB_POSIXCHECK for gnulib's own vasnprintf.c.
(excluded_files): Remove m4/printf-posix.m4.
* tests/atlocal.in (LIBS): As for LDADD in src/local.mk, add
lib/libbison.a so gnulib libraries can be linked.
* etc/prefix-gnulib-mk (prefix): Adjust regex for make file
targets so that gnulib's new arg-nonnull.h and link-warning.h
are matched.
* gnulib: Update.
* src/Sbitset.c (Sbitset__new_on_obstack): Use Sbitset instead
of char*.
(Sbitset__isEmpty): Use Sbitset instead of char*.
* src/Sbitset.h (Sbitset): Make it a pointer to unsigned char
instead of char. This helps to avoid casting errors.
(Sbitset__or): Use Sbitset instead of char*.
That is, use CHAR_BIT and UCHAR_MAX instead of 8 and 0xff.
* src/Sbitset.h (Sbitset__nbytes): Here.
(Sbitset__byteAddress): Here.
(Sbitset__bit_mask): Here.
(Sbitset__last_byte_mask): Here.
(Sbitset__ones): Here.
(SBITSET__FOR_EACH): Here.
* src/parse-gram.y: Clean up sorting of declarations.
Use types to simplify %printer declarations where possible.
Provide %printer for BRACKETED_ID and symbol.prec.
* src/symtab.c: Whitespace change.
Reported by Michael Raskin at
<http://lists.gnu.org/archive/html/bug-bison/2009-09/msg00001.html>.
* THANKS (Michael Raskin): Add.
* tests/output.at (AT_CHECK_OUTPUT_FILE_NAME): Fix. Cygwin used
to fail at least for file names containing ":" or "\".
(cherry picked from commit 2ceb8c617c)
Conflicts:
THANKS
Except when memory reallocation is required, this change
eliminates the need to invoke yysyntax_error twice and thus to
repeat the collection of lookaheads. It also prepares for
future extensions that will make those repetitions more
expensive and that will require additional memory management in
yysyntax_error. Finally, it fixes an obscure bug already
exercised in the test suite.
* data/yacc.c (yysyntax_error): Add arguments for message
buffer variables stored in the parser. Instead of size, return
status similar to yyparse status but indicating success of
message creation. Other than the actual reallocation of the
message buffer, import and clean up memory management code
from...
(yyparse, yypush_parse): ... here.
* tests/regression.at (parse.error=verbose overflow): No longer
an expected failure.
* tests/atlocal.in (NO_WERROR_CFLAGS): New cpp macro.
* tests/regression.at (parse.error=verbose and
YYSTACK_USE_ALLOCA): New test group.
(parse.error=verbose overflow): New test group that reveals an
obscure bug. Expected fail for now.
* etc/bench.pl.in (generate_grammar_triangular): Do not activate
%debug by default. It can affect the timings even if yydebug=0.
(generate_grammar_calc): For consistency with other grammars,
use YYDEBUG environment variable to set yydebug.
Reported by Jim Meyering.
* src/Sbitset.h (SBITSET__INDEX__CONVERSION_SPEC): New, "zu"
because Sbitset__Index is size_t.
* src/Sbitset.c (Sbitset__fprint): Use it instead of %d.
* src/location.h: (location_print): Add return value.
* src/location.c: (location_print): Return number of printed
characters.
* src/complain.h: Two new functions (complain_at_indent,
warn_at_indent).
* src/complain.cpp: Implement the alignment mechanism. Add new
static variable (indent_ptr). Use and update it (error_message,
complain_at_indent, warn_at_indent).
* src/scan-code.l: Fix strings for translations. Use new *_indent
functions (parse_ref, show_sub_messages).
* NEWS (2.5): Add an announcement about named references.
* doc/bison.texinfo (Decl Summary): Document the "lex_symbol" and
"variant" %define variables.
(C++ Semantic Values): Split into...
(C++ Unions, C++ Variants): these.
The latter is new.
(C++ Parser Interface): Fix type names.
Document parser::syntax_error.
Document the fact that locations are not mandatory.
(C++ Scanner Interface): Split into...
(Split Symbols, Complete Symbols): these.
The later is new.
(Calc++ Parsing Driver): Use variants.
Add more comments.
Adjust style.
(Calc++ Parser): Declare all the tokens, no
longer accept raw characters.
Remove %union.
Adjust types and printers.
Remove destructors.
(Calc++ Scanner): Use make_<SYMBOL> functions.
Use strerror in error message.
* data/c++.m4, data/lalr1.cc (parser::symbol_type): Change the
constructor to take a token_type instead of the (internal) symbol
number.
Call yytranslate_.
* data/variant.hh (b4_symbol_constructor_define_): Therefore,
don't call yytranslate_ here.
Add "symbol not found in production" error message when
an "invalid reference" is detected in named references
resolution.
* src/scan-code.l: Update "invalid reference" case.
* tests/named-ref.at: Adjust test-cases.
* data/yacc.c: Clean up M4 for readability, and make output
whitespace more consistent. For the main parse function
comment, instead of saying "yyparse or yypush_parse", say either
"yyparse" or "yypush_parse" depending on which it actually is.
It is common to use sort of factories in the user actions. These
factories may check some "syntactic" constraints that are not
enforced by the grammar itself. This is possible using YYERROR
within the action itself. Provide the user with a means to throw
a syntax_error exception.
* data/c++.m4 (b4_public_types_declare, b4_public_types_define):
Declare and define yy::parser::syntax_error.
* data/lalr1.cc: Include stdexcept.
(yy::parser::parse): Wrap the user action within a try/catch.
* data/glr.cc: Include stdexcept.
* NEWS (2.6): Document %param, %lex-param, and %parse-param
changes.
* doc/bison.texinfo: Document that %lex-param and %parse-param
are n-ary.
Changes some examples to demonstrate it.
(Calc++ Parser): Use %param.
Provide a means to factor lex-param and parse-param common
declarations.
* src/parse-gram.y (param_type): New.
Define a %printer for it.
(add_param): Use it.
(%parse-param, %lex-param): Merge into...
(%parse): this new token.
Adjust the grammar to use it.
* src/scan-gram.l (RETURN_VALUE): New.
(RETURN_PERCENT_FLAG): Use it.
(RETURN_PERCENT_PARAM): New.
Use it to support %parse-param, %lex-param and %param.
* src/output.c: Don't include assert.h.
(output_skeleton): Use aver not assert.
* src/system.h (aver): In documentation of why, add links to
Paul Eggert's explanations in the mailing lists.
in a symbolic reference resolution. Remove .expr and -expr from
the shown reference when the reference is unresolved.
* src/scan-code.l: Change the error message, adjust location columns,
rename variable "exact_mode" to "explicit_bracketing".
* tests/named-ref.at: Adjust existing tests and add a new one.
* data/bison.m4 (b4_error_verbose_flag): Don't examine value of
%define variable parse.error unless b4_error_verbose_flag is
actually expanded in a skeleton.
And skip tests if perl is not available. This is better than
playing tricks with shell portability. Suggested by Akim
Demaille.
* tests/input.at (Bad character literals): Use it here for
omitting final newlines.
(Bad escapes in literals): Use it here for special characters.
* data/bison.m4 (b4_integral_parser_tables_map): Don't mention
zero values in the YYTABLE comments.
* data/glr.c (yytable_value_is_error): Don't check for zero
value.
* data/lalr1.cc (yy_table_value_is_error_): Likewise.
* data/yacc.c (yytable_value_is_error): Likewise.
* data/lalr1.java (yy_table_value_is_error_): Likewise.
(yysyntax_error): Fix typo in code: use yytable_ not yycheck_.
* src/tables.h: In header comments, explain why it's useless to
check for a zero value in yytable.
* data/bison.m4 (b4_integral_parser_tables_map): Fix YYTABLE
comments: zero indicates syntax error not default action.
* data/c.m4 (b4_table_value_equals): Comment that YYID must be
defined.
* data/glr.c (yyis_pact_ninf): Rename to...
(yypact_value_is_default): ... this.
(yyisDefaultedState): Update for rename.
(yyis_table_ninf): Rename to...
(yytable_value_is_error): ... this, and check for value zero
besides just YYTABLE_NINF.
(yygetLRActions): Check for default value from yypact. It
appears that this check is always performed before this function
is invoked, and so adding the check here is probably redundant.
However, the code may evolve after this subtlety is forgotten.
Also, update for rename to yytable_value_is_error. Because that
macro now checks for zero, a different but equivalent branch of
the if-then-else here is evaluated.
(yyreportSyntaxError): Update for rename to
yytable_value_is_error. The zero condition was mishandled
before.
(yyrecoverSyntaxError): Update for renames. No behavioral
changes.
* data/lalr1.cc, data/lalr1.java (yy_pact_value_is_default_):
New function.
(yy_table_value_is_error_): New function.
(parse): Use new functions where possible. No behavioral
changes.
(yysyntax_error_, yysyntax_error): Use yy_table_value_is_error_.
The zero condition was mishandled before.
* data/yacc.c (yyis_pact_ninf): Rename to...
(yypact_value_is_default): ... this.
(yyis_table_ninf): Rename to...
(yytable_value_is_error): ... this, and check for value zero
besides just YYTABLE_NINF.
(yysyntax_error): Update for rename to yytable_value_is_error.
The zero condition was mishandled before.
(yyparse): Update for renames. No behavioral changes.
* src/tables.h: Improve comments about yypact, yytable, etc.
more. Most importantly, say yytable value of zero means syntax
error not default action.
* src/output.c (ARRAY_CARDINALITY): New macro, copied from
argmatch.h.
(output_skeleton): Use it.
Suggested by Akim Demaille.
