| 23 Jan 2022 |
 toonn | Why do you suspect that one? File contents isn't a problem. The normalization only happens on file names. | 16:53:42 |
 Vladimír Čunát | What I posted is some non-ASCII filename. (link) | 16:54:52 |
| Vladimír Čunát | * What I posted is some non-ASCII filename. (link) | 16:55:14 |
| toonn | Oh, I didn't see the accent. Yeah if it's a combining character it's a probably culprit. | 20:28:22 |
| 24 Jan 2022 |
 Winter (she/her) | In reply to @vcunat:matrix.org
Perhaps it's file cmd/panic/internal/utf8/ùtf8.go in this case, but I wonder if we could improve this generally somehow. Why is that even named that way; what package is it from? | 02:58:26 |
| Winter (she/her) | * In reply to @vcunat:matrix.org
Perhaps it's file cmd/panic/internal/utf8/ùtf8.go in this case, but I wonder if we could improve this generally somehow. Why is that even named that way; what package is it from? | 02:58:31 |
| Vladimír Čunát | Well yes, we could choose the way of trying to convince every upstream that it's a bad idea to use such filenames, but I didn't feel like it, especially when I didn't have any kind of connections to those projects. | 06:55:45 |
| Winter (she/her) | In reply to @vcunat:matrix.org
Well yes, we could choose the way of trying to convince every upstream that it's a bad idea to use such filenames, but I didn't feel like it, especially when I didn't have any kind of connections to those projects. I didn’t mean that, I was just curious as to why they would be using such a filename, considering it’s seemingly a source code file. | 13:54:08 |
| Vladimír Čunát | I'm not sure, but it seemed intentional: https://github.com/maruel/panicparse/commit/53e9d3ee100e0cf54d18eeae09dda6e09a29d090#diff-8173d0da9e11c70bd796508e9b88bbfd3b52d7131b42d7925f0e8e41103f192aR1825 | 13:56:10 |
| toonn | Can the error here be spurious? I didn't expect the changes to affect webkitgtk. https://hydra.nixos.org/build/164731843/nixlog/1 | 17:29:26 |
 jonringer | both clang and gcc like to segfault on my build server when given many threads | 17:33:25 |
| jonringer | To anwser your question directly. It's not common, but it can | 17:34:18 |
| Vladimír Čunát | I would tip this one as OOM killer. | 18:21:52 |
| toonn | Ok, thanks. Could I ask Ofborg to try building webkitgtk? | 18:23:32 |
| Vladimír Čunát | Oh, I tried to restart it but it's again the case that can't be. (instantly aborted by Hydra) | 18:23:40 |
| Vladimír Čunát | Let me build it locally. | 18:24:17 |
| Vladimír Čunát | Have you force-pushed the commit away? | 18:26:24 |
| toonn | Could be but that would've been a week ago. | 18:27:19 |
| Vladimír Čunát | Anyway, better build the current commit, I suppose (dc56a7922). | 18:28:34 |
| toonn | Hmm, it doesn't look like webkitgtk_4_1 depends on LLVM in any way. libsoup_3 succeeds and webkitgtk (without _4_1) succeeds so I am not expecting a failure. | 18:28:59 |
| Vladimír Čunát | It succeeded for me. | 19:13:55 |
| toonn | Thank you! I think that means the LLVM bump for Linux is ready for review then : ) | 20:08:58 |
| 25 Jan 2022 |
 hexa | can we get the nixpkgs:python-unstable jobset renamed to python-updates and matched to the new branch name (python-updates) and made visible again? | 11:51:04 |
| hexa | https://hydra.nixos.org/jobset/nixpkgs/python-unstable | 11:51:22 |
| Vladimír Čunát | In reply to @hexa:lossy.network
can we get the nixpkgs:python-unstable jobset renamed to python-updates and matched to the new branch name (python-updates) and made visible again? done | 11:55:02 |
| Vladimír Čunát | Hopefully without mistakes. I also checked the "Visible" checkbox, though I'm not sure what exactly it does. | 11:55:43 |
| hexa | makes it appear here (https://hydra.nixos.org/project/nixpkgs) again | 11:56:01 |
| Vladimír Čunát | I assume that's desirable. | 11:56:59 |
| hexa | yep, otherwise it can only be visited by deep-linking | 11:58:28 |
 FRidh | Right | 13:31:56 |
