[00:12:39]
<Émy - OniriCorpe> ```
123 ERROR There is not enough disk space left to install this application
Error: Installation failed.
```
wth https://ci-apps-dev.yunohost.org/ci/job/13602
[00:24:01]
<Aleks (he/him/il/lui)> meh, maybe another test in parallel eating up the host's system space
[04:12:45]
<Émy - OniriCorpe> If someone could explain to them the automatic patching system, it could be nice ^w^
https://github.com/YunoHost-Apps/outline_ynh/pull/102#issuecomment-1947639341
[08:49:00]
<Yunohost Git/Infra notifications> App concrete5 failed all tests in job [#23574](https://ci-apps.yunohost.org/ci/job/23574) :(
[08:53:53]
<Yunohost Git/Infra notifications> App drupal failed all tests in job [#23575](https://ci-apps.yunohost.org/ci/job/23575) :(
[08:58:47]
<Yunohost Git/Infra notifications> App grr failed all tests in job [#23576](https://ci-apps.yunohost.org/ci/job/23576) :(
[09:03:35]
<Yunohost Git/Infra notifications> App digiwords failed all tests in job [#23577](https://ci-apps.yunohost.org/ci/job/23577) :(
[09:13:26]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u pushed 1 commit to testing: 16.9.0 ([d958d656](https://github.com/YunoHost-Apps/gitlab_ynh/commit/d958d6565b1670b65dcb6aa544576c3e3fd91ca1))
[09:13:32]
<Yunohost Git/Infra notifications> [gitlab_ynh] @yunohost-bot pushed 1 commit to testing: Auto-update README ([c1373c6c](https://github.com/YunoHost-Apps/gitlab_ynh/commit/c1373c6c059e2db9a73409a168dd11a1d03a7b14))
[09:13:49]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u edited [pull request #242](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242): Testing
[09:13:53]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u [commented](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242#issuecomment-1948012487) on [issue #242](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242) Testing: testme
[09:13:53]
<Yunohost Git/Infra notifications> [gitlab_ynh] @yunohost-bot [commented](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242#issuecomment-1948012566) on [issue #242](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242) Testing: :stuck_out_tongue_winking_eye:
[[Test Badge](https://img.shields.io/endpoint?url=https://ci-apps-dev.yunohost.org/ci/ap...
[09:16:53]
<tituspijean> https://aria.im/_matrix/media/v1/download/pijean.ovh/318e48f94837cb7f1b9200cf56a3c14ff36bfb2bff1beaa668d787f1e51b0186
[09:20:05]
<tituspijean> https://aria.im/_matrix/media/v1/download/pijean.ovh/3d059d767ecda0f8ad21e9e81cc48c70f96150f3bcfc04077aed58446268ed04
[09:21:29]
<Mateusz Szymański> we could run apt proxy on CI
[09:21:32]
<tituspijean> Mmmh, interesting read: https://github.com/oerdnj/deb.sury.org/issues/2074
[09:23:16]
<Mateusz Szymański> > <@titus:pijean.ovh> Mmmh, interesting read: https://github.com/oerdnj/deb.sury.org/issues/2074
why did he start rotating keys? oO
[09:23:27]
<Yunohost Git/Infra notifications> App excalidraw failed all tests in job [#23579](https://ci-apps.yunohost.org/ci/job/23579) :(
[09:38:58]
<Yunohost Git/Infra notifications> App librarian goes down from level 8 to 7 in job [#23580](https://ci-apps.yunohost.org/ci/job/23580)
[10:34:04]
<Yunohost Git/Infra notifications> App dotclear2 failed all tests in job [#23583](https://ci-apps.yunohost.org/ci/job/23583) :(
[11:00:25]
<Yunohost Git/Infra notifications> App garradin failed all tests in job [#23585](https://ci-apps.yunohost.org/ci/job/23585) :(
[11:03:27]
<Yunohost Git/Infra notifications> App gitlab-runner failed all tests in job [#23586](https://ci-apps.yunohost.org/ci/job/23586) :(
[11:26:51]
<Yunohost Git/Infra notifications> App humhub failed all tests in job [#23588](https://ci-apps.yunohost.org/ci/job/23588) :(
[11:44:23]
<Yunohost Git/Infra notifications> App jellyfin failed all tests in job [#23590](https://ci-apps.yunohost.org/ci/job/23590) :(
[11:56:14]
<Yunohost Git/Infra notifications> Job [#23570](https://ci-apps.yunohost.org/ci/job/23570) for ghost failed miserably :(
[12:06:40]
<Dante> Hi all! 🙂
[12:07:05]
<Dante> could someone explain to me what `autoupdate.strategy = "latest_github_release"` does specifically?
