Time |
Nick |
Message |
00:17 |
|
emanuele6 joined #minetest |
00:19 |
|
Verticen joined #minetest |
00:38 |
|
AliasAlreadyTake joined #minetest |
01:52 |
|
grouinos joined #minetest |
02:43 |
|
proller joined #minetest |
03:00 |
|
erle_sleepy joined #minetest |
03:00 |
erle_sleepy |
Jonathon, thanks for pointing to https://git.minetest.land/MineClone2/MineClone2/commit/9d9e21301271da166248a2356b81e0225d738fde – MineClone2 is still broken (both contentdb and git master), but i was able to figure out the missing dependency. do you have a ticket or a PR for the commit at |
03:00 |
erle_sleepy |
https://git.minetest.land/MineClone2/MineClone2/commit/9d9e21301271da166248a2356b81e0225d738fde ? |
03:07 |
erle_sleepy |
https://irc.minetest.net/minetest-dev/2021-11-22#i_5900227 <Jonathon> or minetest.after(0, dofile) provided you dont run into any of the minetest silently failing issues |
03:08 |
erle_sleepy |
can anyone tell me about minetest.after silently failing? |
03:09 |
MTDiscord |
<Jonathon> no idea, i think krock summited that as a git patch and it got merged in. was forums or irc, dont recall |
03:09 |
MTDiscord |
<Jonathon> at least i recall him talking about it |
03:09 |
MTDiscord |
<Jonathon> minetest.after doesnt silently faily |
03:09 |
MTDiscord |
<Jonathon> . s/fail/faily |
03:12 |
erle_sleepy |
Jonathon what i do not understand about https://git.minetest.land/MineClone2/MineClone2/commit/9d9e21301271da166248a2356b81e0225d738fde is that it says “pull/1847/head” but the actual PR 1847 is something entirely different: https://git.minetest.land/MineClone2/MineClone2/pulls/1847 |
03:12 |
erle_sleepy |
i could find no PR that actually matches the commit |
03:13 |
erle_sleepy |
Jonathon, what did you mean with “provided you dont run into any of the minetest silently failing issues” then? |
03:13 |
erle_sleepy |
Jonathon i really want a solution to relying on the leading underscore, bc it is bound to break eventually |
03:14 |
MTDiscord |
<Jonathon> there is a lot of things minetest just silently fails and keeps running on either normally or when functions are called in specific instances |
03:14 |
erle_sleepy |
yeah ok but if that code silently failing on mod loading time, would it be different? |
03:14 |
erle_sleepy |
or is it that during runtime silently failing is more common? |
03:15 |
erle_sleepy |
than during load time |
03:15 |
MTDiscord |
<Jonathon> most of the issues i have seen are runtime related |
03:18 |
erle_sleepy |
Jonathon do you know other mods than _mcl_autogroup who have the requirement to run really late? |
03:18 |
erle_sleepy |
granted, this is a not well formed inquiry |
03:18 |
erle_sleepy |
it is impossible for two mods to be “the last one” |
03:21 |
MTDiscord |
<Jonathon> yes |
03:21 |
MTDiscord |
<Jonathon> ultimately the problem with load last is what i posted in dev irc |
03:21 |
MTDiscord |
<Jonathon> SX — Today at 9:12 PM just seen too many mods use that and it really makes it hard to override things or determine loading order, it is worst tool for this. on_mods_loaded is good. wsor — Today at 9:13 PM its great till you run into 5 mods using it, and mod 2 really needs to actually be after mod 5 |
03:21 |
MTDiscord |
<Jonathon> . idk any good way to figure that out |
03:22 |
erle_sleepy |
well inside mineclone2 or mineclonia it is fine |
03:22 |
MTDiscord |
<Jonathon> well, at this point in time, since im fcused on other things |
03:23 |
erle_sleepy |
thx Jonathon, i will try to use minetest.after then i guess if the current solution breaks. |
03:24 |
erle_sleepy |
_mcl_autogroups is a horrible hack, but i think it needs to have access to all nodes and/or all tools that all mods registered, that is why it wants to come last |
03:24 |
erle_sleepy |
and given that other mods can add tools or nodes, there we have the problem |
03:24 |
MTDiscord |
<Jonathon> ....insert facepalm |
03:24 |
erle_sleepy |
? |
03:24 |
erle_sleepy |
i didn't write it |
03:24 |
MTDiscord |
