- Boost-users - lists.preview.boost.org

Boost.Python Feature Request
by Jenna Louis 16 Jan '04

16 Jan '04

Are there any patches or plans to add the following features to Boost.Python/Pyste? 1. A major usability problem of Boost generated proxies is that IDEs are not able to show any details about the expected parameters. Since Boost already knows everything about the function then it makes sense to at least make some information known about it via a docstring and then these tools can at least show that. 2. Another feature would append a docstring with the descriptive text to the specified Class, Function, or Method. 3. I often generate modules that require a very specific order of initialization for the proxies it contains. I often have to rearrange the _main.cpp's BOOST_PYTHON_MODULE code as depicted below to get my module to work. This: BOOST_PYTHON_MODULE(Sim) { Export_SzCoordPos(); Export_SzCoordAng(); Export_SzCoord(); Export_SzMath(); Export_SzPolyUtils() } To This: BOOST_PYTHON_MODULE(Sim) { Export_SzCoord(); Export_SzMath(); Export_SzPolyUtils() Export_SzCoordPos(); Export_SzCoordAng(); } It would be nice if this list would be generated in the order in which the pyste files are specified on the command line. 4. I would like a way to add pickling code to my class from pyste. I realize its not practical to generate this code automatically but I would at least like to be able to inline this in my pyste file. I can already insert code into the proxy with the declaration_code() feature to handle the pickle struct but I also need to be able to add a pickle method i.e.(".def_pickle(world_pickle_suite())" to the class definition I specify. Thanks, Jenna --------------------------------- Do you Yahoo!? Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes

2 1

RE: [Boost-users] Re: interest in a glob_iterator? (directory_ite rator with regex)
by Jim.Hyslop 15 Jan '04

15 Jan '04

Angus Leeming [mailto:angus.leeming@btopenworld.com] wrote: > Jeff Flinn wrote: > > And a more descriptive, less jargon-ish name. > > Why? Rich is proposing something that would iterate over the files > returned by the unix 'glob' function. As much as you might hate to admit it, you must recognize that the world does not revolve around UNIX (and before anyone claims otherwise, nor does the world revolve around Windows). To my knowledge, "glob" is not a standard computer term, nor a standard C++ term, it is a standard UNIX term. Now, I could be wrong, but my understanding of the ISO standardization process is make everything platform-independent (and keep in mind that one of the original goals of Boost was to be included in the Standard). The *concept* of the "glob" is pretty much platform independent, but the term is not, so it should have (to paraphrase Jeff) a more descriptive, less platform-specific name. -- Jim Hyslop Senior Software Designer Leitch Technology International Inc. (<http://www.leitch.com/>) Columnist, C/C++ Users Journal (<http://www.cuj.com/experts>)

3 2

boost::scoped_ptr and std::auto_ptr interoperability
by Sam Partington 15 Jan '04

15 Jan '04

Hi all, Is there any reason why boost::scoped_ptr::reset doesn't provide a std::auto_ptr overload? Take the following code : class A {}; std::auto_ptr<A> factory(); // ... boost::scoped_ptr<A> a; a.reset(factory().release()); wheras a.reset(factory()); would be neater. (Trivial difference I know, but it does add to readability) Sam --LongSig ****************** This e-mail has been sent from Imagination Technologies Limited. PowerVR, Metagence, Ensigma and PURE Digital are divisions of Imagination Technologies Limited. The information contained in this e-mail, including any attachment, is confidential and may be legally privileged. It is intended solely for the addressee(s) and access to this e-mail by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying or distribution or use of the information contained in this e-mail, is prohibited and may be unlawful. If you have received this e-mail in error, please notify the sender by return e-mail and then delete it from your system. Internet communications cannot be guaranteed to be secure, error or virus-free. The sender does not accept liability for any errors or omissions which arise as a result. Any views expressed in this message are those of the author, except where the author specifies and, with authority, states them to be the views of Imagination Technologies Ltd.

