| 12 May 2023 |
 Growpotkin | Are y'all still doing the meeting? | 15:46:44 |
| Growpotkin | I was in a meeting for work. | 15:46:51 |
 infinisil | We were at the time of your message | 16:36:56 |
| infinisil | You did say you were available at this time though 😅 | 16:37:40 |
| infinisil | If you can't make a meeting I really recommend letting people know in advance! | 16:38:55 |
| infinisil | Robert Hensing (roberth): What do you think of using a pinned version of the lazy trees builtins.filterPath (that's what it's called) in CI to ensure the reference requirement of RFC 140? | 17:38:02 |
| infinisil | It's a bit dangerous considering that lazy trees is very unstable and unreviewable, but considering it's only for a limited part of CI, and pinned, it might be okay, it would simplify the implementation a lot | 17:39:35 |
| infinisil | Oof, actually the error message would be terrible that way.. Since lazy paths have a separate filesystem root, trying to import e.g. ../../../../default.nix (the default.nix in the root) from pkgs/unit/he/hello leads to
error: getting status of '/home/tweagysil/na/nixpkgs/pkgs/unit/he/hello/README.md': No such file or directory
| 17:55:14 |
| infinisil | Oh and in fact it's even worse, because if you try to import ../../../../package.nix, it actually works, but it imports package.nix in the local directory, so ./package.nix! That's really confusing, maybe I should comment this in the lazy trees PR | 17:56:40 |
| infinisil | The cause of these problems is that the parent of a filesystem root is the filesystem root itself, so /.. == /. | 17:57:38 |
| infinisil | I reported this here: https://github.com/NixOS/nix/pull/6530#issuecomment-1546110770 | 18:14:32 |
|  @federicodschonborn:matrix.org changed their profile picture. | 19:34:21 |
| @federicodschonborn:matrix.org changed their display name from Federico Schonborn to Federico Damián Schonborn. | 19:35:02 |
| 15 May 2023 |
| Growpotkin | I had no idea that the date/time had been set | 14:25:49 |
| infinisil | I sent out this message and email invites more than a week in advance 😅 | 14:26:57 |
| Growpotkin | Damn, for whatever reason my calendar missed it. Sorry y'all | 14:33:34 |
| infinisil | Can you check if it's in your spam folder? I just got a report that DMARC is apparently messed up for tweag.io.. | 15:09:56 |
| infinisil | * Can you check if it's in your spam folder? I just got a report that DMARC for google calendar invites is apparently messed up for tweag.io.. | 15:10:21 |
 Wanja Hentze | oh yeah, that sort of thing has bitten me before | 15:45:56 |
| Growpotkin | I'll check | 15:47:24 |
| Wanja Hentze | I feel like ideally, .. on the root should be made to fail (not in unix in general, just in lazy trees) | 15:49:18 |
| Growpotkin | yeah it hit my spam folder | 17:02:39 |
| infinisil | Ah darn, sorry for that :( | 17:03:11 |
| infinisil | I reported it, should get fixed at some point | 17:03:44 |
| 16 May 2023 |
| Growpotkin | Redacted or Malformed Event | 12:48:29 |
| Growpotkin | Redacted or Malformed Event | 12:48:30 |
 Robert Hensing (roberth) | In reply to @infinisil:matrix.org
It's a bit dangerous considering that lazy trees is very unstable and unreviewable, but considering it's only for a limited part of CI, and pinned, it might be okay, it would simplify the implementation a lot Keep calm and carry on | 13:26:07 |
| Robert Hensing (roberth) | Lazy trees isn't going anywhere until it works properly. I'll fight for that on the team | 13:27:31 |
| Robert Hensing (roberth) | not that I expect a fight actually | 13:27:49 |
| Growpotkin | infinisil: who among us doesn't love a fresh rewrite? ;) | 13:27:52 |
