- Boost-users - lists.boost.org

[Asio] io_service and queuing priority
by Dylan Klomparens 22 Dec '09

22 Dec '09

I have a program that uses an io_service to queue and execute function calls sequentially within its own thread. Other threads call io_service::post to queue small tasks, but every so often a large task is posted to the io_service. In my case, this large task is time-sensitive and must take priority over all other tasks. Is there a way to push that large task to the front of the io_service's queue, so I can guarantee it gets completed as soon as possible? A simpler way of asking my question: is it possible to prioritize an io_service queue? How? -- Dylan

2 1

[boost-graph]remove_vertex gives segmentation fault
by Suresh Kumar 22 Dec '09

22 Dec '09

Hi, I get segmentation fault when I call remove_vertex in boost graph. The graph is defined as adjacency_list<listS, listS, undirectedS, Cluster, Cluster> Graph; clear_vertex(leaf,g); remove_vertex(leaf,g); I call clear_vertex() and then remove_vertex(). remove_vertex gives a segmentation violation. How do I go about in debugging this? Thanks suresh -- R. Suresh Kumar PhD Student, Vislab #216, Engineering II Bldg University of California, Riverside, CA 92521

1 1

Re: [Boost-users] mpl/fusion: metafunction to 'rangify'
by Hicham Mouline 22 Dec '09

22 Dec '09

Prettified source code at http://codepad.org/knN4xmIR

1 0

BOOST_THROW_EXCEPTION old style cast
by Sam Miller 22 Dec '09

22 Dec '09

I am trying to build code using the BOOST_THROW_EXCEPTION macro with the gcc warning flags -Werror -Wold-style-cast. The macro appears to cast __LINE__ into an integer using an old-style cast instead of a static_cast. Any idea why? Sam

2 1

Compile error with boost_1_41
by Steve Ong 22 Dec '09

22 Dec '09

Hi, First of all, I don't know whether this is the right group to submit this particular issue; if not, please direct me to the right one... ...The issue: I just downloaded the source from the current Boost release (i.e., boost_1_41_0). I compiled this load into a static library and a dynamic libray (on both Solaris 10 and CYGWIN). After that I I try to compile the "basic_client" and "basic_server" (Source from the boost.asio project). Previously, these compiled just fine under boost_1_40_0, but now I running into the following compile errors: g++ -o chat_client chat_client.cpp -I/home/song/MyProjects/LibDist/boostS_1_41/include -L/home/song/MyProjects/LibDist/boostS_1_41/lib -lposix4 -lboost_thread -lboost_system -pthreads -lsocket In file included from /home/song/MyProjects/LibDist/boostS_1_41/include/boost/thread/future.hpp:12, from /home/song/MyProjects/LibDist/boostS_1_41/include/boost/thread.hpp:24, from chat_client.cpp:16: /home/song/MyProjects/LibDist/boostS_1_41/include/boost/exception_ptr.hpp:43: error: looser throw specifier for `virtual boost::exception_ptr::~exception_ptr()' /home/song/MyProjects/LibDist/boostS_1_41/include/boost/exception/detail/exception_ptr_base.hpp:27: error: overriding `virtual boost::exception_detail::exception_ptr_base::~exception_ptr_base() throw ()' make: *** [chat_client_] Error 1 ...Any advice to fix this would be appreciated. Steven

2 2

[typeof] Determining if a type has been registered or not at compile time
by Edward Diener 22 Dec '09

22 Dec '09

Is there any way one can determine whether a type has been registered or not with Boost.Typeof at compile time so that the information can be used in TMP ? I am thinking, of course, of the use of typeof in a template, where a template parameter may or may not be aregistered type when the template has been instantiated. I would like to provide my own TMP fallback code if the type has not been registered rather than see an immediate compiler error generated. Something like: boost::mpl::eval_if < boost::typeof::registered<T>, // this would be nice to have boost::mpl::identity<BOOST_TYPEOF_TPL(some_expression)>, boost::mpl::identity</* use another facility to get the type I want into a metafunction*/> >

