Time Nick Message 05:53 lissobone No, testing is part of minetest. 05:53 lissobone The only alternative name (but still with 'test') I can think of is rubenwardy's Nubetest. 05:54 lissobone Or even ",omryrdy". 06:39 lissobone I think that you'll like my new mod. It's fun. 06:39 lissobone It is recommended to turn off any acceleration and gravity before using it for best gravityless experience. 07:50 Abdou-31 Hi folks! 07:51 Abdou-31 is there something called `timpstamp`or it's just a typo? 07:56 MTDiscord most likely not, that would be "timestamp" 07:56 MTDiscord (as in, yes that's a typo) 07:56 Abdou-31 nice! thx dude! 07:57 Abdou-31 how are you doing today btw? 07:57 MTDiscord I'm doing good, thanks 07:57 Abdou-31 nice 08:01 MTDiscord can I change the avatar that shows on Discord whenever I type from irc? it's ugly 08:01 MTDiscord that green thingy 08:14 MTDiscord >* minetest.register_on_chatcommand(function(command, params)) * Called always when a chatcommand is triggered, before minetest.registered_chatcommands is checked to see if that the command exists, but after the input is parsed. 08:14 MTDiscord should I remove "that" or "the" ? 08:22 lissobone My mod induces nausea. 08:39 lissobone Potentionally dangerous memetic properties are yet to be discovered. [REDACTED]. 09:12 rubenwardy Abdou-31: is checked to see whether the command exists 09:12 rubenwardy hmm I suppose that needs an "or not" 09:13 rubenwardy is checked to see if the command exists 10:34 MTDiscord is checked to see if the command exists? 10:34 MTDiscord Ok 10:38 MTDiscord In doc/menu_lua_api.txt: core.show_path_select_dialog(formname, caption, is_file_select) ^ shows a path select dialog ... ^ -if dialog was canceled "_cancelled" 10:40 MTDiscord "_cancelled" shouldn't be replaced by "_canceled" right? 10:45 MTDiscord that sentence sounds off 10:46 MTDiscord -if dialog was canceled "_cancelled" ^ will be added to fieldname value is set to formname itself 10:46 MTDiscord I don't get the last part: is set to... 11:09 MTDiscord yeah don't fix typos in the API itself, it won't reflect in the actual source code and will lead to incorrect documentation 11:28 MTDiscord Gotcha 11:28 MTDiscord Thx 11:29 MTDiscord But what about that last part that sounds off? 11:29 MTDiscord (the path select dialog is due for a nuking though but that's besides the point 8)) 11:29 MTDiscord what about it? 11:29 MTDiscord Doesn't sound odd? 11:31 MTDiscord yeah does sound odd, might need a comma or something 11:31 MTDiscord btw while you're at it, menu_lua_api.txt is written in this weird not-markdown format, you could go ahead and reformat it while you're at it 12:08 MTDiscord Might do it, but not now 12:13 MTDiscord Why are there those ^ at the beginning of lines? 12:16 MTDiscord some weird thing to signify the function a description is for 12:17 MTDiscord hmm 12:17 MTDiscord ok 12:18 MTDiscord so I'm about to push a commit that fix typos in files inside /doc directory 12:18 MTDiscord I've spell-checked all the files inside that directory 12:19 MTDiscord is Fix typos in doc files a good commit message? 12:19 MTDiscord don't think so 12:19 MTDiscord I need a better one 12:20 MTDiscord help 12:21 MTDiscord btw, the README.md file is typo-free! 12:21 MTDiscord πŸ‘ 12:26 MTDiscord "Fix typos in doc files" looks good to me, especially if it's a whole bunch of typos everywhere and you can't be more specific. 12:27 MTDiscord "typos" always looks like it should be "typoes" but then the latter looks even MORE wrong πŸ˜„ 12:27 MTDiscord X) 12:27 MTDiscord oh! I almost forgot!! 