October 2005 - Boost-users - lists.preview.boost.org

Boost Graph Library and Boehm Garbage Collector
by FBSL 08 Sep '08

08 Sep '08

Hello I am having trouble to garbage-collect adjacency_list<>. I have instanciated it with correctly garbage-collected vectors and lists (I have copied vecS and listS definitions, and added Boehm GC's default allocator so that they are garbage-collected). In my application I need to dynamically allocate many graphs, but it does not work, there is a (big) memory leak. Unfortunately, there is no template argument in adjacency_list<> to specify Boehm GC's allocator. Does anyone have a hint ? Snippet code "main.cpp" follows, compiled with Boost Graph Library 1.32, Boehm GC 6.4 and G++ 3.3.5, with command line : $ g++ -o main main.cpp -lgc -lgccpp -- #include <boost/graph/adjacency_list.hpp> #include <gc/gc_allocator.h> #include <gc/gc_cpp.h> #include <iostream> #include <vector> #include <list> using namespace std; using namespace boost; namespace boost { struct collected_vecS {}; struct collected_listS {}; template <class ValueType> struct container_gen<collected_vecS, ValueType> { typedef vector<ValueType,gc_allocator<ValueType> > type; }; template <class ValueType> struct container_gen<collected_listS, ValueType> { typedef list<ValueType,gc_allocator<ValueType> > type; }; template <> struct parallel_edge_traits<collected_vecS> { typedef allow_parallel_edge_tag type; }; template <> struct parallel_edge_traits<collected_listS> { typedef allow_parallel_edge_tag type; }; namespace detail { template <> struct is_random_access<collected_vecS> { enum { value = true }; typedef true_type type; }; } } typedef adjacency_list<collected_vecS, collected_vecS, directedS, no_property, no_property, no_property, collected_listS> graph; int main() { // leaks... while(true){ graph * g=new (UseGC) graph; for(unsigned int i=0; i<100; ++i){ add_vertex(*g); } for(unsigned int i=0; i<100; ++i){ for(unsigned int j=0; j<100; ++j){ add_edge(i,j,*g); } } } }

3 4

[statechart] history transitions make state-local-storage "un-cool"
by Mick Hollins 26 Jun '08

26 Jun '08

Apologies if this has been discussed before, but I couldn't find any mention of it: I am currently using the history facility of boost::fsm. I've discovered that when I re-enter a state, X, via a history transition I don't actually re-enter the same instance of X that I was in before, but instead enter a newly constructed instance of X. I just checked the UML spec regarding history transitions and it says that "Any necessary entry actions are performed", so given that boost::fsm uses constructors to represent entry actions I guess a new instance of X has to be constructed to ensure UML compliance. My actual concern is the effect this has on the "coolness" of state-local-storage (see http://boost-sandbox.sourceforge.net/libs/statechart/doc/faq.html#Whats_so_…) It seems to me that state-local-storage, which I agree is generally cool, is particularly un-cool (and bug-inducing) in the presence of history transitions. I'm not sure what a good alternative is, but just thought I'd raise the discussion. cheers, mick PS: does the history facility display the same semantics in boost::statechart? I plan to move to statechart real soon now ...

