We expect users to use the generic STRING->FAT-UUID procedure.
This is consistent with how we already treat FAT32 vs FAT16.
It is not consistent with how we export 8 different aliases for
STRING->DCE-UUID, but I'm unconvinced that would be better.
* gnu/system/uuid.scm (%uuid-parsers, %uuid-printers):
Add the ‘exfat’ file system type.
Change-Id: Ia31482716e4395f9f10f794f49fb31c9f330a2e3
* gnu/packages/file-systems.scm (exfatprogs)[arguments]:
Use G-Expressions. Customise 'check phase to run the tests.
Change-Id: I645444eb1b70bb71666ce6fe905e0229007f5a54
* doc/contributing.texi (Commit Access): Split one trademark Guix
megagraph™ into three of your human paragraphs. No other change.
Change-Id: Ifd2a9166b1dfe1a22dcd0811044a408325cc22cb
This ensures ‘guix publish’ does not re-compress zst files.
* guix/utils.scm (compressed-file?): Add ‘tzst’ and ‘zst’.
Change-Id: Iae39e8826f1fc242d2e15d86c82a8ff84797002e
* gnu/packages/julia-xyz.scm (julia-genericlinearalgebra)[arguments]:
When building for aarch64-linux add a phase to skip two tests.
Change-Id: I9aedee8fc0a4aec77181a3b3c6ab25412882f077
* gnu/packages/python-xyz.scm (python-scikit-build)[arguments]: Add a
phase to mark more network tests as isolated to easily skip them.
Change-Id: I2f955182f411a68550fbd3c356bfda5584091428
It has no effect, probably due to the '+' in the pattern not being escaped.
We also don't want to refer to /var/guix/profiles/per-user/root/bin/guix
anymore, which may not exist (for Guix System users that have never ran 'guix
pull' has root). The 'guix' command used is that specified via the build
system, which is $(bindir)/guix.
* gnu/packages/package-management.scm (guix-daemon) [phases]: Delete
change-default-guix phase.
Change-Id: If4703334f34aac3486ff655af7d8d4c2ccf0a0eb
* gnu/packages/qt.scm (kdsingleapplication)[arguments]: Use ON instead of true
for cmake #:configure-flags.
[license]: Note files covered by each license.
Change-Id: Ia10b43e881e457bb80f0b773a3a64a40a1d8becb
Signed-off-by: Maxim Cournoyer <maxim.cournoyer@gmail.com>
This is a follow-up to 7c5d4031c9.
* gnu/packages/haskell-apps.scm (git-annex)[arguments]: Fix a typo in
the bash completion directory.
Change-Id: Ibbf807203c8305910acf3a2a0dd2d6086ffb6dfd
* gnu/packages/fonts.scm (font-culmus): Update to 0.140.
[arguments]: Update 'build phase to use included scripts.
Change-Id: I6dc1d3d06c57a7d6305467db23bcff220961c014
[[PGP Signed Part:No public key for DC518FC87F9716AA created at 2024-09-10T13:50:15+0900 using EDDSA]]
The attached patch updates vcmi to 1.5.7 and works with QT6 instead of
QT5 (maybe due to fixes in vcmi, maybe due to core-updates merge).
Did more play testing that I had planned...
live well,
vagrant
[2. text/x-diff; 0001-gnu-vcmi-Update-to-1.5.7.patch]
From 72c5e9e79d7f9b51508c97d5d9d3b5fda7e02a0f Mon Sep 17 00:00:00 2001
From: Vagrant Cascadian <vagrant@debian.org>
Date: Mon, 9 Sep 2024 17:02:02 -0700
Subject: [PATCH] gnu: vcmi: Update to 1.5.7.
* gnu/packages/games.scm (vcmi): Update to 1.5.7.
[native-inputs]: Use qtbase and qtools instead of the QT5 variants.
Signed-off-by: Maxim Cournoyer <maxim.cournoyer@gmail.com>
The Minetest engine and the "Minetest Game" have been decoupled upstream
in Minetest 5.9.0:
https://dev.minetest.net/Changelog#Minetest_Game
* gnu/packages/minetest.scm (minetest): Update to 5.9.0.
[arguments]: Adjust #:configure-flags and #:phases to account for the
removal of the Minetest Game from the Minetest codebase.
[native-search-paths, synopsis, description]: Likewise.
[inputs]: Remove IRRLICHT-FOR-MINETEST.
(minetest-data): Update to 0.0.0-0.88ecab3. Rename to MINETEST-GAME and
deprecate MINETEST-DATA.
[synopsis, description]: Adjust to account for the decoupling of the game from
the engine.
* gnu/packages/games.scm (irrlicht-for-minetest): Remove variable.
* guix/build-system/minetest.scm (default-minetest-game): New variable.
(%default-modules): Use it.
Change-Id: I0117bd33a2ce5eb1c79fdb29878682081f61dd4e
Signed-off-by: Leo Famulari <leo@famulari.name>