12:28 MTDiscord there are those centred/centered and behaviour/behavior changes that need to be done 12:28 MTDiscord Oh gross, en_GB... 12:28 MTDiscord lol 12:28 MTDiscord wait 12:29 MTDiscord Sorry I guess that should be "grouss" πŸ™„ 12:29 MTDiscord not files everywhere, but only files inside the /doc folder 12:29 MTDiscord I think Fix typos in files inside doc directory would be clearer 12:29 MTDiscord because doc files is ambiguous 12:31 MTDiscord Houw dou en_GB users knouw where to sprinkle the extra u's? Or wheer re should be ervresed? It boggles my mind. 12:31 MTDiscord T__T 12:32 MTDiscord *bouggles my midn 12:37 MTDiscord in doc/fst_api.txt, there is: on_change = function(type,old_tab,new_tab) -- called on tab chang, does it need to be "change" or "changing"? 12:39 MTDiscord I'd say "called on tab change" 12:39 MTDiscord thx a lot! 12:52 MTDiscord https://github.com/minetest/minetest/pull/12882 12:52 MTDiscord My second baby 13:10 MTDiscord Should I add [NO SQUASH]? 14:24 MTDiscord Test 14:42 srifqi Abdou-31: I do not think that is needed as those commits can be squashed as "fix typo" or something similar. 14:45 MTDiscord agreed, no-squash seems like it's mainly for larger, complex cases where arguably something should have been multiple PRs, but couldn't because of complex prerequisite relationships that would entail, or something. 14:46 MTDiscord Ok 14:47 MTDiscord A quick question: why the lines in lua_api.txt don't end with a trailing space? I'm reading that in markdown, that's how you make line breaks 14:47 MTDiscord So if processed with a md processor, many seperate lines will show as one long line 14:48 MTDiscord A trailing space here means two or more spaces 14:48 MTDiscord Even one trailing whitespace won't make markdown create a line break 14:49 MTDiscord First of all, trailing spaces are 🀒 and the markdown spec doesn't have the authority to make them acceptable. 14:50 MTDiscord Secondly, we don't want to force line breaks in a markdown render, we want long lines to flow and wrap automatically. If we wanted specific line-breaks then we'd probably use preformatted blocks, or plaintext... 14:50 srifqi I am also guessing that those line breaks are due to the line limit and not because new paragraph. All paragraphs inside that file is separated by an empty line IIRC. 14:51 MTDiscord Does markdown support wrap word? 14:51 MTDiscord or whatever that feature is called 14:52 MTDiscord But they say something about using nothing to indicate new lines: >And at least a couple lightweight markup languages don’t require anything at the end of the line β€” just type return and they’ll create a line break. 14:52 MTDiscord But it's the least compatible 14:53 MTDiscord Markdown is written IN plain text, so it's the author's responsibility to manually wrap the markdown source if they want to ensure that all editors consistently fit the text on the display for other doc editors. Markdown's default and correct behavior is to UNWRAP those lines, and then allow them to REWRAP to fit the actual display at render time. 14:53 MTDiscord O.O 14:53 MTDiscord I don't understand 14:53 MTDiscord But that's fine 14:53 MTDiscord I'm still reading about it 14:54 MTDiscord There is actually very little practical use for simple line breaks that aren't paragraph breaks (i.e. indicated by separating whitespace or an indentation) because those line breaks may disappear if the display width is close enough to the point where the line above would have wrapped naturally. 14:54 MTDiscord I agree with you that trailing spaces are weird 14:54 MTDiscord You almost always either want paragraph breaks or no breaks. 14:54 MTDiscord I see 14:54 MTDiscord Wait 14:55 MTDiscord If I put a single line break in the source, that's because I wanted a line break in the source, NOT in the rendered output. If I put 2 line breaks in the source, then that's an indication that I want actual separation in the output. 14:55 MTDiscord So you mean even if there is NO trailing space at the end of a line, a md processor will wrap the lines so they don't go outside the screen? 