| Time |
Nick |
Message |
| 00:42 |
|
mannerism joined #minetest |
| 00:51 |
|
grouinos joined #minetest |
| 01:06 |
|
v-rob joined #minetest |
| 01:16 |
|
specing joined #minetest |
| 02:47 |
|
organizedglobals joined #minetest |
| 03:16 |
asdflkj_sh |
I'm trying to get paradust's https://minetest.dustlabs.io/ to load from my local filesystem (file:///home/asdf/gr/mt/minetest.dustlabs.io/mt/index.html... I did wget -r https://minetest.dustlabs.io/mt, then manually downloaded https://minetest.dustlabs.io/mt/minetest.data, then https://github.com/paradust7/minetest-wasm/raw/main/static/.htaccess) |
| 03:16 |
asdflkj_sh |
I really have no knowledge of browsers, cross origin requests, access control checks, etc., is it possible to modify it to make this work? |
| 03:17 |
MTDiscord |
<Jonathon> @paradust ^ |
| 03:17 |
erle |
asdflkj_sh which browser? |
| 03:17 |
asdflkj_sh |
erle: luakit (webkit based) |
| 03:17 |
erle |
asdflkj_sh if you have python3 installed, you can use python to get a local webserver that serves a directory |
| 03:18 |
erle |
python3 -m http.server 8080 |
| 03:18 |
erle |
i think |
| 03:18 |
MTDiscord |
<paradust> it needs certain headers: https://github.com/paradust7/minetest-wasm/blob/main/static/.htaccess |
| 03:18 |
erle |
paradust you only need those for cross-origin access though |
| 03:18 |
asdflkj_sh |
actually I'm trying have a working PoC that a friend can run on a Chromebook |
| 03:19 |
erle |
so why would they be needed |
| 03:19 |
erle |
if you host it locally |
| 03:20 |
asdflkj_sh |
they have termux so they can run a web server, but I think it's too sandboxed to be accessed from Chrome |
| 03:22 |
asdflkj_sh |
(it's a managed chromebook so it's basically a padded prison cell) |
| 03:22 |
paradust |
I believe the cross-origin headers are needed because webassembly w/ pthreads is a security risk for a website with cookies (due to spectre, etc) |
| 03:22 |
paradust |
or hosting other private data |
| 03:23 |
paradust |
it also needs to be hosted on https, I believe |
| 03:23 |
paradust |
(not 100% sure of that) |
| 03:24 |
asdflkj_sh |
how feasible would it be to embed the data into the js or html so it wouldn't have to be requested separately? |
| 03:24 |
paradust |
embed what data? |
| 03:24 |
paradust |
minetest.data ? |
| 03:24 |
paradust |
that's generated by the emscripten file packager |
| 03:25 |
paradust |
it can be modified, if you're willing to do the work |
| 03:27 |
paradust |
erle: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/SharedArrayBuffer#security_requirements |
| 03:28 |
erle |
oh i see |
| 03:29 |
erle |
paradust as far as i understand though, webassembly with pthreads is a general security risk then? |
| 03:29 |
paradust |
oh certainly. but it is enabled by default :D |
| 03:30 |
|
grouinos joined #minetest |
| 03:32 |
asdflkj_sh |
I'm willing to work, and capable if it involves, say, pasting a snippet of JS someone provides for me and the base64 encoding of mintest.data into index.html or minetest.js, not capable if it's much more complicated (I don't know JS) |
| 03:32 |
paradust |
they must feel confident that separate processes are isolated, but because the webassembly (and even the js) lives in the same process (and even on the same thread, in some cases) as the browser threads, there is no way to mitigate it |
| 03:32 |
|
sys4 joined #minetest |
| 03:33 |
paradust |
asdflkj_sh: what's your overall goal here? |
| 03:35 |
asdflkj_sh |
paradust: to enable my friend to use minetest to learn lua and coding on a managed chromebook that blocks your site |
