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

Code stopped compile after migrating 1.42->1.45
by Pavel Pervov 07 Dec '10

07 Dec '10

I'm compiling my codebase with Microsoft Visual C++ 8/9 (Visual Studio 2005/2008) Here is an example code which I've crafted from what I have in my codebase: //------- code start -------------------------------- #include <algorithm> #include <map> #include <string> #include <iostream> #include <boost/lambda/lambda.hpp> #include <boost/lambda/bind.hpp> typedef std::map<int, int> MyMap; std::ostream& operator << (std::ostream& outs, const MyMap& a) { std::for_each(a.begin(), a.end(), outs << boost::lambda::bind(&A::value_type::second, boost::lambda::_1) << "\n"); return outs; } int main() { MyMap a; std::cout << a; } //------- code end --------------------------------- This code stopped compiling after I've migrated from boost version 1.42 to version 1.45. Any ideas would be appreciated. Regards, Pavel --- This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. Please refer to http://www.db.com/en/content/eu_disclosures.htm for additional EU corporate and regulatory disclosures.

2 1

[Fusion] Flatten a sequence of sequences
by Clinton Mead 07 Dec '10

07 Dec '10

Hi All Lets say I have something like the following: make_vector(1, make_vector('b', 3, make_vector(4, 5.5), "six"), 7, 8) I want to produce an function f such that f(make_vector(1, make_vector('b', 3, make_vector(4, 5.5), "six"), 7, 8)) -> [1,'b',3,4,5.5,"six",7,8] i.e. a flattened version of the sequence. I don't mind if this is a view of the original sequence or an actual vector. I don't mind a solution in C++0x if it can compile on GCC 4.5.1. Thanks

3 2

[interprocess]/[thread] - named_recursive_mutex spurious wake using scoped_lock
by Daniel Biagini 07 Dec '10

07 Dec '10

I'm running boost 1.43 on linux and seeing an issue where the scoped_lock<named_recursive_mutex> timed lock is returning immediately when another thread holds the lock. The lock is being stored in shared memory Am I doing something wrong? The code in question is the azLog function below, (I included some other relevant snips as well). thanks! Dan #define LOCK_WAIT_TIME 5; //seconds static named_recursive_mutex *globalLock = create_logger_lock(); static named_recursive_mutex *create_logger_lock () { named_recursive_mutex *ptr = NULL; if (globalLock) { return(globalLock); } deadlock_count = 0; lock_timeout = LOCK_WAIT_TIME; try { ptr = new named_recursive_mutex(open_or_create, (getShareName() + GLOBAL_LOCK).c_str()); } catch (const interprocess_exception &ie) { cerr << "FATAL: Couldn't initialize the logging subsytem, IE err:" << ie.what() << endl; throw; } catch (const std::exception &se) { cerr << "FATAL: Couldn't initialize the logging subsytem, std err:" << se.what() << endl; throw; } return(ptr); } void azLog(AZ_LOG_SEVERITY sev, AZ_LOG_COMPONENT comp, const string& log_str) { ptime to = second_clock::local_time() + seconds(lock_timeout); scoped_lock<named_recursive_mutex> lock(*globalLock, to); while(!lock.owns() && (second_clock::local_time() < to)) { cerr << pthread_self() << ": SPURIOUS WAKE!!" << to_simple_string(second_clock::local_time()) << " <=:=> " << to_simple_string(to) << endl; lock.timed_lock(to); } .....

1 0

[BGL] vertex, edge, and face properties of graph?
by Anders Wallin 07 Dec '10

07 Dec '10

Hi All, I'm implementing a half-edge data structure(*) based on adjacency_list. I'm storing vertex and edge properties as bundled properties (classes VertProps and EdgeProps), and can read/write them conveniently with e.g.: graph[vertexDescriptor].vertProp_a = 2 and x = graph[edgeDescriptor].edgeProp_b etc. I also need to keep track of the faces of the planar graph, and I'm wondering if there's a built-in way of doing this that would allow me to read/write face-properties via the same syntax: graph[faceDescriptor].faceProperty Could I derive my half-edge-class from adjacency_list, store and update a property-map for face-data, and provide a new overload for operator[] that takes a faceDescriptor type? or is there a better way? any examples available? thanks, AW (*) see e.g. http://www.holmes3d.net/graphics/dcel/

2 1

Re: [Boost-users] Code stopped compile after migrating 1.42->1.45
by Pavel Pervov 07 Dec '10

07 Dec '10

Small correction to the code: //------- code start -------------------------------- #include <algorithm> #include <map> #include <string> #include <iostream> #include <boost/lambda/lambda.hpp> #include <boost/lambda/bind.hpp> typedef std::map<int, int> MyMap; std::ostream& operator << (std::ostream& outs, const MyMap& a) { std::for_each(a.begin(), a.end(), outs << boost::lambda::bind(&MyMap::value_type::second, boost::lambda::_1) << "\n"); return outs; } int main() { MyMap a; std::cout << a; } //------- code end --------------------------------- Pavel --- This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. Please refer to http://www.db.com/en/content/eu_disclosures.htm for additional EU corporate and regulatory disclosures.

