Linking error for libboost_graph in OSX with Xcode

goftAppageJef

New Member
I have been working on some graph code for a homegrown problem, and using my own text files to store the data had reached it's limit. So I had been looking for either a database or flat file solution to store my graph in. Eventually I decided on the Boost::Graph library graphml reader. This is where I started to run into problems.I have compiled all of the binaries and they are located at /usr/local/lib. My source files compile fine, and I have the graphml.hpp file included, but I get the linker error:\[code\]Undefined symbols for architecture x86_64: "boost::read_graphml(std::__1::basic_istream<char, std::__1::char_traits<char> >&, boost::mutate_graph&, unsigned long)", referenced from: void boost::read_graphml<boost::adjacency_list<boost::vecS, boost::vecS,boost::undirectedS, vertex_p_t, boost::property<boost::edge_weight_t, double, boost::no_property>, boost::no_property, boost::listS> > (std::__1::basic_istream<char, std::__1::char_traits<char> >&, boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS, vertex_p_t, boost::property<boost::edge_weight_t, double, boost::no_property>, boost::no_property, boost::listS>&, boost::dynamic_properties&, unsigned long) in GraphAnalyzer.old: symbol(s) not found for architecture x86_64clang: error: linker command failed with exit code 1 (use -v to see invocation)\[/code\]Now, in the xcode project, I do tell it to link to the framework library at /usr/local/lib/libboost_graph-d.a. I have the header include paths set to /usr/local/include/ and the library include path set to /usr/local/lib/.I would still really like to use the graphml reader, however, if it is going to be a problem, then I suspect that I will have to look into other options. Does anybody have any experience with using one of the graph-centric databases out there? What about a nosql type database? I just need a DB that can store the nodes and edges, and the relations (names, weights, etc, more labels) and then have an easy way to construct the graph quickly at runtime.
 
Top