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

Re: [Boost-users] [Test] How to treat memory leaks as errors?
by Василий Старостин 27 Sep '09

27 Sep '09

Hi! >> However, Boost.Test doesn't return non-zero error code when leaks >> detected (as it does this when tests fail). >> How can I make my testing program to return non-zero error code if memory leaked? >Because Boost.Test does not contain anything that can test for memory >leaks. You might want to use something like Google's tcmalloc and so >forth as it can tell you about all that, and you can just add a >Boost.Test testcase at the end of the tests to check for memory leaks >using it. Boost.Test does contain something to test memory leaks since I get such warnings when I loose memory in my Boost.Test-based program :) I know this is _CRT functions (I use MSVC9). And the function, which Boost.Test uses to inform about leaks under Windows, SIGNALS that we have m-leaks in a program. Moreover, Boost.Test can control this function using (to use or not to use). So I expected that Boost.Test can use this function return code and increase errors counter for leaks. If it cannot, this can be improvement for the future. WBR, Vasily P.S. _CrtDumpMemoryLeaks?

3 6

Re: [Boost-users] Sockets and structures?
by Brian Wood 27 Sep '09

27 Sep '09

Daniel F. Savarese writes: > To the OP, I would advise you not take statements such as the above as > gospel without performing measurements yourself. My experience[1] is > directly contrary to the above. I find boost::serialization execution > time to be negligible (for binary archives) and entirely suitable for high > performance networking applications even if that was not the creator's > original objective (a credit to Robert Ramey's skill). With respect for the library and Ramey, I'd like to point out that results obtained with the Boost Serialization library are frequently less efficient than those from the C++ Middleware Writer -- http://webEbenezer.net/comparison.html<http://webebenezer.net/comparison.html> . Additionally, the C++ Middleware Writer automates the job of writing serialization functions for structs/classes unless it is directed not to do so. Regards, Brian Wood Ebenezer Enterprises www.webEbenezer.net "Then Samuel took a rock and set it up between Mizpah and Shen. He named it Ebenezer [Rock of Help] and said, 'Until now the L-RD has helped us.'"

2 1

[Test] Compiler error/missing entry point?
by Michael Spece Ibanez 26 Sep '09

26 Sep '09

I installed the Boost.Test library binary using bootstrap.sh. Is the following a valid program? If not, what is wrong? Is there a missing entry point and, if so, how do I run the test? Source: #define BOOST_TEST_MODULE ExampleTestModule #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_CASE( my_test ) { BOOST_CHECK(1); } Error: /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text+0x155): undefined reference to `boost::unit_test::framework::master_test_suite()' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text+0x232): undefined reference to `boost::unit_test::unit_test_log_t::set_checkpoint(boost::unit_test::basic_cstring<char const>, unsigned int, boost::unit_test::basic_cstring<char const>)' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text+0x317): undefined reference to `boost::test_tools::tt_detail::check_impl(boost::test_tools::predicate_result const&, boost::unit_test::lazy_ostream const&, boost::unit_test::basic_cstring<char const>, unsigned int, boost::test_tools::tt_detail::tool_level, boost::test_tools::tt_detail::check_type, unsigned int, ...)' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text+0x4e0): undefined reference to `boost::unit_test::ut_detail::auto_test_unit_registrar::auto_test_unit_registrar(boost::unit_test::test_case*, unsigned long)' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text$_ZN5boost9unit_test14make_test_caseERKNS0_9callback0INS0_9ut_detail6unusedEEENS0_13basic_cstringIKcEE[boost::unit_test::make_test_case(boost::unit_test::callback0<boost::unit_test::ut_detail::unused> const&, boost::unit_test::basic_cstring<char const>)]+0x54): undefined reference to `boost::unit_test::ut_detail::normalize_test_case_name(boost::unit_test::basic_cstring<char const>)' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text$_ZN5boost9unit_test14make_test_caseERKNS0_9callback0INS0_9ut_detail6unusedEEENS0_13basic_cstringIKcEE[boost::unit_test::make_test_case(boost::unit_test::callback0<boost::unit_test::ut_detail::unused> const&, boost::unit_test::basic_cstring<char const>)]+0xa6): undefined reference to `boost::unit_test::test_case::test_case(boost::unit_test::basic_cstring<char const>, boost::unit_test::callback0<boost::unit_test::ut_detail::unused> const&)' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text$_ZN5boost9unit_test15unit_test_log_tD1Ev[boost::unit_test::unit_test_log_t::~unit_test_log_t()]+0xb): undefined reference to `vtable for boost::unit_test::unit_test_log_t' /cygdrive/c/DOCUME~1/M/LOCALS~1/Temp/ccIao0DX.o:TestExample.cc:(.text$_ZN5boost9unit_test15unit_test_log_tC1Ev[boost::unit_test::unit_test_log_t::unit_test_log_t()]+0x21): undefined reference to `vtable for boost::unit_test::unit_test_log_t' /usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../libcygwin.a(libcmain.o):(.text+0xab): undefined reference to `_WinMain@16' collect2: ld returned 1 exit status

