- Boost-users - lists.boost.org

Regex lib and thread safety
by Leon Mergen 27 Jan '05

27 Jan '05

Hello, I am sorry if this is a stupid question, or simply inappropriatie for this list. I am currently camping with this bug for way too long, so I decided to ask this mailinglist. I am currently in the process of finding out where a segfault is, and how to fix it. Now, I am using the boost regex library to parse my regexes, and for some reason when multiple threads use this regex library, it seems to mess up and generate segfaults (it works perfectly with only one thread). Since it also doesn't always occur on the same place, and doesn't even occur /all/ the times, I have a strong feeling this has something to do with thread safety. On the Boost website is was able to find out that the boost regex library should be thread safe when BOOST_HAS_THREADS is defined; I've tested this, and this worked. ( http://www.boost.org/libs/regex/doc/thread_safety.html ) However, I also was able to find a mailinglist message that provides some instructions on how to make certain boost regex functions thread-safe : ( http://lists.boost.org/MailArchives/boost/msg59110.php ) Now, I am confused; is the boost regex library thread-safe or not ? The solution provided in the mailinglist message ( move the regex_replace () function inside its own scope ) can't be applied here... :( I've posted the stacktrace when the problem occurs and the source of the code 'that matters' below. It also occurs in another function, I've posted the other function too. I hope anyone has any experience with this, or is able to help. Thanks in advance! Regards, Leon Mergen One stacktrace: --- #0 std::string::compare(char const*) const (this=0x0, __s=0x40338b81 "C") at /root/gcc-3.3.2/i686-pc-linux-gnu/libstdc++-v3/include/bits/basic_string.h:257 #1 0x4004b101 in boost::c_regex_traits<char>::update() () from /usr/local/lib/libboost_regex-gcc-1_31.so.1.31.0 #2 0x40084cd9 in boost::reg_expression<char, boost::regex_traits<char>, std::allocator<char> >::set_expression(char const*, char const*, unsigned) () from /usr/local/lib/libboost_regex-gcc-1_31.so.1.31.0 #3 0x080648aa in unsigned boost::reg_expression<char, boost::regex_traits<char>, std::allocator<char> >::set_expression<std::char_traits<char>, std::allocator<char> >>(std::basic_string<char, std::char_traits<char>, >>std::allocator<char> > const&, unsigned) (this=0x413d29cc, >>p=@0x413d2adc, f=34055) at basic_regex.hpp:110 #4 0x080647bf in reg_expression<std::char_traits<char>, std::allocator<char> > (this=0x413d29cc, p=@0x413d2adc, f=33031, a=@0x413d29bc) at basic_regex.hpp:114 #5 0x0806474c in basic_regex<std::char_traits<char>, std::allocator<char> > (this=0x413d29cc, p=@0x413d2adc, f=33031, a=@0x413d29bc) at basic_regex.hpp:361 #6 0x080633ee in RegexEngine::parseReplaceRegex(std::string*, std::string*, std::string*) (this=0x80d1160, regex=Cannot look up value of a typedef) at RegexEngine.cc:8 #7 0x0808070a in ModuleDigitsToText::applyModule(std::string*, std::string) (this=0x80d3190, data=0x80d16e0, language= {static npos = 4294967295, _M_dataplus ={<allocator<char>> = {<No data fields>}, _M_p = 0x80d244c "en"}, static _S_empty_rep_storage = {0, 0, 77, 0}}) at ModuleDigitsToText.cc:565 #8 0x08080270 in ModuleDigitsToText::applyModule(InputData*) (this=0x80d3190, inputData=0x80d16d8) at ModuleDigitsToText.cc:527 #9 0x08062200 in ModuleParser::parse(InputData*) (this=0x80d1150, inputData=0x80d16d8) at ModuleParser.cc:105 #10 0x0805e1db in Normaliser::normalise() (this=0x80d1698) at Normaliser.cc:107 #11 0x080831f4 in Client::operator()() (this=0x80ca808) at Client.cc:44 #12 0x080542bf in boost::detail::function::void_function_obj_invoker0<Client, void>::invoke(boost::detail::function::any_pointer) (function_obj_ptr= {obj_ptr = 0x80ca808, const_obj_ptr = 0x80ca808, func_ptr = 0x80ca808, data = "\b"}) at function_template.hpp:128 #13 0x400a69f4 in boost::thread_group::join_all() () from /usr/local/lib/libboost_thread-gcc-mt-1_31.so.1.31.0 #14 0x401cff60 in pthread_start_thread () from /lib/i686/libpthread.so.0 #15 0x401d00fe in pthread_start_thread_event () from /lib/i686/libpthread.so.0 #16 0x402f7327 in clone () from /lib/i686/libc.so.6 --- And are the two functions where the segfault occurs. Line 8 of RegexEngine.cc is the "boost::regex expression(*regex);" line of the parseReplaceRegex() function. --- std::string RegexEngine::parseReplaceRegex (const std::string * regex, const std::string * replace, const std::string * subject) { // Create an expression out of the regex boost::regex expression(*regex); // Return the replaced regex { return boost::regex_replace(*subject, expression, *replace, boost::match_default | boost::format_sed); } } bool RegexEngine::parseMatchRegex (std::string * regex, std::string * subject, RegexResultSet * what) { #ifdef DEBUG std::cout << "+ RegexEngine::parseMatchRegex[long] ( " << *regex << ", " << *subject << " )" << std::endl; #endif // DEBUG // Prepare result set boost::match_results <std::string::const_iterator> results; // Create string iterators std::string::const_iterator start = subject->begin(); std::string::const_iterator end = subject->end(); // Initialise boolean to know wether a match was reached bool success = false; // Create an expression out of the regex boost::regex expression(*regex); std::vector < std::string > regexResults; // Iterate over the results while (boost::regex_search(start, end, results, expression, boost::match_default)) { regexResults.clear(); for (boost::match_results<std::string::const_iterator>::iterator i = results.begin(); i != results.end(); ++i) { regexResults.push_back(*i); } // Push back a result what->push_back(regexResults); // Set new string iterator start = results[0].second; // Mark that a match was reached success = true; } #ifdef DEBUG std::cout << "- RegexEngine::parseMatchRegex ()" << std::endl; #endif // DEBUG // Return wether a match was reached return success; } ---

