GNU bug report logs - #75188
persist.el should recursively copy hash tables

Previous Next

Full log

View this message in rfc822 format

From: help-debbugs <at> gnu.org (GNU bug Tracking System)
To: Stefan Kangas <stefankangas <at> gmail.com>
Cc: tracker <at> debbugs.gnu.org
Subject: bug#75188: closed (persist.el should recursively copy hash tables)
Date: Thu, 13 Feb 2025 09:55:02 +0000

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

Your message dated Thu, 13 Feb 2025 01:54:11 -0800
with message-id <CADwFkm=QsKmwuUU7rWGOZP+jujYMf1FP10CZX-pOhVgSaC=44w <at> mail.gmail.com>
and subject line Re: bug#75188: persist.el should recursively copy hash tables
has caused the debbugs.gnu.org bug report #75188,
regarding persist.el should recursively copy hash tables
to be marked as done.

(If you believe you have received this mail in error, please contact
help-debbugs <at> gnu.org.)


-- 
75188: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=75188
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems

[Message part 2 (message/rfc822, inline)]

From: Joseph Turner <joseph <at> breatheoutbreathe.in>
To: bug-gnu-emacs <at> gnu.org
Subject: persist.el should recursively copy hash tables
Date: Mon, 30 Dec 2024 01:05:30 -0800

[Message part 3 (text/plain, inline)]

Hello!

This patch demonstrates a bug in persist.el, which may be the cause of
https://github.com/alphapapa/activities.el/issues/109 and a similar bug
in hyperdrive.el which causes the `persist' variable file to be deleted.
We have not yet found a way to reproduce the bug in either activities.el
or hyperdrive.el, so this `persist' bug may be unrelated.

[0001-Test-that-persist-copy-recursively-copies-hash-table.patch (text/x-diff, inline)]

From c28644c957d9fabf883990ff145cec19ed12ff10 Mon Sep 17 00:00:00 2001
From: Joseph Turner <joseph <at> breatheoutbreathe.in>
Date: Mon, 30 Dec 2024 00:40:33 -0800
Subject: [PATCH] Test that persist-copy recursively copies hash tables

---
 test/persist-tests.el | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/test/persist-tests.el b/test/persist-tests.el
index 6bf2ed3708..adf37a5bf0 100644
--- a/test/persist-tests.el
+++ b/test/persist-tests.el
@@ -151,3 +151,14 @@ (ert-deftest test-persist-reset ()
      (should-not (eq persist--test-reset-variable initial-value))
      (should-not (eq initial-value
                      (persist-default 'persist--test-reset-variable))))))
+
+(ert-deftest test-persist-copy-hash-table ()
+  "`persist-copy' should copy hash keys and values."
+  (let* ((hash (make-hash-table))
+         (rec (record 'a))
+         (_ (puthash 'foo rec hash))
+         (copy-hash (persist-copy hash)))
+    (setf (aref (gethash 'foo copy-hash) 0) 'b)
+    (should-not (persist-equal hash copy-hash))
+    (should-not (eq (gethash 'foo hash)
+                    (gethash 'foo copy-hash)))))
-- 
2.46.0

[Message part 5 (text/plain, inline)]

IIUC, the solution requires rewriting `persist-copy-tree' (which is
copied from Emacs 30's `copy-tree') so that it handles hash tables.  If
you all agree, I can do this in `persist.el`.

However, would it make sense for `copy-tree' to handle hash tables?

Forgive me if this has been discussed and settled before.  I did find
this thread about comparing hash tables with `equal':

https://yhetil.org/emacs-devel/871qvz4kdw.fsf <at> localhost/

Thank you!

Joseph

[Message part 6 (message/rfc822, inline)]

From: Stefan Kangas <stefankangas <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: yantar92 <at> posteo.net, mail <at> daniel-mendler.de,
 Joseph Turner <joseph <at> breatheoutbreathe.in>, monnier <at> iro.umontreal.ca,
 adam <at> alphapapa.net, 75188-done <at> debbugs.gnu.org, phillip.lord <at> russet.org.uk
Subject: Re: bug#75188: persist.el should recursively copy hash tables
Date: Thu, 13 Feb 2025 01:54:11 -0800

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Joseph Turner <joseph <at> breatheoutbreathe.in>
>> Cc: 75188 <at> debbugs.gnu.org,  mail <at> daniel-mendler.de,  yantar92 <at> posteo.net,
>>   monnier <at> iro.umontreal.ca,  adam <at> alphapapa.net,
>>   phillip.lord <at> russet.org.uk
>> Date: Mon, 27 Jan 2025 18:08:57 -0800
>>
>> Eli Zaretskii <eliz <at> gnu.org> writes:
>>
>> >> From: Joseph Turner <joseph <at> breatheoutbreathe.in>
>> >> Cc:  Daniel Mendler <mail <at> daniel-mendler.de>,
>> >> 	Ihor Radchenko <yantar92 <at> posteo.net>,
>> >> 	Stefan Monnier <monnier <at> iro.umontreal.ca>,
>> >> 	Adam Porter <adam <at> alphapapa.net>, Eli Zaretskii <eliz <at> gnu.org>,
>> >> 	phillip.lord <at> russet.org.uk
>> >> Date: Tue, 31 Dec 2024 14:12:00 -0800
>> >>
>> >> > Hello!
>> >> >
>> >> > This patch demonstrates a bug in persist.el
>> >>
>> >> [...]
>> >>
>> >> This issue also relates to the bug#63513, in which the suggestion was
>> >> made to replace persist-copy-tree with (compat-call copy-tree) now that
>> >> Compat 30 is available:
>> >>
>> >> https://yhetil.org/emacs-bugs/80479897-500e-fe60-6586-0a44ccb5993b <at> daniel-mendler.de/
>> >
>> > Thanks, would you like to submit a patch along those lines?
>>
>> Please see patch:
>
> LGTM.  Phillip, would you please install?

I'm not Phillip, but I installed the patch now.  Closing.

Previous Next

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