| Time |
Nick |
Message |
| 05:00 |
|
MTDiscord joined #minetest-docs |
| 12:44 |
MTDiscord |
<mark.wiemer> Custom 404 should definitely be possible, can you create an issue? I can look into it |
| 12:59 |
rubenwardy |
Just make a 404.html file |
| 13:01 |
MTDiscord |
<theidealist> on neocities it's not_found.html |
| 13:33 |
|
Desour joined #minetest-docs |
| 13:34 |
MTDiscord |
<siliconsniffer> GH pages will pick up a 404.html in the source directory. https://docs.github.com/en/pages/getting-started-with-github-pages/creating-a-custom-404-page-for-your-github-pages-site |
| 17:15 |
MTDiscord |
<mark.wiemer> I'm back 🙂 created https://github.com/minetest/dev.luanti.org/issues/133 (Add Hugo as a project dependency) and https://github.com/minetest/dev.luanti.org/pull/132 (Fix readme link) |
| 17:20 |
MTDiscord |
<wsor4035> ik, dunno if that will fight with hugo or not. also dont want to completely reimplement colors, etc |
| 17:21 |
MTDiscord |
<wsor4035> commented on pr |
| 18:46 |
MTDiscord |
<mark.wiemer> https://dev.luanti.org/worlds/ mentions "Schem" files, but it's the only page that does--is this a typo? If so, what's the better name? |
| 18:48 |
MTDiscord |
<luatic> Schematic files, or schematics for short |
| 18:48 |
MTDiscord |
<wsor4035> cba to look atm, but schematics, mts, etc |
| 18:50 |
MTDiscord |
<wsor4035> just going to poke you mark about implementing https://github.com/minetest/dev.luanti.org/issues/72 since thats a you thing |
| 18:58 |
MTDiscord |
<mark.wiemer> what does cba mean? |
| 18:59 |
MTDiscord |
<mark.wiemer> Yeah I'll add the new a11y check eventually, focusing on simple spell check today 🙂 |
| 18:59 |
MTDiscord |
<wsor4035> cant be assed |
| 18:59 |
MTDiscord |
<wsor4035> or cant be arsed |
| 19:01 |
MTDiscord |
<wsor4035> interesting, urban dictionary only list the latter 🤷 https://www.urbandictionary.com/define.php?term=cba |
| 19:08 |
MTDiscord |
<mark.wiemer> I'm assuming "coredev" is generally used as one word? I'll keep it for now, but curious if folks want to maybe change it to two words for consistency with other projects? I don't see any need for it to be special in the long term |
| 19:33 |
MTDiscord |
<mark.wiemer> Any reason "unittests" is occasionally written as one word in prose? |
| 19:36 |
MTDiscord |
<josiah_wi> No good reason, if you ask me. But I'm not going to complain as long as we write it the same way consistently. xD |
| 19:38 |
MTDiscord |
<josiah_wi> Our unit tests are organized into a directory called unittests in our file hierarchy. Maybe that influences our documentation? |
| 19:51 |
MTDiscord |
<mark.wiemer> Yeah I noticed that, but I think in prose when we talk about or mention "unit tests" it should be written in sentence case. Was just curious if there was some other project-specific reason I was missing |
| 19:52 |
MTDiscord |
<rollerozxa> unit tests should definitively be written as two words in prose |
| 20:02 |
MTDiscord |
<mark.wiemer> Cool, will update the issue with the same Looks like CI is still panicking on a warning about native HTML, I thought we removed that, I'll check it out. In any case, https://github.com/minetest/dev.luanti.org/pull/136 is up as a draft: Fix ~800 more spelling issues |
| 20:03 |
MTDiscord |
<mark.wiemer> Ah, I see https://github.com/minetest/dev.luanti.org/issues/96 was closed as not planned. Can we re-open it? I think native HTML is just fine and personally I'd encourage it where valuable (like comments, for example). |
| 20:04 |
MTDiscord |
<wsor4035> no, because its a valid warning that you should fix |
| 20:04 |
MTDiscord |
<wsor4035> (generall) |
| 20:05 |
MTDiscord |
<mark.wiemer> The warning is "native HTML", which isn't a problem if you know what the HTML is and you intended for it to be there. The warning was added to Hugo because Hugo is often used to host untrusted content--we don't do that, so we shouldn't be worried about it. See the issue for details |
| 20:06 |
MTDiscord |
<wsor4035> otherwise stuff like <person> message is going to slip through |
| 20:06 |
MTDiscord |
<mark.wiemer> Ah, maybe, but I think the build itself might be wonky, or at least the rendered page wouldn't look good. I think now that we've imported stuff, we'll be manually reviewing the actual look and feel of most changes, right? |
| 20:07 |
MTDiscord |
<wsor4035> if you disable that warning, your not going to even know |
| 20:07 |
MTDiscord |
<mark.wiemer> I'm sure there's a middle ground, but I'm OK with letting things slip through if it makes it easier for us to do our work in other ways |
| 20:08 |
MTDiscord |
<mark.wiemer> we'll know because we look at the page and make sure it contains the content we intended, including the string <person> as text on the page. If we're not doing that, we're being a bit too sloppy! |
| 20:09 |
MTDiscord |
<wsor4035> the irony is you made the pr fixing the stuff https://github.com/minetest/dev.luanti.org/pull/78/files |
