From: Matt Armstrong <matt@rfc20.org>
To: Stefan Monnier <monnier@iro.umontreal.ca>
Cc: 59707@debbugs.gnu.org
Subject: bug#59707: 29.0.50; Seeking a more robust `package-quickstart'
Date: Wed, 30 Nov 2022 11:13:52 -0800 [thread overview]
Message-ID: <87fse08dcf.fsf@rfc20.org> (raw)
In-Reply-To: <87cz95rnhf.fsf@rfc20.org>
[-- Attachment #1: Type: text/plain, Size: 1181 bytes --]
Matt Armstrong <matt@rfc20.org> writes:
>>> 1) Harden Emacs such that signaled errors from
>>> "package-quickstart.elc" don't prevent startup (but are somehow
>>> saved and logged, maybe as warnings?).
>>
>> Agreed. I suspect it should also do things like delete the `.elc`
>> file (and/or the `.el` file), or at least suggest doing it, so as to
>> help diagnose/circumvent the problem.
>
> Ok, let's keep it on the list of possibilities.
This (attached) approach seems to work in my manual testing. If you
like the general idea I can polish it off (try to make a test, etc.).
One thing that bothers me: because I fall back to
`package--activate-all' even package that successfully activate in the
quickstart file can have their activation code run twice, and they
aren't necessarily idempotent operations. Do you think this is a
significant problem?
One possibility is to update `package-activated-list' and
`Info-directory-list' incrementally in the quickstart file, so any
signaled errors leave `package-activated-list' mostly-correct. This
way, `package--activate-all' will attempt to activate only one package
twice -- the one that signaled from the quickstart file.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Make-package-activate-all-resilient-to-quickload-err.patch --]
[-- Type: text/x-diff, Size: 2451 bytes --]
From 0ec2a7408eb248eef21c0a431eaf9c23cd5e99d3 Mon Sep 17 00:00:00 2001
From: Matt Armstrong <matt@rfc20.org>
Date: Wed, 30 Nov 2022 10:56:59 -0800
Subject: [PATCH] Make `package-activate-all' resilient to quickload errors.
If `package-activate-all' fails Emacs fails to start, so if
quickloading fails fall back to per-package activation. Note that
per-package activation already has logic to report package level
actiation errors with `message' and continue.
---
lisp/emacs-lisp/package.el | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)
diff --git a/lisp/emacs-lisp/package.el b/lisp/emacs-lisp/package.el
index 8d44fae30a..95921256d6 100644
--- a/lisp/emacs-lisp/package.el
+++ b/lisp/emacs-lisp/package.el
@@ -1714,16 +1714,23 @@ package-activate-all
package-quickstart-file))))
;; The quickstart file presumes that it has a blank slate,
;; so don't use it if we already activated some packages.
- (if (and qs (not (bound-and-true-p package-activated-list)))
- ;; Skip load-source-file-function which would slow us down by a factor
- ;; 2 when loading the .el file (this assumes we were careful to
- ;; save this file so it doesn't need any decoding).
- (let ((load-source-file-function nil))
- (unless (boundp 'package-activated-list)
- (setq package-activated-list nil))
- (load qs nil 'nomessage))
- (require 'package)
- (package--activate-all)))))
+ (or (and qs (not (bound-and-true-p package-activated-list))
+ ;; Skip load-source-file-function which would slow us
+ ;; down by a factor 2 when loading the .el file (this
+ ;; assumes we were careful to save this file so it
+ ;; doesn't need any decoding).
+ (let ((load-source-file-function nil))
+ (unless (boundp 'package-activated-list)
+ (setq package-activated-list nil))
+ (condition-case err
+ (load qs nil 'nomessage)
+ ;; If quickstart activation fails fall through to
+ ;; `package--activate-all' activation.
+ (error (message "Error loading %s: %s"
+ qs (error-message-string err))))))
+ (progn
+ (require 'package)
+ (package--activate-all))))))
;;;###autoload
(defun package--activate-all ()
--
2.35.1
prev parent reply other threads:[~2022-11-30 19:13 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-30 0:14 bug#59707: 29.0.50; Seeking a more robust `package-quickstart' Matt Armstrong
2022-11-30 2:37 ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-11-30 5:59 ` Matt Armstrong
2022-11-30 13:11 ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
2022-11-30 19:13 ` Matt Armstrong [this message]
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=87fse08dcf.fsf@rfc20.org \
--to=matt@rfc20.org \
--cc=59707@debbugs.gnu.org \
--cc=monnier@iro.umontreal.ca \
/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).