2 5

[Test] catch_system_errors default value ...
by Georg Sauthoff 26 Sep '09

26 Sep '09

Hi, the default of 'catch_system_errors = yes' when using the boost unit testing framework is a bit problematic. Apparently, it installs a signal handler for SIGCHILD. I didn't notice this under Linux. But under Solaris I've got an interrupted waitpid system call in one test case[1]. Well, and the trace looks really funny, if waitpid is interrupted by ... SIGCHILD! (the testcase basically does a fork/exec/waitpid) Thus, the default of this parameter is at least a bit counter-intuitive. Perhaps, you[2] consider to change this default in the future. Best regards Georg Sauthoff [1] I assume, that utf does this on Linux in such a way, that waitpid is automatically resumed [2] I mean the utf maintainer, using gmane, I am not able to CC ...

2 1

[Intrusive] List Manipulation and Management
by Freddie Witherden 26 Sep '09

26 Sep '09

Hi all, I am moderningising/porting some old Pascal code. The code makes extensive use of an intrusive singly-linked structure termed a node. In order to replicate this idiom in C++ I have been investigating the Intrusive library. However, I have several questions about list manipulation and memory management. Firstly, in the code being ported all nodes can be thought of as being heap allocated. The following is extremely common (pseudo-code): Node* afunction(...); [...] Node* tail; link(tail, afunction(...)); This is rather nice as in O(1) you can append a returned list to your current list. However, I am unsure how I would accomplish this in Boost Intrusive. I can not seem to find a way to create the somewhat fragile intrusive::list from just a node (be it singly or doubly linked). The list class also seems to take T& as opposed to T* when pusing to the front/back of the list. My experience of intrusive structure in C was that you'd always take a pointer when adding a new link. Secondly, when dealing with heap allocated items, which need to be deleted, the current code uses a while loop to call each nodes destructor. What is the intrusive equivalent? Of course, I am open to ideas/suggestions about better ways to accomplish the above tasks :) Regards, Freddie.

2 4

[random]: several issues
by Thomas Mang 26 Sep '09

26 Sep '09

HI, After some time I came back to the boost random library, and several things I noticed earlier have not changed since years ago. I hereby ask what the present state of ideas regarding these ideas is. Here's my list: a) The library itself provides files for random deviates of distributions not given in the documentation (e.g. poisson, gamma) etc. I find it truly sad that implementation and documentation is that out of synchronisation b) Several distributions require the engine to return a uniform deviate between [0,1), while other's don't have this prerequisite. I find this extremely error prone and purely documented (it is documented, but it should be so clearer, and especially, louder). Worse, I find it even harder (or impossible) to find out which range the engines return. I am not sure if there is any engine returning this range per se, or if I have always to go through uniform_01. Worst of all, there is neither a compile time check nor a runtime check if that [0,1) result-requirement of the engine holds (as far as I have seen the code) - I absolutely fail to see why such a critical assert is completely missing, also given the poor state of the documentation. The poisson and gamma for example also fall in this category, but are not documented at all. I consider this highly dangerous. c) Random numbers are tightly linked to statistical distributions, offered by the library of it's own. Wouldn't it be convenient to try to integrate the whole distribution part of the random numbers more closely into that library? Presently they are too confusingly standalone. d) Is there anyone actually responsible for the random library at the moment? Is it still under active development / maintained, or is old code simply propagating from release to release ? [No offense intended, but above issues let me have doubts regarding active management). thanks, Thomas