4 11

time_period constructor question
by Stuart Siegel 27 Jan '05

27 Jan '05

Can anyone tell me why the line: time_period t4(t2); in the attached cpp file doesn't compile but the line: time_period t3(t1); does? it seems like i should be calling the same copy constructor in both cases but g++ 3.3.4 complains: error: no matching function for call to ` boost::date_time::period<boost::posix_time::ptime, boost::posix_time::time_duration>::period(boost::posix_time::time_period (&)(boost::posix_time::ptime, boost::posix_time::ptime))' /usr/include/boost/date_time/period.hpp:64: error: candidates are: boost::date_time::period<boost::posix_time::ptime, boost::posix_time::time_duration>::period(const boost::date_time::period<boost::posix_time::ptime, boost::posix_time::time_duration>&) /usr/include/boost/date_time/period.hpp:109: error: boost::date_time::period<point_rep, duration_rep>::period(point_rep, duration_rep) [with point_rep = boost::posix_time::ptime, duration_rep = boost::posix_time::time_duration] /usr/include/boost/date_time/period.hpp:99: error: boost::date_time::period<point_rep, duration_rep>::period(point_rep, point_rep) [with point_rep = boost::posix_time::ptime, duration_rep = boost::posix_time::time_duration] //--------------------------------cut here------------------- #include "boost/date_time/posix_time/posix_time_types.hpp" #include "boost/date_time/gregorian/gregorian_types.hpp" int main() { using namespace boost::posix_time; using namespace boost::gregorian; //---------------------------------------------- date dmin( min_date_time) ; ptime tmin( dmin ); date dmax( max_date_time) ; ptime tmax( dmax ); time_period t1( tmin, tmax ); //--------------------------------------------- time_period t2( ptime(min_date_time), ptime(max_date_time) ); //why does this compile (on g++ 3.3.4) time_period t3(t1); //but this doesn't? i.e., why isn't a time_period a time_period a time_period? time_period t4(t2); return 0; } __________________________________ Do you Yahoo!? All your favorites on one personal page � Try My Yahoo! http://my.yahoo.com

