The symptoms are similar to those exhibited by shell mode without (setq
process-connection-type t).
Any ideas?
Thanks,
Michael
What version of OS X and GDB are you using?
Set gdb-enable-debug-log to t, do M-x gdb, and please post the resulting value
of gdb-debug-log to emacs-pr...@gnu.org (this is the right mailing list
for Emacs in CVS, currently 22.0.50).
Nick
I'm running OS X 10.4.2, and gdb 6.1-20040303 (Apple version gdb-413).
Thanks for replying!
> The variable wasn't recognized when using M-x set-variable, and so I
> set it manually with a setq.
gdb-enable-debug-log should be autoloaded so you should see it even if you
haven't loaded the file. Are you really using 22.0.50? Is it Carbon Emacs
from the CVS repository at savannah or a derived version like Aqua Emacs?
After you typ M-x gdb does Emacs prompt with:
Run gdb (like this): gdb --annotate=3 `yourprog'?
> I couldn't find a gdb-debug-log after
> doing M-x gdb subsequently, though --- am I doing something wrong?
> Sorry for not following instructions and posting to emacs-pretest-bug
> just yet; I wanted to make sure I got the output right first.
Something doesn't add up but there's no harm in posting to emacs-pretest-bug
even if you are doing something wrong. Its a forum for such matters not
an IQ contest.
> I'm running OS X 10.4.2, and gdb 6.1-20040303 (Apple version gdb-413).
I've had reports for 10.2 and 10.3 where, I think, M-x gdb works now. Apple
have their own version of GDB so they might have removed the underlying
annotations in 10.4 (Tiger?) that are needed for GDB to work in Emacs. You
could look at the output of "gdb --annotate=3 yourprog" from the command line
ans see if it marks up the output with frames-invalid, pre-prompt, prompt etc.
Without access to OS X 10.4 (or even a Mac!), I can't really help you.
If you want to dig deeper please report any progress you make to
emacs-devel.
Nick
Then, on a hunch, I turned off viper mode (I forgot to mention I was
using it), and everything started working just fine! Is this a known
issue with viper? (Note: using C-z to enter emacs mode temporarily
didn't fix the problem, nor did M-x viper-go-away --- I had to actually
disable it and restart emacs)
I wasn't aware of it but I've never used viper-mode. At first sight, I would
think it shouldn't interfere. If you had used M-x report-emacs-bug it might
have shown that you had turned viper-mode on. I suggest again that you send
your report to emacs-pr...@gnu.org where someone might, possibly Michael
Kifer, author of Viper, might recognise the problem.
Nick
Perhaps disabling viper-auto-indent in the relevant mode hook would do
the trick?
Ted
--
Edward O'Connor
hob...@gmail.com
Ense petit placidam sub libertate quietem.
Glad to hear it. Which mode hook did you use?
Here's where/how I disable viper-auto-indent:
(setq-default viper-auto-indent t)
(defun ted-disable-viper-auto-indent ()
(when viper-mode
(setq viper-auto-indent nil)))
(mapc (lambda (hook)
(add-hook hook 'ted-disable-viper-auto-indent))
'(eshell-mode-hook nslookup-mode-hook wikipedia-mode-hook
css-mode-hook))
Please M-x report-emacs-bug describing your problem and the workaround
you've found. Hopefully the problem can be fixed so you don't need
the workaround.
Stefan