| 03:36 |
paradust |
huh. do you know why it blocks the site? |
| 03:36 |
paradust |
why is the .data file a problem? |
| 03:36 |
paradust |
you should be able to copy all the files and host it somewhere else (but it needs to be on HTTPS) |
| 03:37 |
asdflkj_sh |
it blocks everything not on a whitelist to keep the managing school legally unliable from accidentally letting kids download porn, malware, and whatnot |
| 03:39 |
asdflkj_sh |
the whitelist is pretty arbitrary but blocks most small independent sites so if I host it it'll likely still be blocked |
| 03:40 |
paradust |
but it lets you run commands from the terminal? |
| 03:41 |
asdflkj_sh |
yes, they have a few whitelisted Android apps available, Termux and MineCraft EE being the only two my friend uses |
| 03:43 |
erle |
what is in minecraft EE that is not normally in minecraft? |
| 03:43 |
asdflkj_sh |
erle: https://en.wikipedia.org/wiki/Minecraft#Minecraft:_Education_Edition |
| 03:45 |
erle |
asdflkj_sh just wait until notch implements lua API lmao |
| 03:48 |
paradust |
the place to start would be to try to setup a local webserver, and see if it even works |
| 03:48 |
paradust |
can you install the netcat tool? |
| 03:50 |
paradust |
if you can, run in terminal: nc -v -l 5000 |
| 03:50 |
paradust |
and then see if you can surf to http://localhost:5000/ |
| 03:50 |
paradust |
if you can, netcat will spit out some http junk |
| 03:53 |
asdflkj_sh |
they're in a different timezone and asleep right now, will ask them to try that though I think it won't work (think we tried something like that in the past but not quite as simple and good a test) |
| 03:55 |
erle |
is your friend able to afford an old thinkpad? minetest runs on everything from the last 15 years or so, but you don't get much more than 30 fps on some. |
| 03:55 |
erle |
if your friend has the money, it might be a better investment of their time. if not, is jailbreaking the chromebook a possibility? |
| 03:55 |
erle |
or would that be detected |
| 03:56 |
asdflkj_sh |
great suggestions, unfortunately both fairly unviable |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:20 |
paradust |
erle: Looks like Minecraft EE disables network play except for within an "Office 365 tenant" |
| 04:20 |
erle |
LOL |
| 04:28 |
|
Verticen joined #minetest |
| 04:31 |
|
sys4 joined #minetest |
| 05:32 |
|
v-rob joined #minetest |
| 05:36 |
|
lemonzest joined #minetest |
| 05:36 |
|
debiankaios joined #minetest |
| 05:44 |
|
Talkless joined #minetest |
| 05:46 |
|
calcul0n joined #minetest |
| 05:57 |
|
definitelya joined #minetest |
| 05:58 |
|
Lesha_Vel_ joined #minetest |
| 06:11 |
|
Sven_vB joined #minetest |
| 06:57 |
|
Wuzzy joined #minetest |
| 07:39 |
Wuzzy |
rubenwardy: What needs to change in Tutorial for acceptance? |
| 07:47 |
MTDiscord |
<ROllerozxa> mod name refactoring I assume? |
| 07:58 |
|
orwell96 joined #minetest |
| 08:01 |
|
proller joined #minetest |
| 08:13 |
|
Hawker joined #minetest |
| 09:03 |
|
Fixer joined #minetest |
| 09:06 |
Wuzzy |
yes but like which mods? |
| 09:24 |
|
___nick___ joined #minetest |
| 09:41 |
|
kabou joined #minetest |
| 10:09 |
|
jvalleroy joined #minetest |
| 10:09 |
|
jvalleroy joined #minetest |
| 10:11 |
|
clavi joined #minetest |
| 10:11 |
|
clavi joined #minetest |
| 10:32 |
|
Flabb joined #minetest |
| 10:37 |
|
grouinos joined #minetest |
| 11:30 |
|
grouinos joined #minetest |
| 13:05 |
|
kabou joined #minetest |
| 13:14 |
|
specing_ joined #minetest |
| 13:56 |
MinetestBot |
