bug#41715: The program '/gnu/store/foobar/compute-guix-derivation' failed to compute the derivation for guix

[o . rojon]

Hej Ludo,

crazy, I wouldn't have thought it would go as deep as gdb as quickly ;-)

I followed the steps you mentioned, the results you find here: 
https://www.mediafire.com/file/g5yz8f3pput8f3w/gdb-output/file  . The 
only part I omitted from the gdb output is the first lines mentioned the 
license. (Will look at IPFS, maybe next time :))


If you need anything else, dont hesitate to specify. Also, should a 
reinstall become necessary, that would be annoying but not the end of 
the world. So if we hit a dead end thats fine by me.


Thanks a lot for your help. So far, I received only good from the guix 
community :)


Greetings,
Olivier

On 09.06.2020 16:30, Ludovic Courtès wrote:

Hi,

o.ro...@posteo.net skribis:


here is the log uploaded on mediafire:
http://www.mediafire.com/file/ldqoi68y88rzrn9/log.bz2/file (note that
if you can recommend another uploading service, feel free to!)


I don’t know, maybe you could run IPFS.

The log reads:

--8<---cut here---start->8---
clone(child_stack=NULL,
flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD,
child_tidptr=0x7f3d44cfee50) = 1498
close(17)   = 0
read(16, "", 1) = 0
close(16)   = 0
wait4(1498, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGBUS}], 0, NULL) = 
1498

--8<---cut here---end--->8---

… which means the ‘compute-guix-derivation’ process crashed with 
SIGBUS.


Could you run:

  ulimit -c unlimited
  guix pull

That should fail again, but this time there should be a ‘core’ file in
the current directory (or ‘core.’ followed by digits).

Then you can run:

  gdb --core=./core

and at the GDB prompt, type:

  thread apply all bt

Could you let me know what that returns?

Thanks,
Ludo’.

bug#41498: sed fails to build on kernels with selinux

[Chris Marusich]

Hi,

With the attached patch, sed builds on my Fedora machine.  Yay!  There
was a small mistake in my prior email to this bug report, so you can
ignore that patch.

Chris Marusich  writes:

> diff --git a/gnu/packages/base.scm b/gnu/packages/base.scm
> index 279fe9e3d8..f075ee8f74 100644
> --- a/gnu/packages/base.scm
> +++ b/gnu/packages/base.scm
> @@ -15,6 +15,7 @@
>  ;;; Copyright © 2018 Tobias Geerinckx-Rice 
>  ;;; Copyright © 2018, 2019 Ricardo Wurmus 
>  ;;; Copyright © 2020 Vitaliy Shatrov 
> +;;; Copyright © 2020 Chris Marusich 
>  ;;;
>  ;;; This file is part of GNU Guix.
>  ;;;
> @@ -141,7 +142,21 @@ including, for example, recursive directory searching.")
>  ".tar.gz"))
>  (sha256
>   (base32
> -  "0alqagh0nliymz23kfjg6g9w3cr086k0sfni56gi8fhzqwa3xksk"
> +  "0alqagh0nliymz23kfjg6g9w3cr086k0sfni56gi8fhzqwa3xksk"))
> +;; Remove this patch once upstream releases a fixed version.
> +(patches
> + (list
> +  (origin
> +(method url-fetch)
> +(uri (string-append
> +  "https://debbugs.gnu.org/cgi/bugreport.cgi?;
> +  "att=1;msg=16;bug=36150;filename=0001-tests-"
> +  "Export-CONFIG_HEADER-to-test-scripts.patch"))
> +(file-name
> + "sed-Export-CONFIG_HEADER-to-test-scripts.patch")
> +(sha256
> + (base32
> +  
> "0q78qzc0mv4bmsf7wnqj9fjbrwb71xz18v07h0jz2cwnbbj1nwm9")))
> (build-system gnu-build-system)
> (synopsis "Stream editor")
> (native-inputs
> -- 
> 2.26.2

Even though this patch is appropriate for the sed maintainers, it is not
appropriate for use in Guix's sed package definition.  This is because
the patch changes testsuite/local.mk, which is used only when creating
the release distribution of sed (via an include directive in
Makefile.am), and Guix builds sed from a release distribution tarball.
As a result, even if Guix modifies the file, it does not have the
desired effect, and the test still fails for the same reason as before.
One way to work around this is to modify the Makefile.in, which is
generated by Automake and included in the release distribution.

