bug#59793: 29.0.60; subr.elc is not compiled correctly

[Akib Azmain Turja via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>]

[-- Attachment #1: Type: text/plain, Size: 1792 bytes --]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Akib Azmain Turja <akib@disroot.org>
>> Cc: Feng Shu <tumashu@163.com>,  59793@debbugs.gnu.org
>> Date: Sat, 03 Dec 2022 23:16:47 +0600
>> 
>> > Am I missing something, or do you redefine a function and then expect it to work like you never redefined it?
>> 
>> Unexpected: bug59593-yank just inserts the killed text.
>> Expected: bug59593-yank intercepts and shows the killed text in echo
>> area.
>
> You have redefined a subr.el function with cl-letf*, so how is what happens
> as result a bug in Emacs?  Does the original subr.el function not do what
> it's supposed to do, before you replace it?
>
> I understand you didn't expect the result of cl-letf*, and were surprised by
> what you saw, but I don't understand why you expect the Emacs development to
> do something about your surprise.
>

With subr.el (source), the bug59793-yank function overrides
'insert-for-yank' successfully.

With subr.elc (compiled byte-code), the function can't override
'insert-for-yank'.  Just try to replace the 'cl-letf' (not 'cl-letf*')
form with an 'error' call, bug59793-yank still just inserts text, which
is, obviously, unexpected.

>> > So I'm still confused...
>> 
>> Is it clear now?  (Somehow, I think, no.)
>
> See above: I understand what you are saying, but not why this is submitted
> as a bug to the Emacs development team.

Both source and compiled Emacs Lisp code should show the same behavior,
except performance.  But 'subr.el' and 'subr.elc' show different
behaviors.  Isn't this a byte-compiler bug?

-- 
Akib Azmain Turja, GPG key: 70018CE5819F17A3BBA666AFE74F0EFA922AE7F5
Fediverse: akib@hostux.social
Codeberg: akib
emailselfdefense.fsf.org | "Nothing can be secure without encryption."

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]