4 6

[Test] Test suite timeout.
by Dmitry V. Krivenok 26 Sep '09

26 Sep '09

Hello! I have a question about timeouts implementation in Boost::Test. Please take a look at a simple example I wrote to demonstrate the problem: krivenok@develop2 16:50:52 /tmp/boost_test $ cat main.cpp #include <boost/test/unit_test.hpp> using namespace boost::unit_test; using boost::unit_test::framework::master_test_suite; //////////////////////////////////////////////////////////////////////////////// void some_test_case() { BOOST_TEST_CHECKPOINT("About to enter an infinite loop!"); while(1); } ////////////////////////////////////////////////////////////////////////////// void another_test_case() { BOOST_TEST_CHECKPOINT("About to enter an infinite loop 2!"); while(1); } ////////////////////////////////////////////////////////////////////////////// boost::unit_test::test_suite* create_test_suite() { boost::unit_test::test_suite *ts = new boost::unit_test::test_suite("Some test suite"); if(!ts) return 0; ts->add(BOOST_TEST_CASE(&some_test_case), 0, 5); ts->add(BOOST_TEST_CASE(&another_test_case), 0, 10); return ts; } /////////////////////////////////////////////////////////////////////////////// test_suite* TestSuitesInit(int argc, char* argv[]) { master_test_suite().add(create_test_suite(), 0, 3); return 0; } /////////////////////////////////////////////////////////////////////////////// int main(int argc, char** argv) { return unit_test_main(&TestSuitesInit, argc, argv); } /////////////////////////////////////////////////////////////////////////////// krivenok@develop2 16:50:53 /tmp/boost_test $ I compiled it as follows: krivenok@develop2 16:56:12 /tmp/boost_test $ g++-4.4.1 -Wall -I/usr/local/dev/boost-1.40.0/include main.cpp /usr/local/dev/boost-1.40.0/lib/libboost_unit_test_framework-mt.a krivenok@develop2 16:56:15 /tmp/boost_test $ And the run: krivenok@develop2 16:58:17 /tmp/boost_test $ time ./a.out Running 2 test cases... unknown location(0): fatal error in "some_test_case": signal: SIGALRM (timeout while executing function) main.cpp(10): last checkpoint: About to enter an infinite loop! unknown location(0): fatal error in "another_test_case": signal: SIGALRM (timeout while executing function) main.cpp(18): last checkpoint: About to enter an infinite loop 2! *** 2 failures detected in test suite "Master Test Suite" real 0m15.002s user 0m14.920s sys 0m0.030s krivenok@develop2 16:58:36 /tmp/boost_test $ According to timings measurement I concluded that Boost::Test supports timeouts on per-test-case basis. Test suite timeout (3 sec) is ignored. Is it a bug? P.S. I also found that timeouts are implemented using alarm system call: krivenok@develop2 16:58:36 /tmp/boost_test $ strace -fF ./a.out 2>&1 | grep alarm alarm(0) = 0 alarm(5) = 0 alarm(0) = 0 alarm(0) = 0 alarm(10) = 0 alarm(0) = 0 krivenok@develop2 17:02:03 /tmp/boost_test $ As I know you can't start two or more timers using alarm since alarm uses exactly one internal timer. The following simple program demonstrates the problem: krivenok@develop2 17:06:38 /tmp/boost_test/alarm $ cat main.c #include <unistd.h> int main() { alarm(2); alarm(5); while(1); return 0; } krivenok@develop2 17:06:42 /tmp/boost_test/alarm $ gcc-4.4.1 main.c krivenok@develop2 17:06:44 /tmp/boost_test/alarm $ time ./a.out Alarm clock real 0m5.001s user 0m4.990s sys 0m0.000s krivenok@develop2 17:06:51 /tmp/boost_test/alarm $ strace -fF ./a.out 2>&1 | grep alarm alarm(2) = 0 alarm(5) = 2 krivenok@develop2 17:07:00 /tmp/boost_test/alarm $ This makes the implementation of mixed timeouts difficult and unreliable. For example, consider the following test tree: MasterTestSuite (timeout = T0) -- TestSuite1 (timeout = T1) -- TestCase11(timeout = T11) -- TestCase12(timeout = T12) -- TestSuite2 (timeout = T2) -- TestCase21(timeout = T21) -- TestCase212(timeout = T22) Actual timeout of any test case depends on timeouts of all nodes in the path. -- Sincerely yours, Dmitry V. Krivenok Orange System Co., Ltd. Saint-Petersburg, Russia work phone: +7 812 332-32-40 cellular phone: +7 921 576-70-91 e-mail: krivenok(a)orangesystem.ru web: http://www.orangesystem.ru skype: krivenok_dmitry jabber: krivenok_dmitry(a)jabber.ru icq: 242-526-443