14:57 MTDiscord That's up to the viewer (e.g. it's often rendered to HTML and then viewed in a browser) but that's the standard behavior across all viewers I know 14:57 MTDiscord I see 14:58 MTDiscord So the lines on yhe viewer will be different from those line breaks on the source, that we see now? 14:58 MTDiscord Weird 15:00 srifqi Markdown is not WYSIWYG (like LibreOffice Writer/Office Word). What you see from the markdown file may be different than the rendered file. 15:00 srifqi Have you tried to open lua_api.txt as a markdown file? You can see the effects of how it was written immediately. 15:00 MTDiscord Oh You're right! I've never tried it 15:00 MTDiscord Thx guys 15:01 MTDiscord But as warr said, the "manual" line break thing is rarely used, we can just paragraphs or let the rendrer decide 15:02 MTDiscord https://cdn.discordapp.com/attachments/749727888659447960/1032670095371149322/lua_api-preview.png 15:02 srifqi That's an example between markdown file and its rendered view. 15:03 MTDiscord Nice! Is it vscode? 15:03 MTDiscord I didn't know it had a md processor 15:04 srifqi Almost! I use VSCodium instead. 15:04 srifqi You can tell the IDE to treat lua_api.txt as Markdown file (Ctrl+K then M). The preview button is at the top-right corner. 15:04 srifqi (the magnifying glass) 15:05 MTDiscord Really, that rendered view is the MAIN reason why most people use MD in the first place (even though the code could be considered readable and theoretically I guess some of the markup in it does disambiguate stuff) 15:05 srifqi It can be read as is or by using a Markdown processor. 15:06 MTDiscord Nice! I'll try it later! 15:06 MTDiscord What's vscodium? Vscode's brother? 15:07 MTDiscord Vscode but demicrosofted 15:07 srifqi According to its website, VSCodium = VSCode - telemetry. 15:08 MTDiscord Good markdown should always strive to be readable both with and without the processor/viewer. If "readable without" weren't one of the goals, then there are saner ways the text could be structured to be easier to process. 15:22 MTDiscord πŸ‘ 15:23 MTDiscord But I've read that you can just turn off the telemetry option on the menu and that's it. 15:23 MTDiscord I think i've read that on Wikipedia 15:28 MTDiscord the only thing that can turn off telemetry in a proprietary program is a strong firewall drop rule 15:33 muurkha sometimes not even that 15:56 MTDiscord Wow 16:45 MTDiscord A question: In README.md there is: Table of Contents --------------------- 1. [Further Documentation](#further-documentation) ... Further documentation ---------------------------- In the rendering, the table of content's entry is a clickable link that takes directly to the h1 header. My question is: how are they linked? Shouldn't be a {#further-documentation} beside the header? 16:47 MTDiscord [URL Text](url link) 16:47 MTDiscord https://docs.github.com/en/get-started/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax#relative-links 16:50 rubenwardy Markdown automatically gives headers anchors 17:08 muurkha no, Markdown doesn't 17:08 muurkha but Github does 17:08 rubenwardy markdown doesn't have a standard 17:08 rubenwardy most markdown variants add header anchors (

` 17:09 rubenwardy *implementations 17:10 muurkha markdown certainly does have a standard, even if not a standard implementation 17:10 rubenwardy it has an info page but it's very vague on a lot of things 17:10 rubenwardy there's no one standard markdown version 17:10 muurkha and I don't think it's true that most markdown variants add header ids (which are not anchors; the tag is an anchor) 17:11 ecraven rubenwardy: even worse, there's *multiple* "standards" :P 17:11 muurkha no, there's just one 17:11 MTDiscord I'm not sure if "being a mess" is exactly what one normally thinks of when they hear "standard", but I guess whatever works 🀷 17:12 muurkha fortunately Gruber was able to browbeat the Commonmark people out of calling their effort "Markdown" 17:12 muurkha the standard is https://daringfireball.net/projects/markdown/, though it does leave a lot to be desired 17:12 rubenwardy I'm well aware of how HTML and works, I was simplifying 17:14 muurkha