I have gone ahead and committed the sed patch to core-updates in
a48a3f0640d76cb5e5945557c9aae6dabce39d93.  I have committed the guix
lint improvement to master in 21887021b9acf60157b1b0a39c16f2ec6498021b.
I am closing this bug report.

-- 
Chris
From 24edab1ef4ecedd2d9971a96a516d800d6933201 Mon Sep 17 00:00:00 2001
From: Chris Marusich 
Date: Tue, 9 Jun 2020 23:35:56 -0700
Subject: [PATCH] gnu: sed: Make it build on SELinux-enabled kernels.

Fixes: .

* gnu/packages/base.scm (sed)[origin][snippet]: New field.  This adds a
snippet, equivalent to the patch submitted upstream, which fixes an
issue that prevents sed from building on SELinux-enabled kernels.
---
 gnu/packages/base.scm | 15 ++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/gnu/packages/base.scm b/gnu/packages/base.scm
index 279fe9e3d8..2c7b4e6882 100644
--- a/gnu/packages/base.scm
+++ b/gnu/packages/base.scm
@@ -15,6 +15,7 @@
 ;;; Copyright © 2018 Tobias Geerinckx-Rice 
 ;;; Copyright © 2018, 2019 Ricardo Wurmus 
 ;;; Copyright © 2020 Vitaliy Shatrov 
+;;; Copyright © 2020 Chris Marusich 
 ;;;
 ;;; This file is part of GNU Guix.
 ;;;
@@ -141,7 +142,19 @@ including, for example, recursive directory searching.")
 ".tar.gz"))
 (sha256
  (base32
-  "0alqagh0nliymz23kfjg6g9w3cr086k0sfni56gi8fhzqwa3xksk"
+  "0alqagh0nliymz23kfjg6g9w3cr086k0sfni56gi8fhzqwa3xksk"))
+;; Remove this snippet once upstream releases a fixed version.
+;; This snippet changes Makefile.in, even though the upstream
+;; patch changes testsuite/local.mk, since we build sed from a
+;; release tarball.  See: https://bugs.gnu.org/36150
+(snippet
+ '(begin
+(substitute* "Makefile.in"
+  (("^  abs_srcdir='\\$\\(abs_srcdir\\)'.*" previous-line)
+   (string-append
+previous-line
+"  CONFIG_HEADER='$(CONFIG_HEADER)'\t\t\\\n")
+(modules '((guix build utils)
(build-system gnu-build-system)
(synopsis "Stream editor")
(native-inputs
-- 
2.26.2



signature.asc
Description: PGP signature

bug#41822: Bug in compute-guix-derivation?

[elaexuotee--- via Bug reports for GNU Guix]

Not exactly sure what to report, but:

$ guix pull
Building from this channel:
  guix  https://git.savannah.gnu.org/git/guix.git   9c6ea3c
Computing Guix derivation for 'x86_64-linux'... \guix pull: error: You 
found a bug: the program 
'/gnu/store/94yr3a44i33635b90lhbji5rixkyijvg-compute-guix-derivation'
failed to compute the derivation for Guix (version: 
"9c6ea3cb5617a1928073f41f174765ab14adbaa9"; system: "x86_64-linux";
host version: "2971ed57345ee4e00058efeaf27c6a0790fdc9dd"; pull-version: 1).
Please report it by email to .


signature.asc
Description: PGP signature

bug#41791: [Shepherd] loses track of Tor

[Ludovic Courtès]

Hi Tobias,

Tobias Geerinckx-Rice  skribis:

> ~ λ while :; do sudo herd restart tor; done
> Service tor is not running.
> Service tor has been started.
> Service tor has been stopped.
> Service tor could not be started.
> Service tor is not running.
> Service tor could not be started.
> ^C
> ~ λ pgrep tor
> 22926
> 23004

Oops, this one is fixed by b6b95685d0b478698d648f13afd927b1e1700d99.
It’s a regression introduced when switching to the Shepherd 0.8.1.