1 0

boost on Mac OS X 10.3 (Panther)
by Vineet Mehta 15 Jan '04

15 Jan '04

Has anyone been able to successfully build and install boost on Mac OS X 10.3 (panther). I am using gcc 3.3, the latest xcode. The python variables were configured to: export PYTHON_ROOT=/System/Library/Frameworks/Python.framework/Versions/2.3 export PYTHON_VERSION=2.3 I have bjam installed. I used bjam "-sTOOLS=darwin" in the boost root directory. However at the yields a number of warning and error messages, and ends with : ...failed updating 70 targets... .. .. Any suggestions ? -Vineet

8 9

Problems with Reverse Cuthill-Mckee?
by Kris Vorwerk 14 Jan '04

14 Jan '04

Hello, I'm a fairly new Boost user, so please forgive me if I'm completely out of whack here. I'm presently using Boost's reverse Cuthill-Mckee ordering to reduce the bandwidth of some very large symmetric, positive definite sparse matrices that I'm investigating. Generally, the cuthill_mckee_ordering() has worked quite well, although a short while ago, I began noticing that it occasionally gave an invalid ordering (i.e., the permutations spit out by the algorithm would have multiple vertexes all the same, which, when used to permute my matrices, would destroy my problem). It took me a while to figure out what was going on, but I've finally distilled a small enough example that exhibits the problem. Consider the following matrix: Matrix: 6x6 [ [3.66667,0,-0.5,0,0,0], [0,2.66667,0,-0.222222,0,0], [-0.5,0,1.66667,0,0,-0.222222], [0,-0.222222,0,3.33333,-2.22222,0], [0,0,0,-2.22222,4.66667,0], [0,0,-0.222222,0,0,4.66667] ] Using fairly straightforward code, I loop through the non-zeros, and add edges to an undirected Boost graph (basically the same as the cuthill_mckee_ordering.cpp example in the docs). Here is the graph that gets created for the above problem, in Graphviz format: graph G { 0; 1; 2; 3; 4; 5; 0--0 ; 0--2 ; 1--1 ; 1--3 ; 2--2 ; 2--5 ; 3--3 ; 3--4 ; 4--4 ; 5--5 ; } And, if I run this example using the cuthill_mckee_ordering technique (or if I modify cuthill_mckee_ordering.cpp to use these vertices and edges), I get the following permutation (using the pseudo-peripheral heuristic): Inv_perm: 0 0 0 0 2 5 You can see, from the Graphviz output, that this graph has two disconnected sets of vertices, which is where I think the problem lies. It may be a problem in my understanding of how to setup the Boost graph, or perhaps a problem in the reordering technique. You'll notice that, in my example, I've eliminated the duplicate edges (such as 2-5 being the same as 5-2) -- I've tried adding them in, too, but it hasn't changed the result. I've noticed that, if one of my matrices produces a graph in which there are no disconnected sets of vertices, the RCM ordering works fine. It's only when the graph is disconnected in this fashion that the ordering fails. I've also noticed that, when using this kind of disconnected graph, if I sometimes select a random starting vertex (instead of the pseudo-peripheral pair heuristic), the cuthill_mckee_ordering segfaults. I've attached some sample code below that seems to exhibit the problem. I'm certainly open to the possibility that it's a problem in my understanding of how to use the library. Also, my Boost version is: 1.30.2 (Redhat RPM) Any thoughts would be appreciated. regards, Kris ============================================================== (borrowed and slightly modified from cuthill_mckee_ordering.cpp) #include <boost/config.hpp> #include <vector> #include <iostream> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/cuthill_mckee_ordering.hpp> #include <boost/graph/properties.hpp> #include <boost/graph/bandwidth.hpp> int main( int, char *[] ) { using namespace boost; using namespace std; typedef adjacency_list < vecS, vecS, undirectedS, property < vertex_color_t, default_color_type, property < vertex_degree_t, int > > > Graph; typedef graph_traits < Graph >::vertex_descriptor Vertex; typedef graph_traits < Graph >::vertices_size_type size_type; typedef std::pair < std::size_t, std::size_t > Pair; //graph G {0;1;2;3;4;5;0--0 ;0--2 ;1--1 ;1--3 ;2--2 ;2--5 ;3--3 ;3--4 ;4--4 ;5--5 ;} Pair edges[] = { Pair( 0, 0 ), Pair( 0, 2 ), Pair( 1, 1 ), Pair( 1, 3 ), Pair( 2, 2 ), Pair( 2, 5 ), Pair( 3, 3 ), Pair( 3, 4 ), Pair( 4, 4 ), Pair( 5, 5 )}; Graph G( 6 ); for( int i = 0; i < 10; ++i ) { std::cout << edges[i].first << ", " << edges[i].second << std::endl; add_edge( edges[i].first, edges[i].second, G ); } graph_traits < Graph >::vertex_iterator ui, ui_end; property_map < Graph, vertex_degree_t >::type deg = get( vertex_degree, G ); for( boost::tie( ui, ui_end ) = vertices( G ); ui != ui_end; ++ui ) { deg[*ui] = degree( *ui, G ); } property_map < Graph, vertex_index_t >::type index_map = get( vertex_index, G ); std::cout << "original bandwidth: " << bandwidth( G ) << std::endl; std::vector < Vertex > inv_perm( num_vertices( G ) ); std::vector < size_type > perm( num_vertices( G ) ); // reverse cuthill_mckee_ordering cuthill_mckee_ordering( G, inv_perm.rbegin(), get( vertex_color, G ), make_degree_map( G ) ); cout << "Reverse Cuthill-McKee ordering:" << endl; cout << " "; for( std::vector < Vertex >::const_iterator i = inv_perm.begin(); i != inv_perm.end(); ++i ) { cout << index_map[*i] << " "; } cout << endl; for( size_type c = 0; c != inv_perm.size(); ++c ) { perm[index_map[inv_perm[c]]] = c; } std::cout << " bandwidth: " << bandwidth( G, make_iterator_property_map( &perm[0], index_map, perm[0] ) ) << std::endl; return 0; }