2 1

unit test: BOOST_FIXTURE_TEST_CASE_TEMPLATE
by Hicham Mouline 26 Sep '09

26 Sep '09

Hello, I would like to request for the addition of this macro to the boost unit tests library. I have a mpl::vector_c<....> and I would like to run a test case for the type being each of the mpl integrals. and test case instantiations would have the same fixture. I don't see a workaround but for now I'll just use a runtime loop. Rds,

2 1

fork in boost?
by pedro chaparro 26 Sep '09

26 Sep '09

hi, there is some function in boost libraries equivalent to "fork" in linux? thanks -- Ing Pedro Alonso Chaparro Valero Ciudad Politecnica de la Innovación iTEAM - Mobile Communications Group Polytechnic University of Valencia C\ Camino de Vera S/N, Edificio 8G 46022 Valencia, Spain

3 2

Argument error after upgrading to Snow Leopard
by s seelenluft 26 Sep '09

26 Sep '09

Hi, after upgrading to Snow Leopard I get following error message compiling even a simple Hello World program with bjam: *** argument error * rule SHELL ( command : * ) * called with: ( ) * missing argument command (builtin):see definition of rule 'SHELL' being called ... The .cpp file is: #include <iostream> int main() { cout "Hello World!" << endl; return 0; } The Jamroot file contains: exe HelloWorld : HelloWorld.cpp ; I have: - bjam installed in /usr/local/bin - boost-build libraries in /usr/share/boost-build - in /usr/share/boost-build/user-config.jam the line: using darwin; This is as far as I know exactly the same as I had it under Leopard. What am I missing here? Thanks. Markus

3 5

[python] Passing pointers as arguments without copying.
by Germán Diago 26 Sep '09

26 Sep '09

Hello. The question is simple, but reading the documentation I cannot understand how to do this: class MyClass : boost::noncopyable { void f(MyClass2 * c2); }; MyClass2 is not copyable either. So I want to be able to wrap that function, but avoid the copying of MyClass2 when called from python. I want to pass the c++ object directly, not a copy of it. I'm not sure if this can be done with ptr() or some kind of call policy, but I don't understand the documentation 100%. Any help will be appreciated. Thanks.

3 3

Week of the Year calculations?
by Ramon F Herrera 26 Sep '09

26 Sep '09

I looked at the Date_Time library but couldn't find this. Need to convert a date into the corresponding week of the year and vice versa. I realize that such thing varies and plan to use the standard most common in the USA. TIA, -Ramon