> This has (this year) happened to nginx in tight edit.conf/restart 
> loops too, but I can't say if the two are at all related.

The nginx issue may be different.  The Shepherd 0.8.0 included fixes
that should help for multi-process services like nginx.  Let me know if
it happens again.

Thanks for reporting the issue… and apologies for grumpily pushing you
to do so on IRC.  It’s a positive outcome at any rate!

Ludo’.

bug#41794: Thoughts on using Zstd compression instead of lzip

[Ludovic Courtès]

Hi,

Tobias Geerinckx-Rice  skribis:

> It's not clear where you want to see zstd used.  Here's a trivial
> patch to add Zstd to ‘guix pack’, if that's what you mean.

I suppose that was for substitutes, in reference to:

  https://guix.gnu.org/blog/2019/substitutes-are-now-available-as-lzip/

The idea is worth discussing, but IMO, romulasry, discussions should
happen on guix-devel rather than guix-patches… unless the discussion
comes with a patch, of course.  :-)

Closing!

Ludo’.

bug#41800: guix lint doesn't catch apache mirror-url

[Ludovic Courtès]

Hello,

Efraim Flashner  skribis:

> There is a mirror://apache download option which isn't caught by 'guix
> lint -c mirror-url'.

Do you have an example to reproduce it?

Thanks,
Ludo’.

bug#41669: Cross-compiled powerpc64-linux bootstrap-tarballs not reproducible

[Jack Hill]

On Tue, 9 Jun 2020, Chris Marusich wrote:


- Try building with different kernel versions on the same machine, to
 see if they differ.


I've done the rebuild after updated from Linux 5.4.41 to 5.4.45, and go 
identical results to my previous build. I am using a different kernel 
configuration than Guix's default kernel, but it was the same between the 
builds.


Best,
Jack

bug#41811: `guix search` shows control sequences

[Lars-Dominik Braun]

Hi,

when using `guix search python2-xcffib` on a foreign distribution (Gentoo) not
all control sequences are stripped by the pager (see location and license):

---snip---
name: python2-xcffib
version: 0.6.0
outputs: out
systems: x86_64-linux i686-linux
dependencies: libxcb@1.14 python2-cffi@1.13.2 python2-six@1.14.0
location: 
8;;file://xxx/gnu/store/1ccswh99papcvkj028s9myy8kq26y26g-guix-module-union/share/guile/site/3.0/gnu/packages/python-xyz.scmgnu/packages/python-xyz.scm:5437:28;;
homepage: https://github.com/tych0/xcffib
license: 8;;http://directory.fsf.org/wiki/License:ExpatExpat8;;
synopsis: XCB Python bindings
description: Xcffib is a replacement for xpyb, an XCB Python bindings.
It adds support for Python 3 and PyPy.  It is based on cffi.
relevance: 20
---

`guix search python2-xcffib | less` works fine, `unset`ting the environment
variable $LESS or setting it to 'FrX' (i.e. what guix does when it’s missing)
fixes the issue too, so I’m a little puzzled. Some more information:

---snip---
$ env | grep -i less
LESS=-R -M --shift 5
LESSOPEN=|lesspipe %s
PAGER=/usr/bin/less
$ less --version
less 551 (PCRE2 regular expressions)
[…]
$ LANG=C guix describe
Generation 15   Jun 11 2020 17:47:13(current)
  guix cbd9581
repository URL: https://git.savannah.gnu.org/git/guix.git
branch: master
commit: cbd9581acc41cd49eb81c2432452cad4de805cbd
---snap---

Cheers,
Lars

bug#41796: Grafts don't handle outputs other than out

[Ludovic Courtès]

Hi Jakub,

Thanks a lot for the reduced test case, much appreciated!

