Bug#392579: Processed: Also seen here
Rene Engelhard wrote:
> tag 392579 + unreproducible
> [ please also always Cc the bugnr, so that your text actually gets sent
> to the maintainer / appears in the buglogi. In this hase I just got
> the message from the BTS ]
>> Same here. 2.0.3 worked perfectly fine.
>> Selecting complex patterns also causes a crash, whereas right-clicking
>> and changing the image's "properties" does work.
> And I cant reproduce that. really. I don't have that here.
> The tag is perfectly right on there, anyway. From The BTS docs:
> This bug can't be reproduced on the maintainer's system. Assistance from third parties is needed in diagnosing the cause of the problem."
I would add "moreinfo", in this case.
> I can't reproduce this on none of my machines (sid, though), and a person I
> asked about that (who id using etch) doesn't have it either...
It didn't happen on Etch's version last week, 2.0.3-x
> So unless you have more, suitable info...
Most "complex selection" operations cause OOo to crash, with a segment
Of course, this is unrelated with the fact that the oowriter script
complains about dirname being called improperly.
$ LOCALE=C oowriter
dirname: missing operand
Try `dirname --help' for more information.
Since upgrading yesterday night, this seems to be a bit more stable than
it used to be (when dealing with images). It does crash suddenly for a
reason I can't fathom, though
Yesterday night it even output a SEGV to the console it was started
from, but no such information today (just tested)
I'll keep testing to see if I can get a precise way of reproducing it
Otherwise, we'll just consider it a by-product of some library
(that's the joy of running Sid.... ;) though it also happened to me when
OOo 2.0.4~rc3 transitioned into testing, all other things being equal.
OOo 2.0.3 was rock-solid in that environment)