4 19

BGL: INTERNAL COMPILER ERROR in VC7.1
by Yariv Tal 10 Jun '07

10 Jun '07

This one was already posted by Nicholas Burlett awhile back (see http://lists.boost.org/MailArchives/boost-users/msg06534.php) but I have some new info that may help. The problem: ========= When compiling code that uses the boost.graph library the VC7.1 compiler shouts: c:\boost_1_32_0\boost\mpl\aux_\preprocessed\plain\apply_wrap.hpp(48) : fatal error C1001: INTERNAL COMPILER ERROR As Nicholas already noted, this seems to be due to some compiler problem with the boost.graph out_edge_iterators since it is caused by the following lines in my code: (FilteredGraphGenerator::FilteredGraph is a typedef of a boost::filtered_graph) typedef boost::graph_traits<FilteredGraphGenerator::FilteredGraph> ::out_edge_iterator oedge_iterator; oedge_iterator edgesItr, edgesEnd; boost::tie(edgesItr, edgesEnd) = boost::out_edges(itr->first, m_g.m_filteredGraph); while (edgesItr != edgesEnd) { // do anything! } In fact, it seems to be related to the operator!= between iterators (if the while loop is commented out it compiles). How ever, the error persists even after changing the operator!= to operator== : while (edgesItr == edgesEnd) { // do anything! } ========================================================================== Some findings: =========== * We used to use boost_1_30_0 and the code used to compile PERFECTLY. The problem only appeared after migrating to boost_1_32_0. * Calling directly to the "equal" function in iterator_core_access instead of the operator== works. typedef boost::graph_traits<FilteredGraphGenerator::FilteredGraph> ::out_edge_iterator oedge_iterator; oedge_iterator edgesItr, edgesEnd; boost::tie(edgesItr, edgesEnd) = boost::out_edges(itr->first,m_g.m_filteredGraph); while (boost::iterator_core_access::equal(edgesItr, edgesEnd, boost::mpl::true_())) { // do somthing } * The error message location is at the mpl::apply_wrap2 function. Any relation to the mpl::apply2 function used for declaring the operator== in the BOOST_ITERATOR_FACADE_INTEROP_HEAD macro? Have these functions changed since v1.30.0?

9 8

[BGL] how to actually search with DFS/BFS?
by martin f krafft 22 Nov '06

22 Nov '06

This may seem like a silly question, but how does one actually search with DFS or BFS? Say I want to find the first node for which a predicate returns true in a BFS traversal starting at node N... The function signatures don't have a parameter for "criteria" or "predicate", so I guess it's up for a Visitor to implement such a check. What is not clear to me is how the visitor functions can break out of the search once a condition is true. Thanks, -- martin; (greetings from the heart of the sun.) \____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck invalid/expired pgp subkeys? use subkeys.pgp.net as keyserver! spamtraps: madduck.bogus(a)madduck.net "information superhighway" is just an anagram for "i'm on a huge wispy rhino fart".

9 12

Building boost for 64-bit on AIX
by Patrik Jonsson 17 Aug '06

17 Aug '06

Hi, I'm trying to build boost (1.32.0) on an IBM SP2 under AIX5, using the 64-bit ABI. This is troublesome. I do: bjam "-sTOOLS=gcc" "-sGXX=g++ -maix64 -pthread" --prefix=$HOME --libdir=$HOME/lib64-gcc --includedir=$HOME/include-gcc install But it doesn't work. I get loads of errors like: gcc-Link-action bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/libboost_date_time-gcc-d-1_32.so ld: 0711-738 ERROR: Input file bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/greg_month.o: XCOFF32 object files are not allowed in 64-bit mode. collect2: ld returned 8 exit status set -e LIBPATH=bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true:$LIBPATH export LIBPATH "g++" -maix64 -pthread -g -fPIC -shared -o "bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/libboost_date_time-gcc-d-1_32.so" "bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/greg_month.o" "bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/greg_weekday.o" "bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/date_generators.o" ...failed gcc-Link-action bin/boost/libs/date_time/build/libboost_date_time.so/gcc/debug/shared-linkable-true/libboost_date_time-gcc-d-1_32.so... ...skipped <@boost!lib64-gcc>libboost_date_time-gcc-d-1_32.so for lack of <@boost!libs!date_time!build/libboost_date_time.so/gcc/debug/shared-linkable-true>libboost_date_time-gcc-d-1_32.so... It seems it's using -maix64 ok when trying to build the library, but not when compiling... Can anyone tell me what the proper way of building boost under these conditions are? Thanks, /Patrik Jonsson

