- Boost - lists.boost.org

[local] "true" lambdas without C++11 but only on GCC
by Lorenzo Caminiti 17 Dec '11

17 Dec '11

Hello all, A curiosity: On GCC compilers (but _only_ on GCC compilers) it is possible to combine Boost.Local and the non-standard GCC "expression statements" feature to implement lambda functions. In other words, it is possible to define a Boost.Local closure within a GCC expression statement: std::vector<int> v(3); v[0] = 1; v[1] = 2; v[2] = 3; int factor = 10; std::for_each(v.begin(), v.end(), LAMBDA(int x, const bind& factor) { std::cout << x * factor << std::endl; } LAMBDA_END ); This is just a curiosity because it is NOT standard C++03 and it will ONLY work with GCC compilers. #include <boost/local/function.hpp> #include <vector> #include <algorithm> #include <iostream> // WARNING: This macros only work on GCC!!! #define LAMBDA_WITH_RESULT(result_type, ...) \ ({ /* open GCC statement expression */ \ result_type BOOST_LOCAL_FUNCTION_PARAMS(__VA_ARGS__) #define LAMBDA(...) \ LAMBDA_WITH_RESULT(void, __VA_ARGS__) #define LAMBDA_END \ BOOST_LOCAL_FUNCTION_NAME(BOOST_PP_CAT(lambda, __LINE__)) \ BOOST_PP_CAT(lambda, __LINE__); \ }) /* close GCC statement expression */ int main() { std::vector<int> v(3); v[0] = 1; v[1] = 2; v[2] = 3; int factor = 10; std::for_each(v.begin(), v.end(), LAMBDA(int x, const bind& factor) { std::cout << x * factor << std::endl; } LAMBDA_END ); return 0; } I might mention this code example as a curiosity in Boost.Local docs, currently it's not mentioned there. Thanks. --Lorenzo

2 3

[log] sourceforge trunk broken with clang
by Michael Caisse 17 Dec '11

17 Dec '11

Changeset 633 fails to compile with clang. The trivial logger example in tutorial will trigger multiple errors. My guess is that simply including trivial.hpp will do the same. -- Michael Caisse Object Modeling Designs www.consultomd.com

2 1

New libraries implementing C++11 features in C++03
by Jeffrey Lee Hellrung, Jr. 17 Dec '11

17 Dec '11

Within the discussion for the review of the proposed Boost.Local library, Hartmut Kaiser raised a concern that I think should be addressed more broadly by the community, if possible. I quote Hartmut: > Ok. However this raises a more serious question. Should we as the > Boost community still encourage solutions and libraries solely for > portability with ancient compilers? I'd say no, but YMMV. Boost will > be still around 2, 5, or 10 years from now. What's the utility of > adding such a _solely_ backwards oriented library from this POV? Indeed, I ask the community, do such libraries belong in Boost? If so, is the bar for acceptance of such libraries automatically and necessarily higher than libraries that introduce genuinely new capabilities in C++11? If so, what additional criteria must such a library meet? I add two additional notes to help put the discussion in context: - Within the proposed Boost.Local review discussion, several individuals pointed out that they will likely have to work professionally on C++03 compilers for at least a few more years. - Recall that Boost has several (recent) libraries that could easily be labelled as "transitional" libraries: Move, Atomic (at least proposed, not sure if accepted), Container. Further, several older libraries are now part of the C++11 standard (e.g., Thread (right?)). - Jeff

20 45

Speeding up set and map insertion
by Tweety 16 Dec '11

16 Dec '11

Hi, I think it would speed up inserting range of elements into set and map, when in tree.hpp we replace: template <class InputIterator> void insert_equal(InputIterator first, InputIterator last) { //Insert with end hint, to achieve linear //complexity if [first, last) is ordered const_iterator end(this->cend()); for( ; first != last; ++first) this->insert_equal(end, *first); } with: template <class InputIterator> void insert_equal(InputIterator first, InputIterator last) { //Insert with hint, to achieve linear //complexity if data is even partially ordered const_iterator it(this->cend()); for( ; first != last; ++first) it = this->insert_equal(it, *first); } I will make almost linear complexity when data is ordered ascending, descending, and even partially ordered. Because when inserting with end hint (btw. wasn't here a bug? because "end" wasn't updated) it's only linear if we insert elements in ascending order.

