Interesting...today's update brought in the emacs 22 pretest snapshot. This is OK with me - I've been building my own version for a little bit now. But I do find it curious, given that emacs 22 final before the Fedora 7 release seems, um, unlikely. There are certain things in emacs 22 (gnus being the main example) which do not go back to version 21 in any sort of graceful way. Does anybody know why this went in now?
jon
On Thursday 25 January 2007 11:00, Jonathan Corbet wrote:
Interesting...today's update brought in the emacs 22 pretest snapshot. This is OK with me - I've been building my own version for a little bit now. But I do find it curious, given that emacs 22 final before the Fedora 7 release seems, um, unlikely. There are certain things in emacs 22 (gnus being the main example) which do not go back to version 21 in any sort of graceful way. Does anybody know why this went in now?
The emacs maintainer felt it was OK to put in, ergo it went in.
Jesse Keating wrote:
On Thursday 25 January 2007 11:00, Jonathan Corbet wrote:
Interesting...today's update brought in the emacs 22 pretest snapshot. This is OK with me - I've been building my own version for a little bit now. But I do find it curious, given that emacs 22 final before the Fedora 7 release seems, um, unlikely. There are certain things in emacs 22 (gnus being the main example) which do not go back to version 21 in any sort of graceful way. Does anybody know why this went in now?
The emacs maintainer felt it was OK to put in, ergo it went in.
FWIW, I had been using 22, built from cvs with the xft patches last year. I found it to be a bit unstable with periodic and unpredictable crashes with some level of interaction noted when Firefox was used. Perhaps related to the FF memory leak issue?
I began to use 23, also from cvs, late last year and despite it's Alpha status, has been rock solid. I don't know if the fact that the xft patches were moved into the mainstream Unicode trunk or other issues facilitate the better stability in 23.
There are changes in some of the ~/.emacs configs that do need to be considered with both 22 and 23.
Also, I am using ECB and had to install the latest CEDET (1.0pre3) and ECB (cvs-snapshot) packages to get them to work properly.
Emacs with xft is a thing of beauty on a high-res LCD panel. :-)
Anyway, since the topic came up, I thought that I would put in my $0.02.
Regards,
Marc Schwartz
Jonathan Corbet wrote:
Interesting...today's update brought in the emacs 22 pretest snapshot. This is OK with me - I've been building my own version for a little bit now. But I do find it curious, given that emacs 22 final before the Fedora 7 release seems, um, unlikely. There are certain things in emacs 22 (gnus being the main example) which do not go back to version 21 in any sort of graceful way. Does anybody know why this went in now?
Hmmm, I'm having all sorts of problems with this version. :) For starters, if I try to resize the Emacs window, it hangs my X11 server, and it has to be restarted. This doesn't happen with any other windows, just Emacs.
There's also a lot of minor things that's not working, like "space" is no longer (by default) bound to do autocompletion in Find-File (I know, I can turn that on). But what's more annoying is that none of my "local variables" comments I have at the end of source files to configure Mode settings (like, C-mode) works. This is actually a "show stopper" for me, anyone know why this isn't working, and how I can get get it functional again?
I'd actually prefer to downgrade to Emacs v21 at this point, is there an easy way to do that with yum?
Cheers,
-- Leif
On Thu, Feb 08, 2007 at 11:53:07AM -0700, Leif Hedstrom wrote:
Jonathan Corbet wrote:
Interesting...today's update brought in the emacs 22 pretest snapshot.
Hmmm, I'm having all sorts of problems with this version. :) For starters, if I try to resize the Emacs window, it hangs my X11 server, and it has to be restarted. This doesn't happen with any other windows, just Emacs.
See https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=224611
You will find there also possible workarounds. Yes, multiple.
Out of curiosity - is your system i386 or x86_64? If not x86_64 then please add a corresponding note there. I would be somewhat surprised to find that issue architecture dependent but so far known instances are reported for x86_64.
This is actually a "show stopper" for me, anyone know why this isn't working, and how I can get get it functional again?
Bugzilla is that ----> way!
Michal
On 2007-02-08, Leif Hedstrom said:
Jonathan Corbet wrote:
Interesting...today's update brought in the emacs 22 pretest snapshot. This is OK with me - I've been building my own version for a little bit now. But I do find it curious, given that emacs 22 final before the Fedora 7 release seems, um, unlikely. There are certain things in emacs 22 (gnus being the main example) which do not go back to version 21 in any sort of graceful way. Does anybody know why this went in now?
Hmmm, I'm having all sorts of problems with this version. :) For starters, if I try to resize the Emacs window, it hangs my X11 server, and it has to be restarted. This doesn't happen with any other windows, just Emacs.
Never happens here.