1 1

[exception] nested exceptions: lost information?
by Pascal Z. 26 Sep '09

26 Sep '09

Hello I'm trying to nest exceptions using boost.Exception, but it seems like some information (in this case the message returned via what() ) is lost on the way. #include <iostream> #include <boost/exception.hpp> typedef boost::error_info<struct tag_nested_exception, boost::exception_ptr> nested_exception; struct file_open_error : public boost::exception, public std::exception {}; int open_file(const std::string& filename) { try { // ... throw std::runtime_error("Some internal error."); } catch(const std::exception& error) { throw file_open_error() << nested_exception(boost::copy_exception(error)); } } int main(int argc, char* argv[]) { try { int file = open_file("foo.txt"); file++; } catch(const boost::exception& error) { try { boost::rethrow_exception(*boost::get_error_info<nested_exception>(error)); } catch(const std::exception& nested) { std::cout << "the nested error is: " << nested.what() << std::endl; } } return 0; } I except the output to be "the nested error is: Some internal error." but it outputs "the nested error is: std::exception". What can I do to make it work? Greetings Pascal

2 2

std::accumulate with condition?
by Archie14 26 Sep '09

26 Sep '09

If a given class defines operator + and there is a vector of instances of that class: how I can use std::accumulate to count sum against this vector for the elements that satisfy certain conditions? class A { public: A(useinsum) : _useinsum(useinsum) {}; bool useinsum() {return _useinsum;} int operator + (int a) {return _i + a;} private bool _useinsum; int _i; } std::vector<A> lst; will that work? std::accumulate(lst.begin(), lst.end(), 0, boost::lambda::bind(&A::useinsum, boost::lambda::_1) == false)

2 1

[graph] Optimizing graph definition
by Maxime van Noppen 26 Sep '09

26 Sep '09

Hi, I'm using a quite large boost::graph (~ 100 000 vertices with 8 to 10 edges each) and am trying to get the best of the structure to be able to increase vertices & edges. The graph is mainly read-only. I add 99.99% of the vertices & edges in a big loop at the beginning. I never delete vertices or edges. After that I mainly use dijkstra and A* algorithms with some full traversals. I use very often the edge(v1, v2, g) function to get the edge (if it exists) between two vertices. My priority is to optimize the use of the graph rather than the creation but I'd like to keep the creation time reasonable. For now I use : typedef boost::adjacency_list< boost::vecS, boost::vecS, boost::directedS, my_vertex, my_edge > graph_t; I didn't find enough informations about complexity of functions and algorithms regarding to the graph definition to find out be myself what would be the best parameters for my use case. Is something improvable here ? Thanks ! -- Maxime

2 1

Re: [Boost-users] Sockets and structures?
by OvermindDL1 26 Sep '09

26 Sep '09

On Fri, Sep 25, 2009 at 6:49 PM, Daniel F. Savarese <dfs(a)savarese.org> wrote: > > In message <3f49a9f40909251225p28d497b6i6d73c2dd53e3ba3e(a)mail.gmail.com>, Overm > indDL1 writes: >>Boost.Serialization is good if you are going to serialize up vast >>hierarchies with lots of pointers that can be self referencing and >>such, but it tends to be very costly in execution speed. Something >>like Boost.Spirit2.1 can serialize things up to a binary format as >>well, and *very* fast. Although something like Google's open source >>message system could serialize up messages even better (we need a >>Boost implementation of that system... hmm...). > > To the OP, I would advise you not take statements such as the above as > gospel without performing measurements yourself. My experience[1] is > directly contrary to the above. I find boost::serialization execution > time to be negligible (for binary archives) and entirely suitable for high > performance networking applications even if that was not the creator's > original objective (a credit to Robert Ramey's skill). > > No tool is suitable for all purposes. So figure out your requirements > and do some testing to see what works best before choosing a solution. > Just because boost::serialization works great for me in a particular > application doesn't mean it's going to work great for someone else with > a different set of requirements. Also, talking about performance > cost/overhead is pointless without precise quantification. Run your > own tests in the context of your anticipated usage and see how it > works out. > > daniel > > [1] www.igfip.com: All interprocess communication and object serialization > uses boost::serialization. Play a game of Reversi or Chess (against the > computer or another player) or chat with another player and see if > boost::serialization in any way impacts the real-time/instantaneous > response to your actions. Behind the scenes, boost::serialization's > impact on CPU time isn't even measurable. It is quite slow when you need speed though. You will not notice it in those times, but compared to other methods it tends to be many many orders of magnitude slower. It is the registery that slows it down though (I have profiled it, AMD Code Analyst). It is, however, *wonderful* for like save/load functionality, I still use it for that, but for heavy network utilization on never idle programs (say a real-time 3d action game), I never would. Use the proper tools for the proper purposes. :)