Signed-off-by: Eric Blake <[email protected]>
(cherry picked from commit 1266b63674)
* submodules/autoconf: Update to autoconf 2.64.
* configure.ac (M4_GNU_OPTION): New define.
* src/output.c (output_skeleton): Use it to resolve FIXME.
* NEWS: Mention this.
Signed-off-by: Eric Blake <[email protected]>
(cherry picked from commit b9ad39c1fb)
Discussed starting at
<http://lists.gnu.org/archive/html/bison-patches/2009-08/msg00036.html>.
* src/scan-gram.l (SC_ESCAPED_STRING, SC_ESCAPED_CHARACTER):
For a \0 and similar escape sequences meaning the null
character, report an invalid escape sequence instead of an
invalid null character because the latter does not actually
appear in the user's input.
In all escape sequence complaints, don't escape the initial
backslash, and don't quote when the sequence appears at the end
of the complaint line unless there's whitespace that quotearg
won't escape.
Consistently say "invalid" not "unrecognized".
Consistently prefer "empty character literal" over "extra
characters in character literal" warning for invalid escape
sequences; that is, consistently discard those sequences.
* tests/input.at (Bad escapes in literals): New.
This patch simplifies "variables: rename namespace as api.namespace",
commit 6750106107.
Suggested by Joel E. Denny in
http://lists.gnu.org/archive/html/bison-patches/2009-07/msg00006.html
* src/muscle-tab.c (muscle_percent_variable_update): New.
(muscle_percent_define_insert): Use it in replacement of the
previous tr invocation.
Remove variable_tr, no longer needed.
* data/bison.m4 (b4_percent_define_copy_, b4_percent_define_copy):
Remove.
* data/c++.m4: No longer handle namespace -> api.namespace.
* tests/input.at (%define backward compatibility): Check that
namespace is treated as api.namespace.
doc: %initial-action to initialize yylloc.
Reported by Bill Allombert.
* doc/bison.texinfo: Set fill-column to 76.
(Location Type): Document the use of %initial-action to initialize
yylloc.
yystate and yystack_[0].state are equal, keep only the latter.
The former was also used as a temporary variable to compute the
post-reduction state. Move this computation into an auxiliary
function.
* data/glr.c (yyLRgotoState): Fuse variable definition and first
assignment.
* data/lalr1.cc (yy_lr_goto_state_): New.
(yyparse): Use it.
Replace remaining uses of yystate by yystate_[0].state.
Remove the former.
* data/lalr1.cc (yyreduce): Compute the resulting state before
running the user action so that yylhs is a valid symbol.
(yyerrorlab): Since yylhs is complete (it knows its type), we can
simply call yy_destroy_ to destroy $$ on YYERROR invocations.
* tests/c++.at (AT_CHECK_VARIANTS): Test YYERROR with variants.
* gnulib: Update.
* .x-update-copyright (COPYING): Add as it's no longer implied
when .x-update-copyright is present.
* cfg.mk (update-copyright-local): Remove, now ignored.
(update-copyright): Declare update-b4-copyright as a dependency.
Bison was uselessly requiring the formatstring macros from
gettext, which resulted in mo files not being installed on systems
that perfectly supported Bison mo files. Lower the requirement.
http://lists.gnu.org/archive/html/bug-bison/2009-08/msg00006.html
* configure.ac: Require need-ngettext instead of
need-formatstring-macros.
Reported by Martin Jabocs.
Suggested by Bruno Haible.
* INSTALL: Restructure.
(Internationalization): New.
(cherry picked from commit b9e42bb439)
Continues Akim's work from his 2009-06-10 commits.
* src/reader.c (check_and_convert_grammar): Don't add any
symbols after the first symbols_do invocation.
* src/symtab.c (symbols_sorted): New static global.
(user_token_number_redeclaration): Update comments.
(symbol_from_uniqstr): If a new symbol is being created, assert
that symbols_sorted hasn't been allocated yet.
(symbols_free): Free symbols_sorted.
(symbols_cmp, symbols_cmp_qsort): New functions.
(symbols_do): Sort symbol_table into symbols_sorted on first
invocation.
* tests/input.at (Numbered tokens): Recombine tests now that the
output should be deterministic across multiple numbers.
Some of our targets use "bison --help", but they can't depend on
"bison" itself (to avoid additional requirements on the user), so
they used to call "make src/bison" in the commands. Then
concurrent builds may fail: one make might be aiming one of its
jobs at compiling src/bison, and another job at generating the man
page. If the latter is faster than the former, then we have two
makes that concurrently try to compile src/bison.
This might also be a more convincing explanation for the failure
described in the patch "build: fix paths".
* Makefile.am (SUFFIXES): Initialize.
* build-aux/move-if-change: New, symlink to gnulib's.
* build-aux/local.mk: Ship it.
* doc/common.x: Remove, merged into...
* doc/bison.x: here.
* doc/local.mk (doc/bison.help): New.
($(CROSS_OPTIONS_TEXI)): Depend on it.
Use src/bison.
(.x.1): Replace with...
(doc/bison.1): this explicit, simpler, target.
(common_dep): Remove, inlined where appropriate.
(SUFFIXES, PREPATH): Remove, unused.
* configure.ac (gl_PREFIXED_LIBOBJS): Don't rename it, rather,
change the value of...
(gl_LIBOBJS): this.
Adjust more variables.
* etc/prefix-gnulib-mk (prefix_assignment): Don't rename
gl_LIBOBJS.
(prefix): Also transform rules whose targets have slashes.
Use $prefix liberally.
Map @MKDIR_P@ to $(MKDIR_P).
Prefix directories that are mkdir'd.
When using $(top_builddir) inconsistently, Make (including GNU
Make) is sometimes confused. As a result it may want to build
lib/libbison.la and $(top_builddir)/lib/libbison.la (the same
file, different names) concurrently, which, amusingly enough,
might end with:
ranlib lib/libbison.a
ranlib lib/libbison.a
make[2]: *** [lib/libbison.a] Segmentation fault
on OS X.
* doc/local.mk, src/local.mk: Do not use $(top_builddir) when not
needed.
* src/reader.c (reader): Move %define front-end variable
defaults and checking into...
(prepare_percent_define_front_end_variables): ... this new
function.
* src/scan-gram.l (INITIAL): For consistency with string
literals, don't store open quote on character literal. It's
discarded before returning anyway.
(SC_ESCAPED_CHARACTER): Similarly, don't store close quote.
Make length test more readable, and make the character stored
for an empty literal more obvious while consistent with the
previous behavior.
* src/symtab.c, src/symtab.h: Rename USER_NUMBER_ALIAS to
USER_NUMBER_HAS_STRING_ALIAS throughout.
* src/symtab.c (symbol_make_alias): Remove comment from symtab.c
that is repeated in symtab.h. Improve argument names to make it
clear which side of the symbol-string alias pair is which.
(symbol_check_alias_consistency): Improve local variable names
for the same purpose.
* src/symtab.h (struct symbol): Make comments about aliases
clearer.
(symbol_make_alias): Improve comments and argument name.
* src/output.c (token_definitions_output): Update for rename to
USER_NUMBER_HAS_STRING_ALIAS and improve comments about aliases.
* build-aux/update-b4-copyright: In warnings, report line
numbers rather than character positions.
* cfg.mk (update-copyright-local): Set to update-b4-copyright so
that update-copyright runs it.
* gnulib: Update.
* build-aux/update-b4-copyright: Do not accept 2-digit
UPDATE_COPYRIGHT_YEAR, which was not handled correctly.
Don't accept a `[' in a b4_copyright argument.
Format code more consistently.
Don't assume b4*copyright never occurs.
Reported by Alex Rozenman. This warning shows up with gcc-4.3.0
and later.
* src/scan-code.l: Add "%option noinput", which I cannot find in
the Flex manual, but which Flex has supported since at least as
far back as 2.5.4. However, if any of our developers still use
Flex 2.5.4, they'll need to stop configuring with
--enable-gcc-warnings because "%option noinput" didn't work
correctly until Flex 2.5.6.
* src/scan-gram.l: Likewise.
* src/scan-skel.l: Likewise.
(cherry picked from commit 42f8609bbd)
* NEWS (2.5): Document.
* src/scan-gram.l (INITIAL): Remove comment that we don't check
the length.
(SC_ESCAPED_CHARACTER): Warn if length is wrong.
* tests/input.at (Bad character literals): New test group.
(cherry picked from commit ac9b0e954b)
* src/named-ref.c: Add a pointer check (named_ref_free).
* src/scan-code.l: New function (variant_table_free). Called in
code_scanner_free.
* src/symlist.c: Call to named_ref_free (symbol_list_free).
Suggested by Eric Blake at
<http://lists.gnu.org/archive/html/bison-patches/2009-04/msg00083.html>.
* tests/existing.at (_AT_TEST_EXISTING_GRAMMAR): Do not
complicate the code by distinguishing between a missing value
and an empty string value for an optional argument. This fix is
allowed by the similar fix in AT_TEST_TABLES_AND_PARSE below.
* tests/local.at (_AT_TEST_TABLES_AND_PARSE): Merge into...
(AT_TEST_TABLES_AND_PARSE): ... this now that the special
arguments are not needed because of the following changes.
Fix stale comments.
Bison developers should use GNU M4 and should not use
POSIXLY_CORRECT when building the test suite, so do not
complicate the code by avoiding $10 and above.
Do not quote an empty string value for an optional argument, and
do not distinguish between a missing value and an empty string
value.
(cherry picked from commit cba975069a)
* src/named-ref.h: Add comments.
* src/parse-gram.y: Readability and style changes.
* src/reader.c: Factoring: assign_named_ref function.
