Stefan Reichör <[EMAIL PROTECTED]> writes:
> Mark Triggs <[EMAIL PROTECTED]> writes:
>
>> (defun tla-changes (&optional arg)
>> "Run tla changes."
>> (interactive "P")
>> (tla-save-some-buffers)
>> (let* ((root (tla-tree-root))
>> (buffer (tla--get-buffer-create 'changes root)))
>> (with-current-buffer buffer
>> (let (buffer-read-only) (erase-buffer))))
>> (tla-run-arch (if arg '("changes") '("changes" "--diffs")) t ; t runs
>> asynchronously
>> :finished (lambda (output error status)
>> (message "No changes in this working copy"))
>> :killed (lambda (output error status)
>> (message "Command killed unexpectedly!"))
>> :error (lambda (output error status)
>> (let ((tla-process-buffer output))
>> (tla-show-changes-buffer)))))
>
> I like that approach. Please commit it, when you think that it is
> stable enough.
Okay, I've been talked into it :o) I'll first have to work out how it
fits in with Matthieu's new multi-process stuff. Currently my
"tla-run-arch" is creating new output and error buffers for each
command run, but hopefully I can combine it with Matthieu's new
approach.
I suppose the least painful way to do it would be to add a new function
that will ultimately replace tla-run-arch and gradually change the
commands over to using that instead? I'll add a note to HACKING
explaining the whole thing.
Cheers,
Mark
--
Mark Triggs
<[EMAIL PROTECTED]>
