GNU bug report logs - #24057
25.1.50; ffap interprets comments beginning with "//" as file path

Previous Next

Full log

Message #20 received at 24057 <at> debbugs.gnu.org (full text, mbox):

From: Kaushal Modi <kaushal.modi <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 24057 <at> debbugs.gnu.org, Noam Postavsky <npostavs <at> users.sourceforge.net>
Subject: Re: bug#24057: 25.1.50; ffap interprets comments beginning with "//"
 as file path
Date: Sat, 23 Jul 2016 11:56:40 +0000

[Message part 1 (text/plain, inline)]

On Sat, Jul 23, 2016 at 3:34 AM Eli Zaretskii <eliz <at> gnu.org> wrote:

> I see no reason to assume that file names cannot appear in comments.
>

I have already tested with these use cases and now it seems to
do-the-right-thing:

In the below table, 'x' represents the point (the cursor would be on the
character to its right).
The second column shows the value that ffap-string-at-point is set to on
doing C-x C-f (with the ido setup explained in the first email).


|-----------------------------------+---------------------------------|
| Example string in `c-mode' buffer | Returned `ffap-string-at-point' |
|-----------------------------------+---------------------------------|
| x//tmp                            | "tmp"                           |
| //xtmp                            | "tmp"                           |
| x////tmp                          | "tmp"                           |
| ////xtmp                          | "tmp"                           |
| x// //tmp                         | ""                              |
| // //xtmp                         | "//tmp"                         |
|-----------------------------------+---------------------------------|


To try this out, paste the below in a new buffer and M-x c-mode.

//tmp
////tmp
// //tmp

Now do C-x C-f with cursor at those 6 different points and you will see
that an attempt to read /tmp happens only in the case of "// //tmp" when
the point is anywhere in the "//tmp" portion following "// ".


I could agree with a user option to disable this behavior, but
> disabling it unconditionally is IMO a mistake.
>

Of course, if this patch returns in unintended ffap-string-at-point, I will
add a defcustom to enable this behavior.
-- 

Kaushal Modi

[Message part 2 (text/html, inline)]

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.