[Xquartz-dev] XQuartz 2.6.1_rc1

[Jeremy Huddleston]

On Mar 18, 2011, at 14:16, Peter Dyballa wrote:

> 
> Am 06.03.2011 um 03:59 schrieb Jeremy Huddleston:
> 
>> I've just uploaded 2.6.1_rc1 for both Leopard and Snow Leopard.
> 
> 
> I am trying it right now.

You might as well try the final release (2.6.1).

> As before the fonts in text panes or widgets in GNU Emacs become unreadable in the end. This means: GNU Emacs 23.3 (stable) and GNU Emacs 24.0.50 (unstable) start in some default mode. Then they reset themselves to using whatever X resources or settings in the init files dictates. This means: In the beginning all text is readable, after the reset it's not readable.

So the first launch is ok, but not subsequent ones?  That sounds like you probably have a bad value in ~/.Xresources and emacs is winning the race the first time.  Please try moving aside ~/.Xresources and see what happens.

FTB of the list, you did mention this to me yesterday, indicating that it had been a problem for a year or so, but this was the first time I've heard about it...


> It makes no difference whether I use GNU Emacs 24.0.50 compiled with X libraries from MacPorts or with some (more) up-to-date libraries from Fink,

... ? Fink doesn't include X libraries in their distribution...  I'm guessing you mean XQuartz X libraries and fink additional libraries.


> the final and the start-up look are equal (initially readable, afterwards distorted). A slight difference exists in the 24.0.50 *Calendar* buffer: The MacPorts "variant" refreshes the complete "window" (widget in the GNU Emacs "frame" which actually is the X window on the screen) while in the Fink "variant" seems to accumulate redraws. I'll send you some screenshots (2 MB or such) better privately.
> 
> The good news: blackbox 0.71 accepts, i.e. does not crash, when LC_CTYPE and/or LANG are set to some UTF-8 value:

Well, I'm glad blackbox got a bugfix, you should mention this in your MacPorts ticket.

What was the last version of XQuartz that worked for you?  Let's try installing the server version from that release to see if the problem goes away.  If it does, then we know it's a server issue... if not, then it's a client issue...

--Jeremy