[PATCH] message-elide-region.

[PATCH] message-elide-region.

[Matthieu Moy]

Hi ! 

You  all should  know the  C-c  C-e shortcut  to elide  a region  when
replying. 

it works when the region is a set of full lines, but if I call it with
regions beginning in the middle of a line, everything goes wrong:

Example :

> it works when  the region is a set  of full lines, but if  I call it
> with  regions beginning  in the  middle of  a line,  everything goes
> wrong:

Now, I want to elide `beginning'. I get this.


> it works when  the region is a set  of full lines, but if  I call it
> with  regions 
[...]

 in the  middle of  a line,  everything goes
> wrong:


What I want is this :


> it works when  the region is a set  of full lines, but if  I call it
> with  regions 

[...]

> in the middle of a line, everything goes wrong:


The function to accomplish this is this one:

;;;###autoload
(defun message-elide-region (b e)
  "Elide the text in the region.
An ellipsis (from `message-elide-ellipsis') will be inserted where the
text was killed."
  (interactive "r")
  (kill-region b e)
  (message-newline-and-reformat)
  (delete-region
   (progn (previous-line 1) (point))
   (progn (next-line 3) (point)))
  (insert message-elide-ellipsis))

Anyone to commit this ?

-- 
Matthieu

By  the  way,  I've  just  discovered   M-RET  to  cut  a  line  in  a
citation. It's very usefull.

Re: [PATCH] message-elide-region.

[Kai Großjohann]

Matthieu Moy <Matthieu.Moy@imag.fr> writes:

> it works when the region is a set of full lines, but if I call it with
> regions [...] in the middle of a line, everything goes wrong:

As you can see, it works for me.  (I also elided "beginning".)  But I
have (setq message-elide-ellipsis "[...]").  Does that make a
difference?

kai
-- 
Silence is foo!

Re: [PATCH] message-elide-region.

[Matthieu Moy]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Matthieu Moy <Matthieu.Moy@imag.fr> writes:
>
>> it works when the region is a set of full lines, but if I call it with
>> regions [...] in the middle of a line, everything goes wrong:
>
> As you can see, it works for me.  (I also elided "beginning".)  But I
> have (setq message-elide-ellipsis "[...]").  Does that make a
> difference? 

Yes, and unfortunately, your example break my code :-(

The   old   message-elide-region    kills   the   region,   and   puts
message-elide-ellipsis instead. It doesn't  take any care of citation,
which   is   very   enoying   when   message-elide-ellipsis   contains
"\n"'s. (which is the default)

My  modification  was  to  add  (message-newline-and-reformat),  which
breaks a cited line properly. 

OK, lets try an improved version :

(defun message-elide-region (b e)
  "Elide the text in the region.
An ellipsis (from `message-elide-ellipsis') will be inserted where the
text was killed."
  (interactive "r")
  (kill-region b e)
  (if (string-match "\n.*\n" message-elide-ellipsis)
      (progn
	(message-newline-and-reformat)
	(delete-region
	 (progn (previous-line 1) (point))
	 (progn (next-line 3) (point))))
    )
  (insert message-elide-ellipsis))

This tests wether message-elide-ellipsis  begins and ends by "\n", and
do what it has to do then.

-- 
Matthieu

Re: [PATCH] message-elide-region.

[Karl Pflästerer]

Kai Großjohann <- Kai.Grossjohann@CS.Uni-Dortmund.DE wrote:
> As you can see, it works for me.  (I also elided "beginning".)  But I
> have (setq message-elide-ellipsis "[...]").  Does that make a
> difference?

If `message-elide-region' should be changed should we also mention the
possible changes we discussed in dcsg? (It was the possibility to change
the ellipsis to some text when C-C C-e had been pressed) eg. change
[...] to [SHORT_DESCRIPTION]

bye
   KP

-- 
    'Twas brillig, and the slithy toves
        Did gyre and gimble in the wabe;
    All mimsy were the borogoves,
         And the mome raths outgrabe.   "Lewis Carroll" "Jabberwocky"

Re: [PATCH] message-elide-region.

[Kai Großjohann]

sigurd@12move.de (Karl Pflästerer) writes:

> If `message-elide-region' should be changed should we also mention the
> possible changes we discussed in dcsg? (It was the possibility to change
> the ellipsis to some text when C-C C-e had been pressed) eg. change
> [...] to [SHORT_DESCRIPTION]

Changing message-elide-region was proposed before, but Lars said that
it doesn't make sense: C-w [ x x x ] is not any longer than C-u C-c
C-e xxx RET.  Why make a keybindig that essentially inserts text into
the buffer?  It only makes sense if something more happens.

I didn't like it, but I had to agree with Lars...

kai
-- 
Silence is foo!