5 16

case insensitive less_than string comparison
by Matthias Kaeppler 07 Mar '06

07 Mar '06

Hi, I am aware that string_algo has an if_iequal functor. Is there also an equivalent for a less-than comparison? I have to compare strings ignoring the case but taking locales into account. Right now I'm calling boost::algorithm::to_lower on two copies of the original strings. I'm not so sure if this is very efficient when I'm comparing thousands of strings. Here's the functor as it is now: struct FileLessNameA : std::binary_function<const File&,const File&,bool> { bool operator() (const File& lhs, const File& rhs) const { std::string path1( lhs.get_name().c_str() ); std::string path2( rhs.get_name().c_str() ); boost::algorithm::to_lower( path1 ); boost::algorithm::to_lower( path2 ); return path1 < path2; } }; Any suggestions for improvements appreciated :) Regards, Matthias -- Matthias Kaeppler

7 17

MultiArrays of Std::Lists: why const iterators?
by Joshua Schpok 12 Nov '05

12 Nov '05

I want to make a 3D array of std::lists: boost::multi_array<std::list<CSomething>,3> listbin; Then I'd like to actually use those lists: /// Walk through list at x,y,z for(std::list<CSomething>::iterator iter = listbin[x][y][z].begin(); reti != iter[x][y][z].end(); iter++) { iter->m_foo += 42; } But multi_array is giving me back *const* list iterators, producing the error: error: conversion from `std::_List_const_iterator<CSomething>' to non-scalar type `std::_List_iterator<CSomething>' requested Why is this so? I can trick the compiler by doing a cast like: std::list<CSomething>::iterator iter = ((std::list<CSomething>)listbin[x][y][z]).begin(); but this seems to poop out at runtime. Is there a proper way to do this? Joshua Schpok

3 2

[BGL] a.k.a. [Graph] -- bundled properties not compiling for me under MS Dev
by Goldman, Jon 08 Nov '05

08 Nov '05

Hello, I am using Microsoft Development Environment 2002. I am trying to compile a Graph and use bundled properties. The error I am getting: \src\vtkSNL\Applications\RouteViz\rtrGraph.cxx(37) : error C2676: binary '[' : 'SimpleGraph' does not define this operator or a conversion to a type acceptable to the predefined operator \src\vtkSNL\Applications\RouteViz\rtrGraph.cxx(37) : error C2228: left of '.i' must have class/struct/union type The code snippet below exhibits the problem: struct nodeProp { int i; }; struct edgeProp { int i; }; typedef adjacency_list<boost::listS, boost::vecS, boost::bidirectionalS, nodeProp, edgeProp> SimpleGraph; void someFunc() { SimpleGraph someGraph(3); Graph::vertex_descriptor v = *vertices(someGraph).first; someGraph[v].i = 7; // This line is where the error is occuring } Anybody familiar with this problem? Thanks for looking this over, -Jon

2 1

[iostreams] problem with gzip_compressor
by Tiago de Paula Peixoto 07 Nov '05

07 Nov '05

Hello. I've been experiencing the following problem with the gzip_compressor filter: It always writes data down the stream, regardless if any data has been written to the filtering_stream. Consider the following program: -------------------------------------------------------- #include <iostream> #include <fstream> #include <boost/iostreams/filtering_stream.hpp> #include <boost/iostreams/filter/gzip.hpp> using namespace std; using namespace boost; int main() { ofstream file; file.open("hello.gz", ios_base::out | ios_base::app | ios_base::binary); iostreams::filtering_stream<boost::iostreams::output> stream; stream.push(iostreams::gzip_compressor()); stream.push(file); return 0; } -------------------------------------------------------- If compiled as "test", the following behavior is observed: $ ./test; ls -l hello.gz -rw-r--r-- 1 count0 users 8 2005-10-30 19:51 hello.gz $ ./test; ls -l hello.gz -rw-r--r-- 1 count0 users 16 2005-10-30 19:51 hello.gz $ ./test; ls -l hello.gz -rw-r--r-- 1 count0 users 24 2005-10-30 19:51 hello.gz That is, it always adds 8 bytes to the file, despite the fact that nothing was written to the stream. To top things off, the data inside the file isn't even recognizable by gzip: $ gunzip hello.gz gunzip: hello.gz: not in gzip format When bzip2_compressor is used, this strange behavior doesn't happen, and the file remains with 0 bytes. I guess this is a bug, right? I'm using boost 1.33.0 with GCC 4.0.1 in a GNU/Linux system. Thanks for any help. -- Tiago de Paula Peixoto <tiago(a)forked.de>

