- Boost-users - lists.boost.org

iterator_facade and gcc-4.3
by Benoit Hudson 08 Nov '07

08 Nov '07

"For example, when defining a constant lvalue iterator, the user can pass a const-qualified version of the iterator's value_type as iterator_facade's Value parameter and omit the Reference parameter which follows." This no longer works in current gcc 4.3 (previous issue I mentioned was a compiler bug, now fixed): it works out to generating a const lvalue in operator[], which makes gcc-4.3 give "error: type qualifiers ignored on function return type" This is either a bug in boost or a bug in gcc; not being clear on which it really is, I'll file it under boost. -- Benoît

3 3

Template type to constant string
by Robert Dailey 07 Nov '07

07 Nov '07

Hi, As you know, with Microsoft VS preprocessor operators you can convert a type passed in to a macro to a string. For example: #define TO_STRING( type ) #type I would like to do the same thing with a template type: template< typename t_type > void TO_STRING() { // Do something here with t_type to make it a string and then return it } Is there something in boost that can facilitate the behavior I'm describing? Thanks.

4 9

[boost-users] [filesystem, format] ambgious overload error.
by chun ping wang 07 Nov '07

07 Nov '07

{ boost::scoped_ptr<boost::filesystem::ofstream> myfile(new boost::filesystem::ofstream(fileOutput)); BOOST_FOREACH(int element, (getKeys<std::vector, int, int>(pMap))) { if (pMap[element] >= min_sup) { std::vector<int> row(1, element); itemsContainer.push_back(row); // (*myfile) << "(" << element << ":" << pMap[element] << ")\n"; (*myfile) << boost::format("(%1%, %2%)\n") % element % pMap[element]; } } } ./cs522/c++/aprioriMain.cpp(79) : error C2593: 'operator <<' is ambiguous C:/boost_msvc/include\boost/format/free_funcs.hpp(33): could be 'std::basic_ostream<_Elem,_Traits> &boost::operator <<<char,std::char_traits<char>,std::allocator<_Ty>>(std::basic_ostream<_Elem,_Traits> &,const boost::basic_format<Ch> &)' [found using argument-dependent lookup] with [ _Elem=char, _Traits=std::char_traits<char>, _Ty=char, Ch=char ] C:/boost_msvc/include\boost/filesystem/path.hpp(617): or 'std::basic_ostream<_Elem,_Traits> &boost::filesystem::operator <<<boost::basic_format<Ch>>(std::basic_ostream<_Elem,_Traits> &,const Path &)' [found using argument-dependent lookup] with [ _Elem=char, _Traits=std::char_traits<char>, Ch=char, Path=boost::basic_format<char> ] while trying to match the argument list '(boost::filesystem::ofstream, boost::basic_format<Ch>)' with [ Ch=char ] note the commented out version works.. this only happens on msvc 2005. it passes on mingw devc++.

1 1

Compile issue with boost::archive::xml_oarchive
by Gerald Fordyce 07 Nov '07

07 Nov '07

Hi, I am attempting to use boost::serialization to serialize a simple derived class to an XML archive. I have no problem creating a text archive - code compiles and the text archive contains both base class and derived class members. If I switch to the boost::archive::xml_oarchive it no longer compiles. Here is the code creating the archive: const DerivedClass obj; std::string filename = "fooBar.xml"; std::ofstream ofs(filename.c_str()); assert(ofs.good()); { boost::archive::xml_oarchive oa(ofs); oa << obj; } Here is the gcc-4.1.2 compiler output (Ubuntu): /usr/local/include/boost-1_34/boost/archive/basic_xml_oarchive.hpp: In member function ‘void boost::archive::basic_xml_oarchive<Archive>::save_override(T&, int) [with T = const DerivedClass, Archive = boost::archive::xml_oarchive]’: /usr/local/include/boost-1_34/boost/archive/detail/interface_oarchive.hpp:79: instantiated from ‘Archive& boost::archive::detail::interface_oarchive<Archive>::operator<<(T&) [with T = const DerivedClass, Archive = boost::archive::xml_oarchive]’ mkevents.cpp:81: instantiated from here /usr/local/include/boost-1_34/boost/archive/basic_xml_oarchive.hpp:83: error: invalid application of ‘sizeof’ to incomplete type ‘boost::STATIC_ASSERTION_FAILURE<false>’ So I'm wondering if the requirements for using txt_oarchive are different than for xml_oarchive? It seems they should both work the same. Thanks Gerald

