Re: GWorkspace seg fault

[Fred Kiefer]

Enrico Sersale schrieb:
> On 2007-02-10 14:52:48 +0200 Fred Kiefer <fredkiefer@gmx.de> wrote:
> 
>> Stefan Bidigaray schrieb:
>>> On 2/9/07, *Enrico Sersale* <enrico@fibernet.ro
>>> <mailto:enrico@fibernet.ro>> wrote:
>>>
>>>      Yes, this is the normal behavior.
>>>      At this point I'd really need a gdb backtrace; you must only use
>>>      "debugapp" instead of "openapp" and send me the results...
>>>
>>>
>>> Keeping in mind I've never done anything like this before, here's the
>>> backtrace and gdb's output...  From GDB's output it looks like it's
>>> something to do with the fact that I'm using the cairo backend, which I
>>> guess I didn't mention before since I didn't think it was relevant
>>> seeing as everything else worked OK.
>>>
>>
>> The only thing that could trigger this problem in the cairo backend is
>> that _ct was not set, when calling this GSCurrentCTM method. I protected
>> all matrix calls from having an undefined _ct. (And submitted some more
>> stuff I had lying around) Could you please retest with current SVN?
>>
>> Thanks
>> Fred
> 
> Now, after a lot of "Font not found Helvetica", I get this:
> 
> <gw-trace>
> 
> and this:
> 
> <syspref-trace>
> 
> is from SystemPreferences selecting the "Indexing" modules.
> 
> In both the cases, it seems that there is something related with
> -GSReadRect: because both the apps try to create some resized imges
> using [NSBitmapImageRep initWithFocusedViewRect:].

In that case you better give up using the cairo backend for GWorkspace.
GSReadRect: is currently not supported for this backend. I only just
started to add some code here. I will look into improving on this, but
as far as I can currently see there is no way to extract data from a
cairo surface with cairo 1.2.4 and I don't plan to move to a more recent
cairo release any time soon.

Cheers,
Fred