2 6

Translating an absolute path to a path relative to the current directory
by Hermann Rodrigues 06 Nov '05

06 Nov '05

Hi, I am running into the following problem these days: I need to translate an absolute path to a path relative to the current directory on Windows. The translation process need to include all required ".." directory specifiers in the path to make sure it can be accessed from the current directory. The following path samples may clarify the problem. -------------------------------------------------------------------------------- Current Directory: C:\x\y\z Absolute Path: C:\x\y\z\k\w.txt Translated Path: k\w.txt (C:\x\y\z is a common parent) Current Directory: C:\x\y\z Absolute Path: D:\a\b\c\d.txt Translated Path: D:\a\b\c\d.txt (no common parent found) Current Directory: C:\x\y\z Absolute Path: C:\x\k\w.txt Translated Path: ..\..\k\w.txt (C:\x is a common parent) -------------------------------------------------------------------------------- Is there any built-in support in the boost filesystem library to achieve such translation? At first I thought the path::relative_path could solve my problem but it does not seem to take into account the current directory, only the path itself. Any sugestion? Thanks in advance. Hermann

2 2

design question for boost::program_options and dynamic options
by boost＠nogga.de 02 Nov '05

02 Nov '05

Hello, I'm new to the boost::program_options library and tried to use it for a command pattern based software. The program is launched with the following commandline: > program command [Options] <Files> The program can use different output formatters with --formatter={XML|binary|text}. Every command and formatter specify their own set of options. I first tried to collect all options form the commands and formatters and parse the command line only once. This lead to the problem, that due to the different positional options information the commandline parser always threw an exception due to wrong command line formatting. Since I couldn't find a way to add mutliple optional option groups to the parser, I moved the parsing code into my command base class and query the derived commands for the different required information, like: class Command { public: vritual options_description GetOptionsDescription (); virtual options_description GetHiddenOptionsDescription (); virtual positional_program_options GetPositionalProgramOptions (); virtual Run (variables_map const & vm) = 0; int Execute (std::vector<std:.string> args) { options_description desc; descr.add (GetOptionsDescription ()).add (GetHiddenOptionsDescription ()); variables_map vm; store(command_line_parser(args) .options(desc). .positional(GetPositionalProgramOptions ()) .run(), vm); notify(vm); Run (vm); } } The execute function is called form the program main after the "command" has been stripped from the commandline and the concrete command was created by the factory. So far so good. This works for the different commands, but not for the formatters. How can I parse the command line dynamically? E.g. if I encounter the "--formatter=xml" argument, a new XMLFormatter is created and the options description from this formatter are added to list of valid options? I know, that there is the notify mechanism, but this only works after the commandline has been successfully parsed. And also the custom parsers do not solve my problem. Any idea how to solve this issue? Best regards Dirk

2 2

configuring a library to work with multiple versions of boost
by Ross Boylan 01 Nov '05

01 Nov '05

I have a library that uses boost and also uses a pre-release version of the ptr_container (header only) library that eventually made it into boost 1.33 (I think). I would like people to be able to compile my code with various versions of boost. I've had it working with 1.31 and 1.32, and expect to have it working with 1.33 when that reaches my distribution. I have a subdirectory with the pre-release ptr_container library and some supporting code from elsewhere in boost. This is in a "boost" subdirectory of my project, which has included the directory above that unconditionally in my compilations. With 1.33 I think I want to ignore my local stuff completely. I'm using autoconf and libtool. Are there autoconf macros available for detecting the boost libraries? Do people have any suggestions about the best way to proceed? One possibility that occurs to me is to test for the version of the library at configure time, and make the include path depend on the results. Thanks. Ross Boylan

