On Mon, Nov 19, 2007 at 09:34:05AM +0800, zou lunkai wrote: > > I belive since this commit I get this result runnign key_event_testrunner > > with a debugging-build of Gnash cvs: > > Not sure, I tried again with local source(not updated) here, couldn't > catch the bug. has it been fixed now?
Fixed. > of smart pointers, I thought no arbitrary code could invalidate the > iterators or pointers in that execution context with respecting the GC > mechanism(collect only once per frame). Pointed-to objects can't be invalidated, iterators themselves can. Consider a list of two elements, execution of arbitrary code deleting both from the container. --strk; _______________________________________________ Gnash-commit mailing list Gnash-commit@gnu.org http://lists.gnu.org/mailman/listinfo/gnash-commit
