GNU bug report logs - #71928
[PATCH] gnu: Throw error when extending services that do not use extensions.

Previous Next

Package: guix-patches;

Reported by: Richard Sent <richard <at> freakingpenguin.com>

Date: Wed, 3 Jul 2024 22:14:02 UTC

Severity: normal

Tags: moreinfo, patch

Full log


View this message in rfc822 format

From: Richard Sent <richard <at> freakingpenguin.com>
To: 71928 <at> debbugs.gnu.org
Cc: ludo <at> gnu.org, Richard Sent <richard <at> freakingpenguin.com>
Subject: [bug#71928] [PATCH v2] services: Warn when extending services that do not use extensions.
Date: Wed, 15 Jan 2025 12:08:00 -0500
* gnu/services.scm (fold-services): Add warning when extending a service that
does not use extensions.

Fixes: https://issues.guix.gnu.org/71887
Change-Id: Ic8d631674bfddde495c93952d9e6cd5649bb287d
---
Hi all. Better late than never, right?

> Currently the “#t means ensure the service is present” convention is not
> enforced; is it even widely followed in the Guix repo?  But then, what
> about repos out there that provide services?

This convention is used for zabbix-front-end-service-type,
cuirass-service-type, and cgit-service-type. There might be
other services that use a different value in the Guix repo, but I
haven't found them yet.

> It would seem to me that this is something that cannot be changed
> overnight without running the risk to break people’s code.  So perhaps
> the first step (and maybe last step?) would be to print a warning.

Done.

This patch has the following limitations:

1. If services A, B, and C all extend unextendable service D, and one
of the services doesn't use #t as the value, the warning is printed
for A, B, and C. I can't imagine this situation being common and it
seems awkward to resolve without needlessly recomputing the extenion
value.

2. Extendable services can not be extended with #t to ensure the
service is instantiated because #t will be passed to compose/extend
and likely throw an error. Not sure of how this should be resolved
because a service could always expect to receive a #t from an
extension and do some special behavior. In my opinion this is a
separate issue.

3. The "#t to ensure an unextendable service is instantiated"
convention isn't documented. Given 2 I couldn't find a way to phrase
it without it turning into word salad.

 gnu/services.scm | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/gnu/services.scm b/gnu/services.scm
index 8a4002e072..2e012ff028 100644
--- a/gnu/services.scm
+++ b/gnu/services.scm
@@ -9,6 +9,7 @@
 ;;; Copyright © 2022 Tobias Geerinckx-Rice <me <at> tobias.gr>
 ;;; Copyright © 2023 Brian Cully <bjc <at> spork.org>
 ;;; Copyright © 2024 Nicolas Graves <ngraves <at> ngraves.fr>
+;;; Copyright © 2025 Richard Sent <richard <at> freakingpenguin.com>
 ;;;
 ;;; This file is part of GNU Guix.
 ;;;
@@ -1250,12 +1251,26 @@ (define* (fold-services services
                      (params     -> (service-value sink))
                      (service
                       ->
-                      ;; Distinguish COMPOSE and EXTEND because PARAMS typically
-                      ;; has a different type than the elements of EXTENSIONS.
-                      (if extend
-                          (service (service-kind sink)
-                                   (extend params (compose extensions)))
-                          sink)))
+                      (begin
+                        (unless (or (null? extensions)
+                                    ;; Extending an unextendable service with
+                                    ;; a value of #t is a convention for "make
+                                    ;; sure the service is present."
+                                    (every (cut eq? #t <>) extensions)
+                                    (and extend compose))
+                          (warning
+                           ;; XXX: This warning lists all dependents even if
+                           ;; only one dependent extends with a non-#t value.
+                           (G_ "Extensions are not supported in \
+~a yet the following service types extend it:~{~%   ~a~}~%")
+                           (service-kind sink)
+                           (map service-kind dependents)))
+                        ;; Distinguish COMPOSE and EXTEND because PARAMS typically
+                        ;; has a different type than the elements of EXTENSIONS.
+                        (if extend
+                            (service (service-kind sink)
+                                     (extend params (compose extensions)))
+                            sink))))
                   (mbegin %state-monad
                     (set-current-state (vhash-consq sink service visited))
                     (return service))))

base-commit: b696658ee8e0655b17f5d26e024956b5148e36d6
-- 
2.47.1





This bug report was last modified 153 days ago.

Previous Next


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