| 12 Nov 2024 |
 sterni (he/him) | you can open a PR it's probably just a paiin to build things on that branch | 19:36:23 |
| sterni (he/him) | you can ask John probably iirc he did rts configure | 19:36:45 |
 alexfmpe | In reply to @sternenseemann:systemli.org
you can open a PR it's probably just a paiin to build things on that branch Yeah took a couple hours on spare machine | 21:23:44 |
| 13 Nov 2024 |
 maralorn | sterni: What do you think about closing haskell-updates and just do our work on staging? | 08:57:26 |
| maralorn | By the way. Is there any reason rn that we shouldn’t merge PRs into haskell-updates? Or can we just do it, only problem being that it will take a while to reach master? | 08:59:08 |
| sterni (he/him) | we can absolutely merge stuff it’s just annoying to build stuff on the branch | 09:36:02 |
| sterni (he/him) | if we keep our branch we at least don’t get our cache (locally) nuked by changes on staging, otherwise yeah | 09:36:44 |
| sterni (he/him) | as long as it doesn’t regress eval of staging | 09:37:00 |
| maralorn | In reply to @sternenseemann:systemli.org
if we keep our branch we at least don’t get our cache (locally) nuked by changes on staging, otherwise yeah Yeah. I am wondering if we can somehow couple that in a sensible way. | 09:53:05 |
| maralorn | Basically we would want our branch to be separate from staging but be merged into it right before moving to staging-next? | 09:54:21 |
| sterni (he/him) | ah you mean in the future? | 09:54:43 |
| maralorn | Ah, yeah. | 09:54:56 |
| maralorn | Sorry | 09:54:59 |
| sterni (he/him) | I want to prepare a PR soon with updated workflows and HACKING.md soon, but basically I'd say the best solution would be: Target staging, base the branch on master (or staging-next, but that's a little more disruptive, I guess), adjust the workflow to merge in the merge-base of the branch we are based on (master or staging-next) and the branch we are targeting (staging), so we would benefit from populated cache, but also would never be ahead of our target branch in any way. Then we'd merge when we are done, to reduce time to deliver this should be timed with staging-next, yes | 09:57:38 |
| sterni (he/him) | I think staging-next and our speed lines up relatively well, so I don't think the frequency of updates would go down, but obviously there's going to be an extra delay between merging a change to haskell-updates and it being delivered to channels | 10:01:15 |
| sterni (he/him) | We'll need to be more liberal with merging low rebuild changes to master, and maybe even high rebuild, urgent fixes to staging. | 10:01:54 |
| maralorn | Yeah, that’s all what I was also thinking. | 10:02:40 |
| sterni (he/him) | But I don't think the delivery time is a problem, really, it'll just solidify our identity as a well-tested, „stable“ package set since we already have trouble delivering the latest versions due to Stackage LTS holding packages back. | 10:02:44 |
| sterni (he/him) | I think cabalfreeze2nix is a good opportunity to integrate a sort of alternative solution that's easy to use into nixpkgs itself | 10:03:50 |
| maralorn | I think at this point in time it won’t actually affect our latency much. Maybe the best-case latency a bit, but the effect on average latency will be negligable. | 10:04:09 |
| maralorn | I just realized that sending smaller fixes to master would actually be benificial to newcomers because they don’t get weird noise from out of date hackage-packages.nix in their PRs. | 10:24:34 |
| maralorn | Redacted or Malformed Event | 11:31:35 |
| 14 Nov 2024 |
 shapr | How do I sign up as official NixPkgs maintainer for my own library on hackage? | 20:54:52 |
| alexfmpe | https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/haskell-modules/configuration-hackage2nix/main.yaml#L113 | 21:01:10 |
| alexfmpe | Might also need to add yourself as nixpkgs maintainer, not sure | 21:01:32 |
| alexfmpe | As in, https://github.com/NixOS/nixpkgs/blob/master/maintainers/maintainer-list.nix | 21:01:59 |
| maralorn | In reply to @shapr:cofree.coffee
How do I sign up as official NixPkgs maintainer for my own library on hackage? Love it. More maintainers, more better! | 21:15:32 |
| shapr | In reply to @alexfmpe:matrix.org
As in, https://github.com/NixOS/nixpkgs/blob/master/maintainers/maintainer-list.nix Thank you! | 21:23:44 |
| 15 Nov 2024 |
 hellwolf | Is haskellPackages.hlint_3_8 broken? | 12:47:01 |
| hellwolf | (I am probably 2w behind nixpkgs-unstable) | 12:47:39 |