| 20:10 |
MTDiscord |
<wsor4035> humans are humans and going to make mistakes |
| 20:12 |
MTDiscord |
<mark.wiemer> Yes, we will make mistakes. I'm ok with a typo or two getting to the website. I'm frustrated by this tool that's stopping me from adding comments to HTML files, especially because those comments are spell-checking directives Humans make mistakes, but tools aren't perfect. We can use informed judgement to decide which tools to use |
| 20:12 |
MTDiscord |
<mark.wiemer> and now that the stuff is fixed, we don't need the warning 😉 |
| 20:12 |
MTDiscord |
<wsor4035> till more gets added |
| 20:12 |
MTDiscord |
<mark.wiemer> Do you mean more imported content or more manually written content? |
| 20:12 |
MTDiscord |
<wsor4035> either or, doesnt matter |
| 20:12 |
MTDiscord |
<wsor4035> cant expect everyone to know |
| 20:13 |
MTDiscord |
<mark.wiemer> Don't you think it's reasonable for us to manually review the rendered page of any future changes? |
| 20:13 |
MTDiscord |
<mark.wiemer> "us" being PR authors, at least |
| 20:13 |
MTDiscord |
<wsor4035> so your going to trust john doe that made a pr to look at the output? |
| 20:13 |
MTDiscord |
<wsor4035> your going to checkout and run every pr? |
| 20:13 |
MTDiscord |
<mark.wiemer> If the PR is small, I trust a reviewer to ensure there isn't anything suspicious |
| 20:14 |
MTDiscord |
<mark.wiemer> If the PR is big, I trust a reviewer to checkout and run it, or ask for it to be broken into smaller pieces |
| 20:14 |
MTDiscord |
<wsor4035> your comments would be slightly less insanty if we had something that generated a preview site for prs that where easy to look at, but we dont |
| 20:15 |
MTDiscord |
<wsor4035> example of where it saved me https://github.com/minetest/dev.luanti.org/pull/123. course i may be an idiot 🙃 |
| 20:15 |
MTDiscord |
<mark.wiemer> insane is a strong word 😦 I'm just saying that we have a tool that is currently slowing us down in practice, right now. You're saying it's saving us from hypothetical situations. I know I fixed previously-imported files, but those were exactly that: files imported in a rush in a situation where we knew we'd be coming back and cleaning them up. Now that things are clean, I think we're in a very good state that we can trust native |
| 20:15 |
MTDiscord |
HTML. Maybe the compromise is to show the warnings, but not block PRs on them? Or I should just find a way to allow HTML comments and nothing else? |
| 20:16 |
MTDiscord |
<mark.wiemer> Thanks for providing a concrete example. But I think "sanity lost" is enough for us to agree "we won't be doing things this way in the long term", does that sounds reasonable? |
| 20:16 |
MTDiscord |
<wsor4035> > hypothetical situations this is false, as demonstrated by the above links |
| 20:16 |
MTDiscord |
<mark.wiemer> Yes, I was mid-sentence when you posted that. I am mostly referring to the future now. I don't think this warning is helping us any more |
| 20:17 |
MTDiscord |
<mark.wiemer> And note I enabled panic on ALL warnings, not just native HTML ones. I'm happy to keep all other warnings, I just want to remove this native HTML one (at least for comments) |
| 20:17 |
MTDiscord |
<mark.wiemer> How about, for now, I find a way to allow just native HTML comments? Because without that I am struggling to disable the spell-checker for certain parts of the file |
| 20:18 |
MTDiscord |
<wsor4035> that seems fine. better to say open a small hole than the front door |
| 20:22 |
MTDiscord |
<wsor4035> looks like cspell supports /* cSpell:disable */, which is also a golang comment syntax. not sure if that helps at all https://cspell.org/configuration/document-settings/ |
| 20:23 |
MTDiscord |
<mark.wiemer> yeah I'm looking into other comment forms now, see also https://discourse.gohugo.io/t/warning-when-mixing-html-and-markdown/52280 for other forms and overall native HTML warning background 🙂 |
| 20:45 |
MTDiscord |
<mark.wiemer> The weirdest thing is that there is no succinct comment syntax. The best I'm finding is https://stackoverflow.com/questions/45224766/how-to-comment-out-content-in-hugo, and even that is: {{% comment %}} hello world {{% /comment %}} vs <!-- hello world --> absolutely wild. Plus Hugo 0.137.1 isn't even on NixOS lol |
| 20:59 |
MTDiscord |
<luatic> maybe that many core devs are german and like to fuse words (me included) lol |
| 20:59 |
MTDiscord |
<rollerozxa> this is almost certainly why you see so many fuse words there |
| 21:01 |
MTDiscord |
<mark.wiemer> Makes sense! nbd for now. In any case, we need to merge https://github.com/minetest/dev.luanti.org/pull/137 before the spell-check currently |
| 21:01 |
MTDiscord |
<mark.wiemer> (for IRC folks, that PR fixes comment syntax) |
| 21:13 |
MTDiscord |
<mark.wiemer> https://github.com/minetest/dev.luanti.org/pull/136 - fix ~800 spelling issues - also ready for review 🙂 |
