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 <akim@lrde.epita.fr>
* 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.