[12:07:23]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u merged [pull request #242](https://github.com/YunoHost-Apps/gitlab_ynh/pull/242): Testing
[12:07:23]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u pushed 5 commits to master ([d499f2f31830...3be74724da5b](https://github.com/YunoHost-Apps/gitlab_ynh/compare/d499f2f31830...3be74724da5b))
[12:07:35]
<Dante> does it skip/replaces the workflows from Github?
[12:07:57]
<Yunohost Git/Infra notifications> [gitlab_ynh] @kay0u [commented](https://github.com/YunoHost-Apps/gitlab_ynh/issues/240#issuecomment-1948272656) on [issue #240](https://github.com/YunoHost-Apps/gitlab_ynh/issues/240) Bad admin account: Should be fixed in 16.9.0, feel free to reopen this issue if needed
[12:09:04]
<Aleks (he/him/il/lui)> > <@thardev:matrix.org> could someone explain to me what `autoupdate.strategy = "latest_github_release"` does specifically?
it means that everyday we have a script checking if the app upstream has a new release published on their github, and if so it will automatically create a pull request to update the app accordingly
[12:09:32]
<Aleks (he/him/il/lui)> for example: https://github.com/YunoHost-Apps/sutom_ynh/pull/18
[12:10:20]
<Dante> hmmm right, so it replaces the logic that is usually on those `updater.sh` scripts
[12:10:25]
<Aleks (he/him/il/lui)> yup
[12:12:03]
<Dante> I see, so I'll need to then remove that logic from there but is it possible to still keep this [part](https://github.com/YunoHost-Apps/mautrix_whatsapp_ynh/blob/master/.github/workflows/updater.sh#L113-L148) that updates the config file of the project? would that work? 🤔
[12:13:35]
<Aleks (he/him/il/lui)> hmnope that's not handled right now ...
[12:13:52]
<Yunohost Git/Infra notifications> App misskey failed all tests in job [#23593](https://ci-apps.yunohost.org/ci/job/23593) :(
[12:16:22]
<Dante> oh I see 😭 I think I'll have to stick to use the `update.sh` file for now then... that update of the config file saves a lot of effort on each upgrade
[12:17:53]
<Dante> or is there a way that I can maybe have both things like the workflow that handles only the `config.yaml` update and also the `autoupdate.strategy = "latest_github_release"` thing
[13:22:36]
<Salamandar> > <@thardev:matrix.org> or is there a way that I can maybe have both things like the workflow that handles only the `config.yaml` update and also the `autoupdate.strategy = "latest_github_release"` thing
yes you completely can
[13:22:59]
<Salamandar> the yunohost uatoupdater will update your sources, and the github workflow will call update.sh
[13:23:22]
<Salamandar> although in the future we might try to handle this kind of cases in the autoupdater
[13:23:54]
<Salamandar> but IMHO we should finish the migration to packagingv2, see which apps still use update.sh, and then try to handle those few cases (in the most generic way)
[13:31:44]
<Yunohost Git/Infra notifications> App pufferpanel failed all tests in job [#23596](https://ci-apps.yunohost.org/ci/job/23596) :(
[14:00:09]
<Aleks (he/him/il/lui)> > <@titus:pijean.ovh> Mmmh, interesting read: https://github.com/oerdnj/deb.sury.org/issues/2074
don't tell me this means every yunohost instance gonna have upgrades blocked because of this T_T
[14:02:09]
<Aleks (he/him/il/lui)> hmkay i see errors when opening the Upgrade view but at least it's not blocking the upgrade
[14:02:28]
<Aleks (he/him/il/lui)> but the regenconf probably won't add the new key because the file already exist so we still gotta fix this
[14:07:39]