2 1

[crc] Does anyone know who did issue #2492?
by Daryle Walker 16 Dec '11

16 Dec '11

I ask because I don't see any bug; his/her example works just fine for me. I need a complete (minimal!) program that shows the error. See more at [https://svn.boost.org/trac/boost/ticket/2492]

1 0

[graph] running test leda_graph_cc.cpp fails - header not found
by Jens Müller 15 Dec '11

15 Dec '11

Hi all, I tried to run the tests for the LEDA adapter in the Boost Graph library. I checked out the latest trunk from SVN. I made the Boost.Build v2, works fine. When running ./../../tools/build/v2/bjam in libs/graph/test, all tests run fine. I installed LEDA (free version) in /home/jens/devel/leda-free/LEDA-6.3-free-fedora-core-8-64-g++-4.1.2-mt After successfully running all other tests, the one for LEDA fails: $ ../../../tools/build/v2/bjam -s LEDA=/home/jens/devel/leda-free/LEDA-6.3-free-fedora-core-8-64-g++-4.1.2-mt/ Performing configuration checks - has_icu builds : yes ...patience... ...patience... ...patience... ...found 6503 targets... ...updating 2 targets... gcc.compile.c++ ../../../bin.v2/libs/graph/test/leda_graph_cc.test/gcc-4.6.1/debug/leda_graph_cc.o In file included from leda_graph_cc.cpp:10:0: ../../../boost/graph/leda_graph.hpp:20:30: fatal error: LEDA/graph/graph.h: No such file or directory compilation terminated. "g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -fPIC -DBOOST_ALL_NO_LIB=1 -I"../../.." -c -o "../../../bin.v2/libs/graph/test/leda_graph_cc.test/gcc-4.6.1/debug/leda_graph_cc.o" "leda_graph_cc.cpp" ...failed gcc.compile.c++ ../../../bin.v2/libs/graph/test/leda_graph_cc.test/gcc-4.6.1/debug/leda_graph_cc.o... ...failed updating 1 target... ...skipped 1 target... Note that I edited boost/graph/leda_graph.hpp, fixing a syntax error and making it point to the now location of the LEDA header files (which contains an additional "graph" sub-directory). But the problem seems to be that the include directory is not at all passed to the compiler. What is going wrong here? Best, Jens

2 2

[graph] LEDA adapter
by Jens Müller 15 Dec '11

15 Dec '11

Hi all, are there still people who want to use the LEDA adapter in the Graph library? Back in 2007, I contributed adapters for some property maps. The checkin unfortunately introduced a typo (let's call it a transmission error ...) which is still there. So apparently no-one used it since then?! There also have been some discussions on removing the support, and patch requests were denied due to a lack of maintainer ... If there is _any_ interest in still maintaining leda_graph.hpp, I'd do it, as Douglas had already offered to me in 2007. I have more time and a better understanding now, and I'd really like to dive a bit more into Boost development. Besides, there apparently now is a free (as in beer) version of LEDA available, which should facilitate more widespread testing. Regards, Jens

3 5

Request: sanity check for a jamfile
by Daryle Walker 15 Dec '11

15 Dec '11

I just added a new test file for Boost.CRC on the SVN trunk. I also tried to adapt the test jamfile to accommodate two test programs. I'm not sure that I did it right, so can anyone check that it works. (I use XCode to build and test my programs.) The new file, "crc_test2.cpp", uses the unit-testing apparatus. The old file, "crc_test.cpp", uses the header-only minimal Testing facility. (I'm not sure that its link to the Test Execution monitor, in the jamfile, is needed.) I added a new testing file because I want to add a test case for a bug and Boost.Test has advanced so much that the first file is way too obsolete to be safely extended. It's change #75959 on the trunk, a.k.a. [https://svn.boost.org/trac/boost/changeset/75959]

