tortoisehg0.4rc1, was: hgk graph layout

[Steve Borho]

On Wed, 2008-04-30 at 14:49 +0100, TK Soh wrote:
> On Wed, Apr 30, 2008 at 9:06 AM,  <Jens.Wulf at sew-eurodrive.de> wrote:
> > So I uninstalled 0.3 and installed 0.4rc1.
> >
> >   - changelog-browser-summary-line-bug and changelog-browser-crashing-bug are gone
> 
> good.

This, I'm stunned about.  There's no reason I can think of for this bug
to go away on it's own.

> >   - compared to 0.3, it slows down windows explorer up to the point where it is unusable,
> >    although I had full overlay icons enabled with 0.3. So for now I have to disable overlay icons completely.
> 
> I am very surprised. I don't observe any perceivable slow down on my
> system with the new overlay icons. And so far, you appear to be the
> first to report such issue.
> 
> >   - when opening the changelog browser, it doesn't show the graph column initially; I have to click refresh
> 
> Is this consistent? Do you open the browser in the subdir or root dir
> of the repos?

This is expected behavior if you open the log viewer in a non-root
directory.  It filters the log based on the directory you are in and
filtered logs do not get graphed.  It's debatable whether this should
still be the default behavior.

> >   - it is still possible to make 'hg vi' work with 0.4rc1, that's nice.
> >   - hgtk log from the command line results in:
> >
> >  Traceback (most recent call last):
> >   File "hgtk", line 100, in <module>
> >   File "hgtk", line 93, in main
> >   File "hggtk\history.pyc", line 566, in run
> >   File "mercurial\hg.pyc", line 60, in repository
> >   File "mercurial\localrepo.pyc", line 2135, in instance
> >   File "mercurial\localrepo.pyc", line 90, in __init__
> >   File "mercurial\extensions.pyc", line 86, in loadall
> >   File "mercurial\ui.pyc", line 445, in warn
> >   File "mercurial\ui.pyc", line 388, in write_err
> >   File "mercurial\util.pyc", line 919, in __getattr__
> >  AttributeError: 'Blackhole' object has no attribute 'closed'
> 
> I'll look into this. BTW, do you get this traceback on other hgtk commands too?

This was a known problem in 0.4rc1, will be fixed in rc2

-- 
Steve Borho (steve at borho.org)
http://www.borho.org/~steve/steve.asc
Key fingerprint = 2D08 E7CF B624 624C DE1F  E2E4 B0C2 5292 F2C6 2C8C