2 1

[boost] [review] Exception - result
by Tobias Schwinger 07 Nov '07

07 Nov '07

I'm pleased to announce that the Exception library has been accepted into boost. Thanks to all formal and informal reviewers and to Emil Dotchevski for the library! Summary of possible improvements: ================================= Design ------ Nearly all reviewers made suggestions for syntactic improvements. The "tag(value) syntax" (as suggested by John Torjo) for instance provides sufficient syntactic sugar and seems lightweight to implement. Suggested features in this area are (those are implementable on top of the current design - but will keep users from reinventing the wheel if provided by the library): - Support for grouping information, and - support for generating stack traces. I personally still have a bad feeling about 'enable_error_info' - let's just revisit and discuss it on the list before changing boost::throw_exception. The related discussions about clone/rethrow support is still ongoing. Documentation ------------- Best practices for the catch-site should be clarified in the documentation. Same goes for using the same tag multiple types with one exception object. Other comments should be addressed (I won't put much energy into summarizing documentation comments, guessing most of them might have been taken care of, already). Deployment ---------- There was a suggestion (from Pavel Vozenilek) to put #ifdef around the code of boost::exception that could be put into a separate translation unit. There should probably also be a 'build' directory and a Jamfile so there is smooth integration with the Boost build system. Of course, Jamfiles for test/examples should be added. Implementation -------------- Several reviewers noted that the library is rather allocation-heavy. It pretty much comes down to shared_ptr overuse (causing many unnecessary allocations of small objects and synchronized instructions to be executed) though not in performance critical code possibly affecting other threads and future allocations. -- Tobias Schwinger - Review Manager - _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/ listinfo.cgi/boost

2 1

[xpressive] un-used variables
by Jeff Flinn 07 Nov '07

07 Nov '07

We compile our app with -Wunused-variables on the Mac with gcc 4.0. We get several unused variable warnings due to <boost/xpressive/regex_primitives.hpp>. Is there some facility for quieting these from xpressive itself? We currently are using #pragma unused ( eol, ...) to quiet the list of about 15 unused variables. Thanks, Jeff Flinn

2 4

[BGL] property_map not compiling because it's not l-value?
by Ferng, Andrew D 07 Nov '07

07 Nov '07

Hi, I'm trying to use the concepts in the Kevin Bacon example in my code and am having trouble with compiling. http://www.boost.org/libs/graph/doc/kevin_bacon.html When I compiled the example by itself, it ran fine. However when I changed it to my data, then it gave me a compile error of "error C2106: '=' : left operand must be l-value" I don't know why this is, other than variable name, I just changed the vertex type from string to index. Nothing that I can see would cause this. The line not compiling is "grid_id[u] = gid;" See below for code. ************************************************************************ **************************** typedef property_map<ConnectionGraph, vertex_index_t>::type vertexID_MapType; typedef property_map<ConnectionGraph, edge_index_t>::type edgeID_MapType; vertexID_MapType grid_id = get(vertex_index, interfaceGraph); edgeID_MapType connectingElement_id = get(edge_index, interfaceGraph); typedef graph_traits<ConnectionGraph>::vertex_descriptor Vertex; typedef std::map<long,Vertex> NameVertexMap; NameVertexMap edges; for(CELAS2::iterator itr = CELAS2s.begin();itr != CELAS2s.end(); ++itr ) { long gid = (*itr)->getSideA_1Grid(); NameVertexMap::iterator pos; bool inserted = false; Vertex u, v; tie(pos, inserted) = edges.insert(std::make_pair(gid, Vertex())); if (inserted) { u = add_vertex(interfaceGraph); grid_id[u] = gid; <<<<-------- ************ this is the line that VS is saying non l-value. **************** pos->second = u; } else u = pos->second; } ************************************************************************ *************************** Any ideas why this is? I'm confused as to what's causing this. Any other way to populate the property map? I'm using VS2005 and boost 1.33.1. Thanks!

2 3

ublas::vector<A>
by e r 07 Nov '07

07 Nov '07

hi, is it possible to do ublas::vector<A> vec; where A is a class that i define, not a built in type? if so, what are the requirements on A?

1 0

boost.python switching from v2.5 to v2.4
by desai3＠cooper.edu 06 Nov '07

06 Nov '07

Hi all: I have used Boost.Python successfully to wrap around C++ API in python2.5. I must say that boost.python is an amazing library. However, if I want to create the same API for python2.4 do I have to rebuild the boost with python2.4 and then run bjam again, or is there any simple solution than that? The reason for asking this is because changing the Jamrules file did not help much. Thank you for your help. Maulik

2 1

[filesystem] Apparent non-compliant code in current operations.hpp, ...
by Marty Fried 06 Nov '07

06 Nov '07

While building 1_34_1 on various platforms we support, I came across some compile errors for the AIX compiler, xlC/vacpp that weren't caught by gcc or MSVC, possibly due to being hidden by a level of indirection. The errors were mostly undefined classes where the template class was fully defined at the time of instantiation. I apologize if my explanation isn't clear or totally correct - I'm not an expert on the C++ standard, and got a lot of the actual explanation from coworkers that know it better than I do. But it's up to me to do the actual reporting and to try to get it to work for us on all the platforms we support. My original post was a few days ago, with the subject "[filesystem] vacpp errors building 1_34_1 filesystem library on AIX". I am because it may look like a vacpp problem, which is an unsupported compiler, but it's not really. It just happens that xlC was the only compiler that caught the error, I think. My original post can also be read at this link: http://permalink.gmane.org/gmane.comp.lib.boost.user/31028 The problem is that the class basic_directory_iterator is not fully defined until after it is used by remove_all(). There were a few other similar errors which I can provide if someone takes an interest in this. If I am wrong, or if there is a reason this won't be fixed, it would be helpful to know, so we can decide what to do. -- Marty Fried Senior Software Developer, Aldon USA http://www.aldon.com

3 7