Subject: commit -- fixed: top ruler / tabs problem
From: Stephen Hack (shack@uiuc.edu)
Date: Thu Dec 02 1999 - 21:07:24 CST
I found what the problem was. Not in the use of the wrong *this pointers (via
creating/destroying the rulers), but in the gtk resize call. It would set the
height of the widget, only if it was != getHeight (which would ALWAYS return
32, the s_iFixedHeight. You could set the height via setHeight, but never
retrieve the value.
There are two ways to fix this.
1) Don't allow setHeight
2) Set the default height to s_ifixedHeight, then allow changes
I implemented the 2nd.
I was seeing this problem when I would show the rulers. This is because it
receives two resize signals, one with the height of 1 (which would set the
setHeight because 1 != getHeight (which was 32)), then a second signal w/
height of 32 (which it would NOT set because according to getHeight, the
height was already 32 (instead of 1).
-shack
On Thu, Dec 02, 1999 at 03:40:49PM -0600, Stephen Hack wrote:
> I've started to make some headway in tracking down my problems. I'm not
> registering my newly created widgets with the right user_data. The height of
> the ruler is "-2", so any clicks are not on the ruler, and therefore, deleted.
>
> Nobody's changed tab code yet, (that's the next POW i'm going to attempt), but
> I have been messing w/ the rulers quite a bit.
>
> -shack
>
> On Thu, Dec 02, 1999 at 01:26:49PM -0800, Paul Rohr wrote:
> > At 03:16 PM 12/2/99 -0600, sterwill@abisource.com wrote:
> > >Stephen Hack wrote:
> > >> I have gotten the rulers to display, but I'm now having a problem w/ the
> > tabs
> > >> not working. When was the last time you tried setting tabs? It looks
> > like it
> > >> deleting the tabs, not setting them.
> > >> (_setTabStops is called with bDelete=TRUE)
> >
> > Sounds like you've got a coordinate problem. Clicking in the white area
> > adds new tabs (for me). Dragging existing ones around moves them. Dragging
> > them "off" is what triggers a delete.
> >
> > It's worked that way for me since March, and still should, since nobody's
> > touched that code since then. Feel free to confirm this with a 0.7.6
> > binary, though.
> >
> > >Hmm... I've never been a very tabby kind of guy (at least with custom
> > >tabs). Paul wrote lots of that code, he's probably much more familiar
> > >with it than I am.
> >
> > Actually, I just put finishing touches on it. Jeff (as usual) did most of
> > the initial low-level work.
> >
> > Paul
This archive was generated by hypermail 2b25 : Thu Dec 02 1999 - 21:08:06 CST