2 3

Re:[Boost-users] interest in a glob_iterator? (directory_ite
by rlyon＠comgroup.net.au 14 Jan '04

14 Jan '04

I can use it. Hopefully you will use regex++ from boost to handle the regular expressions. Regards ... ____________________Reply Separator____________________ Subject: [Boost-users] interest in a glob_iterator? (directory_iterat Author: Rich Johnson <rjohnson(a)dogstar-interactive.com> Date: 13/01/2004 6:45 PM Folks-- Is there general interest in a "globbing" iterator? If so I've got one I'm willing to re-package for submission. (Or at least discuss with folks for improvement)

1 0

tuple problems
by programmer blabla 14 Jan '04

14 Jan '04

Dear Booster's I have a problem compiling a tuple containing an enum in a namespace. For example: namespace TestTuple { enum ETestTuple { eTestTuple0 = 0, eTestTuple1 = 1 }; } void Testtuple5() { typedef boost::tuple<TestTuple::ETestTuple> Tuple; Tuple tpl; } does not compile. I am using MSVC 2003.net (7.1 version), just native code. Is this a M$ problem or is tuple broken? Also I get problem with the combination of boost::bind and tuple, for example: int i; boost::tuple<int> data; i = data.get<0>(); //ok i = boost::tuples::get<0>(data); //ok i = boost::bind(&boost::tuples::get<0>, _1)(data); //error Has anyone a clue? Btw searching lists.boost.org is very hard, because google is out of sync with the list. With kind regards, Me.

