[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: NTEmacs21.0.92.1 Feedback
From: |
Jason Rumney |
Subject: |
Re: NTEmacs21.0.92.1 Feedback |
Date: |
02 Dec 2000 17:54:29 +0000 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.0.93 |
"Eli Zaretskii" <address@hidden> writes:
> > From: Jason Rumney <address@hidden>
> > Date: 02 Dec 2000 15:07:45 +0000
> >
> > The call to set-locale-environment in startup.el seems to be causing a
> > redraw-frame on a frame that hasn't been created yet. Do we have any
> > flags that indicate we are still starting up, to avoid this problem?
>
> Can't you test (e.g. within redraw-frame) whether the frame in
> question is alive, and if not, do nothing?
There was already a check there (which aborts if the frame is not
live), but the test was passing. I think the problem is that we have a
terminal frame at startup, but the W32 code does not handle having
mixed terminal and GUI frames at the same time, so since Emacs is
starting up in GUI mode, it calls the GUI versions of the redisplay.
This has been a problem in the past with the font loading code, but it
has always been worked around by using after-init-hook. I don't think
that is feasible in this case, as the user may make their own changes
to coding systems in their init files which we would not want to
override.
I think I have avoided running the dangerous code now though, by
testing the result of GetDC (which returns a NULL handle if it is
given a pointer to a Window that does not exist).
--
Jason Rumney <address@hidden>