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

RE: regex/string algorithm problem
by The Grumpiest Man You Know 22 Feb '05

22 Feb '05

Boy! You guys are SPOOKY good. I wrestle with something 'til I give up and then I go to write a report and while I formulate the words you start feeding back answers! The reason the code I sent "hangs" is because it's iterating over a zero length match. which consumes 0 characters of input per loop. I must do that once a year in any language I use, you think I'd learn huh? If I can reproduce the segfault and I can't see why it's my fault, can I post that too? Or can we skip that and have y'all start channeling me the clues now. :) Thanks again for your time. -- Blue Skies

3 3

shared_ptr: proper removal from STL Containers?
by Jeremy Hill 21 Feb '05

21 Feb '05

Greetings, I have only recently begun exploring boost, and I must say that I have found it very interesting. In particular I've been focusing on boost::shared_ptr. However, I'm not quite sure how to 'deallocate' shared_ptrs that are in a Container, such as a vector. I'd like to issue some pseudocode to help put things in perspective: class Descriptor { public: bool valid; Descriptor() : valid(true) {} } struct InvalidDesc { bool operator()(Desc_ptr &d) { return (*d).valid; } }; class Desc_broker { private: vector< boost::shared_ptr< Descriptor > > desc_list; public: //These functions are, in this order, called elsewhere: void fun_1() { <conditionally invalidate Descriptors> }; void fun_2() { <iterate through desc_list and do stuff> }; void cleanup() { desc_list.erase( remove_if(desc_list.begin(), desc_list.end(), InvalidDesc()), desc_list.end() ); } }; There is the class 'Descriptor' that is contained within a shared_ptr, which is wrapped in the 'Desc_broker' class. Once a Descriptor is invalidated, such as in fun_1() above, it is no longer needed. I'll need to get rid of invalid Descriptors before the vector is iterated through in fun_2(). However, since the shared_ptr for the invalid Descriptor will not go out of scope, to the best of my knowledge I'll need to help it along by using something along the lines of 'cleanup', which will remove all desc_ptrs whose Descriptors are invalid (via std::remove_if and the InvalidDesc predicate). However, this feels wrong to me because it seems like it is against what shared_ptrs are trying to accomplish. And, in fact, Boost agrees--I receive this on cleanup(): assertion "px != 0" failed: file "/usr/local/include/boost/shared_ptr.hpp", line 253 Aborted (core dumped) What is the correct way to remove the invalid Descriptors? Also, I apologize if this is considered the unacceptable topic "Requests for help interpreting the C++ standard." If that is the case, please tell me so that I may not offend in future posts. Thank you, Jeremy Hill

2 2

Re: problem with lambda and function<>
by Arturo_Cuebas＠bpmicro.com 21 Feb '05

21 Feb '05

In my previous post the sample code was a lot more than what what necessary to ask the question. I could have given this simpler example: #include <boost/function.hpp> #include <boost/lambda/lambda.hpp> #include <boost/lambda/bind.hpp> using namespace boost; using namespace boost::lambda; int f(void) { return 5; } int main(void) { int test = 0; (_1 = bind(&f))(test); // No compile error function<int (void)> functorized_f(&f); (_1 = functorized_f)(test); // Compile error return 0; } This is what my question really boils down to: just like naming delayed variables, is it possible to predefine and name the result of a call to bind() outside a lambda expression? /*what goes here?*/ bound_f = bind(&f); (_1 = bound_f)(test);

1 0

boost::filesystem -- handling dead symlinks
by Matthias Kaeppler 21 Feb '05

21 Feb '05

Hello, I wonder if there is any means of telling boost::filesystem::directory_iterator to skip broken symlinks. It's sort of an annoyance working with them, since operations like is_directory will throw when invoked on a broken symlink. I am keeping a vector of boost::filesystem::pathS, and as soon as I operate on a broken symlink, an exception will be thrown. I do catch it, but that doesn't really help since the program will return from the method where the exception was thrown. I would find it much more convenient to simply skip these broken files, instead of jumping out of the procedure (IMO it's not even exceptional behavior, dead symlinks are pretty common). Any idea what I can do about this? -- Matthias Kaeppler

