- Boost-users - lists.boost.org

[Serialization] un-archiving container of strings surfaces malloc/free bug
by Mark Melton 06 Oct '06

06 Oct '06

The following program archives a vector of one string to the file "/ tmp/foo" and then immediately un-archives "/tmp/foo" into an empty vector of strings. Deep in the bowels of the un-archiving process, malloc reports a bad call to free as noted in the comments at the end of the program. The bad call to free is made during a call to string.reserve(). Varying the program shows that the bad free occurs once for each element in the vector and is always the same address. The behavior is consistent across text, binary and xml archivers as well as vectors and maps of strings. Under different conditions the same error occurs during a call to string.assign(). I have tried gcc3.3/4.0 and 4.1 with no change in the results. My current work- around is to use vector<char> instead of string. I am hoping that someone will be willing to help me debug this problem. I am running on a PPC Mac using the darwin ports development environment: Darwin mars.notlem.com 8.7.0 Darwin Kernel Version 8.7.0: Fri May 26 15:20:53 PDT 2006; root:xnu-792.6.76.obj~1/RELEASE_PPC Power Macintosh powerpc boost @1.33.1_0 gcc41 @4.1.1_0+darwin_8 cheers, mark ======================================================================== = #include <iostream> #include <fstream> #include <vector> #include <string> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/serialization/utility.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/string.hpp> using namespace std; int main(int argc, char const **argv) { typedef vector<string> string_vector_t; string_vector_t output, input; output.push_back("a string"); { ofstream ofs("/tmp/foo"); assert(ofs.good()); boost::archive::xml_oarchive archiver(ofs); archiver << BOOST_SERIALIZATION_NVP(output); } { ifstream ifs("/tmp/foo"); assert(ifs.good()); boost::archive::xml_iarchive archiver(ifs); archiver >> BOOST_SERIALIZATION_NVP(input); } return 0; } // $ ./bug // bug(27658) malloc: *** Deallocation of a pointer not malloced: 0x209c33c; This could be a double free(), or free() called with the middle of an allocated block; Try setting environment variable MallocHelp to see tools to help debug ======================================================================== ======== #0 0x9000621c in free () #1 0x94afb8ec in std::string::_Rep::_M_destroy () #2 0x94afe6d0 in std::string::reserve () #3 0x0007617c in boost::archive::basic_xml_grammar<char>::my_parse () #4 0x000769b8 in boost::archive::basic_xml_grammar<char>::parse_start_tag () #5 0x00076c70 in boost::archive::basic_xml_iarchive<boost::archive::xml_iarchive>::load_s tart () #6 0x00004fc4 in boost::archive::detail::iserializer<boost::archive::xml_iarchive, std::vector<std::string, std::allocator<std::string> > >::load_object_data (this=0x300eb0, ar=@0xbfffe9d8, x=0xbfffed00, file_version=62) at /opt/local/include/boost/archive/ basic_xml_iarchive.hpp:79 #7 0x000638e8 in boost::archive::detail::basic_iarchive::load_object () #8 0x00002d3c in main (argc=3149488, argv=0xbfffe9d8) at /opt/local/ include/boost/archive/detail/iserializer.hpp:375

1 0

Serialization of char*
by Deane Yang 06 Oct '06

06 Oct '06

The code below won't compile for me using VC8. Can someone explain? I'm sure I'm doing something dumb, but I'm too dumb to know what. #include <boost/archive/xml_oarchive.hpp> #include <iostream> #include <fstream> struct Object { char* ptr; }; int main (int argc, char* argv[]) { try { std::cout << "Saving portfolio to archive...\n"; Object object; object.ptr = new char[10]; for (int i= 0; i < 10; ++i) { object.ptr[i] = 'i'; } std::ofstream ofs(filename); assert(ofs.good()); boost::archive::xml_oarchive oa(ofs); const Object& obj(object); oa << BOOST_SERIALIZATION_NVP(obj); } catch (const xml_schema::exception& e) { std::cerr << e << std::endl; return 1; } }

1 2

is there limitation on the file size generated by serialization
by Cheok Yan Cheng 06 Oct '06

06 Oct '06

hello all, we have been using boost serialization library for quite some time in small scale project. so far, we are happy with boost serialization, since it offers several features which are not found in MFC framework. before we migrate boost serialization library is large scale project, can we know 1. is there any limitation on the file size generated by serialization. i.e. what is the maximum file size which can be written by boost serialization? 2. is there any limitation on the object size, which can be loaded/ unloaded by boost serialization? 3. does boost serialization offer any way (checksum?) to ensure the file being written is not corrupted? thank you very much yccheok __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

problem using boost::archive::xml_oarchive to serliaze a map
by Thomas D\\\'Silva 06 Oct '06

06 Oct '06

Hello, I am trying to use boost::archive::xml_oarchive to serialize a vector<map<string, map<string,int> > >. When I examine the serialized xml file, I see that some of the map elements are duplicated which causes an assertion when I try to load the file because a map cannot have two elements with the same key. I would appreciate any insight into this error. Thank you Thomas D'Silva /include/boost/serialization/collections_load_imp.hpp:69: void boost::serialization::stl::archive_input_map<Archive, Container>::operator()(Archive&, Container&) [with Archive = boost::archive::xml_iarchive, Container = std::map<std::string, int, std::less<std::string>, std::allocator<std::pair<const std::string, int> > >]: Assertion `result.second' failed. #include <boost/archive/xml_oarchive.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/map.hpp> #include <boost/serialization/string.hpp> #include <boost/serialization/nvp.hpp> using namespace std; using namespace boost::filesystem; class DistributionTable { private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(neTable); ar & BOOST_SERIALIZATION_NVP(mappedTable); } vector<map<string,map<string,int> > > neTable; vector<map<string,map<string,int> > > mappedTable; public: DistributionTable(){}; }; void saveTable(const DistributionTable &t, string filename) { ofstream ofs(filename.c_str()); boost::archive::xml_oarchive oa(ofs); oa <<BOOST_SERIALIZATION_NVP(t); ofs.flush(); } void loadTable(DistributionTable &t, string filename) { ifstream ifs(filename.c_str()); boost::archive::xml_iarchive ia(ifs); ia >>BOOST_SERIALIZATION_NVP(t); }

