guix-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#27596: [PATCH] guix: lint: Add checker for new upstream versions.


From: Efraim Flashner
Subject: bug#27596: [PATCH] guix: lint: Add checker for new upstream versions.
Date: Mon, 9 Oct 2017 14:17:54 +0300
User-agent: Mutt/1.8.3 (2017-05-23)

On Mon, Oct 09, 2017 at 09:28:04AM +0200, Ludovic Courtès wrote:
> Hello!
> 
> Efraim Flashner <address@hidden> skribis:
> 
> > On Sat, Oct 07, 2017 at 10:50:21PM +0200, Ludovic Courtès wrote:
> 
> [...]
> 
> >> >> +(define (check-for-updates package)
> >> >> +  "Check if there is an update available for PACKAGE."
> >> >> +  (match (package-latest-release package (force %updaters))
> >> >> +    ((? upstream-source? source)
> >> >> +     (when (version>? (upstream-source-version source)
> >> >> +                      (package-version package))
> >> >> +       (emit-warning package
> >> >> +                     (format #f (G_ "can be upgraded to ~a~%")
> >> >> +                             (upstream-source-version source)))))))
> >> >
> >> > I think you can (1) use ‘package-latest-release*’ which ensures that the
> >> > returned version is newer, and (2) add a case for #f since
> >> > ‘package-latest-release*’ can return #f.
> >> >
> >> > Apart from that it LGTM, thank you!
> >> >
> >> > Ludo’.
> >
> > I've been working on it with package-latest-release*, and its nice not
> > having to reimplement the logic, but I can't get it to return the newest
> > version of the upstream-package. I got it to return a useful message
> > with:
> >
> > (define (check-for-updates package)
> >   (if (package-latest-release* package (force %updaters))
> >        (emit-warning package
> >                      (format #f (G_ "can be upgraded to ~a~%")
> >                              (upstream-source-version 
> > (package-latest-release package (force %updaters)))))
> >        #t))
> >
> > but I've already checked if there's an upstream release, I shouldn't
> > need to check a second time. package-latest-release* returns true/false,
> 
> 
> AFAICS ‘package-latest-release*’ returns #f or an <upstream-source>:
> 
> --8<---------------cut here---------------start------------->8---
> scheme@(guile-user)> (package-latest-release* binutils (force %updaters))
> $4 = #<<upstream-source> package: "binutils" version: "2.29.1" urls: 
> ("mirror://gnu/binutils/binutils-2.29.1.tar.gz" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.lz" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.bz2" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.xz") signature-urls: 
> ("mirror://gnu/binutils/binutils-2.29.1.tar.gz.sig" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.lz.sig" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.bz2.sig" 
> "mirror://gnu/binutils/binutils-2.29.1.tar.xz.sig")>
> --8<---------------cut here---------------end--------------->8---
> 
> So you can write:
> 
>   (match (package-latest-release* package (force %updaters))
>     ((? upstream-source? source)
>      …)
>     (#f   ;cannot determine latest release
>      #f))
> 
> HTH!
> 
> Ludo’.

I looked at the final diff between my earlier patch and my final
version, it pretty much added up to the lines you wrote above ;) I
should get myself a rubber duck, should help my debugging and reading
skills.


-- 
Efraim Flashner   <address@hidden>   אפרים פלשנר
GPG key = A28B F40C 3E55 1372 662D  14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]