On Wed, 2012-08-01 at 14:14 +0200, Eike Rathke wrote:
> Please review and cherry-pick to 3-6
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=21cb8210c74e52896ce7fb063f1578b13672f4bd
> that fixes https://bugs.freedesktop.org/show_bug.cgi?id=53012
Done.
> In case we do 3.6.0.5rc we should have it in there as well.
> I'd even say this is a blocker, because application crashes as soon as
> one selects fixed width and from then on will crash on any CSV import
> because the fixed width setting is remembered by the dialog.
Sounds non-ideal :-) makes me wonder if some sort of unit test could be
constructed there, but I guess it's GUI code.
That aside - I've never been a big fan of the "crash on bad input" mode
of OUString::copy() that seems incredibly lame to me. Is there any
convincing performance reason that anyone can work out for not just
returning an empty string in these cases ? ie. create a new rtl _copy
method that does the checking out-of-line in the method: I wonder how
common the case of copy() a full string is that we in-line the check for
that.
ATB,
Michael.
--
michael.me...@suse.com <><, Pseudo Engineer, itinerant idiot
_______________________________________________
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
