Re: [Evolution] gtkhtml very unstable

[Larry Ewing <lewing ximian com>]

On 27 Apr 2001 17:31:43 +0000, dan hensley att net wrote:
> > > Stealth changes like this aren't very nice.  I didn't see
> > > anything about this in the gtkhtml Changelogs, so I never
> > > would have figured this out.  Shouldn't ORBit's version number
> > > have been bumped and gtkhtml's dependencies as well?
> >
> > This isn't a stealth change, this is a bug, a bug in ORBit 0.5.7 that
> > happened by chance to already be fixed in CVS on the orbit-stable-0-5
> > branch.  If you can't deal with things breaking while under development
> > please don't use CVS.
>
> Where can I find an ORBit Changelog?  I can handle things breaking
> in CVS, although it's not very fun.  I'm sure you'd much rather
> hear bug reports on recent work rather than bugs from 0.8.

There isn't one.  Elliot Lee is adamantly opposed to ChangeLogs that are
maintained by hand so ORBit has never had one.  Probably the easiest way
to get something close is to do a query on bonsai at cvs.gnome.org.  I
do really appreciate bug reports, especially when they are on current
code.

> > This looks like problem on your end.
>
> Yes, CVS really sucks rocks sometimes.  I had to move the
> tree I was updating and grab ORBit fresh.  CVS seems to pretty
> frequently update local trees incompletely.  When I got a
> fresh copy, it compiled and installed fine, but the gtkhtml
> problem still exists.

Odd, that crash went away with the ORBit bugfix, although it looks like
Radek commited some highly unstable code earlier today, perhaps that is
what you are seeing now.  I've put in a hack to keep that from cashing
but I'm waiting on Radek for a proper fix.

--Larry