2 2

Possibly Memory Leaks in Boost.Date_Time ?
by Daniel Vallejos 22 Dec '09

22 Dec '09

Hello! Memory Leaks in Boost.Date_Time ? Code: #include <boost/date_time.hpp> #include <iostream> using namespace boost::posix_time; using namespace std; int main() { time_duration interval = seconds(120); cout << "interval = " << interval << endl; } Valgrind profile: $ valgrind ./date_time_test ==8666== Memcheck, a memory error detector ==8666== Copyright (C) 2002-2009, and GNU GPL'd, by Julian Seward et al. ==8666== Using Valgrind-3.5.0-Debian and LibVEX; rerun with -h for copyright info ==8666== Command: ./date_time_test ==8666== interval = 00:02:00 ==8666== ==8666== HEAP SUMMARY: ==8666== in use at exit: 1,595 bytes in 27 blocks ==8666== total heap usage: 48 allocs, 21 frees, 5,650 bytes allocated ==8666== ==8666== LEAK SUMMARY: ==8666== definitely lost: 0 bytes in 0 blocks ==8666== indirectly lost: 0 bytes in 0 blocks ==8666== possibly lost: 619 bytes in 20 blocks ==8666== still reachable: 976 bytes in 7 blocks ==8666== suppressed: 0 bytes in 0 blocks Environment: $ uname -a Linux electroforce 2.6.31-15-generic #50-Ubuntu SMP Tue Nov 10 14:53:52 UTC 2009 x86_64 GNU/Linux $ g++ --version g++ (Ubuntu 4.4.2-5ubuntu1) 4.4.2

1 0

Program executable won't run using boost 1.40
by Carles Fernandez 22 Dec '09

22 Dec '09

Hi everyone, I'm new to boost and I've followed the getting started guide from boost to set it up into my computer. The first example<http://www.boost.org/doc/libs/1_40_0/more/getting_started/windows.html#buil…>compiles and executes correctly. But the second one<http://www.boost.org/doc/libs/1_40_0/more/getting_started/windows.html#link…> fails to execute (but it does compile and link correctly). When I try to execute the program from a command line (like the tutorial indicates) I get the following error: *The system cannot execute the specified program.* * * And if i try to execute it directly from Visual Studio it says: *Unable to start program ...../..../example.exe* *This application has failed to start because the application configuration is incorrect. Review the manifest file for possible errors. Reinstalling the application may fix this problem. For more details, please see the application event log.* I am using Visual Studio 2005 and I have downloaded all the VC8.0 (visual studio 2005) boost 1.40 precompiled libraries using the boostpro installer. Can someone help me please? I'm sure I am doing something wrong, but I don't know what..

2 1

boost::serialization::null_deleter
by Andreas Ronnquist 22 Dec '09

22 Dec '09

Hi! I am using boost::serialization::detail::null_deleter in combination with shared_ptr (on boost 1.34/35) - and I am having a user of my program compiling my program (on boost 1.41) who says he has to change that line to just boost::serialization::null_deleter to get it to work. Should I work around this with pre-processor checks, and in that case - what versions should I look for? Or is there a simpler fix? thanks in advance -- Andreas Ronnquist <andreas.ronnquist(a)meritkonsult.se>

5 4

Re: [Boost-users] storing data
by pete＠pcbartlett.com 22 Dec '09

22 Dec '09

On Tue 22/12/09 05:08 , "Sean Farrow" <sean.farrow(a)seanfarrow.co.uk> wrote: > Hi: > > I am in need to store data. Basically I have an application, when I > have 8 collumns that are named, however thee are 38 possible rows with > differing nameson. I also need to store the screen co-ordinates for > each column name on each row. So for example: > > Column 1, name Item, row 1, x 66, y 85, row 2 x 88 y 121. > > What is the best way of being able to store this data with boost and > get access to the item name, and co-ordinates given a row and column > number? Hi Sean, There are several boost libraries that may be of help - I am thinking of boost.array, boost.multi_array and possibly even boost.multi_index. This is in addition to standard containers such as map and vector. However it is difficult to give very specific advice as I find your mail unclear. The best I can suggest is to think through your problem again, thinking how to make it easier for others to understand - in my experience the very process of thinking how to explain a problem clearly to others may help /you/ solve the problem by yourself. Your mail would also benefit from being passed through a spelling and grammar checker prior to posting. I hope this advice is of help to you, Pete Bartlett

1 0