On 8/15/14, 2:17 PM, Victor Porton wrote: > My testing code for Ada bindings of Raptor at > https://github.com/vporton/redland-bindings/tree/ada2012 (the script > run_all_tests) causes an infinite loop. > > When run under GDB it instead of infinite loop: > > Program received signal SIGSEGV, Segmentation fault. > malloc_consolidate (av=0xf7b43420 <main_arena>) at malloc.c:4151 > 4151 malloc.c: No such file or directory. > > The particular Ada line which causes the infinite loop: > > Assert(To_String(Datatype(Get_Literal(Term_1))) = "http://example.org";, > "Term_1 datatype"); > > My question: May this error be caused by an object not deallocated before > finalizing "world" object? > > In other words: If an object is not deallocated, is it just memory leak, or > may it lead for more serious errors?
I don't know the answer to the specific error here but you have to take care that you are respecting the allocation model. If a function returns a shared string (char*), the caller needs to copy it in order to preserve it. Similarly (but much rarer) an input char* may become owned by a raptor object and so the caller should not deallocate it but I don't think happens too often in raptor I often run the C tests under valgrind and other dynamic and static checkers to ensure all memory and resources are returned on all error paths. Dave _______________________________________________ redland-dev mailing list [email protected] http://lists.librdf.org/mailman/listinfo/redland-dev