<Dante> > <@Salamandar:matrix.org> the yunohost uatoupdater will update your sources, and the github workflow will call update.sh
yes, I thought that as well, it should work fine but I have to check how to commit to the specific PR that the `autoupdate.strategy = "latest_github_release"` creates from the `updater.sh` script since it seems that for that script that was defined on the `updater.yml` file
[14:08:48]
<Dante> > <@Salamandar:matrix.org> but IMHO we should finish the migration to packagingv2, see which apps still use update.sh, and then try to handle those few cases (in the most generic way)
a hook or something similar would be interesting for the packages so then we can add some custom steps after the update is triggered
[15:33:28]
<Salamandar> yes that’s an idea
[15:35:35]
<Yunohost Git/Infra notifications> App jellyfin failed all tests in job [#23590](https://ci-apps.yunohost.org/ci/job/23590) :(
[15:40:39]
<Yunohost Git/Infra notifications> App phpboost failed all tests in job [#23599](https://ci-apps.yunohost.org/ci/job/23599) :(
[15:43:23]
<tituspijean> > <@yunohostinfra:matrix.org> App jellyfin failed all tests in job [#23590](https://ci-apps.yunohost.org/ci/job/23590) :(
wild guess, jellyfin released a new version and deleted the old one. AGAIN.
[15:46:06]
<Yunohost Git/Infra notifications> App weblate goes down from level 8 to 7 in job [#23598](https://ci-apps.yunohost.org/ci/job/23598)
[16:06:59]
<lapineige> Is there a way to trigger the autoupdate manually ? (I'm lazy and want it to do the work for me 😂 but I'm also hasty 😂)
[16:11:52]
<lapineige> Oh wait it just did the autoupdate 🤔
[16:12:00]
<kayou> it's me ;)
[16:15:29]
<Yunohost Git/Infra notifications> App streams failed all tests in job [#23602](https://ci-apps.yunohost.org/ci/job/23602) :(
[16:17:27]
<lapineige> https://yunohost.org/fr/packaging_apps_helpers#ynh-setup-source
If I want to patch `[resources.sources.main]`, I should use a `sources/patches/main-*.patch` file, is that right ?
[16:18:29]
<Salamandar> I think that’s right
[16:18:32]
<lapineige> Oh I see here the syntax : https://github.com/YunoHost-Apps/mopidy_ynh/blob/master/sources/patches/__main__.patch
[16:18:55]
<Salamandar> I would have preferred `patches/main/*.patches` tbh
[16:19:09]
<Salamandar> ’cause 0001-mycommit.patch
[16:20:57]
<lapineige> and here I replace `*` by whatever I want ?
[16:40:19]
<Salamandar> > <@Salamandar:matrix.org> I would have preferred `patches/main/*.patches` tbh
that’s not supported (yet), though, you need main-*.patch
[17:00:35]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot created new branch update_app_levels
[17:00:36]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot opened [pull request #2028](https://github.com/YunoHost/apps/pull/2028): Update app levels according to CI results
[17:00:36]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot pushed 1 commit to update_app_levels: Update app levels according to CI results ([4e9c99cc](https://github.com/YunoHost/apps/commit/4e9c99cc4d332e7259add781c70c27b785e22804))
[17:35:46]
<Yunohost Git/Infra notifications> App itflow failed all tests in job [#23607](https://ci-apps.yunohost.org/ci/job/23607) :(
[17:54:43]
<Tag> grr https://ci-apps-dev.yunohost.org/ci/job/13629 :(
[17:55:16]
<Tag> why does sury fail during the upgrade ?
[17:56:41]
<tituspijean> > <@titus:pijean.ovh> Mmmh, interesting read: https://github.com/oerdnj/deb.sury.org/issues/2074
^.
[17:56:41]
<Émy - OniriCorpe> Because computerz
[17:57:02]
<Tag> > <@titus:pijean.ovh> ^.
ahhh so it will fail for users too ?