[git] x2048 -> minetest/minetest: Run automated tests when lua files change (#12184) 828461c https://github.com/minetest/minetest/commit/828461c193c9dcee1221a367b340084e4ee643ad (2022-04-30T13:54:07Z) |
| 13:59 |
|
cation joined #minetest |
| 14:08 |
|
TomTom joined #minetest |
| 14:11 |
|
cation joined #minetest |
| 14:17 |
|
cation joined #minetest |
| 14:19 |
rubenwardy |
!tell Wuzzy no idea, maybe ask on the thread |
| 14:19 |
MinetestBot |
rubenwardy: yeah, sure, whatever |
| 14:26 |
|
cation joined #minetest |
| 14:49 |
|
smolensket joined #minetest |
| 14:51 |
MinetestBot |
[git] sfan5 -> minetest/minetest: Enable additional warning flags faecff5 https://github.com/minetest/minetest/commit/faecff570c48350c599bcf043a7004a52ddf1b39 (2022-04-30T14:49:41Z) |
| 14:51 |
MinetestBot |
[git] sfan5 -> minetest/minetest: Deal with compiler warnings a89afe1 https://github.com/minetest/minetest/commit/a89afe1229e327da3c397a3912b2d43d2196ea2b (2022-04-30T14:49:41Z) |
| 15:10 |
|
Rafi59 joined #minetest |
| 15:18 |
|
jadedctrl joined #minetest |
| 15:18 |
|
smolensket joined #minetest |
| 15:25 |
|
smolensket joined #minetest |
| 15:32 |
ghoti |
So.. my xban.db reset a couple of days ago, and new bans have been added since then. I have a backup. Is there an easy way to merge the two files? Or do I need to write something? |
| 15:46 |
MTDiscord |
<luatic> How and why did it reset? |
| 15:48 |
MTDiscord |
<luatic> Anyways, the "xban.db" files are just Lua scripts that return a Lua table when run. Simply loading both, merging the tables (possibly recursively), then writing should suffice. |
| 15:55 |
|
sys4 joined #minetest |
| 16:13 |
MTDiscord |
<Jonathon> Xban uses serialize so it loses its mind when it gets to bick |
| 16:13 |
smolensket |
Hello. I am trying to install ubuntu 18.04.1 in VMware Workstation 2019.3 on a Mac. I cannot install the bootloader in UEFI mode, because it cannot install it on the Macintosh EFI System Partition (SAP) (/boot/efi). Is there any solution to this? |
| 16:13 |
MTDiscord |
<Jonathon> *bug |
| 16:13 |
smolensket |
I'm aware of this error message: https://i.imgur.com/6JfPm2d.png |
| 16:13 |
MTDiscord |
<Jonathon> Xban should use json like areas, travel etc does to fix the issue |
| 16:14 |
MTDiscord |
<Jonathon> However the issue was reported and krock closed it as fixed when it isnt |
| 16:15 |
MTDiscord |
<Jonathon> Issue: https://github.com/minetest-mods/xban2/issues/19 |
| 16:17 |
MTDiscord |
<luatic> wsor: that's what my question was aiming at |
| 16:18 |
MTDiscord |
<luatic> and no, Xban shouldn't use JSON |
| 16:18 |
MTDiscord |
<luatic> Minetest's JSON API sucks |
| 16:20 |
|
sinvet joined #minetest |
| 16:21 |
sfan5 |
isn't that just because json itself sucks |
| 16:31 |
Krock |
json and its unicode handling sucks |
| 16:31 |
smolensket |
Any ideas? |
| 16:32 |
Krock |
smolensket: try legacy/MBR boot. that should do the trick. it may be old technology but has proven to be reliable on my side |
| 16:32 |
smolensket |
Krock: Thanks. Will try |
| 16:32 |
Krock |
dude wtf how are you so far to reply |
| 16:32 |
Krock |
s/far/fast/ |
| 16:42 |
MTDiscord |
<Tuxifan> Krock: Some people can type faster than others ? |
| 16:44 |
Krock |
my message was sent at 1651336349 and I go a reply at 1651336350 |
| 16:44 |
Krock |
this makes it a solid 180 words per minute even ignoring the reaction and read time |
| 17:06 |
|
Verticen joined #minetest |
| 17:22 |
MTDiscord |
<Jonathon> Xban should use json unless it wants to keep losing ban lists. |
| 17:23 |
MTDiscord |
<Jonathon> Areas, travelnets, already do this because of the issue |
| 17:27 |
|