This is fixed with 03a70e4c190420e87c0b535285caf8f77260d4ff, which
includes a test inspired by yours.

ecf92194a55188a9c217d76617378749db063453 adds an nghttp2 replacement, as
you suggested on IRC, which is what prompted you to report this bug.
Apparently it works as expected.  Same for 389-ds-base.

Thanks,
Ludo’.

bug#41796: Grafts don't handle outputs other than out

[Ludovic Courtès]

Hi!

I’m trying to estimate the impact of this bug.  As of
a50628bbe0fa4ba3835e311098e4fdf7a1d8a29e, there seems to be only one
package whose replacement could end up not being grafted (here I’m
omitting outputs that, if left ungrafted, won’t affect security):

--8<---cut here---start->8---
scheme@(guile-user)> (fold-packages (lambda (p result)
  (if (and (package-replacement p)
   (> (length (fold delete 
(package-outputs p) '("debug" "doc" "static"))) 1))
  (cons p result)
  result))
'())
$11 = (#)
--8<---cut here---end--->8---

This is because of the “bin” output of ‘nss’.

>From a quick grep, there 3 packages depending on nss:bin: 389-ds-base,
libcacard, and xmlsec-nss.

389-ds-base is affected: it keeps a reference to the ungrafted “bin”:

--8<---cut here---start->8---
$ guix gc --references $(guix build 389-ds-base --no-grafts) |grep nss-
/gnu/store/gfpgqvwrixhf3sf1bnzsfxzvld0nd8b7-nss-3.50
/gnu/store/vvsa5q0g790wi97zadj5qklqpiw1fqc1-nss-3.50-bin
$ guix gc --references $(guix build 389-ds-base) |grep nss-
/gnu/store/588jh89ng8f7ks4wsay6mdm4dxapk2d6-nss-3.50
/gnu/store/vvsa5q0g790wi97zadj5qklqpiw1fqc1-nss-3.50-bin
--8<---cut here---end--->8---

The other two are fine:

--8<---cut here---start->8---
$ guix gc --references $(guix build libcacard --no-grafts) |grep nss-
/gnu/store/gfpgqvwrixhf3sf1bnzsfxzvld0nd8b7-nss-3.50
$ guix gc --references $(guix build libcacard) |grep nss-
/gnu/store/588jh89ng8f7ks4wsay6mdm4dxapk2d6-nss-3.50
$ guix gc --references $(guix build xmlsec-nss --no-grafts) |grep nss-
/gnu/store/fwb0adczsx3nqsdnj92xnv85n93qa17n-xmlsec-nss-1.2.30
/gnu/store/gfpgqvwrixhf3sf1bnzsfxzvld0nd8b7-nss-3.50
$ guix gc --references $(guix build xmlsec-nss ) |grep nss-
/gnu/store/2gzk5rfg86zyxk8d9z6b7x0xkwar95cj-xmlsec-nss-1.2.30
/gnu/store/588jh89ng8f7ks4wsay6mdm4dxapk2d6-nss-3.50
--8<---cut here---end--->8---

Ludo’.

bug#22158: chicken tests do not complete gcc compilation

[Royce Strange]

Hello.

For chicken 4.9.0.1, it looks like this may have been an upstream bug.
Looking at some of Recardo Wurmus's notes
for some tweaks to chicken back then:
https://lists.gnu.org/archive/html/chicken-hackers/2015-04/msg00063.html
Seems to line up with a some chicken bugs reported back then:
https://bugs.call-cc.org/ticket/1120
https://bugs.call-cc.org/ticket/1098
(and have been addressed upstream) 

It has been a while and chicken scheme is now at 5.2.0
It is currently building on cuirass:
https://ci.guix.gnu.org/build/2339572/details

I am also able to build it locally:
-
$ guix --version
guix (GNU Guix) 2971ed57345ee4e00058efeaf27c6a0790fdc9dd

$ guix build chicken --no-substitute
-

Going to go ahead and close.

bug#41801: httpd-config-file should inherit 'package' from httpd-configuration

[Efraim Flashner]

I have a package where I'm testing using httpd-2.2 and I have my custom
httpd-2.2 listed in the 'package' field of httpd-configuration. I
assumed when I specified my custom httpd in package of
httpd-configuration it would use my custom httpd-2.2 also in the
server-root field of httpd-config-file. The manual does say that it
defaults to httpd, but I assumed that it meant httpd, not nginx or
lighttpd or something similar.


