January 2010 - Boost-users - lists.preview.boost.org

[Wave] Modifying the way function-like macros are interpreted
by fractured＠modularsystems.sl 04 Jan '10

04 Jan '10

I am trying to use boost::wave as a preprocessor for LSL, a language used in the virtual world Second Life. There is a data type there for collections of data, but it can only be accessed through functions such as llList2String(list,iterator); I wanted to define macros that would allow me to do something like #define x[y] llList2String(x,y), however, wave as it stands does not seem to have any immediately obvious way to do this.

4 6

[test] 64bit test case deletion errors again
by Alastair Rankine 04 Jan '10

04 Jan '10

This issue has been discussed before on this list, see the thread at gmane.comp.lib.boost.user/51628 I can confirm that the fix in 1.41 does indeed work for most regular test cases. However I believe that there is still an issue when using fixtures. Below is a sample app which crashes for me when compiled with gcc on MacOS X 10.6.2. If someone could confirm that I'm not doing anything particularly stupid, that would be appreciated. #define BOOST_TEST_MODULE autoreg test #include <boost/test/unit_test.hpp> struct Fixture { Fixture() : n(1) {} unsigned n; }; BOOST_AUTO_TEST_SUITE(AutoregTest) BOOST_FIXTURE_TEST_CASE(Test1, Fixture) { BOOST_CHECK_EQUAL(n, 1); } BOOST_FIXTURE_TEST_CASE(Test2, Fixture) { BOOST_CHECK_EQUAL(n, 1); } BOOST_AUTO_TEST_SUITE_END() % ./autoregtest Running 2 test cases... *** No errors detected autoregtest(52796) malloc: *** error for object 0x30001002006d0: pointer being freed was not allocated *** set a breakpoint in malloc_error_break to debug Breakpoint 1, 0x00007fff818f5951 in malloc_error_break () (gdb) bt #0 0x00007fff818f5951 in malloc_error_break () #1 0x00007fff8181f143 in free () #2 0x000000010006dc62 in ~framework_impl [inlined] () at /opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_ports_devel_boost/work/boost_1_41_0/boost/test/impl/framework.ipp:133 #3 0x000000010006dc62 in __tcf_1 () at framework.ipp:225 #4 0x00007fff8182b334 in __cxa_finalize () #5 0x00007fff8182b24c in exit () #6 0x00000001000031e7 in start ()

1 0

asio - interrupting a long process running in server
by Lloyd 04 Jan '10

04 Jan '10

Hi, I am using ASIO to create a server. The clients sends command to the server, the server executes it in it and sends back the result to the client (here the client remotely controls the server). Lets assume that one command implementation at the server side contains only an infinite loop, as this wastes the server resource, the client need to "interrupt" or "terminate" the execution of the infinite loop in the server. Is there any mechanism available in ASIO to accomplish this? What I am doing now is, this long job is created as a thread, collect the thread's id and store it in a container. Then this id is send to the client, if the client wants to interrupt the thread, it sends another command (STOP_THREAD ) with this thread id, the server looks for the thread running with this id, and interrupts it. At the end of execution thread, the thread id has to be removed from the container (thread id,pointer to thread). But this makes it very complicated, and it is not working right. So I am looking for a solution available with ASIO... Please share your ideas... Thanks, Lloyd ___________________ Happy New Year 2010

1 0

[BGL] Free graph - double free or corruption
by Matthieu BOUSSARD 04 Jan '10

04 Jan '10

Hello, I'm using BGL to construct an exploration graph. At each decision step I have to create a new graph. It seems that I'm not freeing correctly the memory, causing a memory leak. The name of a node in the graph is a pointer to a structure, typedef adjacency_list <listS,vecS, directedS, property<vertex_name_t, state_t *, property<vertex_prop_t,s_prop> >, property<edge_weight_t, float, property<edge_name_t,int> > >Graph; I also use a map to find the vertex according to a given state typedef std::tr1::unordered_map<state_t *, vertex_t> state_vertex_map_t; In the destructor of my planner object, every time I try to free anything (from the map or from the vertices of the graph) I'v got an "double free or corruption" error. Could you help me to find the way to free all the memory ? Thank you. Best regards Matthieu

2 4

Re: [Boost-users] Defining Statements with Proto
by David A. Greene 04 Jan '10

04 Jan '10

On Sunday 03 January 2010 18:55:36 you wrote: > On Sunday 03 January 2010 12:15:59 you wrote: > > > There were a few other small problems, but on the whole, you were > > > pretty close. > > You moved the code around a bit so the diff isn't trivial. What were the > other small problems? Ah, I found one. Duplicated template parameters for the expression rules. -Dave

1 0

[test] Unexpected Exceptions
by Florian Winter 04 Jan '10

04 Jan '10

Hi, What is the intended behavior of Boost.Test if an unexpected exception is thrown? Example: void do_not_throw() throw() { throw "unexpected"; } Some of my applications using Boost.Test seem exit with an error code of 0 (successful) if an unexpected exception is thrown. Is this the intended behavior? Best regards, fw