4 5

analogy to mem_fun_ref to member variables
by Chris Goller 27 Jan '05

27 Jan '05

I have a struct that contains two values. struct box { double width; double length; } vector<box> boxes; I want to accumlate all the widths of the boxes. The only solution I've found is to create a member function get_width that will return the width for use in mem_fun_ref eg: accumulate(make_transform_iterator(boxes.begin(), mem_fun_ref(&boxes::get_width)), make_transform_iterator(boxes.end(), mem_fun_ref(&boxes::get_width)), 0.0, plus<double>())); Is there a boost way of not using mem_fun_ref, but rather returns the public member variable. Something that could be named mem_var_ref that would look like this: accumulate(make_transform_iterator(boxes.begin(), mem_var_ref(&boxes::width)), make_transform_iterator(boxes.end(), mem_fun_ref(&boxes::width)), 0.0, plus<double>())); This way I wouldn't need to add extra get logic to a structure that doesn't really need it othewise. Thanks, Chris

3 2

boost/graph 1.32.0 : compilation errors on a SUN machine
by laurent de Vito 26 Jan '05

26 Jan '05

Hello, I cannot compile the boost/graph library 1.32.0 on a SUN machine using Sun C++ 5.5. I get the following error message : boost_1_32_0/boost/graph/depth_first_search.hpp", line 377: Error: Default arguments cannot be added in later declarations of the template function in the same scope. which does make sense since the compiler cannot decide which routine depth_first_visit to use if the second routine (line 371) has an extra template but the the extra template has a default value. So in boost/graph/depth_first_search.hpp, I simply deleted the default argument : template <class IncidenceGraph, class DFSVisitor, class ColorMap, class TerminatorFunc> void depth_first_visit (const IncidenceGraph& g, typename graph_traits<IncidenceGraph>::vertex_descriptor u, DFSVisitor vis, ColorMap color, TerminatorFunc func = TerminatorFunc()) becomes template <class IncidenceGraph, class DFSVisitor, class ColorMap, class TerminatorFunc> void depth_first_visit (const IncidenceGraph& g, typename graph_traits<IncidenceGraph>::vertex_descriptor u, DFSVisitor vis, ColorMap color, TerminatorFunc func) Unfortunately the compilation was afterwards not successful; this time I got "boost_1_32_0/boost/graph/named_function_params.hpp", line 500: Error: Cannot return const boost::bgl_named_params<boost::detail::components_recorder<int*>, boost::graph_visitor_t, boost::no_property> from a function that should return boost::detail::error_property_not_found. "boost_1_32_0/boost/graph/depth_first_search.hpp", line 354: Where: While instantiating "boost::get_param<boost::graph_visitor_t, boost::root_vertex_t, boost::detail::components_recorder<int*>, boost::no_property>(const boost::bgl_named_params<boost::detail::components_recorder<int*>, boost::graph_visitor_t, boost::no_property>&, boost::root_vertex_t)". "boost_1_32_0/boost/graph/depth_first_search.hpp", line 354: Where: Instantiated from non-template code. "boost_1_32_0/boost/graph/depth_first_search.hpp", line 355: Error: Could not find a match for boost::vertices<boost::T>(const boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS, boost::no_property, boost::no_property, boost::no_property, boost::listS>). big thanks for any help, Laurent

3 3

uBLAS, Inverse of Matrices -- Urgent
by Hossein Haeri 26 Jan '05

26 Jan '05

Dear all, Am I missing anything? Isn't there any mechinary in uBLAS which can compute the inverse of a matrix? That being the case, isn't that hilarious? Or, if there is, what's that? All of the Best, --Hossein ___________________________________________________________ ALL-NEW Yahoo! Messenger - all new features - even more fun! http://uk.messenger.yahoo.com

7 34

(?) MPL vs boost.lambda concerning lambda expressions
by Roman 26 Jan '05

26 Jan '05

Hi. Has anybody a suggestion about MPL vs boost.lambda concerning lambda expressions?