* src/scan-code.l: Factoring and style changes. Rename
parse_named_ref to parse_ref. Use "c-ctype.h" from gnulib.
Use "unsigned" type for variant index. Improve readablity.
* src/scan-gram.l: Change error messages and add comments.
* src/symlist.h: symbol_list_null: New function decl.
* src/symlist.c: symbol_list_null: Implement here.
* tests/named-refs.at: Adjust for new error messages.
* examples/calc++/Makefile.am (calc.stamp): New.
Depend on it to create the sources of calc++ so that concurrent
builds don't launch several "extexi" in parallel.
Not only this is inefficient, this also builds incorrect sources
with several extractions mixed together.
* data/bison.m4: Move code related to specific variables after the
definition of the variable-maintaining macros so that we don't
"invoke" b4_percent_define_check_values before it is defined.
Implement, document, and test the replacement of %error-verbose
by %define parse.error "verbose".
* data/bison.m4 (b4_error_verbose_if): Reimplement to track the
values of the parse.error variable.
Make "simple" its default value.
Check the valid values.
* src/parse-gram.y: Use %define parse.error.
(PERCENT_ERROR_VERBOSE): New token.
Support it.
* src/scan-gram.l: Support %error-verbose.
* doc/bison.texinfo (Decl Summary): Replace the documentation of
%define error-verbose by that of %define parse.error.
* NEWS: Document it.
* tests/actions.at, tests/calc.at: Use parse.error instead of
%error-verbose.
Discussed in:
http://lists.gnu.org/archive/html/bison-patches/2009-01/msg00000.htmlhttp://lists.gnu.org/archive/html/bison-patches/2009-02/msg00002.htmlhttp://lists.gnu.org/archive/html/bison-patches/2009-03/msg00009.html
* src/parse-gram.y: Add new syntax (named_ref.opt).
* src/reader.c: Store named refs in symbol lists.
* src/reader.h: New argument for symbol_append and
action_append functions.
* src/scan-code.h: Add new field (named_ref) into
code_props data structure. Keeps named ref of midrule
actions.
* src/scan-code.l: Support for named refs in semantic
action code. New function 'parse_named_ref'.
* src/scan-gram.l: Support bracketed id.
* src/symlist.c: Store named refs in symbol lists.
* src/symlist.h: New field in symbol list: named_ref.
* src/named-ref.h: New file, a struct for named_ref.
* src/named-ref.cp: New file, named_ref_new function.
* src/local.mk: Add two new files.
* tests/testsuite.at: Include new test group:
* tests/named-refs.at: this new file.
Discussed in
http://lists.gnu.org/archive/html/bison-patches/2009-04/msg00033.html
* data/bison.m4 (b4_percent_define_copy_, b4_percent_define_copy):
New.
(b4_percent_define_use): New.
Use it where applicable.
* data/c++.m4: Replace uses of the variable "namespace" by
"api.namespace".
Default the latter to the former.
* doc/bison.texinfo (Decl Summary): Document "namespace" as
obsolete.
Document api.namespace.
Use @samp to document %define uses, keep @code for identifiers.
* NEWS: Likewise.
* tests/c++.at, tests/input.at: Test api.namespace instead of
namespace. (The tests passed with namespace.)
Some consistency checks on symbols are performed after all the
symbols were read, by an iteration over the symbol table. This
traversal is nondeterministic, which can be a problem for test
cases.
Avoid this.
Addresses another form of nondeterminism reported by Joel E. Denny.
http://lists.gnu.org/archive/html/bison-patches/2009-05/msg00023.html
* tests/input.at (Numbered tokens): Split the hexadecimal/decimal
test in two.
Use different file names for the three tests to make the
maintenance easier.
* gnulib: Update to latest.
* lib/.cvsignore, lib/.gitignore, m4/.cvsignore,
* m4/.gitignore: Regen.
* src/symtab.c (symbol_from_uniqstr, semantic_type_from_uniqstr):
Call xalloc_die on hash_insert failures.
Requested by the new __warn_unused_result__ attribute of
hash_insert.
* data/bison.m4 (b4_check_user_names): In header comments, say
"user occurrence" instead of "grammar occurrence".
* src/muscle-tab.h (muscle_percent_define_insert): Likewise.
(muscle_percent_code_grow): Likewise just for consistency.
(cherry picked from commit 246c4efafe)
Conflicts:
src/muscle-tab.h
* data/c++.m4 (b4_namespace_close): Don't let it be printed
among the closing braces here. This fix might make the
generated code easier to debug, but otherwise it should be
insignificant because a trailing `:' is a C++ error already.
(cherry picked from commit 8c221795af)
* doc/bison.simple (Decl Summary): Document token.prefix.
(Calc++ Parser): Various fixes.
Formatting changes.
Use token.prefix.
Introduce a macro TOKEN to shorten the code and make it more
readable.
(Calc++ Scanner): Adjust.
* NEWS (Variable token.prefix): New.
For now, just api.push-pull and lr.keep-unreachable-states.
Maintain old names for backward compatibility.
* NEWS (2.5): Document.
* data/c.m4 (b4_identification): Update comment.
* data/yacc.c: Update access.
* doc/bison.texinfo: Update.
* etc/bench.pl.in (bench_push_parser): Update use.
* src/files.c (tr): Move to...
* src/getargs.c, src/getargs.h (tr): ... here because I can't
think of a better place to expose it. My logic is that, for all
uses of tr so far, command-line arguments can be involved, and
getargs.h is already included.
* src/main.c (main): Update access.
* src/muscle_tab.c (muscle_percent_define_insert): Convert old
variable names to new variable names before assigning value.
* src/reader.c (reader): Update setting default.
* tests/calc.at: Update uses.
* tests/conflicts.at (Unreachable States After Conflict
Resolution): Update use.
* tests/input.at (%define enum variables): Update use.
(%define backward compatibility): New test group.
* tests/push.at: Update uses.
* tests/reduce.at: Update uses.
* tests/torture.at: Update uses.
(cherry picked from commit 812775a039)
Conflicts:
data/c.m4
etc/bench.pl.in
src/parse-gram.c
src/parse-gram.h
tests/conflicts.at
* NEWS (2.5): Add an entry.
* src/getargs.c (usage): Mention IELR(1) and canonical LR(1)
besides just LALR(1) and GLR(1).
* doc/bison.texinfo (Introduction): Likewise.
(Language and Grammar): Bison is no longer limited to LALR(1)
restrictions.
(GLR parsing): Say deterministic or LR(1) rather than LALR(1)
when trying to distinguish from GLR. Talk about LR(1) grammars
rather than LALR(1) grammars.
(Decl Summary): In %define api.push_pull entry, say it applies
to deterministic parsers in C rather than LALR(1) parsers in C.
Add lr.default_rules entry.
Add lr.type entry.
(Mystery Conflicts): Bison is no longer limited to LALR(1)
restrictions.
(Generalized LR Parsing): Same changes as for the previous GLR
section.
(Memory Management): Say deterministic rather than LALR(1).
(Understanding): Correct some bison output.
Index discussion of "accepting state".
Say deterministic rather than LALR(1).
(Bison Options): In --yacc entry, say deterministic rather than
LALR(1).
In --report, --graph, and --xml entries, just don't mention
LALR(1).
(C++ Parsers): Say deterministic rather than LALR(1).
(Table of Symbols): Likewise in YYSTACK_USE_ALLOCA entry.
(Glossary): Add Accepting State, Consistent State, Default Rule,
and IELR(1) definitions.
In Generalized LR (GLR) definition, make same changes as in
previous GLR sections.
In LALR(1) definition, say Bison uses LALR(1) by default rather
than implying Bison is limited to LALR(1).
(LocalWords): Add IELR.
(cherry picked from commit 34a6c2d19b)
Its value can be "LALR", "IELR", or "canonical LR".
* lib/timevar.def (TV_IELR_PHASE1): New var.
(TV_IELR_PHASE2): New var.
(TV_IELR_PHASE3): New var.
(TV_IELR_PHASE4): New var.
* src/local.mk (src_bison_SOURCES): Add AnnotationList.c,
AnnotationList.h, InadequacyList.c, InadequacyList.h, Sbitset.c,
Sbitset.h, ielr.c, and ielr.h.
* src/getargs.h, src/getargs.c (enum trace, trace_args,
trace_types): Add trace_ielr.
* src/lalr.h, src/lalr.c (ngotos): Export it.
(F): Rename to...
(goto_follows): ... this, update all uses, and export it.
(set_goto_map): Export it.
(map_goto): Export it.
(compute_lookahead_tokens): Don't free goto_follows yet. Now
handled in ielr.
(initialize_LA): Export it. Move lookback allocation to...
(lalr): ... here because, for canonical LR, initialize_LA must
be invoked but lookback and much of the rest of LALR isn't
needed.
* main.c (main): Instead of lalr, invoke ielr, which invokes
lalr.
* src/reader.c (reader): Default lr.type to "LALR".
Default lr.default_rules to "accepting" if lr.type is "canonical
LR". Leave the default as "all" otherwise.
Check for a valid lr.type value.
* src/state.h, src/state.c (struct state_list): Add state_list
member.
(state_new): Initialize state_list member to NULL.
(state_new_isocore): New function, exported.
* tests/existing.at (AT_TEST_EXISTING_GRAMMAR): New macro that
exercises all values of lr.type.
(GNU AWK Grammar): Rename test group to...
(GNU AWK 3.1.0 Grammar): ... this, and extend to use
AT_TEST_EXISTING_GRAMMAR.
(GNU Cim Grammar): Extend to use AT_TEST_EXISTING_GRAMMAR.
(GNU pic Grammar): Rename test group to...
(GNU pic (Groff 1.18.1) Grammar): ... this, and extend to use
AT_TEST_EXISTING_GRAMMAR.
