On Wed, Mar 30, 2016 at 11:18 PM, Raymond Toy <toy.raymond(a)gmail.com> wrote:
What are you trying? The c-before-change below?
The 1.98 cc-mode package. The below "fix" makes sure I don't run into
issues with errors in looking-at in non-cc-mode buffers. Due to index
out of bounds or wrong or missing RE. I don't care to debug it
further if I don't run into it again. I added the following "check"
Johann> (unless (or c-buffer-is-cc-mode ;; why is this
happening in non-c-mode?
What do you see? I've had some problems with various things but
quite reproducible. And never at a time where I can look in to it. I
just need to get something done.
What I saw related to the fix is described above. Other issue follows.
The issues with cc-mode I get are quire reproducible, except they do
not manifest in a "fresh" process. So, for example when XEmacs can't
eval t (yes, this happens, see backtrace below). If I restart a new
process and start to edit a file, there are no issues; it only happens
when I have a mixture of modes and/or buffers -- but seems
consistently related to cc-mode or its dependencies.
Can it be because the package is compiled with 21.5 and not 21.4? Is
there an easy way to recompile the package? Or all packages?
Debugger entered--Lisp error: (void-variable t)
custom-declare-variable(comint-input-ignoredups t "*If non-nil,
don't add input matching the last on the input ring.\nThis mirrors the
optional behavior of bash.\n\nThis variable is buffer-local." :type
load-internal("comint" nil t nil binary)
load("comint" nil t nil)
byte-code("..." [require cc-mode comint] 2)
XEmacs-Beta mailing list