Nix Documentation | 404 Members | |
| Discussion about documentation improvements around the Nix ecosystem | 78 Servers |
| Sender | Message | Time |
|---|---|---|
| 11 Apr 2024 | ||
| 22:12:49 | ||
In reply to @fricklerhandwerk:matrix.orgI don't really see a point where splitting would make sense. - the Wayland part is debatable if it is necessary ... but i think it is good because it points at qemu pitfalls. | 22:17:00 | |
| 12 Apr 2024 | ||
In reply to@infinisil:matrix.orgI did get an update. I was sifting through and deleting some emails before getting here so I can't say for sure when, but maybe it was this Monday. Let me know if I can do anything to sort that out! | 17:28:08 | |
| hey all, i recently got it into my head to write a little tool which automatically edits nix files without messing them up, and got it to a point where it can remove e.g. all mentions of i have not opened a pull request with it because that'd just ping a lot of people and i'm not sure if it would be accepted or people here would even be interested, so I thought I'd ask here beforehand | 18:54:15 | |
| * hey all, i recently got it into my head to write a little tool which automatically edits nix files without messing them up, and got it to a point where it can remove e.g. all mentions of i have not opened a pull request with it because that'd just ping a lot of people and i'm not sure if it would be accepted or people here would even be interested, so I thought I'd ask here beforehand | 18:57:21 | |
| (the manual built afterwards & is identical to the manual before the change; i haven't done documentation work on nixpkgs before and am a little unsure how to do further testing/qa on this) | 19:18:45 | |
| 19:21:51 | ||
| Nice! I think we should not just rip out the infinisil, guardian of architectural sanity, please give your blessing. | 20:12:40 | |
| fair ennough (should it be an outright error or just a trace for now?). if nothing else comes up i'd open a PR with this built on then-current master sometime tomorrow | 21:09:05 | |
| * fair enough (should it be an outright error or just a trace for now?). if nothing else comes up i'd open a PR with this built on then-current master sometime tomorrow | 21:09:17 | |
| 13 Apr 2024 | ||
| Hey all I could use some help with understanding how NRD works and how to document library files properly. For reference, I am working on this PR that adds network functions to the nixpkgs library. The manual is failing to build and when I try to build it locally I see this:
which i've determine to be caused by this line in the
This makes sense that it can't find it because if I remove the line, it does build, but there is no documentation for the functions I have created in the manual. Any advice? | 04:10:04 | |
| okay more info I found out. If i change the reference from
to
it builds and renders fine but ipv4 is just a namespace for all my functions inside of the network library. Can we not access functions inside of an attrset or do I need to write things differently so it works? | 04:16:10 | |
| Not sure if that much nesting is ok: https://github.com/NixOS/nixpkgs/blob/master/doc/doc-support/lib-function-locations.nix#L25-L31 | 04:16:49 | |
| If I remove the "ipv4" layer of nesting and adjust the reference, it builds okay. So it seems like the doc infra doesn't allow that much? | 04:18:49 | |
| Well that is unfortunate. Thanks for verifying my results tomberek | 04:24:09 | |
| is your use of doc comment inentional? Starting with "/**"? https://github.com/nix-community/nixdoc | 10:48:13 | |
| Sorry yes obviously it's intentional π sorry for the noise | 10:50:16 | |
| * hey all, i recently got it into my head to write a little tool which automatically edits nix files without messing them up, and got it to a point where it can remove e.g. all mentions of i have not opened a pull request with it because that'd just ping a lot of people and i'm not sure if it would be accepted or people here would even be interested, so I thought I'd ask here beforehand | 13:13:37 | |
In reply to @terru:raccoon.collegeYou can use lib.warn (which gives a trace) :) | 13:36:52 | |
| i know. my question was more: should it give a trace, or should it give a hard (failing) error directly? | 13:38:22 | |
| but i'll use lib.warn if that's preferred (just be a second, I was literally just writing the PR message for this π) | 13:39:15 | |
| Answered on GitHub: https://github.com/NixOS/nixpkgs/pull/299409#issuecomment-2053650225 | 13:41:19 | |
| (it can work) | 13:41:35 | |
| terru: Yeah just a trace for now is my preference | 13:42:04 | |
| fair | 13:45:27 | |
| you should've gotten a ping from github with the PR ;) | 13:45:56 | |
| * you should've just gotten a ping from github with the PR ;) | 13:46:00 | |
grr forgot the final force-push changing the throw to lib.warn, sorry to everyone who now got pinged twice π | 13:48:16 | |
| 16:04:39 | ||
| 14 Apr 2024 | ||
| 17:37:29 | ||