1 0

int64_t in boost/cstdint.h with GCC -std=c++98 or -ansi
by Aaron W. LaFramboise 14 Jan '04

14 Jan '04

Hi, I noticed that boost/cstdint.h does not define int64_t when GCC is in strict mode (-ansi or -std=c++98) on MinGW, Cygwin, and probably other major platforms. This seems to be because boost/cstdint.h refuses to define the type if ULLONG_MAX (or ULONG_LONG_MAX or ULONGLONG_MAX) is not defined, even if BOOST_HAS_LONG_LONG is defined. GCC, quite correctly, will not define these macros in strict mode. However, the library still probably should be defining this type. Ideally, there would be a GCC-specific work-around to prevent warnings on the use of long long in strict mode. In particular, I am using the date_time library which depends on int64_t. It would be very unfortunate and a little silly if I could not use date_time in GCC's strict mode. How would it be best to fix this? Aaron

2 2

d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(31) : error C2039: 'item' : is not a member of 'Node'
by pixellent＠stodge.net 14 Jan '04

14 Jan '04

Hi, I'm trying to use Luabind in my application, and Luabind uses BOOST. I'm getting the following error: d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(31) : error C2039: 'item' : is not a member of 'Node' d:\development\ogre\ogrenew\ogremain\include\ogrenode.h(50) : see declaration of 'Node' d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(31) : error C2146: syntax error : missing ';' before identifier 'type' d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(31) : error C2208: 'struct boost::type' : no members defined using this type d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(32) : error C2039: 'next' : is not a member of 'Node' d:\development\ogre\ogrenew\ogremain\include\ogrenode.h(50) : see declaration of 'Node' d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(32) : error C2146: syntax error : missing ',' before identifier 'next' d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(32) : error C2065: 'next' : undeclared identifier d:\sdk\boost\boost\mpl\list\aux_\iterator.hpp(33) : see reference to class template instantiation 'boost::mpl::list_iterator<Node>' being compiled d:\development\ogre\ogrenew\plugins\ogrelua\luabind\luabind\detail\primitives.hpp(91) : fatal error C1506: unrecoverable block scoping error Error executing cl.exe. Nature.obj - 7 error(s), 0 warning(s) Is this a bug in BOOST or am I doing something silly? Thanks Mike

1 0

Seeking permission to include Boost with book CD
by Matthew Wilson 13 Jan '04

13 Jan '04

Hi chaps / chappettes I am interested in including the latest released copy of Boost with the CD for my forthcoming book, Imperfect C++, to be published by Addison-Wesley this year. I have two questions: 1. Is Boost redistributable in this form? If so, from whom do I get written permission to do so? 2. What is the likely latest version of Boost that will be available around the end of Feb / beginning of March? Thanks in advance -- Matthew Wilson Director, Synesis Software (www.synesis.com.au) STLSoft moderator (http://www.stlsoft.org) Contributing editor, C/C++ Users Journal (www.synesis.com.au/articles.html#columns) -----------------------------------------------------

2 2

re: API documentation ?
by Vikram Anantha 13 Jan '04

13 Jan '04

Is there any place where I can look at the complete API documentation for the boost libraries. Something generated by doxygen or doc++ or something like that. Thanks Vik

2 1

cross compiling boost libs
by Uwe Sander 12 Jan '04

12 Jan '04

Hi, I would like to compile the boost libs for the Win32 platform using mingw on Linux. I have the required stuff installed and I know how to do such a cross compilation if I have a 'configure' based project but I do not know how to tell bjam what I am trying to achieve. Any hint? Thanks Uwe

2 1

Boost Include headers
by jling＠cmmt.ubc.ca 12 Jan '04

12 Jan '04

I need to include the following line in my code. #include "boost/type_traits/object_traits.hpp" My source file is in the root directory of the boost package that I extracted. But, I get errors like: boost/type_traits/object_traits.hpp:19: boost/type_traits/ice.hpp: No such file or directory boost/type_traits/object_traits.hpp:22: boost/type_traits/fwd.hpp: No such file or directory boost/type_traits/object_traits.hpp:25: boost/type_traits/composite_traits.hpp: No such file or directory etc ... I noticed that in say "boost/type_traits/ice.hpp" it also has header files it refers to, that look in "boost/type_traits/" directory, but relative to that header file, ice.hpp, there is no such directory path ... only from the root where my source file is, does that path exist. How does this work and how to I get my code to compile? Thanks, John