[17:57:04]
<Aleks (he/him/il/lui)> somebody broke the internetz
[17:57:04]
<eric_G> https://forum.yunohost.org/t/something-went-wrong-while-updating-the-cache-of-apt-due-to-sury/15024
[17:57:10]
<Tag> mfpffff
[17:57:25]
<Tag> rmbllblblblblbl
[17:57:40]
<Aleks (he/him/il/lui)> i feel ya
[17:57:47]
<Émy - OniriCorpe> Cf xkcd “the entire world infrastructure relies on one random person”
[17:57:49]
<Tag> ye
[17:57:57]
<Tag> right
[17:58:13]
<Tag> soooo good noight
[18:02:04]
<Tag> ah ouais on pourrait faire une release avec `apt-key del 95BD4743; wget -nv -O - "https://packages.sury.org/php/apt.gpg" | apt-key add -` quelque part dans le code ? lul
[18:03:26]
<Émy - OniriCorpe> If broken; then unbroken plz; fi
[18:09:58]
<Yunohost Git/Infra notifications> App gitlab-runner failed all tests in job [#23611](https://ci-apps.yunohost.org/ci/job/23611) :(
[18:13:47]
<Tag> Aleks (he/him/il/lui): dans quoi je le mets ? debian/postinst ?
[18:14:26]
<Tag> (jveux bien faire la PR mais j'ai jamais release encore par contre)
[18:14:27]
<Aleks (he/him/il/lui)> idéalement c'est la regenconf apt qu'il faut fix
[18:14:32]
<Tag> ok
[18:14:47]
<Aleks (he/him/il/lui)> https://github.com/YunoHost/yunohost/blob/dev/hooks/conf_regen/10-apt#L67
[18:15:57]
<Aleks (he/him/il/lui)> genre zmlerp chépa, tester que la clef est expiré c'est super relou, on pourrait tester que la date de modif est supérieur/inférieure à aujourdhui idk
[18:16:12]
<Aleks (he/him/il/lui)> ah ou bien t'as l'ID de la clef
[18:16:23]
<Aleks (he/him/il/lui)> ou la checksum idk
[18:16:26]
<Tag> j'allais faire un truc comme ça :
```
# Check if the key 95BD4743 is present in apt-key
if apt-key list | grep -q "95BD4743"; then
echo "Updating sury apt key..."
apt-key del 95BD4743; wget -nv -O - "https://packages.sury.org/php/apt.gpg" | apt-key add -
fi
```
[18:16:48]
<Aleks (he/him/il/lui)> oui voila
[18:17:11]
<Aleks (he/him/il/lui)> par contre apt-key add c'est déprécié je crois
[18:17:29]
<Aleks (he/him/il/lui)> pas sur mais y'a une histoire comme ça jcrois
[18:17:58]
<Aleks (he/him/il/lui)> lié au fait que dans bookworm c'est maintenant obligé (je sais plus) de mettre [signed-by=/chemin/de/la/clef] dans les sources.list je crois
[18:18:04]
<Tag> mmh
[18:18:19]
<Tag> *j'ai déjà commit*
[18:18:37]
<Tag> *que faire?*
[18:18:52]
<Aleks (he/him/il/lui)> bah tkt on fait une PR on itère
[18:20:20]
<Tag> https://github.com/YunoHost/yunohost/pull/1777
[18:21:01]
<kayou> > <@tag:lostpod.me> (jveux bien faire la PR mais j'ai jamais release encore par contre)
Si jamais ça t'intéresse de savoir comment on fait, c'est expliqué sur une page de doc.yunohost.org
[18:21:32]
<Tag> Oui j'ai lu et ça dit qu'il faut être en forme du coup KO
[18:21:33]
<kayou> Sinon, moi je voyais plus ça comme une migration, mais c'est peut être beaucoup trop lourd pour ce que c'est
[18:22:12]
<Tag> boarf la condition dans la regenconf apt semble pas trop lourde
[18:23:12]
<kayou> > <@tag:lostpod.me> Oui j'ai lu et ça dit qu'il faut être en forme du coup KO
Je crois que c'est une personne avec un pseudo qui commence par `Ale` et qui finit pas `ks` (peut être quelqu'un d'autre ?) qui avait rajouté ça, parce que la personne a déjà fait des release YNH un peu sous l'effet de l'alcool
[18:23:22]
<kayou> Mais j'ai peut être mal compris XD
[18:23:37]
<Tag> cette règle fait sens
[18:23:55]
<kayou> > <@tag:lostpod.me> boarf la condition dans la regenconf apt semble pas trop lourde
Oui en effet ça suffit, c'est juste que c'est un truc à faire une seule fois ^^
[18:24:23]
<Tag> Ouep, en vrai peut être une migration 🤔
[18:24:56]