* tests/reduce.at (AT_TEST_LR_TYPE): New macro that exercises
all values of lr.type.
(Single State Split): New test groups using AT_TEST_LR_TYPE.
(Lane Split): Likewise.
(Complex Lane Split): Likewise.
(Split During Added Lookahead Propagation): Likewise.
Its value describes the states that are permitted to contain
default rules: "all", "consistent", or "accepting".
* src/reader.c (reader): Default lr.default_rules to "all".
Check for a valid lr.default_rules value.
* src/lalr.c (state_lookahead_tokens_count): If lr.default_rules
is "accepting", then only mark the accepting state as
consistent.
(initialize_LA): Tell state_lookahead_tokens_count whether
lr.default_rules is "accepting".
* src/tables.c (action_row): If lr.default_rules is not "all",
then disable default rules in inconsistent states.
* src/print.c (print_reductions): Use this opportunity to
perform some assertions about whether lr.default_rules was
obeyed correctly.
* tests/local.at (AT_TEST_TABLES_AND_PARSE): New macro that
helps with checking the parser tables for a grammar.
* tests/input.at (%define lr.default_rules invalid values): New
test group.
* tests/reduce.at (AT_TEST_LR_DEFAULT_RULES): New macro using
AT_TEST_TABLES_AND_PARSE.
(`no %define lr.default_rules'): New test group generated by
AT_TEST_LR_DEFAULT_RULES.
(`%define lr.default_rules "all"'): Likewise.
(`%define lr.default_rules "consistent"'): Likewise.
(`%define lr.default_rules "accepting"'): Likewise.
* data/bison.m4 (b4_percent_define_if_define_): Also map dashes to
underscores.
* src/scan-gram.l ({id}): Also accept dashes after the initial
letter.
({directive}): Use {id}.
* src/parse-gram.y: Comment and formatting changes.
* doc/bison.texinfo (Symbols): Adjust the lexical definitions of
symbols.
* src/complain.h, src/complain.c (yacc_at): New.
* src/symtab.c (symbol_new): Use yacc_at to report inappropriate
symbol names.
* src/output.c (token_definitions_output): Do not #define token
names with dashes.
* tests/output.at (AT_CHECK_OUTPUT_FILE_NAME): Avoid tripping
autoconf bug related to # in test.
Signed-off-by: Eric Blake <[email protected]>
(cherry picked from commit ab3a683f05)
* src/getargs.h, src/getargs.c (locations_flag): Remove.
* src/getargs.c, src/scan-code.l: Use muscle_percent_define_ensure
to set "locations" to true.
* src/output.c (prepare): Don't output "locations".
* src/scan-gram.l (%locations): Handle it as a %<flag>.
* src/parse-gram.y: It's no longer a token.
Don't handle it.
* data/bison.m4 (b4_locations_if): Define it with
b4_percent_define_if_define.
* data/c.m4, data/glr.cc: Adjust.
Suggested by Joel E. Denny.
* src/muscle_tab.h, src/muscle_tab.c (muscle_percent_define_ensure):
New, extracted from...
* src/parse-gram.y (prologue_declaration: pure-parser): here.
Remove it.
(prologue_declaration: "%<flag>"): Use
muscle_percent_define_ensure.
(%error-verbose, %pure-parser): No longer tokens.
* src/scan-gram.l (pure-parser): Return as a %<flag>.
* build-aux/cross-options.pl: As in --help output, write optional
arguments as [=ARG] not =[ARG].
* doc/bison.texinfo (Bison Options): Add -W/--warnings argument.
If the first m4 in $PATH is wrong, it keeps looking. Moreover, its
requirements for a correct m4 are stricter.
* m4/m4.m4: Make it a symbolic link to submodules/autoconf/m4/m4.m4.
* configure.ac: Update to use AC_PROG_GNU_M4.
Reported by Eric Blake.
* src/parse-gram.y (PERCENT_DEBUG, PERCENT_ERROR_VERBOSE): Token
definitions and associated rules, replaced by....
(PERCENT_FLAG): this new token type, and rule.
* src/scan-gram.l (RETURN_PERCENT_FLAG): New.
Use it for %debug and %error-verbose.
This allows to pass -Derror_verbose on the command line. Better yet, it
allows to pass -Derror_verbose=$(ERROR_VERBOSE), with ERROR_VERBOSE being
defined as false or true.
* data/bison.m4 (b4_percent_define_if_define): Instead of relying
on b4_percent_define_ifdef, for does not check the defined value,
but only whether the symbol is defined, rely on
b4_percent_define_flag_if, so that a value of "false" is processed
as a false.
If not defined, define the flag to "false".
(b4_error_verbose_if): New.
* data/glr.c, data/lalr1.cc, data/yacc.c: Use it instead of
b4_error_verbose_flag.
* src/getargs.h, src/getargs.c (error_verbose_flag): Remove.
* src/output.c (prepare): Don't output it.
* src/parse-gram.y (%error-verbose): Treat as %define error_verbose.
Reported by Akim Demaille at
<http://lists.gnu.org/archive/html/bug-bison/2007-12/msg00001.html>
and discussed again starting at
<http://lists.gnu.org/archive/html/bison-patches/2008-11/msg00102.html>.
* data/bison.m4 (b4_percent_define_default): Leave syncline blank
because location information is bogus.
Use angle brackets to delimit fake file name because square brackets
look like over-quoted m4. Choose a better fake file name.
Use negative line numbers.
* src/muscle_tab.c (muscle_percent_define_default): Likewise.
* src/location.c (location_print): If line for a boundary is negative,
only print that boundary's file name.
* src/location.h: Document that.
* tests/skeletons.at (%define Boolean variables: invalid skeleton
defaults): Update output.
* Makefile.am (AM_CFLAGS): Remove $(WERROR_CFLAGS) because many files
in lib won't compile with it.
* src/local.mk (src_bison_CFLAGS): Use $(WERROR_CFLAGS) here only.
GCC 4.3 now warns about "a || b && c" and asks for explicit
parentheses.
Reported by Alexandre Duret-Lutz.
* data/location.cc: Update copyright years.
(Position::operator==): Use parens to make precedence explicit.
Compare lines and columns first, as they are more likely to be
different, and they are faster to compare.
Yes, this obfuscates the point of this example, variants only. But glr.cc
cannot work (yet?) without locations. This change makes it easier to use
this example with glr.cc.
* examples/variant.yy (assert): %define it.
(locations): Request them.
(yylex): Bind the location to the stage.
* data/c++.m4 (b4_symbol_constructor_declare)
(b4_symbol_constructor_define): New empty stubs.
(b4_public_types_declare, b4_public_types_define): Use them.
* data/lalr1.cc (b4_symbol_constructor_declare)
(b4_symbol_constructor_declare_)
(b4_symbol_constructor_define_, b4_symbol_constructor_define):
Move to...
* data/variant.hh: here.
Remove the "b4_variant_if" parts, as variant.hh is loaded only if
needed.
* data/lalr1.cc: No longer invoke b4_symbol_constructor_define and
b4_symbol_constructor_declare, as it is now done by
b4_public_types_define and b4_public_types_declare.
* data/c++.m4 (b4_public_types_declare): Now define
symbol_type_base and symbol_type.
(b4_public_types_define): New.
In both cases, the definitions are taken verbatim from lalr1.cc.
* data/lalr1.cc: Adjust.
* data/c++.m4 (b4_semantic_type_declare): New.
Factors and generalizes what was in glr.cc and lalr1.cc.
* data/variant.hh (b4_semantic_type_declare): Redefine it for
variants.
* data/lalr1.cc, data/glr.cc: Use it.
* data/lalr1.cc: here.
There is no good reason to keep it private (and it is convenient
to use it from the scanner for instance). It is already public in
glr.cc.
The epilogue has its own ending \n, no need to add another.
* data/glr.c, data/lalr1.java, data/yacc.c: dnl when outputing the
epilogue.
* data/glr.cc: dnl when extending the epilogue.
Remove stray "private:".
This is a temporary band-aid until Bison gets proper alignment handling.
We need it on ARM.
* data/lalr1.cc (variant): Declare the buffer as a union to force
the same alignment as "long double".
Grep on Solaris does not support -q.
Reported by Summum Bonum.
* NEWS: Add a stub for 2.4.2.
* THANKS: Add Summum Bonum.
* tests/atlocal.in (EGREP): New.
(CC, CXX, XSLTPROC): Make it possible to override them via
envvars.
* tests/java.at: Use $EGREP instead of egrep.
Use AT_CHECK's ignore instead of grep's -q.
* TODO: Update.
* data/glr.c (yyreportSyntaxError): Use "switch" instead of
building the error message format dynamically.
* data/lalr1.java: Formatting changes.
* tests/local.mk: Replace "cd && testsuite" by "testsuite -C".
Solves problems when top_srcdir is an absolute path.
Suggested by Eric Blake.
* configure.ac: Require Autoconf 2.62.
* data/lalr1.cc: Comment changes.
* data/yacc.c (yysyntax_error): Rewrite, using a switch as in
lalr1.cc instead of building dynamically the format string.
* tests/local.at (AT_MATCHES_CHECK): New.
Based on Perl instead of Sed. Sed has too many portability
pitfalls, not ever Sed is GNU Sed.
* tests/actions.at (Fix user actions without a trailing semicolon):
Use it.
* data/java.m4 (b4_token_enum): Use the token.prefix.
* tests/local.at (AT_BISON_OPTION_PUSHDEFS): Define AT_TOKEN_PREFIX.
* tests/calc.at (_AT_DATA_CALC_Y): Use it.
Add checks for yacc.c, glr.c, lalr1.cc and glr.cc.
* tests/java.at: Comment changes.