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


signature.asc
Description: PGP signature

bug#41207: accountsservice daemon lacks dbus interfaces

[Ludovic Courtès]

Hi,

L  p R n  d n skribis:

> Ludovic Courtès  writes:
>
 L  p R n  d n skribis:
> [...]
>>> Not exactly, my understanding is that *accountsservice* doesn't find
>>> 'org.freedesktop.Accounts.service' which is provided (and asked) by
>>> lightdm. (I think the error comes from accountsservice)
>>
>> But see:
>>
>> $ (cd /tmp; tar xf $(guix build -S lightdm))
>> $ grep -r "Error updating user" /tmp/lightdm-1.30.0/
>> /tmp/lightdm-1.30.0/common/user-list.c:g_warning ("Error updating 
>> user %s: %s", priv->path, error->message);
>> /tmp/lightdm-1.30.0/common/user-list.c:g_warning ("Error updating 
>> user %s: %s", priv->path, error->message);
>> $ (cd /tmp; tar xf $(guix build -S accountsservice))
>> $ find /tmp/accountsservice-0.6.50/ -name org.freedesktop\*.service.in
>> /tmp/accountsservice-0.6.50/data/org.freedesktop.Accounts.service.in
>> $ find $(guix build accountsservice) -name org.freedesktop\*.service
>> /gnu/store/l9qyf0brhhq7mwcsyhjsh7k0d3ri3ay2-accountsservice-0.6.50/share/dbus-1/system-services/org.freedesktop.Accounts.service
>>
>> That’s why I think it’s lightdm (or dbus-daemon?) that fails to find the
>> .service file.
>>
>> ‘accountsservice-service-type’ extends dbus, such that we have
>> /etc/dbus-1/system-services/org.freedesktop.Accounts.service.  Do you
>> see this as well?
>
> You're right! ;) And probably a dbus-daemon problem from what I
> understand. But I really don't know what I'm talking about here...
> However, just to clarify, I think it's not that it's not finding
> dbus-1/system-services/org.freedesktop.Accounts.service but
> dbus-1/interfaces/org.freedesktop.Accounts(.Users?).xml. In that case,
> using #:env-variable for the dbus service might be a possibility.
> Really, I have quite a lot of files in
> /run/current-system/profile/share/dbus-1/interfaces but I don't know what are
> their purpose... :/

I think we’d need to look at the lightdm service to further debug this.

It’s not upstream yet, right?

Thanks,
Ludo’.

bug#41800: guix lint doesn't catch apache mirror-url

[Efraim Flashner]

There is a mirror://apache download option which isn't caught by 'guix
lint -c mirror-url'.


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


signature.asc
Description: PGP signature

bug#41052: core-updates: LyX: Errors and messages at start up (no textclasses found)

[Efraim Flashner]

This should now be fixed with the recent commits relating to Lyx (I
can't remember which one fixed this! Probably
03a7b67b314286a9185535f5ecbb7cde97ae2c97 with the changes to calling
python in scripts)

