unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Ihor Radchenko <yantar92@gmail.com>
To: Andrea Corallo <akrl@sdf.org>
Cc: 40838@debbugs.gnu.org
Subject: bug#40838: 28.0.50; [feature/native-comp] Function overrides in init.el are not honored after deferred compilation
Date: Sat, 23 May 2020 22:13:06 +0800	[thread overview]
Message-ID: <878shi3fkd.fsf@localhost> (raw)
In-Reply-To: <87o8qk39o5.fsf@localhost>

[-- Attachment #1: Type: text/plain, Size: 669 bytes --]

> If this is what is happening with straight.el, then my recipe is not
> actually reproducing my initial bug report. I have other cases when
> functions redefined in my config are overridden by initial definition
> when .eln is loaded. I think that I still need to investigate for a real
> recipe then.

I have recently tried to find a temporary workaround to this issue with
straight.el+org+native-comp. However, my workaround seems to fail
because of native-comp (at least, I believe so).

The recipe is the same, but I forcefully defined org-version,
org-release, and org-git-version in my init.el before loading org.
See the attached file for details.

Best,
Ihor


[-- Attachment #2: native-fail-recepy.el --]
[-- Type: application/emacs-lisp, Size: 3882 bytes --]

[-- Attachment #3: Type: text/plain, Size: 3531 bytes --]



Ihor Radchenko <yantar92@gmail.com> writes:

>> In summary this is not a problem of the deferred compilation mechanism
>> but is an hack that is not working for this case.
>> 
>> To mitigate this I've added a new customize you can use to define those
>> functions (or whatever) into the compiler environment of the async
>> compilation workers, is called `comp-async-env-modifier-form'.
>
> If this is what is happening with straight.el, then my recipe is not
> actually reproducing my initial bug report. I have other cases when
> functions redefined in my config are overridden by initial definition
> when .eln is loaded. I think that I still need to investigate for a real
> recipe then.
>
> Best,
> Ihor
>
>
> Andrea Corallo <akrl@sdf.org> writes:
>
>> Ihor Radchenko <yantar92@gmail.com> writes:
>>
>>>> thanks this is appreciated because I haven't managed to reproduce it
>>>> myself.
>>>
>>> Finally, I found some reproducible example.
>>> straight.el redefines some org functions before loading org.
>>> It is done in straight--fix-org-function (org-git-version and
>>> org-release are redefined). 
>>> The redefined version works with org.elc, but somehow get overridden
>>> when org.eln is loaded (in my case, the loading is triggered by elfeed-org).
>>>
>>> Steps to reproduce:
>>>
>>> 1. Use the attached file to load emacs. No errors should appear.
>>> 2. Wait until org is native-compiled.
>>> 3. Restart emacs. The following errors appears
>>> (straight--fix-org-function supposed to be a workaround for this error):
>>>
>>> Error (use-package): elfeed-org/:catch: Invalid version syntax: ‘N/A’
>>> (must start with a number)
>>>
>>> 4. Delete org.eln
>>> 5. Restart emacs. The error disappears.
>>
>> Okay I think I've an idea of what is going on here.
>>
>> straight given wants to build org in a way org is not made for is
>> hacking around the problem predefining in the compilation environment
>> `org-release' and `org-git-version'.
>>
>> When org.el is loaded is executing at top level the expansion of
>> `org-check-version' that is supposed to define these two functions,
>> given are already defined by straight.el we should fall in the first if
>> clause an the hacked functions remains.
>>
>> When the eln are compiled by deferred-compilation no-one is hacking the
>> definition of these two functions in the way straight.el would like and
>> so the trouble raise.
>>
>> In summary this is not a problem of the deferred compilation mechanism
>> but is an hack that is not working for this case.
>>
>> To mitigate this I've added a new customize you can use to define those
>> functions (or whatever) into the compiler environment of the async
>> compilation workers, is called `comp-async-env-modifier-form'.
>>
>> 2ac6194585 * Add new customize `comp-async-env-modifier-form' (Bug#40838)
>>
>> I'm for closing this.
>>
>> Bests
>>
>>   Andrea
>>
>> -- 
>> akrl@sdf.org
>
> -- 
> Ihor Radchenko,
> PhD,
> Center for Advancing Materials Performance from the Nanoscale (CAMP-nano)
> State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, China
> Email: yantar92@gmail.com, ihor_radchenko@alumni.sutd.edu.sg

-- 
Ihor Radchenko,
PhD,
Center for Advancing Materials Performance from the Nanoscale (CAMP-nano)
State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, China
Email: yantar92@gmail.com, ihor_radchenko@alumni.sutd.edu.sg

  reply	other threads:[~2020-05-23 14:13 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-25  8:14 bug#40838: 28.0.50; [feature/native-comp] Function overrides in init.el are not honored after deferred compilation Ihor Radchenko
2020-04-25  9:16 ` Andrea Corallo
2020-04-25 14:26   ` Andrea Corallo
2020-04-26  4:55     ` Ihor Radchenko
2020-04-26  8:59       ` Andrea Corallo
2020-04-28 14:40         ` Ihor Radchenko
2020-04-28 15:03           ` Andrea Corallo
2020-05-18 15:23             ` Ihor Radchenko
2020-05-18 20:15               ` Andrea Corallo
2020-05-18 21:01                 ` Andrea Corallo
2020-05-19  3:06                 ` Ihor Radchenko
2020-05-23 14:13                   ` Ihor Radchenko [this message]
2020-05-23 14:39                     ` Andrea Corallo
2020-05-23 15:23                       ` Ihor Radchenko
2020-06-02  7:27                         ` Andrea Corallo
     [not found]                           ` <87zh7ltgc7.fsf@localhost>
2020-07-28  7:24                             ` Andrea Corallo via Bug reports for GNU Emacs, the Swiss army knife of text editors
2020-07-28  9:06                               ` Ihor Radchenko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=878shi3fkd.fsf@localhost \
    --to=yantar92@gmail.com \
    --cc=40838@debbugs.gnu.org \
    --cc=akrl@sdf.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).