Re: Package test service for GNU maintainers

2014-08-20 Thread Ludovic Courtès
John Darrington skribis: > I could give it a go. Cool! > 2. By default, guix does not check that out-of-source builds are functional. >"make distcheck" on the other-hand does. Do we want to duplicate the > function >of "make distcheck"? Perhaps we do, since I have recently seen a rat

Re: [PATCH] Emacs interface for Guix

2014-08-20 Thread Ludovic Courtès
Alex Kost skribis: > From 7641752189cfc4ad3c85a042ea9eeea2b39435b4 Mon Sep 17 00:00:00 2001 > From: Alex Kost > Date: Thu, 14 Aug 2014 00:03:53 +0400 > Subject: [PATCH 1/2] profiles: Add 'manifest-transaction'. > > * guix/profiles.scm (): New record-type. > (manifest-perform-transaction): New

Re: Package test service for GNU maintainers

2014-08-20 Thread Ludovic Courtès
"Jason Self" skribis: > Since GNU maintainers (in theory) already upload their stuff to > ftp.gnu.org or alpha.gnu.org using that file triplet it might be nice > if something could be worked into their existing workflow. Perhaps new > commands in the directive file to do the things you describe?

Re: [PATCH] packages: Remove dead code.

2014-08-20 Thread Ludovic Courtès
Alex Kost skribis: > If I understand correctly, ‘_’ is redundant in “gnu/packages.scm” as it > is already defined in (guix ui). Indeed, applied. > Also I fixed a double space in this commit (not sure if it's acceptable > here). That’s fine, IMO. If this were a wider change, like removing all

Re: [PATCH] Export 'check-package-freshness'

2014-08-20 Thread Ludovic Courtès
Alex Kost skribis: > Here is the patch for that. I'm not sure about the commit message. > Should it be changed? [...] > From a13dbc81b6ca8c16343fd17ec57812f0c24e Mon Sep 17 00:00:00 2001 > From: Alex Kost > Date: Sat, 16 Aug 2014 22:00:34 +0400 > Subject: [PATCH] Move 'check-package-fresh

Re: Emacs interface for Guix

2014-08-20 Thread Alex Kost
Ludovic Courtès (2014-08-20 01:00 +0400) wrote: > Alex Kost skribis: > >> From 7641752189cfc4ad3c85a042ea9eeea2b39435b4 Mon Sep 17 00:00:00 2001 >> From: Alex Kost >> Date: Thu, 14 Aug 2014 00:03:53 +0400 >> Subject: [PATCH 1/2] profiles: Add 'manifest-transaction'. >> >> * guix/profiles.scm ():

[PATCH] profiles: Report about upgrades.

2014-08-20 Thread Alex Kost
Ludovic Courtès (2014-08-16 13:27 +0400) wrote: > Alex Kost skribis: > >> (define* (manifest-show-transaction manifest transaction #:key dry-run?) >> "Display what will/would be installed/removed from MANIFEST by >> TRANSACTION." > > [...] > >> (format (current-error-port) >>

Making 'file' an implicit native input in core-updates

2014-08-20 Thread mhw
Hi, I recently updated libffi, libatomic_ops, and libgc on core-updates. Unfortunately, all three of them fail on MIPS, for the same reason that sudo and gawk currently need workarounds: all of these packages were bootstrapped with a version of 'libtool' that requires /usr/bin/file on MIPS. Given

Re: Making 'file' an implicit native input in core-updates

2014-08-20 Thread Eric Bavier
m...@netris.org writes: > Unfortunately, all three of them fail on MIPS, for the same reason that > sudo and gawk currently need workarounds: all of these packages were > bootstrapped with a version of 'libtool' that requires /usr/bin/file on > MIPS. > > Given that this problem is becoming increa

Re: Making 'file' an implicit native input in core-updates

2014-08-20 Thread Ludovic Courtès
m...@netris.org skribis: > Given that this problem is becoming increasingly common, I think we > should add 'file' as an implicit native-input for the gnu-build-system > in core-updates, and also add a new phase to %standard-phases that > substitutes /usr/bin/file in the usual places with a path t