On Sun, May 03, 2020 at 03:34:53PM +, sirgazil via Bug reports for GNU Guix 
wrote:
> It seems that LyX is missing its default classes, which results in lots of 
> messages when starting the application. 
> 
> In my case, this happens in core-updates using a virtual machine. I can't 
> reproduce this problem in my real machine, maybe because I have some 
> leftovers from previous installations of LyX, I don't know.
> 
> 
> ## Steps to reproduce
> 
> 1. guix system vm os-config.scm (with GNOME and LyX)
> 2. Run the resulting virtual machine
> 3. Log in to the GNOME desktop
> 4. Press Alt + F2, type lyx2.3 and hit Enter to start LyX
> 
> 
> ## Expected result
> 
> LyX starts without errors and displays the "Welcome to LyX" document.
> 
> 
> ## Unexpected result
> 
> A message about a missing a textclass appears:
> 
> https://multimedialib.files.wordpress.com/2020/05/guix-lyx-no-textclasses-message-2020-05-03-1.png
> 
> Clicking on "Reconfigure" results in an error message about an article class 
> not found:
> 
> https://multimedialib.files.wordpress.com/2020/05/guix-lyx-document-class-not-found-error-2020-05-03.png
> 
> Clicking "Ok" results in another message. This time about Cite engine not 
> available:
> 
> https://multimedialib.files.wordpress.com/2020/05/guix-lyx-cite-engine-not-available-message-2020-05-03.png
> 
> Clicking "Ok" here finally takes you to the "Welcome to LyX document".
> 
> https://multimedialib.files.wordpress.com/2020/05/guix-lyx-welcome-document-2020-05-03.png
> 
> At this point, however, I don't want to use the application because I would 
> expect many things to fail. I didn't check if there is any workaround (e.g. 
> installing packages that provide the missing classes and cite engine).
> 
> 
> ## System information
> 
> ```
> $ lyx2.3 --version
> LyX 2.3.3 (2019-06-10)
> CMake Build
> 
> $ LANG=C guix describe
> Generation 78 May 02 2020 18:00:20(current)
>   guix 7b07852
> repository URL: https://git.savannah.gnu.org/git/guix.git
> branch: core-updates
> commit: 7b07852ddb334c92bcef69666f21c599f1f0fa79
>   sirgazil-x ef6ac93
> repository URL: https://gitlab.com/sirgazil/guix-channel-x.git
> branch: master
> commit: ef6ac9331a7847fdec6f0eb199524b1b755ba0df
> ```
> 
> 
> ---
> https://sirgazil.bitbucket.io/
> 
> 
> 
> 
> 
> 

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


signature.asc
Description: PGP signature

bug#37013: LyX can not find refstyle.sty

[Efraim Flashner]

I just made some changes to the Lyx package. Can you test it out now to
see if it works as expected?

On Thu, Aug 22, 2019 at 04:23:18PM +, Xavier Montillet wrote:
> Hi Ricardo,
> 
> LyX seems to build and run fine without this texlive-union. The texlive stuff 
> is only required when one wants to compile a lyx document to a pdf. I'd 
> therefore expect it to not be a dependency (but maybe a recommended package).
> 
> And removing this texlive-union fixes my problem: Now, if texlive is 
> installed in the profile, it finds it and all its components.
> 
> (Note: I use "./pre-inst-env guix environment --pure --container 
> --share=$XAUTHORITY --share=/tmp/.X11-unix --ad-hoc lyx bash coreutils evince 
> -- env XAUTHORITY=$XAUTHORITY DISPLAY=$DISPLAY lyx2.3" to run lyx so that it 
> reconfigures at startup)
> 
> Xavier
> 
> On Mon, Aug 12, 2019, at 9:30 PM, Ricardo Wurmus wrote:
> > 
> > Hi Xavier,
> > 
> > thanks for the report!
> > 
> > > Since it loaded inputenc, the package just before refstyle, from
> > > /gnu/store/wy5865kh09qak63rpq3v47wnxv2gj1y0-texlive-union-49435/share/texmf-dist/tex/latex/base/inputenc.sty,
> > > and /gnu/store/wy5865kh09qak63rpq3v47wnxv2gj1y0-texlive-union-49435/ does 
> > > not contain refstyle, I
> > > think that the problem is that LyX uses the pdflatex from this
> > > textlive-union instead of the one from the full texlive.
> > 
> > The problem is that Lyx is configured to use the texlive-union that was
> > provided during the build.  It should prefer an optional TeX Live
> > installation that the user provides.  Perhaps we can make it
> > (optionally?) look up the TeX Live tools on PATH before falling back to
> > its texlive-union.
> > 
> > > LyX has a reconfigure option (Tools -> Reconfigure) but it does not
> > > seem to help.
> > 
> > Investigating this might be a good starting point.
> > 
> > --
> > Ricardo
> > 
> >
> 
> 
> 

