[Stackless] Stackless and tracing (for debugger)
richard at ccpgames.com
Wed Apr 26 16:09:07 CEST 2006
> From Harry Kalogirou on 26 April 2006 14:02
> I'm implementing a debugger for a project that uses Stackless python
> and I stumbled upon this problem: The Stackless
> implementation of Python
> seems to reset the c_tracefunc at scheduling.c:324 which causes the
> debugger to stop getting info about the executing script after a
> stackless.run() call.
> Why is this implemented this way? Is there a bigger picture
> I'm missing?
> Is there some "tasklet enabled" method for debugging that is used in
> Stackless that deprecated the old debugger method?
Because tracing only applies to the current tasklet. As I understand
it, it stores and resets the tracing settings, then switches to the
next tasklet and when it returns, it restores your tracing values
again. So what you are tracing would just be the main tasklet.
What if you wrapped each tasklet created with code which installed
your tracing hooks for it before the main tasklet code was executed.
Could you then have your debugger provide a Stackless aware overview
of all executing tasklets?
Stackless mailing list
Stackless at stackless.com
More information about the Stackless