bug#51091: [PATCH v3] guix: opam: Do not fail when refreshing.

[Julien Lepiller <julien@lepiller.eu>]

[-- Attachment #1: Type: text/plain, Size: 2388 bytes --]

Forgot to send to the address that actually closes the issue ^^"



-------- Courriel d’origine --------
De : Julien Lepiller <julien@lepiller.eu>
Envoyé : 19 novembre 2021 13:13:08 GMT-05:00
À : zimoun <zimon.toutoune@gmail.com>
Cc : "Ludovic Courtès" <ludo@gnu.org>, Xinglu Chen <public@yoctocell.xyz>
Objet : Re: [bug#51091] [PATCH v3] guix: opam: Do not fail when refreshing.

So, since I pushed Zimoun's patch, I'm closing this one. I received some help on IRC this morning so I should be able to come up with something more generic for all importers at some point. Zimoun if you want to work on that, I'd be happy to collaborate :)

Le 19 novembre 2021 06:30:20 GMT-05:00, zimoun <zimon.toutoune@gmail.com> a écrit :
>Hi Julien,
>
>On Fri, 19 Nov 2021 at 12:21, Julien Lepiller <julien@lepiller.eu> wrote:
>
>> >> I forgot to remove the catch #t around the whole body of the function.
>> >> I noticed that guard* was raising &non-continuable so I tried to fix it
>> >> by using raise-continuable from (ice-9 exceptions). Is this the correct
>> >> solution?
>> >
>> >I suppose, though I’m not sure why it needs to be continuable: you could
>> >just catch the exception and move on to the next package?
>>
>> I don't understand how to catch the exception though, unless you mean wrap everything with catch #t, which kinda defeats the purpose of having a condition in the first pjace. guard* raises &non-continuable unless the condition is continuable, or I'm missing something in the way I use it. I have no idea what a continuable exception is, so let me just push the other patch.
>>
>> (guard* (c ((opam-error? c) #f)))
>>   (raise (condition (&opam-error …))))
>>
>> Doesn't return #f as I expect, but raises &non-continuable.
>
>I sympathize and I had / is still having hard time with similar use
>cases.  That's one of the reasons (among my laziness :-)) that [1] is
>not fixed yet. :-)
>
>1: <1: <http://issues.guix.gnu.org/issue/45984>
>
>
>
>> Hopefully someone smarter than me can figure it out. I'll push the other patch, although I don't like the double warning in the updater.
>
>I agree.  And move all G_ strings to guix/scripts/ is a good idea, IMHO.
>Well, I do not know. :-)
>
>(I secretly hoped that you would be the smarter than me person fixing
>the recursive importers. ;-))
>
>
>Cheers,
>simon

[-- Attachment #2: Type: text/html, Size: 3459 bytes --]