(AT_CHECK_JAVA_MINIMAL): Define the END token.
(Java parser class and package names): Add token.prefix check.
* tests/local.mk (tests/atconfig): The rule was incorrect, but
remove it: now that there is no tests/Makefile.am, the top-level
Makefile properly updates atconfig when needed.
if it seems necessary.
* src/scan-code.l (flex rules section): Flag cpp directive from
any `#' to the first unescaped end-of-line. Semicolon is not
needed after `;', `{', '}', or cpp directives and is needed after
any other token (whitespaces and comments have no effect).
* tests/actions.at (Fix user actions without a trailing semicolon):
New test.
* tests/input.at (AT_CHECK_UNUSED_VALUES): Add semicolons to
to make user actions complete statements.
Adjust column numbers in error messages.
* tests/regression.at (Fix user actions without a trailing semicolon):
Remove. Covered by new test.
* tests/output.at (m4_expand): Don't override in newer autoconf,
where the underlying implementation changed.
* tests/cxx-type.at (_AT_RESOLVED_GLR_OUTPUT)
(_AT_RESOLVED_GLR_OUTPUT_WITH_LOC, _AT_AMBIG_GLR_OUTPUT)
(_AT_AMBIG_GLR_OUTPUT_WITH_LOC, _AT_GLR_STDERR)
(_AT_VERBOSE_GLR_STDERR): Expand to double-quoted strings,
since some of them contain unbalanced ')'.
Signed-off-by: Eric Blake <[email protected]>
* data/bison.m4 (b4_symbol_action_location): New.
* data/c.m4 (b4_symbol_actions): Remove.
Adjust all callers to use by b4_symbol_foreach and the corresponding
b4_symbol_printer/destructor macro.
* data/glr.cc: Adjust.
* data/lalr1.java: Adjust the %destructor sanity check.
* src/output.c (symbol_code_props_output): Remove, we no longer
need the b4_symbol_printers/destructors tables.
Instead of defining complex list of tuples to define various properties of
the symbols, we now prefer to define symbols as "structs" in m4: using the
symbol key (its number), and the property name, b4_symbol gives it value.
Use this to handle destructors and printers.
* src/output.c (CODE_PROP): New.
(prepare_symbol_definitions): Use it to define the printer and
destructor related attributes of the symbols.
* data/lalr1.cc (b4_symbol_actions): Rename as...
(b4_symbol_action): this.
Use b4_symbol instead of 6 arguments.
(b4_symbol_printer, b4_symbol_destructor): New.
Use them instead of b4_symbol_actions.
The real problem is rather gnulib.mk, which itself is extracted from a
Makefile.am that gnulib expects to the "recursive". The tool
prefix-gnulib-mk converts such a gnulib.mk to be non-recursive. Also,
some AC_SUBST variables need to be adjusted.
* etc/prefix-gnulib-mk: New.
* bootstrap (slurp): Use it to convert further gnulib.mk.
* lib/Makefile.am: Rename as...
* lib/local.mk: this.
Adjust to be prefixed.
* Makefile.am, configure.ac: Adjust.
* src/local.mk (AM_CPPFLAGS): Extend it, don't define it.
* data/java.m4 (b4_typed_parser_table): Rename as...
(b4_typed_parser_table_define): this, for consistency.
Accept a comment as $4.
Move $2 into yy*_.
(b4_integral_parser_table): Rename as...
(b4_integral_parser_table_define): this.
* data/lalr1.java: Adjust all uses.
Use b4_parser_tables_define instead of generation by hand.
* data/lalr1.cc (b4_tables_map): Move to...
* data/bison.m4: here.
Update the comment for yytable during the flight.
(b4_tables_declare, b4_tables_define): New.
* data/lalr1.cc: Use them.
* data/c.m4 (b4_table_define): New.
* data/yacc.c: Use b4_tables_define instead of output the tables
by hand.
* tests/regression.at (Web2c Actions): Adjust the expected output,
the order of the tables changed.
These tables are no longer needed in the parsers, and they don't seem to
be useful. They are not documented either.
* src/output.c (prepare_rules): Get rid of rhs and prhs.
Adjust the computation of (yy)r2.
TESTSUITEFLAGS is barely readable.
* tests/local.mk (TESTSUITE_FLAGS): Default to $(TESTSUITEFLAGS)
for backward compatibility.
Use the former instead of the latter.
They were used to get the symbol types, given a rule number, when
displaying the top of the stack before a reduction. But the symbol type
is available from the state stack. This has two be benefits: two tables
less in the parser (making it smaller), and a more consistent use of the
three stacks which will help to fuse them.
* data/yacc.c (yyprhs, yyrhs): Remove.
(YY_REDUCE_PRINT): Pass yyssp to yy_reduce_print.
(yy_reduce_print): Take yyssp as argument.
Use it, together with yystos, to get the symbol type.
* tests/regression.at (Web2c Report): Remove these tables from the
expected output.
The point is to factor the generation of the tables across skeletons.
This is language dependant.
* data/c.m4 (b4_comment_): New.
Should be usable to define how to generate tables independently of
the language.
(b4_c_comment): New.
(b4_comment): Bounce to b4_c_comment.
Now support $2 = [PREFIX] for indentation.
* data/lalr1.cc (b4_table_declare): Don't output a comment if
there is no comment.
Indent it properly when there is one.
Output the ending semicolon.
(b4_table_define): Space changes.
Output the ending semicolon.
(b4_tables_map): New.
Use it twice instead of declaring and defining the (integral)
tables by hand.
* data/lalr1.cc (b4_table_declare): New.
Use it to declare the tables defined with b4_table_define.
(b4_table_define): Declare a third arg to match b4_table_declare
signature.
Move all the comments around invocations of b4_table_define into
the invocations itselves.
Move things around to have the order for declarations and
definitions.
* doc/bison.texinfo (-g, -x): Add space before argument.
(Option Cross Key): Implement FIXME: listing directives also.
* build-aux/cross-options.pl: Read from <STDIN> rather than <>.
(Short Option): Special case -d. Put arguments inside @option.
(Bison Directive): Add column, automatically extracted from
src/scan-gram.l (actual name passed as the first argument)
with special case for %define.
* doc/local.mk (doc/cross-options.texi): Pass src/scan-gram.l
to build-aux/cross-options.pl.
* src/getargs.c (usage): Document limitations of cross-options.pl.
* src/scan-gram.l: Likewise.
Reported by Csaba Raduly at
<http://lists.gnu.org/archive/html/bug-bison/2008-11/msg00048.html>.
* THANKS (Csaba Raduly): Add.
* data/glr.c: Fix overquoting on b4_prefix for yylval and yylloc.
* tests/calc.at (_AT_DATA_CALC_Y): If %defines is specified, generate
lexer in a separate module that includes the defines file.
(AT_CHECK_CALC): From AT_FULL_COMPILE, request compilation of lexer
source.
* tests/local.at (_AT_BISON_OPTION_PUSHDEFS): Push AT_DEFINES_IF.
Adjust AT_LOC and AT_VAL to use AT_NAME_PREFIX.
(AT_BISON_OPTION_POPDEFS): Pop AT_DEFINES_IF.
(AT_DATA_SOURCE_PROLOGUE): New.
(AT_DATA_GRAMMAR_PROLOGUE): Use AT_DATA_SOURCE_PROLOGUE.
(AT_DATA_SOURCE): New.
(AT_FULL_COMPILE): Extend to support an additional source file.
* data/lalr1.cc (b4_subtract): Move to...
* data/bison.m4: here.
* data/glr.c (b4_rhs_data): Use it.
* data/yacc.c (b4_rhs_value, b4_rhs_location): Use it.
* doc/bison.texinfo: Synchronize ``Detail Node Listing''.
Align menus. Adjust word wrapping. Use node names for menu names.
(Examples): Don't abbreviate node names.
(LocalWords): Remove abbreviations.
(Copying): Make description a sentence.
(Java Action Features): Remove period to match the rest of menu.
* src/getargs.c (command_line_location): Set parameters to void.
* src/output.c (symbol_type_name_cmp): Make static.
(symbols_by_type_name): Set parameters to void.
(symbol_definitions_output): Remove unused parameter. Rename as...
(prepare_symbol_definitions): this.
(muscles_output): Move symbol_definitions_output to...
(output): here as prepare_symbol_definitions.
* tests/c++.at (AT_CHECK_VARIANTS): Remove unused parameters of main.
(AT_CHECK_NAMESPACE): Make unused parameter lloc unnamed.
There are two issues to handle: first scanning nested angle bracket pairs
to support types such as std::pair< std::string, std::list<std::string> > >.
Another issue is to address idiosyncracies of C++: do not glue two closing
angle brackets together (otherwise it's operator>>), and avoid sticking
blindly a TYPE to the opening <, as it can result in '<:' which is a
digraph for '['.
* src/scan-gram.l (brace_level): Rename as...
(nesting): this.
(SC_TAG): New.
Implement support for complex tags.
(tag): Accept \n, but not <.
* data/lalr1.cc (b4_symbol_value, b4_symbol_value_template)
(b4_symbol_variant): Leave space around types as parameters.
* examples/variant.yy: Use nested template types and leading ::.
* src/parse-gram.y (TYPE, TYPE_TAG_ANY, TYPE_TAG_NONE, type.opt):
Rename as...
(TAG, TAG_ANY, TAG_NONE, tag.opt): these.
* tests/c++.at: Test parametric types.
This is not sufficient, but we test at least that the make_SYMBOL
interface is not affected by token.prefix. A more general test
will be implemented when the support of token.prefix is generalized
to more skeletons.
* tests/c++.at: One more variant test, using token.prefix.
* tests/Makefile.am: Rename as...
