- Boost - lists.boost.org

Problem with BOOST_NO_CXX11_VARIADIC_TEMPLATES in gcc.hpp
by Richard Hadsell 30 Aug '14

30 Aug '14

I think I found a problem in config/compiler/gcc.hpp in Boost 1.56.0. I was trying to come up with a workaround for a bug having to do with variadic templates, when compiling with g++ 4.5.1 (on Linux Fedora 14). Modifying line 160 to turn on BOOST_NO_CXX11_VARIADIC_TEMPLATES for g++ versions earlier than 4.7 had not effect. I discovered that the version test in line 160 has no effect, because it is in a block (lines 153-164) that only pertains to g++ 4.2 or earlier, if compiled with -std=c++0x. The block for versions 4.3 and later (lines 145-151) always define BOOST_HAS_VARIADIC_TMPL and never BOOST_NO_CXX11_VARIADIC_TEMPLATES. I think that is a mistake, at least partially, because proposal N2555 is only supported in version 4.4. At any rate, someone should look at these lines and make some sense of them. The GCC support table shows general support for variadic templates in 4.3 and support for N2555 (with a long name) in 4.4. Furthermore, the bug that produces a message like this: "sorry, unimplemented: cannot expand 'Args ...' into a fixed-length argument list" was only fixed in g++ 4.7.0 (according to a bug report). That issue could also use a guard macro, as was done in one of the boost libraries ( BOOST_CONTAINER_UNIMPLEMENTED_PACK_EXPANSION_TO_FIXED_LIST). -- Dick Hadsell 203-992-6320 Fax: 203-992-6001 Reply-to: hadsell(a)blueskystudios.com Blue Sky Studioshttp://www.blueskystudios.com 1 American Lane, Greenwich, CT 06831-2560

4 8

[exception] Demangling support not merged to master?
by Andrey Semashev 30 Aug '14

30 Aug '14

Hi, Sadly, I discovered that the support for type name demangling in Boost.Exception did not make it into 1.56. It's still not merged to master. The tests look green, could this change be merged to master?

2 3

[move] boost::move incorrect documentation?
by Mostafa 30 Aug '14

30 Aug '14

Here http://www.boost.org/doc/libs/1_56_0/doc/html/boost/move_idp159681920.html it states: "This function provides a way to convert a reference into a rvalue reference in compilers with rvalue references. For other compilers converts T & into ::boost::rv<T> & so that move emulation is activated." But in <boost/move/utility.hpp> I see: template <class T> inline typename ::boost::move_detail::enable_if_c < enable_move_utility_emulation<T>::value && !has_move_emulation_enabled<T>::value, T&>::type move(T& x) { return x; } Therefore shouldn't the above be: "This function provides a way to convert a reference into a rvalue reference in compilers with rvalue references. For other compilers if T is Boost.Move enabled type then it converts T & into ::boost::rv<T> & so that move emulation is activated, else it returns T &."

2 1

Boost Build and -Wall
by Emil Dotchevski 29 Aug '14

29 Aug '14

I see that in Boost Build <warnings>on translates to -Wall. What value do I specify for <warnings> to get the compiler-default warnings level? -- Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode

3 4

Boost Build Qt5 support on MACOSX?
by Emil Dotchevski 29 Aug '14

29 Aug '14

Hello, I downloaded Qt and installed it into its default location on mac os x, then tried to initialize it with: using qt5 : /users/<user_name>/Qt/5.3/clang_64 ; I don't know if the above is the correct way to initialize the qt5 support, but I get bad include directories, e.g. #include <QtGui/QWindow> won't work. Searching the Qt directories reveals that the file QWindow is found under Qt/5.3/clang_64/lib/QtGui.framework/Versions/5/Headers. So what is the right way to do this? -- Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode

1 0

[units] Interest in UDLs (e.g. _kg) for units?
by Curdeius Curdeius 28 Aug '14

28 Aug '14

Hi all, I was wondering if there is any interest in adding User-Defined Literals (UDLs) to Boost.Units. IMO, it could simplify some code and improve code readability in some cases. For example instead of writing (it's a somehow artificial example) 5.34 * si::milli * si::kilogram; 1234 * si::kilogram; one would write 5.34_g; 1234_kg; I propose to add this feature for compatible compilers (where BOOST_NO_CXX11_USER_DEFINED_LITERALS is not defined). The user would have to pull corresponding namespace (let's call it `udl`) to use it, for example using namespace boost::units::systems::si::udl; // bring _kg, _km, ... into scope What are your feelings about it? In the first place, I would go for `operator ""` taking as arguments `unsigned long long` and `long double`. Later, one could think about the one taking a `const char *` argument which would provide a quantity with an underlying type of arbitrary precision (like boost::multiprecision::cpp_int). Regards, Marek Kurdej -- Ph.D. candidate

5 5

Re: [boost] [units] Interest in UDLs (e.g. _kg) for units?
by Marek Kurdej 28 Aug '14

