GNU bug report logs - #48666
Leaks are still possible in hook--depth-alist

Previous Next

Full log

Message #8 received at 48666 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: jakanakaevangeli <at> chiru.no
Cc: 48666 <at> debbugs.gnu.org, Daniel Mendler <mail <at> daniel-mendler.de>
Subject: Re: bug#48666: Leaks are still possible in hook--depth-alist
Date: Tue, 22 Feb 2022 10:25:35 -0500

jakanakaevangeli <at> chiru.no [2021-05-26 10:57:47] wrote:
>     (with-temp-buffer
>       (add-hook 'post-command-hook (make-symbol "fun-sym") 77 t))
>     (get 'post-command-hook 'hook--depth-alist)
> Perhaps hook--depth-alist could be a hash table with weak keys.

I just pushed to `master` a patch which should hopefully fix this
problem for good (basically, the depth alist is now kept in an(other)
variable (so for functions added to the buffer-local portion of the
hook, the depth is stored in the buffer-local portion of the depth
variable)).

I hope it will also fix the problems seen in
https://github.com/minad/corfu/issues/133

It makes it a bit harder to check if the old problem is still present or
not, OTOH, because the depth-alist isn't just a quick (get <hook>
'hook--depth-alist) away but you need to (symbol-value (get <hook>
'hook--depth-alist)) and that will only show you the global portion or
the portion relevant to the current buffer.


        Stefan

Previous Next

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