* tests/local.mk: this.
* Makefile.am, configure.ac: Adjust.
* Makefile.am (DISTCLEANFILES): Define.
(maintainer-check, maintainer-xml-check, maintainer-push-check):
Remove, we no longer need to bounce to the real targets.
It does not work, and I don't know how it was supposed to work: it seems
to be looking for sources in the build tree. I just moved it at a better
place, fixing it is still required.
* src/local.mk (echo): Remove.
(sc_tight_scope): Move to...
* maint.mk: here.
* src/Makefile.am: Rename as...
* src/local.mk: this.
Prefix all the paths with src/.
(AUTOMAKE_OPTIONS): Build object files in the sub dirs.
(AM_CPPFLAGS): Find find in builddir/src.
(YACC): Move the flags into...
(AM_YFLAGS): here.
* maint.mk (sc_tight_scope): Disable.
It used to bounce to the version in src/Makefile.am which is now
part of this very Makefile.
* Makefile.am, configure.ac: Adjust.
* src/scan-code-c.c, src/scan-code.l: We can no longer rely on
include "..." to find files "here": we are no longer in src/, so
qualify the includes with src/.
* doc/Makefile.am (PREPATH): No longer include the top_builddir
prefix.
(.x.1): Adjust to be able to create src/foo from the top level
Makefile, instead of going bounce to src/Makefile the creation of
foo.
examples/calc++/Makefile.am might be interesting to keep as is, since
it is an example in itself.
* examples/Makefile.am: Rename as...
* examples/local.mk: this.
Adjust.
* Makefile.am, configure.ac: Adjust.
Recursive Makefiles are really way too slow, let's get rid of some of
them.
* build-aux/Makefile.am: Rename as...
* build-aux/local.mk: this.
Adjust paths.
* Makefile.am, configure.ac: Adjust.
* data/location.cc (position::position): Accept file, line and
column as arguments with default values.
Always qualify initial line and column literals as unsigned.
(location::location): Provide convenience constructors.
Using template buys us nothing, and makes it uselessly complex to
construct a symbol. Besides, it could not be generalized to other
languages, while make_FOO would work in C/Java etc.
* data/lalr1.cc (b4_symbol_): New.
(b4_symbol): Use it.
(b4_symbol_constructor_declaration_)
(b4_symbol_constructor_definition_): Instead of generating
specializations of an overloaded template function, just generate
several functions whose names are forged from the token names
without the token.prefix.
(b4_symbol_constructor_declarations): Generate them for all the
symbols, not just by class of symbol type, now that instead of
specializing a function template by the token, we generate a
function named after the token.
(b4_symbol_constructor_specialization_)
(b4_symbol_constructor_specializations): Remove.
* etc/bench.pl.in: Adjust to this new API.
Provide a means to add a prefix to the name of the tokens as output in the
generated files. Because of name clashes, it is good to have such a
prefix such as TOK_ that protects from names such as EOF, FILE etc.
But it clutters the grammar itself.
* data/bison.m4 (token.prefix): Empty by default.
* data/c.m4 (b4_token_enum, b4_token_define): Use it.
* data/lalr1.cc (b4_symbol): Ditto.
This is allows the user to get the type of a token return by
yylex.
* data/lalr1.cc (symbol::token): New.
(yytoknum_): Define when %define lex_symbol, independently of
%debug.
(yytoken_number_): Move into...
(symbol::token): here, since that's the only use.
The other one is YYPRINT which was not officially supported
by lalr1.cc, and anyway it did not work since YYPRINT uses this
array under a different name (yytoknum).
To reach good performances these functions should be inlined (yet this is
to measure precisely). To this end they must be available to the caller.
* data/lalr1.cc (b4_symbol_constructor_definition_): Qualify
location_type with the class name.
Since will now be output in the header, declare "inline".
No longer use b4_symbol_constructor_specializations, but
b4_symbol_constructor_definitions in the header.
Don't call it in the *.cc file.
The constructors are called by the make_symbol functions, which a
forthcoming patch will move elsewhere. Hence the interest of putting them
together.
The stack_symbol_type does not need to be moved, it is used only by the
parser.
* data/lalr1.cc: Move symbol_type and symbol_base_type
constructors into...
(b4_symbol_constructor_definitions): here.
Adjust.
Forthcoming changes will make it possible to use yytranslate_
from outside the parser implementation file.
* data/lalr1.cc (b4_yytranslate_definition): New.
Use it.
* data/lalr1.cc (b4_symbol_constructor_specialization_): No need
to refer to the class name to use a type defined by the class for
arguments of member functions.
This patch is debatable: the tradition expects yylex to return an int
which happens to correspond to token_number (which is an enum). This
allows for instance to return characters (such as '*' etc.). But this
goes against the stronger typing I am trying to have with the new
lex interface which return a symbol_type. So in this case, feed
yytranslate_ with a token_type.
* data/lalr1.cc (yytranslate_): When in %define lex-symbol,
expect a token_type.
The union used to compute the size of the variant used to iterate over the
type of all the symbols, with a lot of redundancy. Now iterate over the
lists of symbols having the same type-name.
* data/lalr1.cc (b4_char_sizeof_): New.
(b4_char_sizeof): Use it.
Adjust to be called with a list of numbers instead of a single
number.
Adjust its caller for new-line issues.
Symbols may have several string representations, for instance if they
have an alias. What I call its "id" is a string that can be used as
an identifier. May not exist.
Currently the symbols which have the "tag_is_id" flag set are those that
don't have an alias. Look harder for the id.
* src/output.c (is_identifier): Move to...
* src/symtab.c (is_identifier): here.
* src/symtab.h, src/symtab.c (symbol_id_get): New.
* src/output.c (symbol_definitions_output): Use it to define "id"
and "has_id".
Remove the definition of "tag_is_id".
* data/lalr1.cc: Use the "id" and "has_id" whereever "tag" and
"tag_is_id" were used to produce code.
We still use "tag" for documentation.
* data/lalr1.cc (_b4_args, b4_args): New.
Adjust all uses of locations to make them optional.
* tests/c++.at (AT_CHECK_VARIANTS): No longer use the locations.
(AT_CHECK_NAMESPACE): Check the use of locations.
* tests/calc.at (_AT_DATA_CALC_Y): Adjust to be usable with or
without locations with lalr1.cc.
Test these cases.
* tests/output.at: Check lalr1.cc with and without location
support.
* tests/regression.at (_AT_DATA_EXPECT2_Y, _AT_DATA_DANCER_Y):
Don't use locations.
* TODO (lalr1.cc/I18n): Remove.
* data/lalr1.cc (yysyntax_error_): Support the translation of the
error messages, as done in yacc.c.
Stay within the yy* pseudo namespace.
* data/lalr1.cc (b4_lex_symbol_if): New.
(parse): When lex_symbol is defined, expected yylex to return the
complete lookahead.
* etc/bench.pl.in (generate_grammar_list): Extend to support this
yylex interface.
(bench_variant_parser): Exercise it.
* data/lalr1.cc (yytranslate_): Handle the EOF case.
Adjust callers.
No longer expect yychar to be equal to yyeof_, rather, test the
lookahead's (translated) kind.
make_symbol provides a means to construct a full symbol (kind, value,
location) in a single shot. It is meant to be a Symbol constructor,
parameterized by the symbol kind so that overloading would prevent
incorrect kind/value pairs. Unfortunately parameterized constructors do
not work well in C++ (unless the parameter also appears as an argument,
which is not acceptable), hence the use of a function instead of a
constructor.
* data/lalr1.cc (b4_symbol_constructor_declaration_)
(b4_symbol_constructor_declarations)
(b4_symbol_constructor_specialization_)
(b4_symbol_constructor_specializations)
(b4_symbol_constructor_definition_)
(b4_symbol_constructor_definitions): New.
Use them where appropriate to generate declaration, declaration of
the specializations, and implementations of the templated
overloaded function "make_symbol".
(variant::variant): Always define a default ctor.
Also provide a copy ctor.
(symbol_base_type, symbol_type): New ctor overloads for value-less
symbols.
(symbol_type): Now public, so that functions such as yylex can use
it.
Test 214 was failing: it greps with a pattern containing [ ]* which
obviously meant to catch spaces and tabs, but contained only tabs.
Tabulations in sources are a nuisance, so to simplify the matter, get rid
of all the tabulations in the Java sources. The other skeletons will be
treated equally later.
* data/java.m4, data/lalr1.java: Untabify.
* tests/java.at: Simplify AT_CHECK_JAVA_GREP invocations:
tabulations are no longer generated.
* NEWS: Document them.
General Java skeleton improvements.
* configure.ac (gt_JAVACOMP): Request target of 1.4, which allows
using gcj < 4.3 in the testsuite, according to comments in
gnulib/m4/javacomp.m4.
* data/java.m4 (stype, parser_class_name, lex_throws, throws,
location_type, position_type): Remove extraneous brackets from
b4_percent_define_default.
(b4_lex_param, b4_parse_param): Remove extraneous brackets from
m4_define and m4_define_default.
* data/lalr1.java (b4_pre_prologue): Change to b4_user_post_prologue,
which marks the end of user code with appropriate syncline, like all
the other skeletons.
(b4_user_post_prologue): Add. Don't silently drop.
(yylex): Remove.
(parse): Inline yylex.
* doc/bison.texinfo (bisonVersion, bisonSkeleton): Document.
(%{...%}): Fix typo of %code imports.
* tests/java.at (AT_JAVA_COMPILE): Add "java" keyword.
Support annotations on parser class with %define annotations.
* data/lalr1.java (annotations): Add to parser class modifier.
* doc/bison.texinfo (Java Parser Interface): Document
%define annotations.
(Java Declarations Summary): Document %define annotations.