<Aleks (he/him/il/lui)> non mais les migrations c'est relou faut créer 15 strings i18n et tout 😅
[18:24:59]
<Aleks (he/him/il/lui)> je teste le truc
[18:25:05]
<Tag> okok
[18:26:48]
<kayou> Anéfé
[18:28:46]
<Tag> bon je vais noter dans ma endless to-do list : créer un tool pour créer des migrations
[18:29:27]
<Tag> flemme d'écrire du code, autant écrire du code qui écrit du code :')
[18:30:40]
<Tag> (avec chatgpt ayaaa) /s
[18:30:40]
<kayou> C'est vrai que créer une migration, c'est ultra relou
[18:30:59]
<Émy - OniriCorpe> > <@tag:lostpod.me> flemme d'écrire du code, autant écrire du code qui écrit du code :')
Ah oui le littéral taff de Bram
[18:31:35]
<Émy - OniriCorpe> Et techniquement c’est ce que font les langages compilés et interprétés mdr
[18:32:58]
<Tag> alors, je me corrige d'un niveau au moins : flemme d'écrire du code qui écrit du code, autant écrire du code qui écrit du code qui écrit du code !
[18:33:57]
<Tag> *et les electrons qui bougent blablalba*
[18:34:11]
<Aleks (he/him/il/lui)> bon zblerg il manquait un espace dans l'ID de la clef chez moi parce que l'output de apt-key list est complètement claquée et du coup je suis tombé sur 3 autres clefs expirée alors j'ai crafté un truc pour cleanup toutes les clefs expirées T_T
[18:34:32]
<Aleks (he/him/il/lui)> halfixingbulb.gif
[18:35:37]
<Aleks (he/him/il/lui)> ```
EXPIRED_KEYS="$(apt-key list | grep -A1 "expired:" | grep -v 'expired\|^-' | sed 's/\s//g')"
for KEY in $EXPIRED_KEYS; do apt-key del $KEY 2>/dev/null; done
```
[18:35:50]
<Aleks (he/him/il/lui)> m'enfin du coup je me questionne de si apt-key affiche le même format partout
[18:39:04]
<Émy - OniriCorpe> Black magic
[18:51:56]
<mathieuw> > <@Alekswag:matrix.org> m'enfin du coup je me questionne de si apt-key affiche le même format partout
Justice nulle part !
Ok, je ->[]
[19:38:24]
<Yunohost Git/Infra notifications> App sharkey rises from level 0 to 7 in job [#23614](https://ci-apps.yunohost.org/ci/job/23614) !
[19:42:02]
<Salamandar> > <@mathieuw:tetaneutral.net> Justice nulle part !
> Ok, je ->[]
apt c'est un peu l'enfer pour ça… zéro stabilité d'api
[19:55:08]
<Aleks (he/him/il/lui)> ouai je comprends pas, y'a pas de truc pour avoir une output en json, et ça te balance un warning en mode "olala apt-key est déprécié" mais y'a pas d'alternative obvious du tout pour gérer les clefs à part genre idk utiliser les commandes gpg qui sont pas du tout obvious non plus
[19:55:28]
<Aleks (he/him/il/lui)> et ça fait genre 5 ans que c'est déprécié alors genre "décidez-vous" quoi x_x
[20:02:51]
<Salamandar> :|
[20:02:52]
<Salamandar> ouais
[20:04:20]
<Salamandar> et la foultitute d'outils
[20:04:25]
<Salamandar> ```
salamandar:salamandar:# compgen -c | grep -e '^apt\|^dpkg' | sort
apt
apt-add-repository
apt-cache
apt-cdrom
apt-config
apt-copy
apt-extracttemplates
apt-file
apt-ftparchive
apt-get
apt-history
apticron
aptitude
aptitude-changelog-parser
aptitude-create-state-bundle
aptitude-curses
aptitude-run-state-bundle
apt-key
apt-listchanges
apt-list-packages
apt-mark
apt_pref_compdef
apt-sortpkgs
dpkg
dpkg-architecture
dpkg-buildflags
dpkg-buildpackage
dpkg-checkbuilddeps
dpkg-deb
dpkg-distaddfile
dpkg-divert
dpkg-fsys-usrunmess
dpkg-genbuildinfo
dpkg-genchanges
dpkg-gencontrol
dpkg-gensymbols
dpkg-maintscript-helper
dpkg-mergechangelogs
dpkg-name
dpkg-parsechangelog
dpkg-preconfigure
dpkg-query
dpkg-realpath
dpkg-reconfigure
dpkg-scanpackages
dpkg-scansources
dpkg-shlibdeps
dpkg-source
dpkg-split
dpkg-statoverride
dpkg-trigger
dpkg-vendor
```
[20:04:28]
<Salamandar> like wtf
[20:04:52]
<Salamandar> "dpkg-deb" ?????
