Federico Beffa skribis:
> 1. GSettings schemas: More than schemas compilation, the problem is that
> the schemas are by default expected to be in $datadir/glib-2.0/schemas
> (with $datadir usually being /usr/share). In spite of this, other
> directories can be specified with the help of the envi
Federico Beffa skribis:
> From 9bea9402675314a9ed5e74bc23919db423795663 Mon Sep 17 00:00:00 2001
> From: Federico Beffa
> Date: Thu, 25 Sep 2014 17:04:13 +0200
> Subject: [PATCH] maxima: Remove Tcl from 'inputs'. Add 'post-install' phase
>
> * gnu/packages/maths.scm(maxima): Remove Tcl from 'in
Manolis Ragkousis skribis:
> Taken care of the FIXME.
Thanks, I’ve merged it with the other commit on wip-hurd, and rebased it
on top of master.
> Just wanted to add that in base.scm, instead of gnu-gettext we needed
> perl. Probably I changed it by mistake while rebasing some local
> commits.
Working Ruby package. Next step is to get gems working locally.
diff --git a/gnu/packages/ruby.scm b/gnu/packages/ruby.scm
new file mode 100644
index 000..a13606c
--- /dev/null
+++ b/gnu/packages/ruby.scm
@@ -0,0 +1,99 @@
+;;; GNU Guix --- Functional package management for GNU
+;;; Copyright ©
Ok, hopefully this is the right patch after rebase and format-patch.
>From d0d9843f80aca29e51f82c2da894369f892c9e3c Mon Sep 17 00:00:00 2001
From: Pjotr Prins
Date: Sat, 27 Sep 2014 14:56:44 +0200
Subject: [PATCH] packages: Add Ruby
* gnu/packages/ruby.scm (Ruby): Added Ruby package
* guix/licen
As much as I'd like to be six years old again in some aspects, I'm
curious why most of the dates under /gnu/store and under ~/.guix-profile
are set to Dec 31 1969? Yes, I know that this is just before the
beginning of the Unix epoch, which must figure into the equation in some
way.
If this is in
Nate Bargmann writes:
> As much as I'd like to be six years old again in some aspects, I'm
> curious why most of the dates under /gnu/store and under ~/.guix-profile
> are set to Dec 31 1969? Yes, I know that this is just before the
> beginning of the Unix epoch, which must figure into the equat
Pjotr Prins writes:
> Ok, hopefully this is the right patch after rebase and format-patch.
>
I can confirm that the package builds and runs. :)
> From d0d9843f80aca29e51f82c2da894369f892c9e3c Mon Sep 17 00:00:00 2001
> From: Pjotr Prins
> Date: Sat, 27 Sep 2014 14:56:44 +0200
> Subject: [PATCH
>> And thread_terminate does exist, that's why it's enough for our needs.
>
> You mean ‘thread_terminate_release’?
Okay, let me explain it a bit better
Here is the part we are interested in
"libpthread/sysdeps/mach/pt-thread-terminate.c"
/* Terminate and release all that's left. */
err = __
* On 2014 27 Sep 08:32 -0500, David Thompson wrote:
>
> This is by design in order to have more deterministic builds. If two
> files have the same contents but have different timestamps, then they
> will have different hashes.
Thanks, David.
That helps me understand the system a bit better.
-
Nate Bargmann skribis:
> As much as I'd like to be six years old again in some aspects, I'm
> curious why most of the dates under /gnu/store and under ~/.guix-profile
> are set to Dec 31 1969?
It should be noted that it’s actually Jan. 1st 1970 UTC. :-)
Ludo’.
l...@gnu.org (Ludovic Courtès) writes:
> Federico Beffa skribis:
>
>> 1. GSettings schemas: More than schemas compilation, the problem is that
>> the schemas are by default expected to be in $datadir/glib-2.0/schemas
>> (with $datadir usually being /usr/share). In spite of this, other
>> directo
Hello Guix of the hackathon! :-)
Every time core packages like libc or Bash are changed, we end up
rebuilding (or re-downloading) the world, and storing multiple copies of
mostly-identical packages. The daemon implements a simple deduplication
strategy, where identical files found in the store a
I am working on the cross-base patch so we can merge it with wip-hurd as
well, but some problems need to be discussed.
First, in glibc/hurd headers we need to redeclare #:configure-flags with
less flags.
