unofficial mirror of bug-guix@gnu.org 
 help / color / mirror / code / Atom feed
From: Maxim Cournoyer <maxim.cournoyer@gmail.com>
To: 73508@debbugs.gnu.org
Cc: Liliana Prikler <liliana.prikler@gmail.com>
Subject: bug#73508: Generic updaters should run last
Date: Fri, 27 Sep 2024 15:49:46 +0900	[thread overview]
Message-ID: <878qvdegl1.fsf@gmail.com> (raw)

Hi,

Currently the updaters are run in the order of their alphabetical
names.  For example, with the following instrumentation to see the
updater names tried in guix/upstream.scm:

--8<---------------cut here---------------start------------->8---
@@ -259,6 +271,7 @@ (define* (package-latest-release package
 one."
   (any (match-lambda
          (($ <upstream-updater> name description pred import)
+          (pk 'trying-updater: name)
           (and (pred package)
                (import package #:version version))))
        updaters))
--8<---------------cut here---------------end--------------->8---

attempting to update gnome-desktop produces:

--8<---------------cut here---------------start------------->8---
./pre-inst-env guix refresh gnome-desktop

;;; (trying-updater: bioconductor)

;;; (trying-updater: composer)

;;; (trying-updater: cpan)

;;; (trying-updater: cran)

;;; (trying-updater: crate)

;;; (trying-updater: egg)

;;; (trying-updater: elpa)

;;; (trying-updater: gem)

;;; (trying-updater: generic-git)

;;; (trying-updater: generic-html)
gnu/packages/gnome.scm:2265:13: gnome-desktop would be upgraded from 44.0 to 44.1
--8<---------------cut here---------------end--------------->8---

It seems to me the generic updaters, being generic thus less
likely to be precise/correct, should be tried last by 'guix refresh'.

This is ensured with the following patch:

--8<---------------cut here---------------start------------->8---
modified   guix/upstream.scm
@@ -48,6 +48,7 @@ (define-module (guix upstream)
   #:use-module (srfi srfi-26)
   #:use-module (srfi srfi-34)
   #:use-module (srfi srfi-35)
+  #:use-module (srfi srfi-71)
   #:use-module (rnrs bytevectors)
   #:use-module (ice-9 match)
   #:use-module (ice-9 regex)
@@ -226,15 +227,26 @@ (define (importer-modules)
 (define %updaters
   ;; The list of publically-known updaters, alphabetically sorted.
   (delay
-    (sort (fold-module-public-variables (lambda (obj result)
-                                          (if (upstream-updater? obj)
-                                              (cons obj result)
-                                              result))
-                                        '()
-                                        (importer-modules))
-          (lambda (updater1 updater2)
-            (string<? (symbol->string (upstream-updater-name updater1))
-                      (symbol->string (upstream-updater-name updater2)))))))
+    (let* ((updaters
+            (sort (fold-module-public-variables
+                   (lambda (obj result)
+                     (if (upstream-updater? obj)
+                         (cons obj result)
+                         result))
+                   '()
+                   (importer-modules))
+                  (lambda (updater1 updater2)
+                    (string<?
+                     (symbol->string (upstream-updater-name updater1))
+                     (symbol->string (upstream-updater-name updater2))))))
+           (generic-updaters rest (partition
+                                   (compose (cut string-prefix? "generic" <>)
+                                            symbol->string
+                                            upstream-updater-name)
+                                   updaters)))
+      ;; Ensure the generic updaters are tried last, as otherwise they could
+      ;; return less accurate results.
+      (append rest generic-updaters))))
 
 ;; Tests need to mock this variable so mark it as "non-declarative".
 (set! %updaters %updaters)
@@ -259,6 +271,7 @@ (define* (package-latest-release package
--8<---------------cut here---------------end--------------->8---

Now, the 'gnome-updater' at least ends up being tried before the
'generic-html':

--8<---------------cut here---------------start------------->8---
;;; (trying-updater: bioconductor)

;;; (trying-updater: composer)

;;; (trying-updater: cpan)

;;; (trying-updater: cran)

;;; (trying-updater: crate)

;;; (trying-updater: egg)

;;; (trying-updater: elpa)

;;; (trying-updater: gem)

;;; (trying-updater: github)

;;; (trying-updater: gnome)

gnu/packages/gnome.scm:2265:13: gnome-desktop would be upgraded from 44.0 to 44.1
--8<---------------cut here---------------end--------------->8---

I'll send a 'git format-patch' version in a follow-up.

-- 
Thanks,
Maxim




             reply	other threads:[~2024-09-27  6:51 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-27  6:49 Maxim Cournoyer [this message]
2024-09-27  7:02 ` bug#73508: [PATCH] upstream: Try the generic importers last Maxim Cournoyer
2024-10-14 11:01   ` Ludovic Courtès
2024-10-19 12:52     ` bug#73508: Generic updaters should run last Maxim Cournoyer

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://guix.gnu.org/

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

  git send-email \
    --in-reply-to=878qvdegl1.fsf@gmail.com \
    --to=maxim.cournoyer@gmail.com \
    --cc=73508@debbugs.gnu.org \
    --cc=liliana.prikler@gmail.com \
    /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/guix.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).