Thoughts about a better Quality Management process for Plasma

[Martin Gräßlin]

On Thursday 17 January 2013 12:30:03 Luca Beltrame wrote:
> In data giovedì 17 gennaio 2013 11:22:04, David Edmundson ha scritto:
> > backtrace it's impossible to see who's at fault. The backtrace only
> > shows some timer firing, or a python binding call or something generic
> > - it's impossible to tell what's at fault. I have no idea how we can
> > fix this.
> 
> Chiming in for the Python parts: at least part of the crashes are due to
> errors in the Python applets themselves. To be more precise, Python uses a
> garbage-collection method based on references, which includes the bindings
> to C++ objects. In such cases I've noticed that the exception makes the
> object marked for garbage collection → deleted at the C++ level as well,
> and if you try accessing it again → crash.
uh that sounds bad. I thought we do these script stuff to get the applet into 
a sandbox so that it cannot crash. Pity :-(
> 
> By errors I mean Python exceptions (IOW, errors in the Python code itself).
> 
> --
> Luca Beltrame - KDE Forums team
> KDE Science supporter
> GPG key ID: 6E1A4E79
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20130117/322ae1bd/attachment.sig>