bug#4194: 23.0.91; Cursor face not changing
Nikolaj Schumacher <[email protected]>
| Newsgroups | gmane.emacs.bugs,gmane.emacs.pretest.bugs |
|---|---|
| Message-ID | <[email protected]> |
Sorry for the delay. I checked this again using a build from today.
On 2009-05-31, at 17:27, Adrian Robert wrote:
> On Apr 3, 2009, at 5:33 AM, Chong Yidong wrote:
>
>>> Although I am able to configure the cursor face using the usual
>>> custumization tool, the real colors/face doesn't change.
>>
>> I can't reproduce this, so it's likely a bug on NS. Adrian, can you
>> reproduce it?
>
> I cannot.
Setting the background color seems to work now. The foreground color,
inverse-video, and others don't stick.
> Nikolaj Schumacher added later that:
>
>> The same seems to happen for the region face. The background color
>> reverts to ns_selection_color sometime after the custom variables are
>> read. (It works when reading the custom settings again, later.)
>
> I looked at the code and cannot see how this could be happening.
> The region face is set to ns_selection_color in ns-win.el, but this
> should be overridden by .emacs read later. This should have been
> cleared up by David Reitter's changes around 2009-03-14, though
> since the ChangeLog entries are incomplete it is difficult to be
> certain when the last broken code was removed.
>
> Anyway, if you still observe these in a recent build let me know and
> I'll go on a more thorough witch-hunt.
I've looked at this further, and it seems to work with an empty .emacs.
It fails with my configuration, because I set the frame's background
color like this:
(setq default-frame-alist
'((background-color . "#040420")
(foreground-color . "wheat")))
If I don't set the background-color, I get my customized region face.
If I manually evaluate the line after the initial frame was created,
any additional frame uses the default ns_selection_color for the
region. The first frame and customize-face keep my color.
Nik