GNU bug report logs - #35564
27.0.50; [PATCH] Tweak dired-do-shell-command warning about "wildcard" characters

Previous Next

Full log

View this message in rfc822 format

From: Kévin Le Gouguec <kevin.legouguec <at> gmail.com>
To: 35564 <at> debbugs.gnu.org
Cc: Stefan Monnier <monnier <at> iro.umontreal.ca>
Subject: bug#35564: [PATCH v2] Tweak dired warning about "wildcard" characters
Date: Sun, 09 Jun 2019 13:08:41 +0200

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

Hello,

Here is my second attempt at solving this issue.

To recap: dired-do-shell-command warns the user about non-isolated '*'
and '?' characters since the function will not substitute them.  It
refers to these characters as "wildcards", which can be incorrect: they
may be quoted or backslash-escaped, in which case the shell will not
interpret them as wildcards.

My main motivation to change this warning is that it trips my brain to
have to answer "yes" ("yes, I want to use wildcards") when no wildcards
are involved.

I could not come up with a simple, self-sufficient rephrasing for the
warning, so I decided to display the command itself as part of the
warning prompt, highlighting the non-isolated characters.

The first patch adjusts y-or-n-p so that it preserves the prompt's text
properties.  The second patch changes dired-do-shell-command's prompt.

[0001-Preserve-text-properties-in-y-or-n-p-prompts.patch (text/x-diff, attachment)]

[0002-Tweak-dired-warning-about-wildcard-characters.patch (text/x-diff, attachment)]

[Message part 4 (text/plain, inline)]

Sample screenshot:

[dired-warning-highlight.png (image/png, attachment)]

[Message part 6 (text/plain, inline)]

I am not sure these patches should be applied as-is.  Some things I
wonder about:

1. About read--propertize-prompt…

    1. Should the function return a copy of its argument instead of
       propertizing it directly?

    2. Is it properly named?  Does it fit in subr.el?  I placed it there
       because I figured other users of read-char in subr.el could use
       it, e.g. read-char-choice.

2. dired-aux.el already contains some logic to detect isolated
   characters; I could not think of a way to re-use it, so I added my
   own functions to find *non*-isolated characters.  I added unit tests
   for these new functions; still, there may be some redundancy there.

WDYT?


PS1: I am still absolutely open to simply rephrasing the prompt…  I just
cannot come up with good alternatives.

PS2: CC'ing Stefan to follow up on the discussion on emacs-devel.
<https://lists.gnu.org/archive/html/emacs-devel/2019-05/msg00339.html>

Previous Next

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