2 1

[python] [concept] CallPolicies implementation
by Lorenzo Caminiti 15 Dec '11

15 Dec '11

Hello all, Does Boost.Python implement the CallPolicies concept? I couldn't find it... http://www.boost.org/doc/libs/1_48_0/libs/python/doc/v2/CallPolicies.html If not, would something like this be OK?? template<class P> struct CallPolicies { BOOST_CONCEPT_ASSERT((ResultConverterGenerator<P::result_converter>)); BOOST_CONCEPT_ASSERT((Metafunction<P::extract_return_type>)) BOOST_CONCEPT_USAGE(CallPolicies) { bool b = x.precall(a); // convertible to bool PyObject* o = x.postcall(a, r); // convertible to PyObject* } private: P x; PyObject* a; PyObject* r; }; I'd have to implement ResultConverterGenerator and Metafunction as well. Thanks. --Lorenzo

2 1

[parameter] compiler errors with operators
by Lorenzo Caminiti 15 Dec '11

15 Dec '11

Hello all, Does Boost.Parameter support operators? The following example is from the docs but it doesn't compile. Looking at the preprocessed code /EP, the errors seem to be because the macro defines a struct template postfixing it with the operator name and symbol "template <class Args> struct boost_param_result_11operator() { ... };". Maybe this worked once before and it's now broken? #include <boost/parameter.hpp> #include <iostream> BOOST_PARAMETER_NAME(arg1) BOOST_PARAMETER_NAME(arg2) struct callable2 { BOOST_PARAMETER_CONST_MEMBER_FUNCTION( (void), operator(), tag, (required (arg1,(int))(arg2,(int)))) { std::cout << arg1 << ", " << arg2 << std::endl; } }; int main ( void ) { callable2 c; c(1, 2); return 0; } On MSVC 8 with latest Boost trunk: c02.cpp c02.cpp(11) : error C2988: unrecognizable template declaration/definition c02.cpp(11) : error C2059: syntax error : 'type' c02.cpp(11) : error C2334: unexpected token(s) preceding '{'; skipping apparent function body c02.cpp(11) : error C2988: unrecognizable template declaration/definition c02.cpp(11) : error C2059: syntax error : 'type' c02.cpp(11) : error C2334: unexpected token(s) preceding ':'; skipping apparent function body c02.cpp(11) : error C2143: syntax error : missing ';' before '<' c02.cpp(11) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int c02.cpp(11) : error C2238: unexpected token(s) preceding ';' c02.cpp(11) : error C2143: syntax error : missing ';' before '<' c02.cpp(11) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int c02.cpp(11) : error C2238: unexpected token(s) preceding ';' c02.cpp(11) : error C2988: unrecognizable template declaration/definition c02.cpp(11) : error C2059: syntax error : '<' c02.cpp(11) : error C3861: 'boost_param_parameters_11operator': identifier not found c02.cpp(11) : error C2039: 'type' : is not a member of '`global namespace'' c02.cpp(11) : error C3861: 'boost_param_parameters_11operator': identifier not found c02.cpp(11) : error C2039: 'type' : is not a member of 'boost::parameter::aux::match' ..\..\..\..\..\boost\trunk.windows\boost/parameter/preprocessor.hpp(99) : see declaration of 'boost::parameter::aux::match' c02.cpp(11) : error C2039: 'type' : is not a member of '`global namespace'' c02.cpp(11) : error C2334: unexpected token(s) preceding '{'; skipping apparent function body c02.cpp(14) : error C2270: 'boost_param_default_11operator' : modifiers not allowed on nonmember functions c02.cpp(14) : error C2091: function returns function c02.cpp(20) : error C2064: term does not evaluate to a function taking 2 arguments Thanks. --Lorenzo

2 1