1 0

Re: [Boost-users] [boost user list] Portal Web Design Support of Boost C++
by OvermindDL1 25 Sep '09

25 Sep '09

Do not send this message to me, send it to boost. :) On Fri, Sep 25, 2009 at 1:51 PM, Ray Burkholder <ray(a)oneunified.net> wrote: >> > >> > Does boost C++ support the design of portal web pages or processing >> of >> > database data from web pages depending on user selection of controls >> on the >> > web page? >> >> That has 'nothing' to do with your chosen language, rather with the >> web library you use, the most popular of which is (Fast)CGI, which >> does support it all (what do you think PHP/Python/Java/etc... are all >> built to use, CGI, it can do it all). >> >> Thus, this is not a boost question. > > You may instead be looking for something like http://www.webtoolkit.eu/wt > which makes use of some boost functionality. I however have been looking for something like that, thanks for the link!

1 0

Sockets and structures?
by pedro chaparro 25 Sep '09

25 Sep '09

hi, i'm wondreing if using asio library can i send a structure and not just string messages using the boost::asiio:write command? or how could i do to send a structure which have many information fields? thanks -- Ing Pedro Alonso Chaparro Valero Ciudad Politecnica de la Innovación iTEAM - Mobile Communications Group Polytechnic University of Valencia C\ Camino de Vera S/N, Edificio 8G 46022 Valencia, Spain

4 4

[boost user list] Portal Web Design Support of Boost C++
by Boost lzw 25 Sep '09

25 Sep '09

Hi there, Does boost C++ support the design of portal web pages or processing of database data from web pages depending on user selection of controls on the web page? Thanks in advance, Robert

2 1

Boost serialization libray 1.32 & 1.39 compatibility issue ( from "To DO" list)
by sam p 25 Sep '09

25 Sep '09

Hi, I am facing backward compatibility problem with boost serialization library 1.39. My setup is something like this, (1) I have one version of my program that is built using boost 1.32 serialization library, lets call it 1_32.exe (2) I have another version of my program that is built using boost 1.39 serialization library, lets call it 1_39.exe (3) These two versions of my program communicate with each other over network and they use boost text serialization to exchange data. (4) Program 1_39.exe is able to deserialize data sent by 1_32.exe but 1_32.exe is not able to deserialize data sent by 1_39.exe While debugging this problem I realized that program 1_39.exe is putting archive version as “5” and 1_32.exe is putting archive version as “3”. This results into 1_32.exe throwing “Unknown version” exception. While going through boost documentation I read about “Back Versioning” description provided at http://www.boost.org/doc/libs/1_39_0/libs/serialization/doc/todo.html#backv… Looks like its not very difficult to address this issue if we modify some code in boost serialization library. Are you planning to address this in near future or can you please advice me on the files those need to be changed? Or is there any simple workaround as a stop gap measure to fix this issue? Thanks, Sameer

3 2

Passing reference params to boost::thread, parameter copies.
by Jason Cipriani 25 Sep '09

25 Sep '09

At the end of this message is a test program that uses boost::thread to create two threads. The first is one that takes a parameter by value, the second takes a parameter by reference. I have a couple of questions about some behavior here: 1) When passing by value, why are so many copies of the data made? The program output indicates that there are 9 (!) objects created (at most 5 exist simultaneously), when ideally there would be only 2 (the original and the copy passed to the thread). 2) When passing by reference, why is the data being copied? The copying behavior when passed by reference is identical to when the param is passed by value (9 objects created, 5 simultaneously). Also, and this isn't shown in the program below, but I would therefore expect the program to crash if the boost::thread outlived the referenced object (of course, it does not, since the object is copied). So, I'm surprised by the number of copies, this seems unreasonable, and also I'm confused about why by value and by reference behave identically. Thanks, Jason === BEGIN EXAMPLE === #include <boost/thread/thread.hpp> #include <cstdio> using namespace std; class test { public: test () { printf("%p\n", (void *)this); } test (const test &t) { printf("%p from %p\n", (void *)this, (void *)&t); } ~test () { printf("~%p\n", (void *)this); } }; void expected_copy (test) { printf("in expected_copy\n"); } void expected_nocopy (const test &) { printf("in expected_nocopy\n"); } int main () { printf("expected_copy:\n"); { test t; boost::thread thr(expected_copy, t); thr.join(); } printf("expected_nocopy:\n"); { test t; boost::thread thr(expected_nocopy, t); thr.join(); } printf("complete\n"); getchar(); } === END EXAMPLE ===

