Re: [Emacs-diffs] /srv/bzr/emacs/trunk r109160: * lisp/vc/vc-dispatcher.el (vc-compilation-mode): Add, based on

[Sam Steingold <sds@gnu.org>]

> * Stefan Monnier <zbaavre@VEB.HZbagerny.PN> [2012-07-19 18:00:44 -0400]:
>
> Could you explain this patch?

it extends the existing vc-bzr functionality to git & hg.

> I do not understand what it's trying to do.  Using compilation-mode on
> the output of those tools doesn't sound very useful.  What's the
> use-case?

C-x v d emacs/trunk RET
lands you in a *vc-dir* buffer
+
starts vc-pull (running "bzr update") and creates a *vc-bzr* buffer with
the command output.
putting it in the compilation-mode gives us 2 features:
-- one can use 'q' and 'z' to get rid of it
-- one can click on a file listed by bzr as modified and the file is visited.


>> +  (let* ((error-regexp-alist
>> +          (vc-make-backend-sym backend 'error-regexp-alist))
>> +         (compilation-error-regexp-alist
>> +          (and (boundp error-regexp-alist)
>> +               (symbol-value error-regexp-alist))))
>
> This is clearly wrong: error-regexp-alist seems to be meant to hold a
> symbol, but its name implies it should hold an alist.

it's a symbol whose value is an alist.
yes, you might argue that it should have been named error-regexp-alist-symbol

-- 
Sam Steingold (http://sds.podval.org/) on Ubuntu 12.04 (precise) X 11.0.11103000
http://www.childpsy.net/ http://think-israel.org http://memri.org
http://palestinefacts.org http://camera.org http://americancensorship.org
Lisp suffers from being twenty or thirty years ahead of time.