2 1

BOOST Filesystem file_size problem
by Claude 07 Dec '10

07 Dec '10

Hi! I have a problem with file_size () If I try to use file_size () like this: MyFileName string ("myFile"); file_size (path (MyFileName)); [...] Undefined reference to `boost:: filesystem2:: detail:: file_size_api (std:: basic_string <char, std:: char_traits <char>, std:: allocator <char>> const &) '[...] What is wrong? -- View this message in context: http://boost.2283326.n4.nabble.com/BOOST-Filesystem-file-size-problem-tp307… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

[statechart] optional<event> - warning
by Igor R 07 Dec '10

07 Dec '10

Hello, When I define optional<my_event>, I get a warning (MSVC10): warning C4291: 'void *boost::statechart::event<MostDerived>::operator new(size_t,void *)' : no matching operator delete found; memory will not be freed if initialization throws an exception In class statechart::event I see 2 overloads of operator new: void * operator new( std::size_t size ) void * operator new( std::size_t, void * p ) But only 1 operator delete: void operator delete( void * pEvent ) i.e., no placement delete. Is this intentional? Thanks.

2 4

Object pool
by Lloyd 07 Dec '10

07 Dec '10

Hi, I have gone through the object pool documentation, but could not find a clear answer to my doubt. I want to make a pool of 10 objects. As and when an object need to be created, the pool will be searched for free objects, then the free ones will be chosen. Otherwise it will wait until any of the object is free. Is this possible with object pool? My questions are... How can I allocate 10 objects in single shot? How can I get the reference to unused object from the pool? Is there a same mechanism for threads as well? Thanks in advance, Lloyd

2 1

Boost 1.39.0 program options.
by Sunil Thomas 07 Dec '10

07 Dec '10

Hi, Has any one else had a *glibc error (invalid free/delete issue)* pointing to boost_program-options in boost 1.39.0? Program exits with glibc error (double free; although nothing obvious stands out). Running valgrind hasn't really helped a lot. Is this a known boost issue in 1.39.0? I am using non-multithreaded versions of the boost libs: program-options, regex, iostreams, python, filesystem, system, date_time, unit_test_framework, and multithreaded version only for boost lib: thread. I've reached my wit's end on this problem that occurs only on linux (red hat 5, GNU compilers, GCC version 4.1.2) and any advice is helpful. Greatly appreciate if you could share your experience on how you fixed it. Thanks! --Tom.

2 1

Building boost on AIX6.
by Pat Suwalski 06 Dec '10

06 Dec '10

Hello, I'm having a lot of difficulty building boost on AIX6 with gcc. Searching for my problem just came up with a bunch of Windows users not settings their variables correctly. That does not seem to be the case for me. I'm dealing with v1.34.1, because apparently that works well on AIX. I had no problem building bjam. But then: -bash-3.2$ pwd /home/pats/boost-1.34.1 -bash-3.2$ ./tools/jam/src/bin.aix/bjam Unable to load Boost.Build: could not find build system. --------------------------------------------------------- /home/pats/boost-1.34.1/boost-build.jam attempted to load the build system by invoking 'boost-build tools/build/v2 ;' but we were unable to find "bootstrap.jam" in the specified directory or in BOOST_BUILD_PATH (searching /usr/share/boost-build, /home/pats/boost-1.34.1/tools/build/v2). Please consult the documentation at 'http://www.boost.org'. -bash-3.2$ ls -la /home/pats/boost-1.34.1/tools/build/v2/bootstrap.jam -rw-r--r-- 1 pats staff 761 Oct 29 2003 /home/pats/boost-1.34.1/tools/build/v2/bootstrap.jam Clearly the file is there. Setting explicit $BOOST_BUILD_ROOT and $BOOST_ROOT changes nothing, other than the error message output not mentioning /usr/share. So, the error message is lying, or there is something less obvious wrong. I had no issues building the same source tree on Suse and Solaris. Any advice? Many thanks, --Pat

1 0