ruby-nokogiri may be non-deterministic

  • Done
  • quality assurance status badge
Details
2 participants
  • Maxim Cournoyer
  • Wiktor Zelazny
Owner
unassigned
Submitted by
Wiktor Zelazny
Severity
normal
W
W
Wiktor Zelazny wrote on 5 Jul 2022 07:09
(name . bug-guix@gnu.org)(address . bug-guix@gnu.org)
20220705050934.bp2p4sbuyc43jenm@nibiowz
Discovered by accident and did not explore further.
W?
W
W
Wiktor Zelazny wrote on 5 Jul 2022 07:15
(name . 56389@debbugs.gnu.org)(address . 56389@debbugs.gnu.org)
20220705051517.5naaqa5cgli2dvm7@nibiowz
On Tue, Jul 05, 2022 at 07:09:37AM +0200, wzelazny@vurv.cz wrote:
Toggle quote (1 lines)
> Discovered by accident and did not explore further.
ruby-nokogiri@1.10.9 to be exact. Did not check @1.12.5. Guix version
4d126e776239cc2e9b2e39718d0a1051f4a9bbc3.
W?
M
M
Maxim Cournoyer wrote on 7 Jul 2022 19:06
Re: bug#56389: ruby-nokogiri may be non-deterministic
(name . Wiktor Zelazny)(address . wzelazny@vurv.cz)(name . 56389@debbugs.gnu.org)(address . 56389@debbugs.gnu.org)
87tu7szwch.fsf@gmail.com
Hi,

Wiktor Zelazny <wzelazny@vurv.cz> writes:

Toggle quote (7 lines)
> On Tue, Jul 05, 2022 at 07:09:37AM +0200, wzelazny@vurv.cz wrote:
>
>> Discovered by accident and did not explore further.
>
> ruby-nokogiri@1.10.9 to be exact. Did not check @1.12.5. Guix version
> 4d126e776239cc2e9b2e39718d0a1051f4a9bbc3.

I checked, and only 1.10.9 is nondeterministic. We should work toward
eliminating 1.10.9.

Thanks,

Maxim
W
W
Wiktor Zelazny wrote on 7 Oct 2022 08:54
(name . Maxim Cournoyer)(address . maxim.cournoyer@gmail.com)(name . 56389@debbugs.gnu.org)(address . 56389@debbugs.gnu.org)
Yz/NHHIFmqXahPNO@319guix
On Thu, Jul 07, 2022 at 01:06:06PM -0400, Maxim Cournoyer wrote:
Toggle quote (8 lines)
>
> Wiktor Zelazny <wzelazny@vurv.cz> writes:
>
> > ruby-nokogiri@1.10.9 to be exact. Did not check @1.12.5. Guix version
> > 4d126e776239cc2e9b2e39718d0a1051f4a9bbc3.
>
> I checked, and only 1.10.9 is nondeterministic. We should work toward
> eliminating 1.10.9.
Hi Maxim,
Sorry for responding so late.
The 1.10.9 comes with the TODO
;; TODO: In the next rebuild cycle, provide texlive a version of ruby-hydra
;; that does not depend on byebug and rspec, so that their dependencies can
;; be updated more freely. For now pin this version to avoid rebuilds.
ruby-nokogiri@1.10.9 is an indirect input of ruby-hydra, which in turn
is an input of texlive-hyphen-package. So nothing can be done until the
“next texlive rebuild cycle”. How long are the cycles, and is the
current one coming to an end soon by a chance?
W?
?
?
?elazny Wiktor wrote on 8 Oct 10:54 +0200
(name . Maxim Cournoyer)(address . maxim.cournoyer@gmail.com)(name . 56389-done@debbugs.gnu.org)(address . 56389-done@debbugs.gnu.org)
ZwTzZY8AIqJVXyax@319guix
On Fri, Oct 07, 2022 at 08:54:25AM +0200, wzelazny@vurv.cz wrote:
Toggle quote (5 lines)
> The 1.10.9 comes with the TODO
>
> ;; TODO: In the next rebuild cycle, provide texlive a version of ruby-hydra
> ;; that does not depend on byebug and rspec, so that their dependencies can
> ;; be updated more freely. For now pin this version to avoid rebuilds.
Apparently, this is what happened, as ruby-nokogiri@1.10.9 is not in
Guix, anymore.
Closing.
W?
Closed
M
M
Maxim Cournoyer wrote on 11 Oct 02:54 +0200
(name . ?elazny Wiktor)(address . wzelazny@vurv.cz)(name . 56389-done@debbugs.gnu.org)(address . 56389-done@debbugs.gnu.org)
87cyk78nml.fsf@gmail.com
Hi,

?elazny Wiktor <wzelazny@vurv.cz> writes:

Toggle quote (11 lines)
> On Fri, Oct 07, 2022 at 08:54:25AM +0200, wzelazny@vurv.cz wrote:
>
>> The 1.10.9 comes with the TODO
>>
>> ;; TODO: In the next rebuild cycle, provide texlive a version of ruby-hydra
>> ;; that does not depend on byebug and rspec, so that their dependencies can
>> ;; be updated more freely. For now pin this version to avoid rebuilds.
>
> Apparently, this is what happened, as ruby-nokogiri@1.10.9 is not in
> Guix, anymore.

Indeed, and the version we now have *is* reproducible:

Toggle snippet (8 lines)
guix challenge ruby-nokogiri

1 éléments du dépôt ont été analysés :
- 1 (100.0 %) étaient identiques
- 0 (0.0 %) étaient différents
- 0 (0.0 %) étaient impossibles à évaluer

Toggle quote (2 lines)
> Closing.

Thanks for the heads-up and close.

--
Maxim
Closed
?
Your comment

Commenting via the web interface is currently disabled.

To comment on this conversation send an email to 56389@debbugs.gnu.org

To respond to this issue using the mumi CLI, first switch to it
mumi current 56389
Then, you may apply the latest patchset in this issue (with sign off)
mumi am -- -s
Or, compose a reply to this issue
mumi compose
Or, send patches to this issue
mumi send-email *.patch