[20:16:02]
<Salamandar> https://aria.im/_matrix/media/v1/download/matrix.org/UvdhKvHPauIJgUbQmwaygTfY
[20:16:56]
<Salamandar> https://aria.im/_matrix/media/v1/download/matrix.org/GatsUFuBognDyhkEsRIqJiBf
[20:17:58]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot pushed 1 commit to add-to-wishlist-ittools: Add ITTools to wishlist ([615622ff](https://github.com/YunoHost/apps/commit/615622ffff54d34e5969ec07f6292354bbe6d9f4))
[20:17:59]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot created new branch add-to-wishlist-ittools
[20:18:01]
<Yunohost Git/Infra notifications> [apps] @yunohost-bot opened [pull request #2029](https://github.com/YunoHost/apps/pull/2029): Add ITTools to wishlist
[20:18:55]
<Salamandar> @oniricorpe:im.emelyne.eu pour ton histoire d'artefacts de job gitlab-ci…
[20:19:10]
<Salamandar> https://git.pleroma.social/pleroma/pleroma/-/releases/v2.6.1
[20:19:22]
<Salamandar> ya même pas de lien quelconque vers le job CI…
[20:19:38]
<Salamandar> donc je vois même pas comment le lien pourrait être fait
[20:21:11]
<Salamandar> mais ça me fait réfléchir à un moyen de rendre l'autoupdater plus générique
[20:21:47]
<Salamandar> Genre, on se fixe une API, et dans le package, le packager implémente des méthodes python qui vont remplacer celles de bases
[20:21:50]
<Salamandar> un truc du style
[20:26:35]
<Bram> > <@oniricorpe:im.emelyne.eu> Ah oui le littéral taff de Bram
ewe
[20:27:10]
<Yunohost Git/Infra notifications> [apps] @OniriCorpe [commented](https://github.com/YunoHost/apps/pull/2029#issuecomment-1949285426) on [issue #2029](https://github.com/YunoHost/apps/pull/2029) Add ITTools to wishlist: Sounds nice
[20:27:31]
<Émy - OniriCorpe> > <@Bram_:matrix.org> ewe
Éoué
[20:29:21]
<Yunohost Git/Infra notifications> [apps] @OniriCorpe [commented](https://github.com/YunoHost/apps/pull/2028#issuecomment-1949288825) on [issue #2028](https://github.com/YunoHost/apps/pull/2028) Update app levels according to CI results: Oh no everything is broken, sury edition ™️
[20:45:19]
<Salamandar> > <@Salamandar:matrix.org> Genre, on se fixe une API, et dans le package, le packager implémente des méthodes python qui vont remplacer celles de bases
let me think again: any github repository could inject code into the autoupdater, that runs on yunohost infra
[20:45:24]
<Salamandar> *ahem* nice idea *ahem*
[20:55:35]
<Aleks (he/him/il/lui)> that's why i was looking at systemd-nspawn which is basically a chroot on steroid
[20:55:41]
<Aleks (he/him/il/lui)> and easy to use
[20:56:44]
<Aleks (he/him/il/lui)> like idk, having arbitrary scripts spitting out json that would be caught by the autoupdater
[20:58:12]
<kayou> I don't understand, but at this point, I'm too afraid to ask...
