GNU bug report logs - #78777
30.1; insert-file-contents should not set buffer-file-name to nil

Previous Next

Full log

Message #47 received at 78777 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Jimmy Yuen Ho Wong <wyuenho <at> gmail.com>
Cc: Kenichi Handa <handa <at> gnu.org>, Eli Zaretskii <eliz <at> gnu.org>,
 78777 <at> debbugs.gnu.org
Subject: Re: bug#78777: 30.1; insert-file-contents should not set
 buffer-file-name to nil
Date: Mon, 16 Jun 2025 11:14:20 -0400

> Can userlock--ask-user-about-supersession-threat be modified to check for
> create-lockfiles and a newly introduced inhibit-file-lock internal variable
> before running userlock--check-content-unchanged?
>
> This change also seems to have been introduced to before-change-functions
> are called at the right time
> <https://lists.gnu.org/archive/html/bug-gnu-emacs/2016-08/msg01029.html>,
> perhaps we should ask Stefan what the best thing to do here is.

No, if you look at the diff, you'll see that this just moved the
`specbind`.  It was introduced earlier by:

```diff
commit d07af40d882673ccb30c267a617e673c22a85ee4
Author: Kenichi Handa <handa <at> gnu.org>
Date:   Mon Oct 23 12:40:32 2006 +0000

    (Finsert_file_contents): On replacing, temporarily bind
    buffer-file-name to Qnil before calling insert_from_buffer.

diff --git a/src/fileio.c b/src/fileio.c
--- a/src/fileio.c
+++ b/src/fileio.c
@@ -4359,8 +4359,12 @@
       inserted_chars
 	= (buf_bytepos_to_charpos (XBUFFER (conversion_buffer),
 				   same_at_start + inserted)
 	   - same_at_start_charpos);
+      /* This binding is to avoid ask-user-about-supersession-threat
+	 being called in insert_from_buffer (via in
+	 prepare_to_modify_buffer).  */
+      specbind (intern ("buffer-file-name"), Qnil);
       insert_from_buffer (XBUFFER (conversion_buffer),
 			  same_at_start_charpos, inserted_chars, 0);
       /* Set `inserted' to the number of inserted characters.  */
       inserted = PT - temp;
```

I'm not sure exactly what was the motivation for the change.
I mean, the comment explains that the purpose is to silence a potential
`ask-user-about-supersession-threat`, but I don't know why we should
care about such a possibility and why it would be right to silence
it here.

Nor do I understand why it was added to the code path where coding
conversion was needed but not to the other one.

I see that back in 2016 I apparently understood the situation a bit
better (or at least I thought) because while moving the `specbind`
I managed to improve the comment with:

             AFAICT we could avoid ask-user-about-supersession-threat by
             setting current_buffer->modtime earlier, but we could still
             end up calling ask-user-about-supersession-threat if the
             file is modified while we read it, so we bind
             buffer-file-name instead.  */

but it's still insufficient for me to understand now what's going on.
Handa?


        Stefan

Previous Next

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