2 1

VC7.1 Internal Compiler error compiling Graphviz 1-32
by Alessandro Cossetto 26 Jan '05

26 Jan '05

I tried to compile graphviz library but compiling graphviz_digraph_parser.cpp (and graphviz_graph_parser.cpp ) VC7.1 gave me an internal compiler error (compiler file 'msc1.cpp', line 2701). The error occours in boost\gaph\properties.hpp at line 203: template <class Graph, class Property> struct property_map { private: typedef typename property_kind<Property>::type Kind; typedef typename detail::property_map_kind_selector<Kind>::type Selector; ==> typedef typename Selector::template bind_<Graph, Property> Bind; typedef typename Bind::type Map; public: typedef typename Map::type type; typedef typename Map::const_type const_type; }; It happens when compiler tries to resolve the graph_traits instance (line 124 in graphviz_digraph_parser.cpp) namespace graphviz { ==> typedef boost::graph_traits<GRAPHVIZ_GRAPH>::vertex_descriptor Vertex; typedef boost::graph_traits<GRAPHVIZ_GRAPH>::edge_descriptor Edge; typedef GRAPHVIZ_GRAPH Subgraph; Using GCC I compiled the library wihtout problem. I was not able to find a workarount. Can anyone help me? Alessandro

1 0

[ANN] Interfaces Library
by Jonathan Turkanis 26 Jan '05

26 Jan '05

I am pleased to announce the much anticipated (by me) release of a preliminary version of my interfaces library. The download page and online docs are here: http://www.kangaroologic.com/interfaces/ (For fun, click on "Online Documentation" and then on the Boost logo.) The library is an implementation of Christopher Diggins's proposal to add interfaces to C++, which was posted on comp.std.c++ early last year and later was described in the C/C++ User Journal. It provides a macro-based Interface Definition Language (IDL) which can be used to define C++ class types called interfaces. An interface is a lightweight value type associated with a set of named function signatures. An interface instance can be bound at runtime to any object which implements the interface, i.e., to any object of a type with accessible non-static member functions having the same name and signature as the set of functions associated with the interface. The functions of the bound object can then be invoked through the interface instance using the "dot" operator. Binding is completely non-intrusive: the object's type need not declare any virtual functions or derive from any particular base class. Current applications include: * Non-intrusive dynamic polymorphism - interfaces can often be used in place of abstract base classes, and are sometimes much faster. * Dynamic inheritance - allows function calls to be forwarded automatically to an object specified at runtime. * Smart Interface Pointers - smart pointers which can manage the lifetime of any object whose type implements a given interface. * Smart References - like smart interface pointers, but the managed object is accessed using the dot operator. * The Object Oriented Template Library, a project in the initial stages of development, by Christopher Diggins. Planned future applications include: * Runtime reflection - will allow an interface's functions to be enumerated and looked up by name at runtime * Aspect Oriented Programming as described in a August 2004 Dr. Dobb's Journal article by Christopher Diggins. Possible future applications include: * Integration with component architectures such as COM and CORBA. * Boost.Names - by abstracting the various uses of identifiers in C++ into preprocessor generated types called names, C++ classes and functions can be constructed in a straightforward manner using template metaprogramming. Best Regards, Jonathan

1 0

what does BOOST_NO_ARGUMENT_DEPENDENT_LOOKUP
by Kevin Lee 25 Jan '05

25 Jan '05

I need some expert's opinion on this: When I try to use the boost 1.32.0 in my work env, if i don't define this flag BOOST_NO_ARGUMENT_DEPENDENT_LOOKUP, the lib/smart_ptr/test/intrusive_ptr_test.cpp fails to compile. if i do define it, my compiler would be happy (HP aCC) on intrusive_ptr_test. My questions are: 1) What is the BOOST_NO_ARGUMENT_DEPENDENT_LOOKUP? 2) What will be the other impacts if I do define it in my code globally besides the intrusive_ptr_test? Thanks in advance! Kevin Lee ----------------------------------------- This message and its attachments may contain privileged and confidential information. If you are not the intended recipient(s), you are prohibited from printing, forwarding, saving or copying this email. If you have received this e-mail in error, please immediately notify the sender and delete this e-mail and its attachments from your computer.

2 1

Weak_ptr lost its operator==
by Brian McCandless 25 Jan '05

25 Jan '05

Hi, I just upgraded from boost 1.29 to boost 1.32, and found that operator== for weak_ptr has been dropped between these versions. My code uses std::find to find a weak_ptr in a container. This no longer compiles (g++-3.4.1) since find uses operator==. Has anyone else encountered this kind of problem? What are some good workarounds? Could operator== be put back? Thanks, Brian

2 1