3 4

MPL iterator category
by Philippe Mori 21 Feb '05

21 Feb '05

In C++ Template Metaprogramming book, there is the expression i::category in tables 5.1, 5.2 and 5.3 that says that the result is convertible to mpl::forward_iterator_tag, mpl::bidirectional_iterator_tag and mpl::random_access_iterator_tag respectively. Since a random access iterator is bidirectional iterator which is in turn a forward iterator, I was expecting that i::category for a bidirectional iterator would also be convertible to mpl::forward_tag which is not the case. So if an algorithm has to works differently for random-access iterator but it the same for forward and bidirectional iterator, it seems to me that I would need some extra logic to uses forward iterator algorithm if I have a bidirectional iterator as theree are no implicit conversion. In fact, in the implementation we derives from int_<n> where n is 0, 1 or 2 depending on the iterator type. Philippe

3 2

boost::normal_distribution in boost 1.32
by Roman 21 Feb '05

21 Feb '05

Hi all. I observe boost::normal_distribution always return "nan" with all RNGs whereas other distributions behave normally. I'm talking about boost 1.32 - in earlier version boost::random has different interface.

2 2

Boost 1.32.0 and VS.NET 2005 under AMD64
by p.klotz＠icoserve.com 21 Feb '05

21 Feb '05

There exist a few problems when trying to compile Boost 1.32.0 under AMD64 with VS.NET 2005: * No toolset exists to compile Boost 1.32.0 native on the AMD64 platform (no cross compile). The attached file vc-8_0-amd64-tools.jam fixes that. * InterlockedExchange() in line 44 of lwm_win32.hpp collides with a macro of the same name in WinBase.h One finds in WinBase.h: #elif defined(_M_AMD64) && !defined(RC_INVOKED) ... #define InterlockedExchange _InterlockedExchange I changed the name of Boost's function to InterlockedExchangeBoost() to avoid this problem. * No type long_type exists when compiling with the Microsoft Compiler. The 64Bit part in lwm_win32.hpp seems to have been written for the Intel Compiler. An additional #ifdef now checks if the 64Bit compiler is from Intel or Microsoft. The InterlockedExchange() and long_type issues are fixed by boost_1_32_0-AMD64.patch. Maybe these changes can be integrated in some form in the next version of Boost. Best regards, Peter.

1 0

Creating a mst graph from another, undirected one
by aitor 21 Feb '05

21 Feb '05

Hello, i'm working with a (huge) undirected graph, with severeal internal properties in both vertices and edges, including weights. At one point, i compute the minumim spanning tree of the graph. The question is simple: how can i create another graph (a tree, so a directed graph) from the original undirected graph plus the list of edges conforming the mst ? And how can i do it _efficiently_ (considering that the original graph will be huge) ? Consider this seudo-code: #include <boost/graph/adjacency_list.hpp> #include <boost/graph/graph_traits.hpp> #include <boost/property_map.hpp> #include <boost/graph/properties.hpp> #include <boost/graph/kruskal_min_spanning_tree.hpp> typedef adjacency_list < boost::listS, boost::vecS, boost::undirectedS, VertexProperties, EdgeProperties> UGraph; typedef adjacency_list < boost::listS, boost::vecS, boost::directedS, VertexProperties, EdgeProperties> MSTree; void createGraph() { UGraph g; add_vertices(); add_edges(); compute_edges_weights(); vector<graph_traits<UGraph>::edge_descriptor> mst_edges(num_vertices(g)); kruskal_minimum_spanning_tree(g, back_inserter(mst_edges)); // At this point i want to create another graph of type MSTree from // g and mst_edges. How can i do it efficiently ? } thank you in advance aitor

