Re: Patches for autopositioning of labels in fltk

[logari81]

On Fri, 2010-12-17 at 22:24 -0500, Ben Abbott wrote:
> On Dec 17, 2010, at 3:58 PM, logari81 wrote:
> 
> > On Tue, 2010-12-14 at 19:25 -0500, Ben Abbott wrote:
> >> On Dec 14, 2010, at 2:42 PM, logari81 wrote:
> >> 
> >>> A very simple solution that I have tried in the frontend is adding a
> >>> __fltk_redraw__()
> >>> just after the set() call in __axis_label__.m . At least on my
> >>> installation it seems to work fine with your test script.
> >>> 
> >>> I could easily implement the same in the backend as well. But it is a
> >>> bit more difficult to achieve finer control of the situation and request
> >>> only the modified axis to be redrawn.
> >>> 
> >>> What do you think? Do you see any performance issues due to excessive
> >>> redraw calls?
> >>> 
> >>> Kostas
> >> 
> >> This is too far out of my sandbox for me to offer an informed opinion.
> >> 
> >> So, I've cc'd some of the developer's who have contributed to the OpenGL 
> >> backend.
> >> 
> >> Ben
> >> 
> > 
> > I have prepared a summary of the changes that we have discussed in this
> > thread and rebased the corresponding changesets onto the current
> > revision of octave. I think that the first two of the attached
> > changesets should be adopted independently of the the third one.
> > 
> > 1. The first changetset sets the default axis labels size from 12 to 10.
> > 
> > 2. The second changeset permits the user to manually set the position
> > and rotation of the axis labels and title in the fltk backend.
> > 
> > 3. The third one fixes:
> >> px = get (xlabel ('hello world'), 'position')
> >> px =   0   0   0
> > This error also occurs with the current octave version, so is not
> > related to the previous two patches, so the adoption of the third patch
> > shouldn't be a blocker for the first two.
> > 
> > Regards
> > 
> > Kostas
> > <1_axes-default-fontsize.changeset><2_no-label-autopositioning-if-manual.changeset><3_redraw-axes-after-label-set.changeset>
> 
> I've pushed the first two changesets. For the 3rd is it no necessary to check 
> that the flkt backend is active?
> 
thanks, I hope to find the time to implement the tightinset calculation
for the fltk backend before 3.4 is released.

> Does something like the following be included in __axis_labels__.m ?
> 
>       if (strcmp (get (ancestor (ca, "figure"), "__backend__"), "fltk"))
>         __fltk_redraw__ ();
>       endif
> 
> Ben
> 
actually, __fltk_redraw__ looks among the figures which use the fltk
backend and as far as I understand redraws only those marked as
modified. So it shouldn't cause great problems. In any case I think your
check adds some more safety, so it would be a good idea to include it.

If you finally commit this change I think we should keep an eye on it
until the release of 3.4 so that we can identify any possible side
effects in time.

Kostas