There's also a lot of minor things that's not working, like "space" is no longer (by default) bound to do autocompletion in Find-File (I know, I can turn that on).
The NEWS file tells you how to get back to the old obsolete behavior.
But what's more annoying is that none of my "local variables" comments I have at the end of source files to configure Mode settings (like, C-mode) works. This is actually a "show stopper" for me, anyone know why this isn't working, and how I can get get it functional again?
It works all the time for me. But look at the manual (info "(emacs)Specifying File Variables")
I'd actually prefer to downgrade to Emacs v21 at this point, is there an easy way to do that with yum?
Certainly. Fedora core 6 and 5 have only Emacs 21.
Cheers,
-- Leif
If every thing are compatible they would just call it 21.5. Of course it takes time getting used to regarding that 21 was released on October 28, 2001.
But please try to look at the huge NEWS file (224k) in data-directory. It will certainly help with some of your questions.
This is a good move and we should stick to it.
On Thu, Feb 08, 2007 at 07:33:07PM +0000, Leo wrote:
On 2007-02-08, Leif Hedstrom said:
Hmmm, I'm having all sorts of problems with this version. :) For starters, if I try to resize the Emacs window, it hangs my X11 server, and it has to be restarted.
Actually no. It does not hang server. It locks emacs window. If you can login from a remote and kill emacs process then everything will work fine after that.
This doesn't happen with any other windows, just Emacs.
Never happens here.
Is "here" a rawhide installation? There is no problem with emacs 22 on FC5 or FC6 or when ATK support is turned off or when metacity for a window manager is not used (I do not know about compiz). Moreover the issue is very hard to repeat if you are running under strace. :-)
Michal
On 2007-02-08, Michal Jaegermann said:
Never happens here.
Is "here" a rawhide installation? There is no problem with emacs 22 on FC5 or FC6 or when ATK support is turned off or when metacity for a window manager is not used (I do not know about compiz). Moreover the issue is very hard to repeat if you are running under strace. :-)
Michal
Wherever it is, it is something else needs fixing ;)
On Fri, Feb 09, 2007 at 12:12:29AM +0000, Leo wrote:
On 2007-02-08, Michal Jaegermann said:
Never happens here.
Is "here" a rawhide installation? There is no problem with emacs 22 on FC5 or FC6 or when ATK support is turned off or when metacity for a window manager is not used (I do not know about compiz). Moreover the issue is very hard to repeat if you are running under strace. :-)
Michal
Wherever it is, it is something else needs fixing ;)
You know that already? Lucky you! https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=224611#c75 suggests that this is not necessarily the case. You may be correct but this is not that clear at this point.
Again - it would be nice to know if "here" a rawhide installation. Could you tell? An arch also would be nice.
Michal
On 2007-02-09, Michal Jaegermann said:
Wherever it is, it is something else needs fixing ;)
You know that already? Lucky you! https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=224611#c75 suggests that this is not necessarily the case. You may be correct but this is not that clear at this point.
Again - it would be nice to know if "here" a rawhide installation. Could you tell? An arch also would be nice.
I try not to touch bugzilla since I usually get a reply like 1 year later when I have completely lost my interest in the original question.
My user-agent says: No Gnus v0.6, Emacs/22.0.93.6 (2007-02-08), Fedora 6 gnu/linux
;)
On Fri, Feb 09, 2007 at 01:27:40AM +0000, Leo wrote:
My user-agent says: No Gnus v0.6, Emacs/22.0.93.6 (2007-02-08), Fedora 6 gnu/linux
For the problem to show up you need emacs 22 compiled with Gtk+ and a rawhide system. There is no surprise that for Fedora 6 you are not seeing it.
Michal
Michal Jaegermann michal@harddata.com wrote:
Hmmm, I'm having all sorts of problems with this version. :) For starters, if I try to resize the Emacs window, it hangs my X11 server, and it has to be restarted.
Actually no. It does not hang server. It locks emacs window. If you can login from a remote and kill emacs process then everything will work fine after that.
It's actually a window manager interaction problem. The least intrusive way I've found to unblock things is to switch out of X (alt-ctrl-F1) and back in. This works a little less well after the most recent update, though, in that it causes the offending emacs window to be moved to the first virtual desktop...
I've seen the window manager hang in other contexts as well. I have six virtual desktops arranged in two rows:
1 2 3 4 5 6
#4, for whatever reason, seems to be especially prone to this. Going for the GNOME menus from the panel (which I have on the bottom of the screen) will often misbehave in that desktop. Weirdness.
(This is with metacity, incidentally, though I believe I've seen the emacs hang with compiz too).
jon
-
Jesse Keating -
Jonathan Corbet -
Leif Hedstrom -
Leo -
Marc Schwartz -
Michal Jaegermann