grouinos joined #minetest |
| 17:34 |
MTDiscord |
<Warr1024> IIRC mod_storage is now actually database-backed, so isn't it becoming more practical now to just store your data in one of those, broken up into separate keys? |
| 17:39 |
MTDiscord |
<Tuxifan> Krock: https://paste.centos.org/view/31b54b06 |
| 17:40 |
MTDiscord |
<Tuxifan> It was a bot |
| 17:40 |
Krock |
it works pretty good tbf |
| 17:42 |
MTDiscord |
<Warr1024> Clever |
| 17:42 |
MTDiscord |
<ROllerozxa> tbh I already thought it was kinda weird someone was asking about installing ubuntu in a VM in #minetest so the fact it was a bot wasn't very surprising :P |
| 17:43 |
Krock |
I wonder where the answer went |
| 17:44 |
Krock |
nowhere? |
| 17:44 |
MTDiscord |
<Warr1024> I could imagine a business angle behind using chatbots to harvest expert answers over IRC and other non-obvious sources, to run a stack-overflow-like help site but without just stealing the same SO answers that everybody else scrapes. |
| 17:45 |
MTDiscord |
<Warr1024> How many times have you searched a technical question, clicked what looked like an answer, and just got the same not-actually-useful result you already found on SO, but on some different site with half-broken layout? For me ... too many times indeed. |
| 17:47 |
jadedctrl |
jesus, those sites are the worst… |
| 17:58 |
MTDiscord |
<GoodClover> there needs to be a collaborative project collecting shit sites like that and removing them from results |
| 18:02 |
MTDiscord |
<Warr1024> The harder you wack, the harder they mole. |
| 18:02 |
|
___nick___ joined #minetest |
| 18:03 |
MTDiscord |
<Tuxifan> Krock: Into some GPT I guess |
| 18:03 |
definitelya |
GoodClover: That implies search engines would care. |
| 18:05 |
|
___nick___ joined #minetest |
| 18:06 |
MTDiscord |
<GoodClover> I'm pretty sure most search engines would be happy to remove bad results |
| 18:06 |
MTDiscord |
<GoodClover> unless they're super anti-censorship |
| 18:06 |
MTDiscord |
<ROllerozxa> they're even worse in other languages. in english they're reasonably comprehensible but in swedish they're just... not even google translate quality, they legit must be using some bargain bin translator service to translate their english content to every other language known to man |
| 18:06 |
|
jess joined #minetest |
| 18:08 |
MTDiscord |
<Warr1024> tbh, after messing around a bunch with weblate, and having to deal with a bunch of languages that are like half-filled-in, it would be nice to be able to just pull in automatic machine translations for everything and use it as a fallback beneath human-translated stuff... |
| 18:10 |
definitelya |
GoodClover: hmm, maybe the FOSS ones, the others would need incentive probably. |
| 18:22 |
|
Sven_vB joined #minetest |
| 18:37 |
|
gera joined #minetest |
| 18:38 |
settl3r[m] |
how do i "convince" my lava source block to produce (cobble-)stone instead of pumice/basalt? |
| 18:39 |
gera |
Hi. I'm new to lua. I want to make a command that takes coordinates, three floats positive or negative like this "-111.5 18 75". How do I parse such string in lua? |
| 18:46 |
Krock |
local parts = inputstring:split(" ") |
| 18:46 |
Krock |
then use tonumber(parts[1]) for each index |
| 18:47 |
Krock |
there's also the minetest.string_to_pos() function which however won't be as easy to integrate into multi-argument chat commands |
| 18:49 |
gera |
I came up with this: x, y, z = string.match("111.5 -2 3", "([+-]?%d+%.?%d*) ([+-]?%d+%.?%d*) ([+-]?%d+%.?%d*)") |
| 18:50 |
gera |
tonumber might be a better idea,ty |
| 19:00 |
calcul0n |