* tests/java.at (Java parser class modifiers): Test annotations.
Do not generate code for %error-verbose unless requested.
* data/lalr1.java (errorVerbose): Rename to yyErrorVerbose.
Make private. Make conditional on %error-verbose.
(getErrorVerbose, setErrorVerbose): New.
(yytnamerr_): Make conditional on %error-verbose.
(yysyntax_error): Make some code conditional on %error-verbose.
* doc/bison.texinfo (Java Bison Interface): Remove the parts
about %error-verbose having no effect.
(getErrorVerbose, setErrorVerbose): Document.
Move constants for token names to Lexer interface.
* data/lalr1.java (Lexer): Move EOF, b4_token_enums(b4_tokens) here.
* data/java.m4 (b4_token_enum): Indent for move to Lexer interface.
(parse): Qualify EOF to Lexer.EOF.
* doc/bison.texinfo (Java Parser Interface): Move documentation of
EOF and token names to Java Lexer Interface.
* tests/java.at (_AT_DATA_JAVA_CALC_Y): Remove Calc qualifier.
Make yyerror public.
* data/lalr1.java (Lexer.yyerror): Use longer parameter name.
(yyerror): Change to public. Add Javadoc comments. Use longer
parameter names. Make the body rather than the declarator
conditional on %locations.
* doc/bison.texinfo (yyerror): Document. Don't mark as protected.
Allow user to add code to the constructor with %code init.
* data/java.m4 (b4_init_throws): New, for %define init_throws.
* data/lalr1.java (YYParser.YYParser): Add b4_init_throws.
Add %code init to the front of the constructor body.
* doc/bison.texinfo (YYParser.YYParser): Document %code init
and %define init_throws.
(Java Declarations Summary): Document %code init and
%define init_throws.
* tests/java.at (Java %parse-param and %lex-param): Adjust grep.
(Java constructor init and init_throws): Add tests.
* TODO (-D): is implemented.
(associativity): Same precedence must have the same associativity.
For instance, how can a * b / c be parsed if * is %left and / is
%right?
(YYERRORCODE, YYFAIL, YYBACKUP): New.
* src/output.c (type_names_output): Document all the symbols,
including those that don't have a type-name.
(symbol_definitions_output): Define "is_token" and
"has_type_name".
* data/lalr1.cc (b4_type_action_): Skip symbols that have an empty
type-name, now that they are defined too in b4_type_names.
Small speedup (1%) on the list grammar. And makes yytranslate_ available
in non member functions.
* data/lalr1.cc (yytranslate_): Does not need to be a instance
function.
* data/c.m4: b4_comment(TEXT): Don't indent empty lines.
* data/lalr1.cc: Don't indent before rule and symbol actions, as
they can be empty, and anyway this incorrectly indents the first
action.
This is just nicer to read, I observed no speedup.
* data/lalr1.cc (yyeof_, yylast_, yynnts_, yyempty_, yyfinal_)
(yterror_, yyerrcode_, yyntokens_): Define as members of an enum.
(yyuser_token_number_max_, yyundef_token_): Move into...
(yytranslate_): here.
* etc/bench.pl.in: Inform compilation-mode when we change the
directory.
(generate_grammar_list): Recognize %define "variant" in addition
to %define variant.
Before we were using tables which lines were the symbols and which
columns were things like number, tag, type-name etc. It is was
difficult to extend: each time a column was added, all the numbers had
to be updated (you asked for colon $2, not for "tag"). Also, it was
hard to filter these tables when only a subset of the symbols (say the
tokens, or the nterms, or the tokens that have and external number
*and* a type-name) was of interest.
Now instead of monolithic tables, we define one macro per cell. For
instance "b4_symbol(0, tag)" is a macro name which contents is
self-decriptive. The macro "b4_symbol" provides easier access to
these cells.
* src/output.c (type_names_output): Remove.
(symbol_numbers_output, symbol_definitions_output): New.
(muscles_output): Call them.
(prepare_symbols): Define b4_symbols_number.
This improves the "list" bench by 2%.
* data/lalr1.cc (variant::build): Add an overloaded version with
an argument.
* tests/c++.at (AT_CHECK_VARIANT): Check it.
Unfortunately it is not possible to reuse the %prec directive. This
is because to please POSIX, we do not require to end the rules with a
semicolon. As a result,
foo: bar %prec baz
is ambiguous: either a rule which precedence is that of baz, or a rule,
and then a declaration of the precedence of the token baz.
* doc/bison.texinfo: Document %precedence.
(Precedence Only): New.
* src/assoc.h, src/assoc.c (precedence_assoc): New.
* src/conflicts.c (resolve_sr_conflict): Support it.
* src/scan-gram.l, src/parse-gram.y (%precedence): New token.
Parse it.
* tests/calc.at: Use %precedence for NEG.
* tests/conflicts.at (%precedence does not suffice)
(%precedence suffices): New tests.
* etc/bench.pl.in (@directive): New.
(&bench_grammar): Use it.
(&bench_list_grammar): New, to provide access to the "variant"
grammar.
Use it.
(getopts): Support -d, --directive.
* data/lalr1.cc (symbol_base_type, symbol_type): New.
(data_type): Rename as...
(stack_symbol_type): this.
Derive from symbol_base_type.
(yy_symbol_value_print_): Merge into...
(yy_symbol_print_): this.
Rename as...
(yy_print_): this.
(yydestruct_): Rename as...
(yy_destroy_): this.
(b4_symbols_actions, YY_SYMBOL_PRINT): Adjust.
(parser::parse): yyla is now of symbol_type.
Use its type member instead of yytoken.
* data/lalr1.cc (b4_symbol_actions): Bounce $$ and @$ to
yydata.value and yydata.location.
(yy_symbol_value_print_, yy_symbol_print_, yydestruct_)
(YY_SYMBOL_PRINT): Now take semantic value and location as a
single arg.
Adjust all callers.
(yydestruct_): New overload for a stack symbol.
To display rhs symbols before a reduction, we used information about the rule
reduced, which required the tables yyrhs and yyprhs. Now use rely only on the
state stack to get the same information.
* data/lalr1.cc (b4_rhs_data, b4_rhs_state): New.
Use them.
(parser::yyrhs_, parser::yyprhs_): Remove.
(parser::yy_reduce_print_): Use the state stack.
* data/lalr1.cc (b4_lhs_value, b4_lhs_location): Adjust to using
yylhs.
(parse): Replace yyval and yyloc with yylhs.value and
yylhs.location.
After a user action, compute yylhs.state earlier.
(yyerrlab1): Do not play tricks with yylhs.location, rather, use a
fresh error_token.
* src/getargs.c (skeleton_prio): Use default_prio rather than 2, and
remove static qualifier because others will soon need to see it.
(language_prio): Likewise.
(getargs): Use command_line_prio rather than 0.
* src/getargs.h (command_line_prio, grammar_prio, default_prio): New
enum fields.
(skeleton_prio): Extern it.
(language_prio): Extern it.
* src/parse-gram.y: Use grammar_prio rather than 1.
* src/getargs.h, src/getargs.c (skeleton_arg, language_argmatch):
The location argument is now mandatory.
Adjust all dependencies.
(getargs): Use command_line_location.
* src/getargs.c (quotearg.h, muscle_tab.h): Include.
(getargs): Define file_name.
* src/main.c (main): Initialize muscle_tab before calling
getargs.
* src/muscle_tab.c (muscle_init): No longer define file_name, as
its value is not available yet.
* build-aux/cross-options.pl: The argument ends at the first
space, not the first non-symbol character.
Use @var for each word appearing the argument description.
* data/lalr1-fusion.cc (yydestruct_): Invoke the variant's
destructor.
Display the value only if yymsg is nonnull.
(yyreduce): Invoke yydestruct_ when popping lhs symbols.
This is used to help the user catch cases where some value gets
ovewritten by a new one. This should not happen, as this will
probably leak.
Unfortunately this uncovered a bug in the C++ parser itself: the
lookahead value was not destroyed between two calls to yylex. For
instance if the previous lookahead was a std::string, and then an int,
then the value of the std::string was correctly taken (i.e., the
lookahead was now an empty string), but std::string structure itself
was not reclaimed.
This is now done in variant::build(other&) (which is used to take the
value of the lookahead): other is not only stolen from its value, it
is also destroyed. This incurs a new performance penalty of a few
percent, and union becomes faster again.
* data/lalr1-fusion.cc (variant::build(other&)): Destroy other.
(b4_variant_if): New.
(variant::built): New.
Use it whereever the status of the variant changes.
* etc/bench.pl.in: Check the penalty of %define assert.
Reported by Sergei Steshenko at
<http://lists.gnu.org/archive/html/bug-bison/2008-11/msg00001.html>.
* THANKS (Sergei Steshenko): Add.
* src/scan-code.l (SC_RULE_ACTION): Fix it.
* tests/regression.at (Fix user actions without a trailing semicolon):
New test case.
* etc/bench.pl.in (&run, &generate_grammar): New.
Rename the grammar generating functions for consistency.
Change the interface so that the list of benches to run is passed
as (optionless) arguments.
(&compile): Use &run.
* data/lalr1-fusion.cc (b4_symbol_variant): Adjust additional
arguments.
(variant::build): New overload for
copy-construction-that-destroys.
(variant::swap): New.
(parser::yypush_): Use it in variant mode.
std::list cannot be copied via memcpy, they are more demanding than
std::string. Use one std::list to strengthen the test.
* examples/variant.yy: Use lalr1-fusion.cc, not lalr1.cc.
Adjust.
Create a list of strings, instead of a single large string.
* data/bison.m4 (b4_copyright): Fix the indentation of the
copyright year paragraph.