Why don't we keep the good old updater.yaml & updater.sh? :)
[20:58:56]
<Émy - OniriCorpe> > <@kayou:matrix.org> I don't understand, but at this point, I'm too afraid to ask...
> Why don't we keep the good old updater.yaml & updater.sh? :)
Because nobody uses it outside of the packager core team lmao
[20:59:04]
<Salamandar> > <@Alekswag:matrix.org> that's why i was looking at systemd-nspawn which is basically a chroot on steroid
yeah, container <3
[20:59:41]
<Salamandar> > <@kayou:matrix.org> I don't understand, but at this point, I'm too afraid to ask...
> Why don't we keep the good old updater.yaml & updater.sh? :)
well tbh i liked the whole "each app has its own ci job"
[21:00:05]
<Salamandar> but the way the code was done (100% in the app side) was unmaintainable. To have code on the "infra" side was required
[21:01:36]
<Aleks (he/him/il/lui)> idk how many app do need custom logic though
[21:01:58]
<Aleks (he/him/il/lui)> though somebody was mentioning earlier the need to also update conf file according to a template fetch from upstream etc
[21:02:12]
<Salamandar> > <@Alekswag:matrix.org> idk how many app do need custom logic though
not many
[21:04:48]
<Salamandar> afaik:
* getting the latest version's configuration file -> they should just do that in the install/upgrade script
* getting an obscure CI job's artifact (so get release -> get its commit -> get its CI pipeline -> get the relevant CI job -> get the artifacts -> check if artifacts match)
* crappy version/tag names
* getting a docker image's content -> yeah maybe we could consider docker hub as a forge ? (from the autoupdater's point of view)
[21:05:40]
<Salamandar> i'm trying to handle the crappy version/tag names today
[21:06:04]
<Émy - OniriCorpe> The new system provides a centralized display of all waiting opened PRs, so that we don't forget them, and we can go and repair them if necessary, that’s awesome
[21:07:27]
<Émy - OniriCorpe> In the past few days we updated many year-outdated apps, that’s really nice
[21:11:52]
<Salamandar> woops cinny is drunk
[21:11:58]
<Salamandar> https://aria.im/_matrix/media/v1/download/matrix.org/NwYxdiYLqyhgLJQdnuSncsdd
[21:12:05]
<Salamandar> i see your both thumbsup separately xD
[21:12:40]
<Émy - OniriCorpe> It’s because 2 anarchists are a scission
[21:13:05]
<Salamandar> o/
[21:13:33]
<Émy - OniriCorpe> https://aria.im/_matrix/media/v1/download/im.emelyne.eu/uLfaIIBgYnujqMjWldOzWtAm
[21:13:49]
<Émy - OniriCorpe> Is there a way to activate this automatically for all repos in the yunohost-app org?
[21:14:25]
<Émy - OniriCorpe> Maybe there is a way to check if the PR comes from `testing`?
[21:14:56]
<Émy - OniriCorpe> To avoid maintainers doing PR directly in `master` like i saw yesterday x)
[21:15:14]
<Salamandar> ye ye i know
[21:15:16]
<Émy - OniriCorpe> To avoid maintainers doing PR directly in master (and not `testing`) like i saw yesterday x)
[21:15:17]
<Salamandar> i have a script for that somewhere
[21:15:35]
<Salamandar> https://github.com/Salamandar/automatic-repositories-rules
[21:15:42]
<Émy - OniriCorpe> > <@Salamandar:matrix.org> woops cinny is drunk
Same on element ios btw
[21:15:49]
<Salamandar> but I wasn't sure about *which* rules we want
[21:16:04]
<Salamandar> i discussed a bit with @Alekswag:matrix.org about that
[21:16:20]
<Salamandar> also i discovered that `github` package for python already handles most of the boilerplate i manually coded…
[21:17:02]
<Émy - OniriCorpe> > <@Salamandar:matrix.org> but I wasn't sure about *which* rules we want
We can at least put basic ones like “plz PR and not commit directly to master”
[21:17:26]
<Salamandar> i like to commitpushforcetomaster
[21:17:32]
<Salamandar> 😬
[21:17:33]
<Salamandar> also
[21:17:42]
<Salamandar> https://www.youtube.com/watch?v=aolI_Rz0ZqY
[21:17:47]
<Salamandar> > git push --force-with-lease
[21:17:58]
<Émy - OniriCorpe> Omg 😨🤓
[21:19:48]
<Émy - OniriCorpe> Our packages need a testing branch among other standardizations for the project to work nicely and people don’t respect that so…
[21:22:03]
<Salamandar> xD
[21:22:11]
<Salamandar> i push to master only when i'm sure it's okay
[21:22:26]
<Salamandar> like a simple autoupdate.strategy tag -> release
[21:22:32]
<Salamandar> but for the rest…
[21:22:53]
<Salamandar> also when things are in testing, no one ever actually test them except for the big apps (hello nextcloud and jellyfin)
[21:23:36]
<Salamandar> we could have a `!send_to_forum` to send an automated message to the forum about the testing PR ?
