| 26 Oct 2024 |
 Tristan Ross | 肉? w | 03:34:40 |
 infinisil | @rosscomputerguy:matrix.org Spell it out :P
| 03:44:45 |
| infinisil | 肉す | 03:45:19 |
| Tristan Ross | Oh, I realize now | 03:46:04 |
| Tristan Ross | Doesn't quite work as I was expecting. | 03:46:27 |
 emily | are you going to put CJK characters in the 25.05 codename to break .NET even more? | 15:03:24 |
| emily | btw, https://github.com/NixOS/nixpkgs/pull/346059 seems a bit stalled out | 15:07:53 |
| Tristan Ross | In reply to @emilazy:matrix.org
are you going to put CJK characters in the 25.05 codename to break .NET even more? Is there something that would work with W that is CJK? | 17:39:00 |
 leona | I'm currently reading https://github.com/NixOS/nixpkgs/pull/349057, which unfortunately contains a breaking change by removing a version of graylog. But this version is EOL since >5 months. I think it makes sense to remove it, but am not sure | 18:32:45 |
| emily | AIUI security trumps the freeze | 18:38:02 |
| emily | does it have a history of security updates / CVEs? | 18:38:07 |
| emily | https://endoflife.date/graylog "Graylog’s release, support and EOL policy is not clearly defined." oh good | 18:39:00 |
| leona | it's SSPL and I don't know it very much, just stumbled over the PR. AFAIK it had quite some CVEs | 18:40:02 |
| leona | not any in newer versions that aren't fixed in 5.1 tho yet | 18:41:07 |
| emily | I think it's sensible to not ship EOL stuff for the new release. the change of default is a little painful: I would be tempted to throw on old stateVersions explaining the situation rather than potentially silently breaking things. | 18:41:46 |
| emily | that's how I handled it in https://github.com/NixOS/nixpkgs/pull/345625 | 18:42:03 |
| leona | same | 18:42:07 |
| emily | (to be clear though, I have no authority over the release process, just my personal take.) | 18:42:41 |
| emily | (IMO "removing a dead package that we anticipate will cause problems over the course of the release cycle" is one of the more sensible late breaking changes) | 18:43:14 |
| Tristan Ross | In reply to @emilazy:matrix.org
I think it's sensible to not ship EOL stuff for the new release. the change of default is a little painful: I would be tempted to throw on old stateVersions explaining the situation rather than potentially silently breaking things. Agreed | 18:43:59 |
| Tristan Ross | In reply to @emilazy:matrix.org
(IMO "removing a dead package that we anticipate will cause problems over the course of the release cycle" is one of the more sensible late breaking changes) Yeah, it's good to get those in | 18:45:15 |
| Tristan Ross | The freeze is mainly to prevent any packages from breaking that could hurt the release but a security issue is imo more harmful | 18:45:49 |
| 27 Oct 2024 |
|  peterwang joined the room. | 04:32:09 |
|  Scrumplex joined the room. | 19:27:49 |
| Scrumplex | Hi there, short question: Is renaming a package a breaking change assuming an alias is added? | 19:28:30 |
 Vladimír Čunát | Hmm. If your configuration disabled aliases (that's those in aliases.nix), it would be breaking, right? | 19:30:08 |
| Vladimír Čunát | Though easy to fix. | 19:30:36 |
| Scrumplex | But I feel like disabling aliases is like subscribing to breakage | 19:30:51 |
| Scrumplex | In my case I am just worried about causing breakage within Nixpkgs. I was pinged in two separate PRs that involve renaming a package and I wasn't sure if I can merge them during the breaking change freeze | 19:31:46 |
| Vladimír Čunát | My gut feeling is that it's still fine at this point. Maybe until the release. | 19:31:53 |