2 5

Boost graph problem removing last vertex
by gast 128 21 Feb '05

21 Feb '05

Dear all, I have a graph, but when I remove the last vertex, the edge isn't removed. This can even lead to a crash. typedef boost::adjacency_list<boost::listS, boost::vecS, //or listS, same problem boost::directedS, boost::property<boost::vertex_index_t, int, boost::property<boost::vertex_name_t, std::string> > > GraphCrash; typedef boost::graph_traits<GraphCrash>::vertex_descriptor vertex_t; typedef boost::graph_traits<GraphCrash>::edge_descriptor edge_t; typedef boost::graph_traits<GraphCrash>::vertex_iterator vertex_iterator; typedef boost::graph_traits<GraphCrash>::edge_iterator edge_iterator; GraphCrash graph; //vertices vertex_t v1 = boost::add_vertex(graph); boost::put(boost::vertex_name, graph, v1, "v1"); vertex_t v2 = boost::add_vertex(graph); boost::put(boost::vertex_name, graph, v2, "v2"); //edges boost::add_edge(v1, v2, graph); //remove last vertex boost::remove_vertex(v2, graph); //should not give any edge, however...: std::pair<edge_iterator, edge_iterator> prEdges; for (prEdges = boost::edges(graph); prEdges.first != prEdges.second; ++prEdges.first) { const edge_t e = *prEdges.first; const vertex_t u = boost::source(e, rGraph); const vertex_t v = boost::target(e, rGraph); } Do I something wrong? I use vc++ 7.1. Wkr, me (btw entering this message through gmane, it bounces back, with error that is is a top posting. Strange...) _________________________________________________________________ Talk with your online friends with MSN Messenger http://messenger.msn.nl/

4 4

regex/string algorithm problem
by The Grumpiest Man You Know 21 Feb '05

21 Feb '05

Hi, I think this is a bug but I'm willing to believe I'm misusing something. As a side point, I'm unsure of how to choose the regex syntax. I think I use boost::regex::perl as the second argument to the constructor but I haven't had much look with either that or the char_classes value. Anyway for me the snippet below the ----s hangs in the second find_all_regex. Ok, it may not hang, it may be building up a nice big core to dump but I only leave it for 20 seconds or so. This started with me trying to use find_regex which segfaulted when I tried to increment the iterator for the first time. I thought that might be me so I tried the safer way of just asking someone else to do all the real work :). This might prompt calls of "known issue." if I haven't messed up. In which case would someone care to hit me with a clue stick as to how I should check before bugging the list? Thanks for your time. Using Debian unstable: ii libboost-regex 1.32.0-2 regular expression library for C++ ii libboost-regex 1.32.0-2 regular expression library for C++ which I'm pretty sure is one for the library and one for the dev headers but I'm a noob there too, so I could be wrong. The compiler is: $ g++ -v Reading specs from /usr/lib/gcc-lib/i486-linux/3.3.5/specs Configured with: ../src/configure -v --enable-languages=c,c++,java,f77,pascal,objc,ada,treelang --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-gxx-include-dir=/usr/include/c++/3.3 --enable-shared --with-system-zlib --enable-nls --without-included-gettext --enable-__cxa_atexit --enable-clocale=gnu --enable-debug --enable-java-gc=boehm --enable-java-awt=xlib --enable-objc-gc i486-linux Thread model: posix gcc version 3.3.5 (Debian 1:3.3.5-8) -------------- #include <string> #include <iostream> #include <boost/algorithm/string_regex.hpp> template<class T> std::ostream& operator<<(std::ostream& cout, const std::vector<T>& v) { std::copy(v.begin(), v.end(), std::ostream_iterator<T>(cout, " ")); return cout; } main() { std::string line1("word -opt 5 -opt 45M -flag word -flag -opt word"); typedef boost::find_iterator<std::string::iterator> fi; for (fi i = boost::make_find_iterator(line1, boost::first_finder("word", boost::is_equal())); ! i.eof(); ++i) { std::cout << boost::copy_iterator_range<std::string>(*i) << std::endl; } std::vector<std::string> result; boost::regex rx1(" [-a-z]*"); boost::find_all_regex(result, line1, rx1); std::cout << result.size() << result << std::endl; result.clear(); boost::regex rx2("[-a-z]*"); boost::find_all_regex(result, line1, rx2); std::cout << result.size() << result << std::endl; std::cout << line1 << std::endl; } -- Blue Skies

2 1