Re: Initial frame displays the wrong colors
-- "Stephen J. Turnbull" <stephen(a)xemacs.org> spake thusly:
>>>>> "Matt" == Matt Tucker
<tuck(a)whistlingfish.net> writes:
Matt> -- "Stephen J. Turnbull" <stephen(a)xemacs.org> spake thusly:
Matt> I would content that "It's always worked this way" and
"This
Matt> is the correct behavior" are two totally different
Matt> statements.
They are.
By "correct behavior" I mean "matches specifications." Since the
specification of visible behavior was never actually written down, I
approximate it with the historical behavior.
If you want to write a new specification that says that it's a bug,
fine with me. But many many releases of XEmacs have been made with
this "bug" unresolved and unremarked by the people planning releases.
We need to get 21.4 promoted to stable. This is not a showstopper.
Therefore it will wait.
I didn't mean to imply that it was anything I thought should be fixed
for 21.4. I don't even necessarily think it's something that needs to
be fixed in 21.5. It sounds like a difficult problem, and there are
certainly plenty of other things to work on. I simply felt that this is
a bug and should be recognized as such, and that the method for dealing
with it is nothing more than a workaround.
If I ever get the itch, I may try to fix this particular problem. Until
then, I'll simply hope that someone else does, and deal with it as best
I can until then. It's really not that difficult, since I'm aware of
it, but I remember being confused by this and can certainly understand
how it would cause problems for others as well.
Matt> Why should anyone have to muck around with a completely
Matt> different facility for only the _first_ frame?
They shouldn't. That's not what I said, what I said is "go ahead and
find a way to make it work in a desirable way". He did. I'm happy.
So users use Michael's workaround[1] for now, and somebody does
something better and more permanent in 21.5. Maybe it gets
backported to 21.4.
Fair enough. Sorry if I came off like a pompous ass in my response. ;-)
Attachments:
- attachment.sig (application/pgp-signature — 232 bytes)