4 3

Using Smart Assert Library with MSVC 7.0 and 7.1
by Ferentinos, Jerry 12 Jan '04

12 Jan '04

Hi All, I'm having some trouble getting John Torjo's Smart Assert library to work with MSVC 7.0 and 7.1. Here's the error I get when compiling the example in boost_root\libs\smart_assert\example\enforce.cpp on MSVC 7.0 (7.1 yields a similar error): --snip-- c:\Program Files\Microsoft Visual Studio .NET\Vc7\include\vector(95) : error C2664: 'std::_Vector_val<_Ty,_Alloc>::_Vector_val(_Alloc)' : cannot convert parameter 1 from 'const std::_Vector_val<_Ty,_Alloc>::_Alty' to 'boost::smart_assert::Private::smart_assert_alloc<type>' with [ _Ty=boost::smart_assert::assert_context::val_and_str, _Alloc=boost::smart_assert::Private::smart_assert_alloc<boost::smart_assert: :assert_context::val_and_str> ] and [ _Ty=boost::smart_assert::assert_context::val_and_str, _Alloc=boost::smart_assert::Private::smart_assert_alloc<boost::smart_assert: :assert_context::val_and_str> ] and [ type=boost::smart_assert::assert_context::val_and_str ] No constructor could take the source type, or constructor overload resolution was ambiguous c:\Program Files\Microsoft Visual Studio .NET\Vc7\include\vector(93) : while compiling class-template member function 'std::vector<_Ty,_Ax>::vector(const std::vector<_Ty,_Ax>::_Myt &)' with [ _Ty=boost::smart_assert::assert_context::val_and_str, _Ax=boost::smart_assert::Private::smart_assert_alloc<boost::smart_assert::as sert_context::val_and_str> ] c:\boost_1_30_0\boost\smart_assert\priv\fwd\smart_assert_context.hpp(155) : see reference to class template instantiation 'std::vector<_Ty,_Ax>' being compiled with [ _Ty=boost::smart_assert::assert_context::val_and_str, _Ax=boost::smart_assert::Private::smart_assert_alloc<boost::smart_assert::as sert_context::val_and_str> ] --snip-- I hand instantiated "const std::_Vector_val<_Ty,_Alloc>::_Alty" and I got "boost::smart_assert::Private::smart_assert_alloc<type>" (where _Ty, _Alloc, and type have been replaced by their actual parameter values) so why is MSVC reporting an error? Has anyone else come across this problem? If so, how did you solve it? I tried searching the archive and came up with nothing, mostly because clicking on a search result brings you to an altogether different message. Any clues? Jerry Ferentinos ------------------------------------------------------------------------------ Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD) that may be confidential, proprietary copyrighted and/or legally privileged, and is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please immediately return this by e-mail and then delete it. ------------------------------------------------------------------------------

1 0

RE: [Boost-users] Boost Include headers
by Ben Hutchings 12 Jan '04

12 Jan '04

jling(a)cmmt.ubc.ca wrote: > I need to include the following line in my code. > > #include "boost/type_traits/object_traits.hpp" > > My source file is in the root directory of the boost package that I > extracted. > > But, I get errors like: > boost/type_traits/object_traits.hpp:19: boost/type_traits/ice.hpp: No > such file or directory <snip> Don't put your source file in the Boost directory. Put it somewhere else and add the Boost directory to the include search path (on the command line this is usually done with a -I option).