2 2

ptr_map
by Oleg Smolsky 01 Nov '05

01 Nov '05

Hello Thorsten, thanks for the ptr_vector clarifications. I have another question about ptr_map: I can declare a map iterator in the following way: ptr_map<string, SomeClass>::iterator i = m.begin(); Now, I'd expect to be able to compile the following, as with a normal map: string s = i->first; i->second.SomeMethod(); ...assuming SomeClass::SomeMethod() exists and is declared as public. However, "first" and "second" don't seem to exist in ptr_map. Could you shed some light on how I can iterate your map please? Best regards, Oleg.

2 3

Visual C++ 2005 (8.0) support
by Ovanes Markarian 01 Nov '05

01 Nov '05

Hi all, did someone test boost 1.33 with Visual C++ 2005(8.0)? I am wondering how I can compile the library using jam. Should I use VC++ 7.1 option? Another question. I derived an iterator from the iterator_facade template. As I have recently found out VC++ 8.0 did not support friend template classes, e.g. template<class ....> class my_type { template <class> friend class my_type; ... }; This will not work in VC++ 8.0. Did you know about this issue? Should this be filed as a bug at MS? With Kind Regards, Ovanes Markarian

5 5

How to build Boost for Cygwin (asked the noob)
by Olumide 01 Nov '05

01 Nov '05

Hi - Would someone please give me some (SIMPLE) pointers on how to build Boost for Cygwin. Thanks, - Olumide

2 1

[threads] read_write_mutex and condtion
by Phil Endecott 01 Nov '05

01 Nov '05

Dear All, Is it possible to use a boost::condition in combination with a read_write_mutex? I've tried but failed. This is with 1.32.0. My test code follows. As you can see it can use either a mutex or a read_write_mutex depending on which bits are commented-out. It works as expected with a mutex, but with a read_write_mutex I get a compiler error, shown below. Perhaps I am missing some fundamental reason why conditions can't work with read_write_mutexes. Can someone explain? Thanks, Phil. #include <string> #include <iostream> #include <boost/thread/mutex.hpp> #include <boost/thread/read_write_mutex.hpp> #include <boost/thread/condition.hpp> #include <boost/thread/thread.hpp> #include <boost/bind.hpp> #include <unistd.h> using namespace std; using namespace boost; struct shared_thing { string data; //typedef mutex mx_t; //typedef mutex::scoped_lock read_lock; //typedef mutex::scoped_lock write_lock; typedef read_write_mutex mx_t; typedef read_write_mutex::scoped_read_lock read_lock; typedef read_write_mutex::scoped_write_lock write_lock; mx_t mx; condition cn; shared_thing::shared_thing(void): //mx() mx(read_write_scheduling_policy::writer_priority) {} void writer(void) { while(1) { sleep(1); { write_lock l(mx); data += "."; cn.notify_all(); } } } void reader(void) { while(1) { read_lock l(mx); cn.wait(l); cout << data << endl; } } }; int main(int argc, char* argv[]) { shared_thing t; thread wr_thread(bind(&shared_thing::writer,&t)); t.reader(); } $ g++ -pthread -O -Wall -o condvar condvar.cc -lboost_thread /usr/include/boost/thread/detail/read_write_lock.hpp: In member function `void boost::condition::wait(L&) [with L = boost::detail::thread::scoped_read_lock<boost::read_write_mutex>]': condvar.cc:51: instantiated from here /usr/include/boost/thread/detail/read_write_lock.hpp:297: error: ` boost::read_write_mutex&boost::detail::thread::scoped_read_lock<boost::read_write_mutex>::m_mutex ' is private /usr/include/boost/thread/condition.hpp:92: error: within this context

