Subject:
|
Re: glut conundrum
|
Newsgroups:
|
lugnet.cad.dev.mac
|
Date:
|
Mon, 18 Feb 2002 01:38:23 GMT
|
Viewed:
|
1724 times
|
| |
|
|
In lugnet.cad.dev.mac, Ray Sanders writes:
> In lugnet.cad.dev.mac, Don Heyse writes:
> > That may be a very old bug. I remember there was a problem with the
> > background color way back when. You can check the logs or browse the
> > current (and old) source at the sourceforge to compare.
>
> Indeed, it is a very old bug, and it appears that someone ran into this more
> recently based on the following comment in the 0.9.5 code ...
>
> // NOTE: I dont understand the colortable stack used in translate_color()
> // so use the default color_table to avoid bad colors (bad stack ptr?).
>
> So someone else saw a different symptom of the same issue
That would be me. Anywhere you see "NOTE" its probably a note to
myself that I want to get back to something and do it right when
I have more time. I don't always get around to revisiting things
though, especially if the cheap fix works well enough...
> What appears to be happening (in 0.5) is that the function 'translate_color' >is called from 'colormenu' and it wants zcolor_table to be valid. But, at
> the conclusion of the previous rendering, znamelist_stack_index is
> decremented and the associated color table is freed, so what
> 'translate_color' sees in memory is detritus left over from the
> previous allocation. It just so happens, that on the
> Mac runtime, that detritus is intact other than the entry for black. >Immediately after 'colormenu' changes the background color, a new
> rendering is invoked, which reallocates the color table stack and
> everything in the rendering looks cool. So the error is transient
> during the instance that 'colormenu' is calling 'translate_color'.
So I guess if someone changes the color table with the 0 COLOR command
they won't get the new background color with my cheap fix. I'm still OK
with this, I think. But if you've got a better solution, I'd be happy
to fix it the right way.
Don
|
|
Message is in Reply To:
 | | Re: glut conundrum
|
| (...) Indeed, it is a very old bug, and it appears that someone ran into this more recently based on the following comment in the 0.9.5 code ... // NOTE: I dont understand the colortable stack used in translate_color() // so use the default (...) (23 years ago, 17-Feb-02, to lugnet.cad.dev.mac)
|
6 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