1 0

globbing with regex??
by Rich Johnson 12 Jan '04

12 Jan '04

I'm new to regex and would like to use it to match filenames, but I'd like to stick with a csh/bash syntax. Is this advisable? If so, would anyone have a set of regex traits for this syntax? If not, could someone point me in the right direction? Thanks, --rich An sample usage is: namespace fs = boost::filesystem char * dirname; /*...glob_regex_traits definition */ boost::regex expr( "*.cpp" , /*...glob_regex_traits...*/ ); for( fs::directory_iterator iter( fs:path( dirname ) ) ; iter != fs::directory_iterator() ; ++iter ) { boost::cmatch match_data if( regex_match( iter->leaf().c_str(), match_data, expr ) ) { /*...do something...*/ } }

2 1

RE: [Boost-users] Weird freeze with boost::thread on Win2K
by Yarwood, Martin (UK) 12 Jan '04

12 Jan '04

Have you done a sleep(0) or similar to give up thread control so the threading control gives the other one some processor time? > -----Original Message----- > From: Viktor Lundström [mailto:d99-vlu@nada.kth.se] > Sent: 09 January 2004 10:47 > To: boost-users(a)lists.boost.org > Subject: [Boost-users] Weird freeze with boost::thread on Win2K > > > *** WARNING *** > > This mail has originated outside your organization, > either from an external partner or the Global Internet. > Keep this in mind if you answer this message. > > Hi! > I'm using boost::thread on MSVC6/Win2k. > When I start two processes which both use boost::thread, > the second process waits for up to a minute or two before starting. > The process doesn't bug, it just hangs there for a while and then > starts. > Is this a known problem? > > Regards, > Viktor Lundström > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > ******************************************************************** This email and any attachments are confidential to the intended recipient and may also be privileged. If you are not the intended recipient please delete it from your system and notify the sender. You should not copy it or use it for any purpose nor disclose or distribute its contents to any other person. ********************************************************************

1 0

Weird freeze with boost::thread on Win2K
by Viktor Lundström 10 Jan '04

10 Jan '04

Hi! I'm using boost::thread on MSVC6/Win2k. When I start two processes which both use boost::thread, the second process waits for up to a minute or two before starting. The process doesn't bug, it just hangs there for a while and then starts. Is this a known problem? Regards, Viktor Lundström

3 2

NEWBIE QUESTIONS ABOUT BGL
by Hugues Joly 09 Jan '04

09 Jan '04

Hi, I've read most of the online documentation about the Boost Graph Library and I've tried a few examples but some questions remain unanswered: Properties: - Once we have a property map for vertices or edges, do we have to assign a value to each vertex (or edge) of the graph? (My preliminary tests tell me that we have to.) - Is the graph the only mean to iterate through all the elements of a property map that it owns? - What is the difference between a graph and a property graph: when would we choose to use a graph interface instead of a property graph? - How can I attach a local property to an element of a subgraph and still have the original property unchanged for the corresponding element on the parent graph? (At the time beeing it seems impossible to do so.) - How can I define a custom property within a specific namespace? PROBLEMS: It seems that there are compilers' limitations regarding the number of properties (and their value type) that can be attached to an adjacency_list (at least with CodeWarrior). Thanks in advance. __________________________________________________________ Lèche-vitrine ou lèche-écran ? magasinage.yahoo.ca

3 2

MAKING A SEARCHABLE ARCHIVE OF THE boost-users' MAILING LIST
by Hugues Joly 09 Jan '04

09 Jan '04

Hi, I think that a searchable archive would be a big help in providing newcomers with answers on basic topics. At least, a downloadable archive would do the job. Who could take care of that? __________________________________________________________ Lèche-vitrine ou lèche-écran ? magasinage.yahoo.ca

2 1

RE: [Boost-users] MAKING A SEARCHABLE ARCHIVE OF THE boost-users' MAILINGLIST
by John Wilkinson 09 Jan '04

09 Jan '04

This list is available through Gmane (http://www.gmane.org) which is an NNTP interface for mailing lists. They have an archive of about 5100 messages for this list. Other boost lists are also available (http://news.gmane.org/index.php?prefix=gmane.comp.lib.boost) John -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Hugues Joly Sent: Friday, January 09, 2004 10:53 AM To: boost-users(a)lists.boost.org Subject: [Boost-users] MAKING A SEARCHABLE ARCHIVE OF THE boost-users' MAILINGLIST Hi, I think that a searchable archive would be a big help in providing newcomers with answers on basic topics. At least, a downloadable archive would do the job. Who could take care of that? __________________________________________________________ Lèche-vitrine ou lèche-écran ? magasinage.yahoo.ca _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