settl3r[m], it's probably the gloopblocks mod which changes lava+water behaviour |
| 19:00 |
calcul0n |
so simple answer: disable it :p |
| 19:01 |
calcul0n |
or make cobble from basalt, there should be a recipe for that |
| 19:01 |
|
Talkless joined #minetest |
| 19:06 |
|
v-rob joined #minetest |
| 19:07 |
|
appguru joined #minetest |
| 19:15 |
MTDiscord |
<luatic> gera: keep in mind tonumber is overly permissive though: it will allow NaNs and inf |
| 19:15 |
MTDiscord |
<luatic> as well as hex, exponents etc |
| 19:28 |
|
Verticen joined #minetest |
| 19:33 |
erle |
luatic that's not overly permissive |
| 19:33 |
erle |
luatic gera Krock the correct thing is to first match via a regular expression, then use tonumber |
| 19:34 |
Krock |
there's no "correct way". many roads lead to rome |
| 19:34 |
erle |
gera if you either use tonumber or use the current player position without checking it, you can run into this bug, so don't https://github.com/minetest/minetest/issues/11742 |
| 19:35 |
erle |
Krock there is a correct way in terms of “state of the art”. and that is to match the input using a context-free, regular or calc-regular grammar. and to perform full recognition before processing. |
| 19:35 |
MTDiscord |
<luatic> erle: no, a regex is only part of the solution |
| 19:35 |
erle |
luatic, yeah, but it needs to be the *first* part of the solution |
| 19:35 |
Krock |
regex will solve your problems and shoot yourself into the knee |
| 19:35 |
erle |
to make sure that tonumber gets no NaN |
| 19:36 |
erle |
or -Inf or so |
| 19:36 |
MTDiscord |
<luatic> to enforce a simple [+-]?%d+%.?%d+ format |
| 19:36 |
MTDiscord |
<luatic> Krock: no |
| 19:36 |
erle |
if you want to enforce a format, you need a regex |
| 19:36 |
MTDiscord |
<luatic> must be %d* |
| 19:36 |
erle |
for a commannd |
| 19:36 |
MTDiscord |
<luatic> Lua patterns are easy to write |
| 19:36 |
erle |
lua patterns also suck if you write them badly lol |
| 19:36 |
MTDiscord |
<luatic> Anyways, you need Lua patterns + range checks here |
| 19:36 |
MTDiscord |
<luatic> Yeah, Lua patterns have poor performance characteristics |
| 19:37 |
erle |
i believe you can hang mcl5 by inputting 4096 times the letter a in the creative inventory search field |
| 19:37 |
erle |
due to backtracking runtime bullshit |
| 20:03 |
MTDiscord |
<luatic> bruh what |
| 20:03 |
MTDiscord |
<luatic> show me the patterns involved please |
| 20:03 |
MTDiscord |
<luatic> but yes, Lua patterns have worst-case exponential runtime |
| 20:04 |
MTDiscord |
<luatic> the sad thing is that they probably even have exponential runtime for some simple patterns that a decent regex engine would be able to match in linear time |
| 20:04 |
MTDiscord |
<luatic> and yes, I am planning to write my own RegEx engine in pure Lua :P |
| 20:09 |
erle |
luatic we fixed it in mineclonia i think, i am not motivated to look it up |
| 20:10 |
MTDiscord |
<luatic> hmm okay |
| 20:18 |
|
Verticen joined #minetest |
| 21:44 |
|
kabou joined #minetest |
| 21:49 |
|
jesopo joined #minetest |
| 21:50 |
|
Gustavo6046 joined #minetest |
| 21:51 |
|
Gustavo6046 joined #minetest |
| 22:00 |
|
Sven_vB joined #minetest |
| 22:16 |
|
grouinos joined #minetest |
| 22:32 |
|
panwolfram joined #minetest |
| 22:45 |
|
fluxionary joined #minetest |
| 22:55 |
|
kabou joined #minetest |
| 23:05 |
|
hlqkj joined #minetest |
| 23:14 |
|
Toothless joined #minetest |
| 23:17 |
|
AliasStillTaken joined #minetest |
| 23:35 |
|
kabou joined #minetest |
| 23:39 |
|
calcul0n joined #minetest |
| 23:43 |
|
v-rob joined #minetest |
