Hi again, > If gcj cannot reliably walk the stack at the moment, I just want to add one comment here. The problem (at least as we found it @aicas) is not to reliably walk the stack, the problem is that the VMStackWalker interface is not well suited for 'walking' the stack. You can pull the whole class context array in one blob (which IMO should be avoided, a kind of iterator over the stackframes makes more sense, otherwise you could just as well ask the current Thread to give you an array of StackFrameElement), or you can get the last class on top of the context, which is not what is commonly needed (i.e. for logging). A clean(er) solution would be to offer a kind of enumeration/iterator - like interface, which can be really walked, without the need to look at the whole stack (which is far too expensive, when you are only interested in the first couple of stackframes). /Roman -- ?Improvement makes straight roads, but the crooked roads, without Improvement, are roads of Genius.? - William Blake -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 191 bytes Desc: Dies ist ein digital signierter Nachrichtenteil Url : http://developer.classpath.org/pipermail/classpath/attachments/20060516/a2ec4352/attachment.pgp