Building from MSVC
by Phil Hornby 09 Jan '04

09 Jan '04

Okay I know that there is this nice build tool called bjam - but I am a simple man and like using my nice friendly IDEs....so I was wondering if anyone has an MSVC 7.1 project that they have set up to build the libraries, i.e. the dlls that bjam produces? If so can I grab a copy as my first quick try didn't work...but then I only put 5 mins into it... -- Phil "Requirements - what are they I just hack something together that does what I think they want" ;)

1 0

boost::signals + pthreads = seg fault on Linux
by Uwe Sander 09 Jan '04

09 Jan '04

Hi, I have a problem with boost::signals when I try to use them in multi threaded projects. I have tried gcc 3.2.3 and gcc 3.3.2, on two different Linux boxes, with the same result. The following test program works fine w/o threads and crashes after the 17th call to connect if I add -pthread to the compile command: #include <boost/signal.hpp> void on_do_something () { } int main () { boost::signal<void ()> sig; for (int i = 0; i < 100; ++i) sig.connect (&on_do_something); } I have tried electric fence, then it crashes immediatly in the first call to pthread_mutex_init, but I can't see any reason for that crash with gdb. Once again, even with electric fence, everything works fine w/o threads, so I see no mem corruption or any reason why it should not work _with_ threads. Any idea? Thanks Uwe

2 3

Test library fails to link on OS X 10.3 Panther
by Scott Lamb 08 Jan '04

08 Jan '04

I've downloaded boost-1.30.2 and boost-jam-3.1.7 to my PowerBook running Mac OS X 10.3 (Panther). After installing boost-jam without problems, I'm attempting to build the boost test library as follows: $ uname -a Darwin spiff.local 7.2.0 Darwin Kernel Version 7.2.0: Thu Dec 11 16:20:23 PST 2003; root:xnu/xnu-517.3.7.obj~1/RELEASE_PPC Power Macintosh powerpc $ gcc_select Current default compiler: gcc version 3.3 20030304 (Apple Computer, Inc. build 1495) $ cd boost-1.30.2/libs/test/build $ bjam -sTOOLS=darwin It chugs for a while, but gets link failures. It gives a warning about undefined symbols and dies. The symbols are cpp_main, test_main, and init_unit_test_suite. It doesn't really look like this is a fatal error - it says it's just a warning and that prebinding is disabled. And I think these are to be defined by the user of the library, so isn't it normal for them to be undefined here? In any case, this is the only error I see other than bjam complaining that the command failed. Any ideas how to fix it? Thanks, Scott Lamb The first of the failed commands did this: darwin-Link-DyLib-action ../../../libs/test/build/bin/libboost_prg_exec_monitor. dylib/darwin/debug/runtime-link-static/shared-linkable-true/ libboost_prg_exec_mo nitor.dylib ld: warning prebinding disabled because of undefined symbols ld: Undefined symbols: cpp_main(int, char**) export DYLD_LIBRARY_PATH ld -dynamic -m -r -d -o "../../../libs/test/build/bin/libboost_prg_exec_moni tor.dylib/darwin/debug/runtime-link-static/shared-linkable-true/ libboost_prg_exe c_monitor.lo" "../../../libs/test/build/bin/libboost_prg_exec_monitor.dylib/dar win/debug/runtime-link-static/shared-linkable-true/execution_monitor.o" "../../. ./libs/test/build/bin/libboost_prg_exec_monitor.dylib/darwin/debug/ runtime-link- static/shared-linkable-true/cpp_main.o" \ && \ c++ -static-libgcc -g -Wl,-dynamic -nostartfiles -Wl,-dylib -Wl,-ldylib1.o -Wl,-dylib_compatibility_version,1.30.2 -W,l-dylib_current_version,1.30.2 -o ". ./../../libs/test/build/bin/libboost_prg_exec_monitor.dylib/darwin/ debug/runtime -link-static/shared-linkable-true/libboost_prg_exec_monitor.dylib" "../../../li bs/test/build/bin/libboost_prg_exec_monitor.dylib/darwin/debug/runtime- link-stat ic/shared-linkable-true/libboost_prg_exec_monitor.lo" \ \ \ && \ rm -f "../../../libs/test/build/bin/libboost_prg_exec_monitor.dylib/darwin/d ebug/runtime-link-static/shared-linkable-true/ libboost_prg_exec_monitor.lo" ...failed darwin-Link-DyLib-action ../../../libs/test/build/bin/libboost_prg_exe c_monitor.dylib/darwin/debug/runtime-link-static/shared-linkable-true/ libboost_prg_exec_monitor.dylib...