Use b4_copyright_years when no years are given.
* data/lalr1.cc, data/lalr1-fusion.cc, data/location.cc
(b4_copyright_years): New.
Use it.
* doc/bison.texinfo (calc++.cc): Propagate failures to the exit
status.
* examples/calc++/test ($me, $number, $exit, run): New.
Use them to propagate errors to the exit status.
* etc/bench.pl.in (variant_grammar): Fix the computation of
$variant.
Generate a grammar file that can work with or without %debug.
Do use the @directive.
(bench_variant_parser): Check impact of %debug.
(@directives): Rename all the occurrences to...
(@directive): this, for consistency.
* etc/bench.pl.in: More doc.
Some fixes in the documentation.
($cflags, $iterations, &help, &getopt): New.
Use them.
(&variant_grammar): Let the number of stages be 10 times what is
specified.
* etc/bench.pl.in ($cxx, &variant_grammar, &bench_variant_parser):
New.
(&compile): Be ready to compile C++ parsers.
(&bench_push_parser): Move debug information to the outermost
level.
* THANKS: Add Michiel De Wilde.
* etc/bench.pl.in (&directives): New.
(&triangular_grammar, &calc_grammar): Use it to format the Bison
directives.
(&triangular_grammar): Do use the directives (were ignored).
(&bench_grammar, &bench_push_parser): Adjust to pass lists of
directives.
In order to make it easy to perform benchmarks to ensure that there are no
performance loss, lalr1.cc is forked into lalr1-fusion.cc. Eventually,
lalr1-fusion.cc will replace lalr1.cc.
Meanwhile, to make sure that lalr1-fusion.cc is correctly exercized by the
test suite, the user must install a symbolic link from lalr1.cc to it.
Instead of having three stacks (state, value, location), use a stack
of triples. This considerably simplifies the code (and it will be
easier not to require locations as currently does the C++ parser),
and also gives a 10% speedup according to etc/bench (probably mainly since
memory allocation is done once instead of three times).
Another motivation is to make it easier to destruct properly
semantic values: now that they are bound to their state (hence
symbol type) it will be easier to call the appropriate destructor.
These changes should probably benefit the C parser too.
* data/lalr1.cc: Copy as... * data/lalr1-fusion.cc: this new
file.
(b4_rhs_value, b4_rhs_location): New definitions overriding those
from c++.m4.
(state_stack_type, semantic_stack_type, location_stack_type)
(yystate_stack_, yysemantic_stack_, yylocation_stack_): Remove.
(data_type, stack_type, yystack_): New.
(YYLLOC_DEFAULT, yypush_): Adjust.
(yyerror_range): Now based on data_type, not location_type.
This is needed to prepare a forthcoming patch that fuses the three
stacks into one.
* data/lalr1.cc (parser::yypush_): New.
(parser::yynewstate): Change the semantics: instead of arriving to
this label when value and location have been pushed, but yystate
is to be pushed on the state stack, now the three of them must
have been pushed before. yystate still must be the new state.
This allows to use yypush_ everywhere instead of individual
handling of the stacks.
* data/lalr1.cc (b4_symbol_actions): New, overrides the default C
definition to use references instead of pointers.
(yy_symbol_value_print_, yy_symbol_print_, yydestruct_):
Take the value and location as references.
Adjust callers.
This patch was inspired by work by Michiel De Wilde. But he used Boost
variants which (i) requires Boost on the user side, (ii) is slow, and
(iii) has useless overhead (the parser knows the type of the semantic value
there is no reason to duplicate this information as Boost.Variants do).
This implementation reserves a buffer large enough to store the largest
objects. yy::variant implements this buffer. It was implemented with
Quentin Hocquet.
* src/output.c (type_names_output): New.
(output_skeleton): Invoke it.
* data/c++.m4 (b4_variant_if): New.
(b4_symbol_value): If needed, provide a definition for variants.
* data/lalr1.cc (b4_symbol_value, b4_symbol_action_)
(b4_symbol_variant, _b4_char_sizeof_counter, _b4_char_sizeof_dummy)
(b4_char_sizeof, yy::variant): New.
(parser::parse): If variants are requested, define
parser::union_type, parser::variant, change the definition of
semantic_type, construct $$ before running the user action instead
of performing a default $$ = $1.
* examples/variant.yy: New.
Based on an example by Michiel De Wilde.
To make future changes easier, no longer rely on ".TYPE" being the
way to get a semantic value.
* data/c.m4 (b4_symbol_value): New.
Use it.
* data/c++.m4, data/yacc.c: Use it.
* data/glr.c: Use b4_symbol_value.
(b4_rhs_data): New.
Use it.
# We always use _deps and _libs instead of .deps and .libs, because
# the latter is an invalid name on 8+3 MS-DOS file system. This makes
# the generated Makefiles good for every DJGPP installation, not only
# the one where the package was configured (which could happen to be
# a Windows box, where leading dots in file names are allowed).
/^rmdir[ ]*\.tst/i\
am__leading_dot=_
# Replace (command) > /dev/null with `command > /dev/null`, since
# parenthesized commands always return zero status in the ported Bash,
# even if the named command doesn't exist
/if ([^|;`]*null/{
s,(,`,
s,),,
/null[ ]*2>&1/s,2>&1,&`,
/null.*null/s,null.*null,&`,
/null.*null/!{
/null[ ]*2>&1/!s,null,&`,
}
}
# DOS-style absolute file names should be supported as well
/\*) top_srcdir=/s,/\*,[\\\\/]* | ?:[\\\\/]*,
# The following two items are changes needed for configuring
# and compiling across partitions.
# 1) The given srcdir value is always translated from the
# "x:" syntax into "/dev/x" syntax while we run configure.
/^[ ]*-srcdir=\*.*$/a\
ac_optarg=`echo "$ac_optarg" | sed "s,^\\([A-Za-z]\\):,/dev/\\1,"`
/set X `ls -Lt \$srcdir/i\
if `echo $srcdir | grep "^/dev/" - > /dev/null`; then\
srcdir=`echo "$srcdir" | sed -e "s%^/dev/%%" -e "s%/%:/%"`\
fi
# Autoconf 2.52e generated configure scripts
# write absolute paths into Makefiles and bison.in
# making them useless for DJGPP installations for
# which the package has not been configured for.
/MISSING=/,/^$/{
/^fi$/a\
am_missing_run=`echo "$am_missing_run" | sed 's%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%${top_srcdir}%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%${top_srcdir}%'`
}
/^install_sh=/a\
install_sh=`echo "$install_sh" | sed 's%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%${top_srcdir}%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%${top_srcdir}%'`
# This will only work if the assumption that the
# testsuite is ran from the following path:
# ${top_srcdir}/tests/testsuite.dir
# holds. The explicit names are of no importance.
# ../../. == ${top_srcdir}
#/^esac$/,/^ac_abs_srcdir=/ {
/^ac_abs_srcdir=/i\
ac_abs_builddir=`echo "$ac_abs_builddir" | sed "s%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../.%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../.%"`\
ac_abs_top_builddir=`echo "$ac_abs_top_builddir" | sed "s%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../..%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../..%"`\
ac_abs_top_srcdir=`echo "$ac_abs_top_srcdir" | sed "s%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../..%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../..%"`
#ac_abs_srcdir=`echo "$ac_abs_srcdir" | sed "s%/dev/.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../.%;s%.:.*/bison[-_0-9]\\{1,1\\}[-.0-9A-Za-z+]*%../../.%"`
/^ac_abs_srcdir=/a\
for dj_dir in $ac_abs_builddir $ac_abs_top_builddir $ac_abs_top_srcdir $ac_abs_srcdir\
do\
ac_abs_builddir=`echo "$ac_abs_builddir" | sed "s%//%/%g"`\
ac_abs_top_builddir=`echo "$ac_abs_top_builddir" | sed "s%//%/%g"`\
ac_abs_top_srcdir=`echo "$ac_abs_top_srcdir" | sed "s%//%/%g"`\
ac_abs_srcdir=`echo "$ac_abs_srcdir" | sed "s%//%/%g"`\
done
# Add DJGPP version information.
/^#define VERSION/s/\$VERSION/& (DJGPP port (r1))/
# We need makeinfo to make the html formated docs.
Copyright (C) 2018-2019 Free Software Foundation, Inc.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with no Front-Cover Texts, and with no Back-Cover
Texts. A copy of the license is included in the "GNU Free
Documentation License" file as part of this distribution.
# LocalWords: mfcalc calc parsers yy ispell american
--->
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Akim Demaille
Eduard StaniloiuandAkim Demaille
Akim DemailleandAkim Demaille
Paul HilfingerandAkim Demaille
JannickandAkim Demaille
Adam SampsonandAkim Demaille
Josh SorefandAkim Demaille
Paul Eggert
Jiahao LiandAkim Demaille
Nate GuerinandAkim Demaille
Andreas SchwabandAkim Demaille
Valentin TolmerandAkim Demaille
Valentin TolmerandAkim Demaille
Dennis HeimbignerandAkim Demaille
Petr MachataandAkim Demaille
Theophile Ranquet
Theophile RanquetandAkim Demaille
Victor SantetandAkim Demaille
Stefano LattariniandAkim Demaille
Tim LandscheidtandAkim Demaille
Akim DemailleandAkim Demaille
Jim MeyeringandAkim Demaille
Joel E. Denny
Joel E. Denny
Angelo BorsottiandJoel E. Denny
Bernd KieferandJoel E. Denny
Alex Rozenman
Paul Hilfinger
Jonathan NiederandJoel E. Denny
Eric Blake
Alex Rozenman
Joel E. Denny
Samuel TardieuandAkim Demaille
Di-an Jan
Akim DemailleandAkim Demaille