3 3

Boost unit_test.cpp problem when try to compile simple program
by Robert Matusewicz 25 Sep '09

25 Sep '09

I tried to use Boost test module in my project but I have a little problem. When I try to compile simple program: #define BOOST_TEST_MAIN #include <boost/test/included/unit_test.hpp> it fails to compile on Visual Studio 2008 with three errors: c:\program files (x86)\boost\boost_1_39\boost\test\impl\exception_safety.ipp(513) : error C2491: 'boost::itest::exception_safety' : definition of dllimport function not allowed c:\program files (x86)\boost\boost_1_39\boost\test\impl\logged_expectations.ipp(228) : error C2491: 'boost::itest::logged_expectations' : definition of dllimport function not allowed c:\program files (x86)\boost\boost_1_39\boost\test\impl\unit_test_main.ipp(174) : error C2491: 'boost::unit_test::unit_test_main' : definition of dllimport function not allowed On Linux machines everything works OK. So, anybody could help me with that? Need more data? Robert

2 3

import boost cstdint into default namespace
by V S P 25 Sep '09

25 Sep '09

Can I somehow import the boost/cstdint definitions into the default namespace without importing any other things from the boost namespace (I am trying to compile a source code that relies on having stdint.h but VC 2008 does not have it so I want to use boost/cstdint but without much modifications to the source at question) thank you -- Vlad P author of C++ ORM http://github.com/vladp/CppOrm/tree/master -- http://www.fastmail.fm - Access your email from home and the web

2 1

BOOST_SYSTEM_NO_DEPRECATED and boost/filesystem/convenience.hpp
by Paul 25 Sep '09

25 Sep '09

Hi, If i compile with -DBOOST_SYSTEM_NO_DEPRECATED and #include <boost/filesystem/convenience.hpp> The compile will fail with the message: boost/filesystem/convenience.hpp: In function ‘typename boost::enable_if<boost::filesystem::is_basic_path<Path>, bool>::type boost::filesystem::create_directories(const Path&)’: boost/filesystem/convenience.hpp:48: error: ‘posix’ is not a member of ‘boost::system’ BOOST_FS_FUNC(bool) create_directories(const Path& ph) { if (ph.empty() || exists(ph)) { if ( !ph.empty() && !is_directory(ph) ) boost::throw_exception( basic_filesystem_error<Path>( "boost::filesystem::create_directories", ph, make_error_code( boost::system::posix::file_exists ) ) ); return false; } The line with boost::system::posix::file_exists is the key problem. In boost/system/error_code.hpp, posix is defined as a namespace # ifndef BOOST_SYSTEM_NO_DEPRECATED namespace posix = errc; namespace posix_error = errc; # endif but ONLY if I don't define NO_DEPRECATED... Maybe time for regression tests with BOOST_SYSTEM_NO_DEPRECATED defined? thanks, Paul

1 0

boost::filesystem::rename() documentation incorrect
by Paul 25 Sep '09

25 Sep '09

Hi all, this documentation is incorrect: Otherwise, if to_p resolves to an existing file, it is removed. This is from Boost 1.40.0, and yet the code for rename() calls rename_api() which then does the following: if ( fs::exists( status_api( to, dummy ) ) ) return error_code( EEXIST, system_category ); ie, if to_p resolves to an existing file, it will always return/throw an error... NOT remove the existing file, as was documented. Or am I reading things wrong? cheers Paul

1 0