3 2

Re: [Boost-users] [Multi-Index] composite_key problems
by JOAQUIN LOPEZ MU?Z 31 Oct '05

31 Oct '05

----- Mensaje original ----- De: Andreas Volz <lists(a)brachttal.net> Fecha: Lunes, Octubre 31, 2005 7:27 pm Asunto: Re: [Boost-users] [Multi-Index] composite_key problems > The questions seems to by stupic, but how > could I > access the data looked up by B alone? What should I assign to the > left? > > ??? = es.get<1>().find(b); > > An iterator or what else? This returns an iterator, yep. Please notice that es.get<1>() returns a reference to an index, which is, in many respects, entirely similar to a regular std::set. So you have find(), equal_range() etc. just like in the STL case, with equivalent syntax and semantics. You might want to give the Boost.MultiIndex tutorial a further read --anyway, don't hesitate to come back here if things aren't still clearer. Joaquin M Lopez Muñoz Telefonica, Investigacion y Desarrollo

2 1

Re: [Boost-users] Translating an absolute path to a path relative
by Gottlob Frege 31 Oct '05

31 Oct '05

> > Date: Mon, 31 Oct 2005 09:55:57 -0300 > From: Hermann Rodrigues <hermann.rodrigues(a)gmail.com> > Subject: [Boost-users] Translating an absolute path to a path relative > to the current directory > To: boost-users(a)lists.boost.org > Message-ID: > <497dfc460510310455j1bd56615peffd2f6040fe849e(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > Hi, > > I am running into the following problem these days: > > I need to translate an absolute path to a path relative to the current > directory on Windows. The translation process need to include all > required ".." directory specifiers in the path to make sure it can be > accessed from the current directory. > > Have you tried the Windows functions such as PathRelativePathTo, and its compliment PathCombine?

2 1

[BUG] missing references in docs on web site (threads)
by Ross Boylan 31 Oct '05

31 Oct '05

While reading the web pages for the threads library I noticed some missing references. I don't know if they reflect a problem with the documentation source or a problem with the processing converting the source into html (if there was any such processing). I did check the html source of teh web pages to verify that it wasn't a display problem with my browser. http://boost.org/doc/html/threads.html#id1357380 In "Potentially non-thread-safe functions" this appears: "It is possible to write thread-safe implementations of these by using thread specific storage (see boost::thread_specific_ptr), and several C++ compiler vendors do just that. The technique is well-know and is explained in ." The reference at the end is missing. I thought I saw another example like this in the threads doc, but can't find it now. Also, the quote from Brinch Hansen at the end of http://boost.org/doc/html/threads/rationale.html doesn't indicate the source (though one could at least guess the article from the bibliography, though he's co-author of another one in the bibliography).

1 0

Re: [Boost-users] [Multi-Index] composite_key problems
by Andreas Volz 31 Oct '05

31 Oct '05

Am Mon, 31 Oct 2005 18:14:00 +0100 schrieb JOAQUIN LOPEZ MU?Z: > // lookup by (A,B); > > es.find(make_tuple(a,b)); > > // lookup by A alone > > es.find(make_tuple(a)); > > // lookup by B alone > // note get<1> retrieves the second index. > > es.get<1>().find(b); > > Is the issue clearer now? If not, please tell me > so and I'll try to explain myself a bit further. A little clearer, but not at all. I'm really new to Boost, so I doesn't understand all. The questions seems to by stupic, but how could I access the data looked up by B alone? What should I assign to the left? > ??? = es.get<1>().find(b); An iterator or what else? There was a print_out_by_name function in docs that I tried to adopt, but I failed. And the compile errors are not very helpfull. As normal with template programming :-( regards Andreas