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


signature.asc
Description: PGP signature

bug#41050: LyX: No icons in toolbars

[Efraim Flashner]

This is now fixed with 7a7a29445ced24af0dcdabfe8011538711fe05d9

On Sun, May 03, 2020 at 02:21:23PM +, sirgazil via Bug reports for GNU Guix 
wrote:
> This happens in both master and core-updates. Also in both real and virtual 
> machines.
> 
> ## Steps to reproduce
> 
> 1. guix install lyx
> 2. Press Alt + F2, type lyx2.3 and hit Enter to start LyX
> 
> 
> # Expected result
> 
> LyX starts and the toolbar buttons should have icons (see 
> https://www.lyx.org/Screenshots).
> 
> 
> # Unexpected result
> 
> The icons are missing (see 
> https://multimedialib.files.wordpress.com/2020/05/guix-lyx-missing-icons-in-toolbars-2020-05-03.png).
> 
> 
> # System information
> 
> ```
> $ lyx2.3 --version
> LyX 2.3.3 (2019-06-10)
> CMake Build
> 
> $ LANG=C guix describe
> Generation 78 May 02 2020 18:00:20(current)
>   guix 7b07852
> repository URL: https://git.savannah.gnu.org/git/guix.git
> branch: core-updates
> commit: 7b07852ddb334c92bcef69666f21c599f1f0fa79
>   sirgazil-x ef6ac93
> repository URL: https://gitlab.com/sirgazil/guix-channel-x.git
> branch: master
> commit: ef6ac9331a7847fdec6f0eb199524b1b755ba0df
> ```
> 
> 
> ---
> https://sirgazil.bitbucket.io/
> 
> 
> 
> 
> 
> 

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


signature.asc
Description: PGP signature

bug#36394: guix.gnu.org/packages lists incorrect sqlite versions

[Royce Strange]

Tobias Geerinckx-Rice  writes:

> Guix,
>
> The package list at  displays
> the wrong versions for sqlite:
>
>  sqlite 3.26.0
>  sqlite 3.26.0
>  sqlite-with-column-metadata 3.26.0
>
> Which should be:
>
>  $ guix pull && guix package -A ^sqlite
>  sqlite   3.26.0  …
>  sqlite   3.24.0  …
>  sqlite-with-column-metadata  3.24.0  …

It looks like the duplicates happen to replaced packages:
https://guix.gnu.org/packages/G/page/9/
(see gnutls, as sqlite is one package now)

Narrowed it down to the code that generates the packages list 
for the website in the guix-artwork repo.

It looks like the original intent was to filter out replaced packages,
instead of keeping them in the list (which is what guix search does).
This will affect https://guix.gnu.org/packages.json which I believe feeds
guix package information in repology.

Attached is a patch to filter out the replaced packages.
If it is more appropriate for the website to be consistent with 
guix search and keep replaced packages in the list, 
I can submit an updated patch.
>From aac774421d5b746e633138c78768dcadb6be1650 Mon Sep 17 00:00:00 2001
From: Royce Strange 
Date: Thu, 11 Jun 2020 00:41:45 -0500
Subject: [PATCH] website: Fix duplicated packages in the package-list

Fixes .

* website/apps/packages/data.scm (%package-list):
Filter out replaced packages.
---
 website/apps/packages/data.scm | 7 +++
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/website/apps/packages/data.scm b/website/apps/packages/data.scm
index f1fa55d..d1bbc92 100644
--- a/website/apps/packages/data.scm
+++ b/website/apps/packages/data.scm
@@ -41,11 +41,10 @@
(sort (parameterize ((%package-module-path (last-pair
(%package-module-path
(fold-packages (lambda (package lst)
-(if (package-superseded package)
+(if (or (package-superseded package)
+(package-replacement package))
 lst
-(cons (or (package-replacement package)
-  package)
-  lst)))
+(cons package lst)))
   '()))
  (lambda (p1 p2)
(string