3 5

[UTF] Accessing global fixture members within tests
by Simon Pickles 04 Jan '10

04 Jan '10

Hi I am using the Unit Test Framework to test my use of boost.python. I am using a global fixture to initialize the python intepreter a single time (rather than before each test). My problem is I would like to use the Python initialising class in my tests, and I can't figure out how to do this. I don't want to use normal fixtures since each test would call initialize the interpreter, causing problems. Here is what I want to do: ////////////////////////////////////////////////////// #include <boost//python.hpp> #include <boost//test//unit_test.hpp> #include "..//pythonManager//pythonManager.h" using namespace boost; // First make a global fixture which sets up python interpreter struct PythonManagerFixture { PythonManagerFixture() { pm.Init(); } PythonManager pm; }; // Set up suite BOOST_AUTO_TEST_SUITE(TestPythonManager) // Only init python interpreter once BOOST_GLOBAL_FIXTURE(PythonManagerFixture); BOOST_AUTO_TEST_CASE(TestInit) { // Get python globals dict python::dict globalDict = python::extract<python::dict>(pm.m_pyGlobals); // COMPILE ERROR 'pm' : undeclared identifier // Has logging module been imported? BOOST_REQUIRE(globalDict.has_key("logging")); } BOOST_FIXTURE_TEST_CASE(TestLogCallback, PythonManagerFixture) // RUNTIME ERROR pm.Init called a second time - bad { function<void(const string&, const string&)> f = pm.GetLogFunction(); // pm is in scope though.... BOOST_REQUIRE(f); } // End the suite BOOST_AUTO_TEST_SUITE_END() /////////////////////////////////////////////////////////////////// I need a way to supply the PythonManager instance, pm, to the test cases. Thanks for any advice Simon

2 3

alignment question
by Slawomir Lisznianski 03 Jan '10

03 Jan '10

I'm having trouble implementing properly aligned storage. The following example shows conceptually what I'm trying to do, but it's NOT a correct way of doing it. template <typename T> struct envelope { size_t len = sizeof(T); // <--- compiler may pad here, hence this doesn't work... T body; }; So, given an arbitrary type T, we want to create an envelope that will have the size of T immediately before it, with no padding between len and body. The user would then be able to: envelope<MyStruct> env; env.body.x = ... send(env); I'm curious if anyone has run into something similar and have a clean solution. Our code currently has to defer to temporary buffers and memcpy'ing data back to it. Thanks, Slaw

3 2

Re: [Boost-users] regex
by ninti＠internode.on.net 03 Jan '10

03 Jan '10

Thanks for your help, linking was exactly the problem. As you can no doubt tell, I'm new not only to Boost but C++ in general. If you have the patience for one more newbie question ... I have the following code (in a FastCGI++ web application). The intention is to sanitise user input by allowing only alphanumeric characters, spaces, full stops and hyphens, simply replacing everything else with underscores: std::string input_name = environment.posts["name"].value; std::string filler = "_"; boost::regex re("[^a-zA-Z0-9\s.-]"); std::string output_name = boost::regex_replace(input_name, re, filler); Compilation fails with the following error: contact.cpp: In member function ‘virtual bool Contact::response()’: contact.cpp:48: error: unknown escape sequence '\s' The regex works fine with the \s removed (but of course replaces all spaces with underscores). These alternatives compile OK but don't work (program still replaces all spaces with underscores): [^a-zA-Z0-9\\s.-] [^a-zA-Z0-9[:space:].-] I know this is going to be something pretty basic, but how can I include whitespace - non-breaking, breaking, tabs, newlines, the lot - in the regex above as symbols to NOT be replaced? Thanks again Mike

2 1

regex
by ninti＠internode.on.net 03 Jan '10

03 Jan '10

Boost 1.40 on CentOS 5.3 64bit: I have #included <boost/regex.hpp> in a file but get the following compile time error: /tmp/ccXM0HkR.o: In function `boost::basic_regex<char, boost::regex_traits<char, boost::cpp_regex_traits<char> > >::assign(char const*, char const*, unsigned int)': contact.cpp:(.text._ZN5boost11basic_regexIcNS_12regex_traitsIcNS_16cpp_regex_traitsIcEEEEE6assignEPKcS7_j[boost::basic_regex<char, boost::regex_traits<char, boost::cpp_regex_traits<char> > >::assign(char const*, char const*, unsigned int)]+0x27): undefined reference to `boost::basic_regex<char, boost::regex_traits<char, boost::cpp_regex_traits<char> > >::do_assign(char const*, char const*, unsigned int)' collect2: ld returned 1 exit status This file definitely exists at /usr/include/boost/regex.hpp. Another Boost #include in the same file (<boost/date_time/posix_time/posix_time.hpp>) produces no error. 'ldconfig -p' lists libboost_regex.so. What am I doing wrong? Thanks Mike

2 1