1 0

[Boost.filesystem] Support of Unicode OS ?
by Marc Viala 31 Oct '05

31 Oct '05

Hi, I would like to know if there are some implementations of "Boost.filesystem" to support Unicode OS like Windows XP? We are involved in a large project, we use Boost libraries intensively, version 1.33, and we are looking forward for this mandatory feature. Thanks in advance for your reply. Marc Viala www.acticm.com <http://www.acticm.com> mailto:mviala@acticm.com <mailto:mviala@acticm.com>

1 0

C++ Connections question
by Elisha Berns 31 Oct '05

31 Oct '05

Ok, So this isn't a Boost.X question, but I have to ask since Dave Abrahams is giving two talks at the C++ Connections conference next week. The question is since the sessions, any session not just Dave's, are either only 60 or 75 minutes max, how informative and in depth can they possibly be? It would appear they are only to whet one's appetite for learning something, but that is precisely my question. Thanks for offering an appraisal whether this conference is really worth the bother. Elisha Berns e.berns(a)computer.org tel. (310) 556 - 8332 fax (310) 556 - 2839

2 1

Re: [Boost-users] downcasting shared pointer
by Bjorn.Karlsson＠readsoft.com 31 Oct '05

31 Oct '05

Hello Faheem, > From: Faheem Mitha [mailto:faheem@email.unc.edu] > > It would be nice if there was some method to downcast the shared_ptr > itself. See http://www.boost.org/libs/smart_ptr/shared_ptr.htm#static_pointer_cast for details on how to do just that. Regards, Bjorn Karlsson

2 1

Implementing lock/unlock with boost::thread
by Johan Torp 31 Oct '05

31 Oct '05

I'm trying to implement a class having a lock and an unlock-method. If several different threads call scoped_lock::lock, what is the effect? class AntiRAII{ public: AntiRAII() : mLock(mMutex, false) {} void Lock() {mLock.lock(); } void Unlock() {mLock.unlock();} private: boost::mutex mMutex; boost::mutex::scoped_lock mLock; }; I cannot find any documentation on the behaviour. The usual usage of scoped_lock is one per thread, which is why I'm sceptical to this approach. How can this be implemented? Is it better to expose the mutex? Regards, Johan Torp

3 2

Re: [Boost-users] [Multi-Index] composite_key problems
by JOAQUIN LOPEZ MU?Z 31 Oct '05

31 Oct '05

----- Mensaje original ----- De: Andreas Volz <lists(a)brachttal.net> Fecha: Domingo, Octubre 30, 2005 8:15 pm Asunto: [Boost-users] [Multi-Index] composite_key problems > Hi, > > I like to build a data structure and like to access it with different > keys, so I try multi_index for this. [...] > I hacked the basic example to get this working. My example compiles, > but if I execute it, I get only warnings: > > probe: /usr/include/boost/multi_index/detail/index_iterator.hpp:107: > typename Node::value_type& > boost::multi_index::detail::index_iterator<Node, > Container>::operator*() const [with Node = > Container>boost::multi_index::detail::ordered_index_node<boost:: > multi_index::detail::index_node_base<learnedMeasure> > Container>>, Container = > Container>>boost::multi_index::detail::ordered_index<boost:: > multi_index::composite_key<learnedMeasure, > Container>>boost::multi_index::member<learnedMeasure, std::string, > Container>>&learnedMeasure::measu > .... more warnings > > I put my example code online. Perhaps someone could take a look in it > and help me to find this problem. > > http://cpp.sourceforge.net/?show=9550 > Hello Andreas, I can't compile your source right now, but from the looks of it I cannot see anything wrong. What compiler are you using? If MSVC++ 6.0 or 7.0, take into account they have non-conformances you must take into account when using Boost.MultiIndex, see http://boost.org/libs/multi_index/doc/compiler_specifics.html Your particular code won't work on either of these compilers due to problems with member<> and composite_key that are addresses in the aforementioned page. Is this the problem? If not, I'll try to get back to you in a couple of days. Best, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1