1 0

Ann: Statistical distributions / Mathematical Special Functions
by John Maddock 06 Oct '06

06 Oct '06

This package consists of 3 related components, and we're now seeking further feedback on these before the final push prior to a review: Statistical Distributions ~~~~~~~~~~~~~~~~~~~~~~~~~ Following feedback from the previous preview, we have re-organised so that distributions are now C++ classes. See the docs and especially the tutorial for illustrations on how this will be useful in real life. We would especially welcome feedback from those with experience of using statistics, and indeed from professional statisticians who are probably not Boosters. Mathematical Special Functions ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ These are currently focused on those functions that are useful for statistical applications: Floating point classification. Gamma/lgamma/digamma/beta/erf/erfc Ratios of gamma functions Factorials The incomplete gamma and beta functions. The inverses of the gamma, beta, and error functions. Derivatives of the incomplete gamma and beta functions. Compared to the previous release, most of the changes involve accuracy and quality of implementation issues. Error handling has now been revised to provide a comprehensive package-wide error handling scheme complete with meaningful error messages. Toolkit ~~~~~~~ Provides tools to assist in the implementation of numerical methods: Infinite series evaluation. Continued fraction evaluation. Polynomial and rational function evaluation. Root finding with derivatives (Newton/Halley/Schroeder methods). Root finding without derivatives. Function minimisation. Misc Tools ~~~~~~~~~~ These are strictly experimental, but are used in the development of approximations: they are provided to hopefully encourage others to develop further special function implementations. Tools include: Polynomial arithmetic and manipulation. Minimax approximations (the Remez algorithm). Helper functions used in testing etc. Documentation and source downloads are available online at www.johnmaddock.co.uk/toolkit Regards, John Maddock.

2 2

How to obtain old files from Boost.Vault?
by Markus Werle 06 Oct '06

06 Oct '06

Hi! In <http://groups.google.com/group/boost-list/browse_thread/thread/ 404f6a4eb1f13ed1/fe070423d5945258?q=SVD++ublas&lnk=nl&hl=en&> Le Farfadet Spatial explains that he has uploaded a file inverse-01262005.tar.gz with code under boost license to the vault (see bottom of page). Maybe I am stupid or maybe the vault is not a place for stuff to last forever: I failed to obtain this file. Also google fails here <http://www.google.com/custom?q=inverse&hq=site%3A www.boost-consulting.com+inurl%3A www.boost-consulting.com%2Fvault&search.x=0&search. y=0&cof=LW%3A277%3BL%3Ahttp%3A%2F%2F www.boost.org%2Fboost.png%3BLH%3A86%3BAH%3A center%3BGL%3A0%3BS%3Ahttp%3A%2F%2Fwww.boost.org% 3BAWFID%3A9b83d16ce652ed5a%3B&sa=Google+Search&domains= www.boost.org%3Blists.boost.org> So I have some questions: 1. Could someone point me to a place where this file still can be found? 2. If the vault is just a bin to put stuff that gets deleted later how can we (as community) avoid to loose the ingenious stuff that enters that place and keep it forever? I haste to download all that might be of interest for me, but sometimes I cannot foresee what might get interesting in some years. Also the quality of the code posted to the vault is rather high, so probably a bit more data persistence is quite appropriate, isn't it? regards, Markus

1 0

boost::regex problem
by boost＠larroy.com 06 Oct '06

06 Oct '06

Hi, I don't get why I'm getting diferent results with this regexp in perl and in boost::regex, is there something I'm doing wrong? regards echo " text/html; charset=iso-8859-1" | perl -wne 'if(m#([^/;]+/[^/;]+)#) { print $1."\n" }' http://rafb.net/paste/results/jrUGyO97.html -- Pedro Larroy Tovar, pedro at larroy dot com | http://pedro.larroy.com/ * Las patentes de programación son nocivas para la innovación * http://proinnova.hispalinux.es/

2 1

Non-intrusive NVP serialization of a base object
by Deane Yang 06 Oct '06

06 Oct '06

How do I do: BOOST_SERIALIZATION_BASE_OBJECT_NVP(my_base_class) in a non-intrusive serialize function?

2 1

[multi_index_container] Thread safety
by Prashant Thakre 05 Oct '06

05 Oct '06

Hi, As per the section "Complexity and exception safety": Iterator and reference validity is preserved in the face of insertions, even for replace and modify operations. Would it be safe to insert() from within two or more threads simultaneously ? Assumption being that the keys inserted will be always be different. -- regards, Prashant Thakre

4 3

graph-tool and boost/random
by Jens Müller 05 Oct '06

05 Oct '06

I cannot compile graph-tool on amd64/gcc4.1.1/Gentoo and Boost 1.33.1, while Tiago can with the same gcc. Maybe someone can take a look? See http://news.gmane.org/gmane.comp.science.graph.graph-tool.general/cutoff=9 ...

1 1