I've read your review and i've corrected formatting on my patch. It's seems correct now. I've also corrected ChangeLog but it was tricky for myself (write correct and clear english and think about what is necessary details that i have to write in ChangeLog). I've wrote documentation for this feature. All of these you can find in attachment. I've run general testsuites in original unpatched gdb-7.2 and save stdout and stderr to separate files. After i've run exactly same tests (make check) in patched gdb-7.2 and save to another separate files. diff between these files shows me that my patch don't bring side effects. Of course i understand that general testsuites can check only for rough mistakes. So i should learn DejaGNU to test my patch. I think on how i can test possible side effects from my patch. I can send 'make check' results if you want. So my patch, doc patch and ChangeLog needs a new review in any case. I'm also send separate email to Tom Tromey about how i should assign copyright for the community. On Wed, Jun 29, 2011 at 2:35 AM, Phil Muldoon wrote: > Tom Tromey writes: >> Whether this one meets the bar, I don't know.  Is basename really the >> obvious transform to apply?  What about just dropping the compilation >> directory? > > Well in this context, user-driven needs, to me, are the best bar.  If > this contributor has written this patch, with a specific need (nota > super-specialised) in mind that is great.  I wish I had the backtrace > Pythonic interface ready, but, OTOH, every use-case is great to mould > that functionality. > > Cheers, > > Phil > I thought on cutting off compilation path but i decided to make it easier to start. If my tiny patch and idea is useful for somebody i glad to improve this. -- With best regards. Eldar Gaynetdinov