| 23 Sep 2023 |
 evils | nvm, missed i have to explicitly declare "commit" as a supported feature | 22:07:32 |
| evils | In reply to @evils:nixos.dev
euhm, does maintainers/scripts/update.nix not use uncommitted changes to an update.sh? seems like it, quite easy to get stuck wondering why local changes have no effect xD | 22:40:33 |
| evils | also odd, if i delete my versions.nix the updating crashes, while manual invocation of update.sh will just regenerate the file | 22:42:25 |
| 24 Sep 2023 |
| evils | an actual question:
update.nix seems to only attempt to update kicad and not kicad-unstable when asked to update by maintainer
is there a specific reason for that
and will r-ryantm attempt to update kicad-unstable? | 04:09:33 |
|  mib 🥐 joined the room. | 12:23:11 |
 Artturin | In reply to @evils:nixos.dev
an actual question:
update.nix seems to only attempt to update kicad and not kicad-unstable when asked to update by maintainer
is there a specific reason for that
and will r-ryantm attempt to update kicad-unstable? $ nix-shell maintainers/scripts/update.nix --argstr package kicad-unstable
this derivation will be built:
/nix/store/sqhsd1a7q02wmsscynhfij3nsh14q3rb-packages.json.drv
building '/nix/store/sqhsd1a7q02wmsscynhfij3nsh14q3rb-packages.json.drv'...
Going to be running update for following packages:
- kicad-unstable-e0d4cf2d5b
Press Enter key to continue...
Running update for:
- kicad-unstable-e0d4cf2d5b: UPDATING ...
seems ok, did not let it finish
| 16:18:49 |
| Artturin | https://r.ryantm.com/log/kicad-unstable/2023-07-23.log | 16:18:54 |
| Artturin |
Do not update GNOME during a release cycle
| 16:19:02 |
| Artturin | which is the issue we fixed previously | 16:19:07 |
| Artturin | so next update should work | 16:19:19 |
| evils | yea, it already attempted an update of stable
but that's wrong due to my update.sh
https://github.com/NixOS/nixpkgs/pull/256786
i've fixed that, but it's built on top of my kicad-testing branch | 16:19:47 |
| evils | In reply to @artturin:matrix.org
$ nix-shell maintainers/scripts/update.nix --argstr package kicad-unstable
this derivation will be built:
/nix/store/sqhsd1a7q02wmsscynhfij3nsh14q3rb-packages.json.drv
building '/nix/store/sqhsd1a7q02wmsscynhfij3nsh14q3rb-packages.json.drv'...
Going to be running update for following packages:
- kicad-unstable-e0d4cf2d5b
Press Enter key to continue...
Running update for:
- kicad-unstable-e0d4cf2d5b: UPDATING ...
seems ok, did not let it finish
well, there you're telling it to update that
doing --argstr maintainer evils only attempts the stable package
| 16:20:18 |
| Artturin | Redacted or Malformed Event | 16:31:39 |
| Artturin | * Try using callPackage instead of override in all-packages.nix | 16:31:45 |
| Artturin | https://github.com/NixOS/nixpkgs/blob/d84cc41f8babd418c295fcbfbd41a1fd4e2adaec/maintainers/scripts/update.nix#L56-L66 | 16:34:01 |
| Artturin | Maybe it's not being counted as a somewhatUniqueRepresant | 16:34:12 |
| Artturin | both the position and updateScript are the same | 16:34:55 |
| 27 Sep 2023 |
| mib 🥐 changed their display name from mib to mib 🥐. | 05:53:09 |
| 3 Oct 2023 |
|  pbsds changed their profile picture. | 21:05:06 |
| 4 Oct 2023 |
| pbsds changed their profile picture. | 22:20:52 |
| 9 Oct 2023 |
|  linj joined the room. | 19:21:42 |
| 22 Oct 2023 |
|  Janik (they/them) joined the room. | 21:04:27 |
| 25 Oct 2023 |
|  Federico Damián Schonborn changed their profile picture. | 00:12:44 |
| 27 Oct 2023 |
| Federico Damián Schonborn changed their profile picture. | 01:24:21 |
| 4 Nov 2023 |
|  shawn8901 joined the room. | 16:09:12 |
| shawn8901 | Hi, i am currently trying do understand why r-ryantm does not create update PRs for portfolio. So the bot does actually get triggered when upstream does create a new release, but then it does not really find an update. (my observation)
When running the updateScript (that i attached the package quite a while ago), it currently does find an update, so my assumption is, that it kind of is working.
The latest logs of r-ryantm are here for the package: https://r.ryantm.com/log/portfolio/2023-11-03.log
There are multiple things confusing me here:
There was no releas of portfolio on the 3rd, but today on the 4th. Repology did update that on 14:45 actual releas was at 9:42 (my timezone https://repology.org/project/portfolio-performance/history & https://github.com/portfolio-performance/portfolio/releases ) As far as i understood nixpkgs-update does react on reading the history. am i right on that?
Beside thath, that on the very first line the version is wrong, that was working according the logs as some point in the past. (portfolio 0 -> 1 vs the actual numbers in the past portfolio 0.64.1 -> 0.64.4)
What i see, when it got the version right it has the github link appended, so what it then triggered by a GH read?
Some time in the past it seemed to have worked (e.G. https://r.ryantm.com/log/portfolio/2023-06-06.log)
Does someone have an idea whats going wrong here or it there actually nothing wrong and i am just to impatient? :( | 16:26:00 |
 ryantm | 0 -> 1 is the fake version numbers it uses when it does an update based on the package having an updateScript | 16:34:00 |
| shawn8901 | In reply to @ryantm:matrix.org
0 -> 1 is the fake version numbers it uses when it does an update based on the package having an updateScript ah! okay that make sense so far then for me :) So it was just running with no further reason the update script? | 16:37:57 |
| shawn8901 | *reason = trigger | 16:38:17 |
| ryantm | Right! Pretty sure the logs will have other info if the trigger was something else | 16:40:48 |