<Jonathon> for what you want to do from my brief look register_on_mods_loaded is best |
03:25 |
erle_sleepy |
ok thx |
03:25 |
MTDiscord |
<Jonathon> not minetest.after(0, func) hack |
03:25 |
erle_sleepy |
i use the minetest.after(0, func) thing for ingame startup self-tests, do you think that is a horrible idea? |
03:27 |
erle_sleepy |
like, i put code in there that would crash the game, so it immediately crashes if a bugfix stops working |
03:27 |
MTDiscord |
<Jonathon> thats fine since i assume you want to run them on startup, there might be cases where you want minetest.after(1, func) but its per case basis |
03:27 |
erle_sleepy |
what is the difference between 0 and 1 at this point? is something not yet done at 0 that is done at 1 that i should know about? |
03:28 |
|
queria^clone joined #minetest |
03:29 |
erle_sleepy |
Jonathon the only reason i can think of rn why register_on_mods_loaded was not used is that it is a 5.0.0 thing ig |
03:29 |
MTDiscord |
<Jonathon> it depends on how the code you are testing is structured |
03:30 |
MTDiscord |
<Jonathon> lol, your just repeating what i said |
03:31 |
erle_sleepy |
? |
03:32 |
erle_sleepy |
the code i am testing does not do anything using minetest.after(0, func) |
03:32 |
erle_sleepy |
but i will keep it in mind |
03:33 |
|
queria^clone joined #minetest |
03:35 |
erle_sleepy |
Jonathon the issue i have with fully accepting the “it's a pre 5.0.0 thing” is that _mcl_autogroup was rewritten 8 months ago https://git.minetest.land/MineClone2/MineClone2/pulls/1285 – i need to do more research |
03:41 |
MTDiscord |
<SX> Also minetest.after(0, fn) executes after loading is complete, minetest.register_on_mods_loaded executes before loading is completed |
03:42 |
erle_sleepy |
yes i realized that when i wrote the self tests |
03:42 |
MTDiscord |
<SX> Basically minetest.after for things that are supposed to be runtime and minetest.register_on_mods_loaded for things that are supposed to be initialization |
03:42 |
erle_sleepy |
SX i need access to the world in my test cases, so minetest.after is the only thing that works |
03:42 |
erle_sleepy |
any earlier and no actual world exists ig? |
03:43 |
MTDiscord |
<SX> Yes, for that it is exactly correct thing to run single shot globalstep |
03:43 |
MTDiscord |
<SX> (it is same as globalstep, just shortcut for single shot and delayed) |
03:44 |
erle_sleepy |
are you aware of other mods than mineclonia that do tests at runtime? |
03:45 |
erle_sleepy |
(i am not) |
03:45 |
MTDiscord |
<SX> I'm only running mineunit.... can't tell :) |
03:45 |
MTDiscord |
<SX> But for minetest after and how it is globalstep see https://github.com/minetest/minetest/blob/master/builtin/common/after.lua |
03:46 |
erle_sleepy |
he, very funny thx |
03:50 |
erle_sleepy |
SX how would i write a test case that tests if the engine crashes in specific circumstances in mineunit anyway? |
03:51 |
erle_sleepy |
i can't exactly check the results of the operation that might crash in lua, can i? |
03:51 |
erle_sleepy |
i mean, not if it crashes |
03:54 |
MTDiscord |
<SX> mineunit is really still both barebones, some incorrect behavior, heavy development and so on but if you're interested take a look at technic mod tests here: https://github.com/mt-mods/technic/tree/master/technic/spec |
03:55 |
MTDiscord |
<SX> Also (already outdated) readable list of things it does for technic https://github.com/mt-mods/technic/pull/208#issuecomment-946919098 |
03:55 |
erle_sleepy |
thx |
03:55 |
erle_sleepy |
ok, so “does not crash” is on the list, i am looking! |
03:56 |
MTDiscord |
<SX> oh yes you can test for expected crash too, some tests do that. simply catch errors (pcall and friends under the hood) |
03:58 |
MTDiscord |
<SX> also mineunit is simply built on to of busted so basic stuff works without mineunit using just busted (which is more for actual unittesting and mineunit for testing things more like running actual world in minetest) |
03:59 |
erle_sleepy |