We shouldn't but because we use the same configure flags as in glibc and
glibc/hurd, we get an
* On 2014 27 Sep 09:35 -0500, Ludovic Courtès wrote:
> It should be noted that it’s actually Jan. 1st 1970 UTC. :-)
D'oh! Indeed. As I am six hours behind (America/Chicago) the system
rightly interprets the beginning of the epoch UTC as the prior day
here. All these years and that never dawned
Nate Bargmann skribis:
> * On 2014 27 Sep 09:35 -0500, Ludovic Courtès wrote:
>> It should be noted that it’s actually Jan. 1st 1970 UTC. :-)
>
> D'oh! Indeed. As I am six hours behind (America/Chicago) the system
> rightly interprets the beginning of the epoch UTC as the prior day
> here. Al
In order to make my cross-base i686-pc-gnu toolchain patch much
simpler, I changed the --host=i686-pc-gnu flag with
--build=i686-pc-gnu in hurd-headers.
The problem is that if we keep --host, we have the same problems we
had with hurd-minimal. But because it's only the headers, --build flag
will d
Happy first day of the Guix hackathon, everyone!
I spent my day working on generalizing the 'guix import' UI to allow for
using a PyPI importer in addition to the pre-existing Nix importer.
It's now at the point where I stop coding and open it up for review. :)
>From b3ec259fd097034631cf311040af7
I should mention that the PyPI import requires the guile-json library.
How should this be handled in our build scripts?
--
David Thompson
Web Developer - Free Software Foundation - http://fsf.org
GPG Key: 0FF1D807
Support the FSF: https://fsf.org/donate
David Thompson skribis:
> I should mention that the PyPI import requires the guile-json library.
> How should this be handled in our build scripts?
I think it’s fine to install it whether or not guile-json is available.
Now, when there are tests ;-), there’ll have to be an Automake
conditional
David Thompson skribis:
> I spent my day working on generalizing the 'guix import' UI to allow for
> using a PyPI importer in addition to the pre-existing Nix importer.
> It's now at the point where I stop coding and open it up for review. :)
Heheh, cool! :-)
Overall looks good to me. Some com
* guix/scripts/lint.scm (check-start-with-package-name): New method.
* tests/lint.scm: Test it.
---
guix/scripts/lint.scm | 9 +
tests/lint.scm| 10 ++
2 files changed, 19 insertions(+)
diff --git a/guix/scripts/lint.scm b/guix/scripts/lint.scm
index ee00e5c..8f49afc 1006
* guix/scripts/lint.scm (check-description-style,
check-synopsis-start-upper-case): New methods
* tests/lint.scm: Test them.
---
guix/scripts/lint.scm | 26 ++
tests/lint.scm| 18 ++
2 files changed, 44 insertions(+)
diff --git a/guix/scripts/lint.
The following five patches improve the "guix lint" command, according to
suggestions from Alex Kost and Ludovic Courtès.
Cyril Roelandt (5):
guix lint: make sure check-patches retrieves patch names.
guix lint: Make sure synopses are not too long.
guix lint: check whether descriptions and sy
* guix/scripts/lint.scm (check-synopsis-length): New procedure.
* tests/lint.scm: test it.
---
guix/scripts/lint.scm | 9 -
tests/lint.scm| 9 +
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/guix/scripts/lint.scm b/guix/scripts/lint.scm
index 0fbec1b..12391
* guix/scripts/lint.scm (check-start-article): use "string-ci=?" instead of
"string=?".
* tests/lint.scm: Add corresponding tests.
---
guix/scripts/lint.scm | 4 ++--
tests/lint.scm| 18 ++
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/guix/scripts/lint.
* guix/lint (check-patches): Test the output of origin-patches.
---
guix/scripts/lint.scm | 1 +
1 file changed, 1 insertion(+)
diff --git a/guix/scripts/lint.scm b/guix/scripts/lint.scm
index 83dde9a..0fbec1b 100644
--- a/guix/scripts/lint.scm
+++ b/guix/scripts/lint.scm
@@ -152,6 +152,7 @@ Run
Pjotr Prins writes:
> Working Ruby package. Next step is to get gems working locally.
I don't have the patches quite ready yet, but I wrote a working Ruby
build system for building gems locally. I'll share the patches when I
wake up.
--
David Thompson
Web Developer - Free Software Foundation
In the section of the manual that first explains how to write a package
recipe, the example didn't work. The fix is trivial, just use
define-public instead of define.
---
doc/guix.texi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guix.texi b/doc/guix.texi
index f73ce52..
I was working off of this example from the manual when trying to
package feh this afternoon, but the example had a mistake. Thanks to
the folks on irc who helped me work this out (and a number of other
issues I was having).
Ian Denhardt (1):
Fix incorrect example package recipe in docs.
doc/gu
30 matches
Mail list logo