2 1

Problems compiling Boost.Function using gcc-3.3 on MacOS X (Jaguar)
by Jens Lindblad 08 Jan '04

08 Jan '04

Hello, I am trying to move a project of mine from gcc-3.1 to gcc-3.3 under MacOS X Jaguar. I'm using Apple's developer tools with the December 2002 updater which included gcc-3.3. Everything compiles just fine under 3.1 but I really need to use precompiled headers which does not work very well. I have managed to precompile my prefix header using 3.3 and I can also compile most of my source files using it. However, as soon as the compiler encounters a source file where Boost.Function is used I get an internal compiler error. So, I wrote a small program (included at the end of this email) which demonstrates the error. It works just fine under 3.1 but not under 3.3. Using preferred or portable syntax makes no difference. Has anyone else experienced this problem, or has it been fixed in Xcode? Perhaps I'm just doing something wrong? I'd really like to know before I file a bug. (I did a couple of searches on lists.boost.org but didn't find anything of interest.) Thanks in advance, Jens ------------------------------------------------------------ #include <iostream> #include <boost/function.hpp> void test( int x ) { std::cout << x << std::endl; } int main( int argc, char* argv[] ) { boost::function<void (int)> __test(test); __test(5); return 0; } ------------------------------------------------------------ Which gives the following error: /Development/Libraries/boost-1.30.2/boost/function/ function_template.hpp: In member function `void boost::function1<R, T0, Allocator>::assign_to(FunctionPtr, boost::detail::function::function_ptr_tag) [with FunctionPtr = void (*)(int), R = void, T0 = int, Allocator = std::allocator<void>]': /Development/Libraries/boost-1.30.2/boost/function/ function_template.hpp:432: instantiated from `void boost::function1<R, T0, Allocator>::assign_to(Functor) [with Functor = void (*)(int), R = void, T0 = int, Allocator = std::allocator<void>]' /Development/Libraries/boost-1.30.2/boost/function/ function_template.hpp:293: instantiated from `boost::function1<R, T0, Allocator>::function1(Functor, typename boost::detail::function::enable_if<boost::type_traits::ice_not<boost::is _same<Functor, int>::value>::value, int>::type) [with Functor = void (*)(int), R = void, T0 = int, Allocator = std::allocator<void>]' /Development/Libraries/boost-1.30.2/boost/function/ function_template.hpp:602: instantiated from `boost::function<R ()(T0), Allocator>::function(Functor, typename boost::detail::function::enable_if<boost::type_traits::ice_not<boost::is _same<Functor, int>::value>::value, int>::type) [with Functor = void (*)(int), R = void, T0 = int, Allocator = std::allocator<void>]' test.cpp:12: instantiated from here /Development/Libraries/boost-1.30.2/boost/function/ function_template.hpp:451: internal compiler error: in c_expand_expr, at c-common.c:4752 Please submit a full bug report, with preprocessed source if appropriate. See <URL:http://developer.apple.com/bugreporter> for instructions.

1 0