[21:24:27]
<Émy - OniriCorpe> https://github.com/YunoHost-Apps/cheky_ynh/pull/38/commits
[21:24:45]
<Émy - OniriCorpe> I have the culprits name
[21:26:07]
<Yunohost Git/Infra notifications> [apps] @Salamandar opened [pull request #2030](https://github.com/YunoHost/apps/pull/2030): Update the autoupdater
[21:26:26]
<Émy - OniriCorpe> https://github.com/YunoHost-Apps/pleroma_ynh/pull/263
[21:26:26]
<Salamandar> > <@oniricorpe:im.emelyne.eu> https://github.com/YunoHost-Apps/cheky_ynh/pull/38/commits
i see my name. did i do something wrong here ? ><'
[21:26:47]
<Émy - OniriCorpe> https://github.com/YunoHost-Apps/framagames_ynh/pull/24
[21:27:12]
<Salamandar> hmmm @oniricorpe:im.emelyne.eu
[21:27:21]
<Yunohost Git/Infra notifications> App weblate goes down from level 8 to 7 in job [#23615](https://ci-apps.yunohost.org/ci/job/23615)
[21:27:22]
<Émy - OniriCorpe> :”3
[21:27:23]
<Salamandar> you know you can do it from cmdline without a PR (rebasing/updating testing)
[21:27:36]
<Émy - OniriCorpe> Yes i know
[21:27:48]
<Émy - OniriCorpe> But using an ipad is… something
[21:28:14]
<Salamandar> ah
[21:28:15]
<Salamandar> …
[21:28:16]
<Salamandar> :(
[21:28:27]
<Émy - OniriCorpe> I’m not at home for now x)
[21:39:56]
<Salamandar> :D
[21:44:33]
<Yunohost Git/Infra notifications> App yourls failed all tests in job [#23617](https://ci-apps.yunohost.org/ci/job/23617) :(
[21:48:36]
<Yunohost Git/Infra notifications> App homarr failed all tests in job [#23618](https://ci-apps.yunohost.org/ci/job/23618) :(
[22:30:56]
<Yunohost Git/Infra notifications> App galette failed all tests in job [#23621](https://ci-apps.yunohost.org/ci/job/23621) :(
[22:33:50]
<Yunohost Git/Infra notifications> App my_capsule failed all tests in job [#23623](https://ci-apps.yunohost.org/ci/job/23623) :(
[22:38:20]
<Yunohost Git/Infra notifications> App owntracks failed all tests in job [#23624](https://ci-apps.yunohost.org/ci/job/23624) :(
[22:43:42]
<Yunohost Git/Infra notifications> App zwiicms failed all tests in job [#23625](https://ci-apps.yunohost.org/ci/job/23625) :(
[23:39:54]
<Yunohost Git/Infra notifications> [apps] @alexAubin pushed 10 commits to master ([0d2d5e5c33fd...01020522dec5](https://github.com/YunoHost/apps/compare/0d2d5e5c33fd...01020522dec5))
[23:39:58]
<Yunohost Git/Infra notifications> [apps/master] Gitlab replaces / with - in tag tarballs - Félix Piédallu