28 Aug '14

Looks much nicer! > Will reduce the user resistance to using Boost.Units - it's a bit verbose > for mere engineers and scientist's taste ;-) > But have you implemented it yet? > (and are there any hidden gotchas? - the proposed C++ time literals > suffixes proved to be a can of worms :-( > Paul I have a partial implementation for mass and length units. You can have a look at it : https://github.com/mkurdej/units/tree/feature/user-defined-literals You guess right, there are some problems to solve with output of quantities created with literals. See lines with FIXME in test/test_literals.cpp For those who were sceptic about "megagrams" and so on: that was just an example. Actually, I implemented tonnes (t), kilotonnes (kt), Gt and so on. Still, someone could possibly use Mg, Gg for uniformity... (but I haven't found such a person yet). Marek Kurdej -- Ph.D. candidate

1 0

Compiling with MPI support on Linux using clang and libc++
by Gonzalo BG 28 Aug '14

28 Aug '14

Dear boost users/devs, has anyone had any luck compiling the latest boost from source with mpi support on linux using clang and libc++? I'm trying to set up a project on travis-ci and am running into ABI issues (cxxabi files not found,etc). I could try to use libstdc++ instead but would prefer to keep on using libc++. If anyone has been successful on setting this up on travis-ci and has a .travis.yml that I could have a look at I would be really grateful. Bests, Gonzalo

1 0

[move] Implicit conversion to rvalues and move ctors.
by Mostafa 28 Aug '14

28 Aug '14

Here's another use case that fails to compile in C++03 but compiles successfully in C++11. (Note, it fails to compile in VS2005 and g++ 4.8.2 -std=c++98, and compiles successfully with g++ 4.8.2 -std=c++11) MOC is a BOOST_MOVABLE_BUT_NOT_COPYABLE type. Foo is an empty with an implicit conversion function to MOC. "MOC m(Foo());" compiles successfully but "MOC m2(makeFoo());" fails to compile. Is this a bug or a limitation of Boost.Move in C++03? //---------- //Start Code //---------- #include <boost/move/core.hpp> #include <boost/move/utility.hpp> struct MOC { MOC() : value(121) {} MOC(BOOST_RV_REF(MOC) rhs) : value(rhs.value) {} int value; private: BOOST_MOVABLE_BUT_NOT_COPYABLE(MOC) }; struct Foo { operator MOC () { return MOC(); } }; Foo makeFoo() { return Foo(); } int main() { // (1) Successfully Compiles. MOC m(Foo()); // (2) Fails to Compile. MOC m2(makeFoo()); return 0; } //-------- //End Code //-------- //-------------------- //Start Error Messages //-------------------- boost_move_implicit_conversion_to_rvalue.cc: In function ‘int main()’: boost_move_implicit_conversion_to_rvalue.cc:32:19: error: no matching function for call to ‘MOC::MOC(Foo)’ MOC m2(makeFoo()); ^ boost_move_implicit_conversion_to_rvalue.cc:32:19: note: candidates are: boost_move_implicit_conversion_to_rvalue.cc:10:3: note: MOC::MOC(MOC&) BOOST_MOVABLE_BUT_NOT_COPYABLE(MOC) ^ boost_move_implicit_conversion_to_rvalue.cc:10:3: note: no known conversion for argument 1 from ‘Foo’ to ‘MOC&’ boost_move_implicit_conversion_to_rvalue.cc:7:3: note: MOC::MOC(boost::rv<MOC>&) MOC(BOOST_RV_REF(MOC) rhs) : value(rhs.value) {} ^ boost_move_implicit_conversion_to_rvalue.cc:7:3: note: no known conversion for argument 1 from ‘Foo’ to ‘boost::rv<MOC>&’ boost_move_implicit_conversion_to_rvalue.cc:6:3: note: MOC::MOC() MOC() : value(121) {} ^ boost_move_implicit_conversion_to_rvalue.cc:6:3: note: candidate expects 0 arguments, 1 provided //------------------ //End Error Messages //------------------

2 3

[review] Formal review period for VMD library begins today, Aug 21, and ends Sat, Aug 30
by Steven Watanabe 27 Aug '14

27 Aug '14

AMDG The formal review of the VMD library by Edward Diener starts today, Aug 21st and is scheduled to continue until Aug 30th. About VMD =============== The Variadic Macro Data library is a library of variadic macros which enhance the functionality in the Boost preprocessor library ( Boost PP ). The main focus of the library is the parsing/identifying of a subset of C++ preprocessor data types and of all Boost PP data types. This allows enhanced preprocessor metaprogramming based on the type of macro input. The library may also be seen as a repository for future enhancements to the Boost PP library functionality using variadic macros. Where to get it =============== The library is available on github at https://github.com/eldiener/variadic_macro_data Review guidelines ================= Reviews should be submitted to the developer list (boost(a)lists.boost.org) with '[vmd]' in the subject. Or if you don't wish to for some reason, you can send them privately to me. If so, please let me know whether or not you'd like your review to be forwarded to the list. For your review you may wish to consider the following questions: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? And finally, every review should answer this question: - Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. In Christ, Steven Watanabe

10 35