| 23 Feb 2023 |
 Winter (she/her) | hmm, i'll give it a try | 17:45:18 |
 Artturin | i tried something here https://github.com/Artturin/ofborg/commits/oneoff (the 2 c commits) | 17:45:32 |
| Artturin | i dont remember what since its been a while | 17:45:45 |
| Winter (she/her) | looks like you just did the API changes | 17:45:55 |
| Winter (she/her) | neat | 17:46:19 |
| Winter (she/her) | will pull down your branch and take a look, thank ykun | 17:46:33 |
| Winter (she/her) | * will pull down your branch and take a look, thank you | 17:46:41 |
| 27 Feb 2023 |
 dotlambda | Can we make ofborg fail if aliases are used? | 16:40:49 |
| dotlambda | Nixpkgs-review keeps failing on me when someone introduces an alias but forgets to update the references treewide. | 16:41:30 |
| dotlambda | e.g. https://github.com/NixOS/nixpkgs/pull/218641 | 16:41:46 |
 cole-h | Well, there is a ofborg-eval-package-list-no-aliases check, but that indeed doesn't catch it. I'll take a look in a few. | 16:47:40 |
| 24 Mar 2023 |
 vcunat | There's another case: https://github.com/NixOS/nixpkgs/pull/222710#issuecomment-1482486968 | 09:18:44 |
| vcunat | The command
nix-env -qa --json --file . --arg config '{ allowAliases = false; }'
doesn't seem suitable for this check.
| 09:23:15 |
|  @raboof:matrix.org joined the room. | 09:39:31 |
 Sandro | In reply to @robert:funklause.de
Nixpkgs-review keeps failing on me when someone introduces an alias but forgets to update the references treewide. It only does that because there shouldn't be aliases allowed 🤡 | 09:39:48 |
 hexa | in https://github.com/NixOS/nixpkgs/pull/219493 tests on aarch64-linux are shown as successful, but following the link the log is 404 | 12:39:38 |
| hexa | which is funny, since apparently the build doesn't even succeed | 12:41:09 |
| hexa | the tests for aarch64-linux were handleTestOn [ "x86_64-linux" ] only | 13:26:28 |
| hexa | 🥁 | 13:26:31 |
| cole-h | In reply to @hexa:lossy.network
which is funny, since apparently the build doesn't even succeed Sigh I should really fix that. Iirc it happens when the logs are completely empty... | 13:45:08 |
| hexa | nix also returned early | 13:47:42 |
| hexa | Thats why there is no log | 13:47:54 |
| cole-h | In reply to @vcunat:matrix.org
The command
nix-env -qa --json --file . --arg config '{ allowAliases = false; }'
doesn't seem suitable for this check.
Do you happen to know a better invocation for this purpose? I should also fix this 😅 | 13:50:42 |
| vcunat | In reply to @cole-h:matrix.org
Do you happen to know a better invocation for this purpose? I should also fix this 😅 Not off the top of my head. The tarball job also does the check in the same wrong way. I heard nixpkgs-review does blow up, so perhaps it could be a good inspiration. | 13:52:40 |
| cole-h | I did happen to conveniently miss sandro's message 😅 OK, will take a look and see what I can do. | 13:53:48 |
| vcunat | Oh right, I forgot about the source of the information :-) | 13:55:25 |
| Sandro | ? | 13:55:44 |
| cole-h | (Your message, that replied to dotlambda saying that nixpkgs-review caught the issue) | 13:56:07 |
| vcunat | I forgot that "I heard" it just a bit above in this thread. | 13:56:24 |
| cole-h | so, funny story about the aliases thing | 22:27:29 |