using the wrong terminology doesn't simplify things 17:14 muurkha https://babelmark.github.io/?text=Title%0A%3D%3D%3D%3D%3D%0A%0Ah21%0A-----%0A%0Ah22%0A----- shows 16 implementations generating headers without IDs 17:14 muurkha 9 implementations generating headers with IDs (including, importantly, pandoc) 17:15 muurkha and one implementation generating headers with anchors (GFM) 17:16 muurkha several implementations that claim to implement GFM omit the IDs 17:17 muurkha oh and gambas is totally broken on that example but it also doesn't generate IDs 17:17 muurkha nine of the backends it tries to hit are broken now 17:20 MTDiscord Honestly we are probably going to have to pick which implementations we support at some point and disavow the rest; it's impractical to expect us to maintain documentation such that it works equally well across all renderers when the spec is so jank. 17:21 MTDiscord MT is supposed to be a net exporter of jank, we don't need no imports. 17:24 muurkha :) 17:25 muurkha my own approach to this for Dercuano was to just use a single renderer for the Markdown, and then mangle the HTML output further with some custom Python 17:26 muurkha but having now written books' worth of Markdown I look askance at renderers that claim to be "Markdown" but corrupt what I've written 17:28 MTDiscord Thx guys for your explanation! 17:34 rubenwardy ContentDB uses python-markdown with a bunch of extensions to try and get close to GFM. There not being a strict standard is a pain for making md portable 17:37 rubenwardy "fenced_code", "tables", "codehilite", "toc", a custom strikethrough plugin, and custom mention plugin 18:36 sofar "Inside the box" server is back up. 18:52 rubenwardy I'd like it if you could release the tutorial boxes at least, inside the box would be a really good singleplayer game. There's a fork that makes it work in singleplayer, but it's missing content 18:56 muurkha there is a strict standard, though, and it doesn't have fenced code blocks, tables, code highlighting, or tables of contents in it 18:57 muurkha for tables of contents I usually use http://canonical.org/~kragen/sw/addtoc.js 18:58 muurkha this has the disadvantage that it's JS though. also it adds IDs to headers 18:59 muurkha the biggest problem I think is that Markdown was designed to *not* be extended, unlike HTML 19:00 muurkha so when people add LaTeX, tables, etc., there's not a good way to do it in a fully backward-compatible way 19:00 muurkha the fenced code blocks in GFM and CommonMark (which are not in Markdown, despite popular belief) do offer a useful extension mechanism 19:01 muurkha used by things like PlantUML 19:01 muurkha if you want a standardized Markdown-like language that supports the GFM features, maybe CommonMark is it? 22:46 HanSaul hi i wonder if minetest downloaded cache is shared between all servers so once minetest_game is downloaded once it wont re-download for other servers? 22:47 MTDiscord I think it's supposed to work that way, yes. Media is identified by its exact content hash, so there should be no issues with servers with different versions. 22:48 MTDiscord That being said, minetest_game itself should be by far not the main cost of media downloading on most servers :-) 23:16 HanSaul thanks wwar 23:25 muurkha can you use that to fingerprint what other servers someone else has been playing on? 23:55 fluxionary muurkha: ish. the server doesn't get a list of which hashes the client has seen before, but the server can query whether the client has seen a particular hash. you can get info that way, but it'd be hard to track an individual, let alone create a system to track thousands of them. 23:58 fluxionary if someone doesn't clear their cache, it's easy (w/ a hacked server) to see whether a player is new to your server (or set of servers), or not. 23:59 fluxionary https://en.wikipedia.org/wiki/Spy_pixel 23:59 fluxionary https://en.wikipedia.org/wiki/Web_beacon