extension, color, black and white causes core dump

[Matt Cargo]

Using Inkscape 0.48.3.1 r9886 (Mar 29 2012) on Ubuntu 12.

Attached file causes core dump when I try to covert it to black and white.

[jazzynico (jazzynico)]

Not reproduced on Ubuntu 11.04, Inkscape 0.48.2, 0.48.3.1 (not the official package!) and trunk revision 11325.

Does it crash with other extensions (in the color menu or another menu)? Does it affect the extension only (crash in the python code, but Inkscape is still alive) or the whole application?

Could you please run inkscape from the console in case it returns an error message before crashing.

[Alvin Penner (apenner)]

- reproduced on Windows 7, Inkscape 0.48.3.1. It is an Inkscape crash with no exit messages in DOS. It also occurs with other extensions like Color->Brighter.
- not reproduced on Windows XP, bzr rev 11275

[Matt Cargo (mcargo)]

The only message from the console is this:

Emergency save completed. Inkscape will close now.
If you can reproduce this crash, please file a bug at www.inkscape.org
with a detailed description of the steps leading to the crash, so we can fix it.

[su_v (suv-lp)]

On OS X 10.7.3 (64bit), default prefs (new ~/.config/inkscape)
- Not reproduced with Inkscape 0.48.2
- Reproduced with Inkscape 0.48.3.1 (consistently)
- Not reproduced with Inkscape 0.48+devel r11325

Backtrace attached (unfortunately I don't have a debug build of the stable 0.48.3.1 release at hand).

[su_v (suv-lp)]

Duplicate of Bug #944077 “inkscape crashed with SIGSEGV in Inkscape::Extension::Implementation::Script::effect()”?
<https://bugs.launchpad.net/inkscape/+bug/944077>

[Kris (kris-degussem)]

Having had a look on the code recently (not trying to simulate the bug myself), the backtrace indicates a null pointer dereference and if so it is a duplicate.

[jazzynico (jazzynico)]

> Not reproduced on Ubuntu 11.04, Inkscape 0.48.2, 0.48.3.1 (not the official package!) and trunk revision 11325.

Correction: I didn't test with 0.48.3.1, but with the 0.48.x branch...
Thus yes, it's a duplicate, and it's already fixed for 0.48.4.