SX i do know too little, how woulda pcall deal with a segfault? |
03:59 |
|
Verticen joined #minetest |
04:00 |
MTDiscord |
<SX> oh, if it is engine crash then no. you cannot do that without catching it in c++ and that's bit cursed thing to do... |
04:01 |
erle_sleepy |
ah, i guess i will keep executing code like that as well |
04:01 |
MTDiscord |
<SX> (your program state is messed up bad by then and recovering from segfault is not simple thing to do) |
04:01 |
erle_sleepy |
i do not want to recover |
04:02 |
erle_sleepy |
SX what API version of minetest is mineunit targeting btw? |
04:04 |
MTDiscord |
<SX> random version... I've started working on actual engine version selection but probably takes some time to finish that thing. also like said it is NOT working exactly like minetest engine for most parts even while it can run technic stuff and somewhat advtrains standalone |
04:04 |
erle_sleepy |
it's pretty cool! |
04:06 |
MTDiscord |
<SX> there's github app to get github workflows easily if wanted, luarocks provides scm dev package |
04:07 |
erle_sleepy |
i am not going to rely on microsoft for my virtual legos ^^ |
04:07 |
MTDiscord |
<SX> (also just added $ mineunit --demo which installs simple demo tests in current mod directory, should give idea) |
04:08 |
erle_sleepy |
may years ago i was banned from github in error, but not just normally, i was hellbanned. i could see my account, other people could not, during work! |
04:08 |
MTDiscord |
<SX> well github app is just shortcut for github users, other platform users can add similar workflow |
04:09 |
erle_sleepy |
they fixed it, but i found this so malicious that this is their way to ban people that i have not put private projects on there since |
04:09 |
erle_sleepy |
(i did not even get any good reason, only that an automated system did it, lol) |
04:11 |
erle_sleepy |
SX regarding interesting mineunit specs, do you know any test cases for dupes? |
04:11 |
erle_sleepy |
(where the test case tests that you can not dupe) |
04:12 |
erle_sleepy |
i have not been able to come with one up myself so far |
04:13 |
MTDiscord |
<SX> dupe what? overwrite lua variables for example? or items in game? |
04:13 |
erle_sleepy |
items in game |
04:14 |
erle_sleepy |
as a player, i want them ofc. but as a game developer i do not. |
04:14 |
erle_sleepy |
and once fixed, it would be good to regression test them. |
04:16 |
MTDiscord |
<SX> well when you're testing things you have to think about what you're exactly testing. usually duplication is because of some complex operation / combination of things, you can test those but instead of allowing and testing all possible combinations you should just test that your interface handles correct data correctly and incorrect data correctly. |
04:16 |
erle_sleepy |
SX the technic mineunit stuff looks amazing thank you for showing it to me |
04:17 |
erle_sleepy |
hmm ok you are right. dupes, like many bugs, are at the boundaries of systems. |
04:19 |
erle_sleepy |
i wish someone had made a test case for this so i knew what was wrong: https://github.com/minetest-mods/moreblocks/commit/f8a7d66403b4b1e2d87eeb4249a9be3edb906861#diff-0fc67bb5079c5d1a67ba1a0a697d0dbcf77ef1324c4e1264c59f0a593ea72ca5 |
04:19 |
MTDiscord |
<SX> many go forward by adding test when bug is found and then fixing code in a way that test passes |
04:19 |
erle_sleepy |
yes i have seen that at work. test driven development. |
04:20 |
erle_sleepy |
sadly most amateur devs i know think it is a waste of time since they “know” their code is correct. |
04:20 |
erle_sleepy |
in the same way that ppl “know” that bedrock can not be broken. ;) |
04:20 |
erle_sleepy |
i.e. a simple, obvious & wrong idea |
04:21 |
erle_sleepy |
SX sadly most of the really interesting bugs i know are engine bugs, but mineunit would help immensely whenever i need a test world. |
04:22 |
erle_sleepy |
like when mcl2 broke minecarts |
04:22 |
erle_sleepy |
good mineunit testing would have figured it out i guess |
04:23 |
MTDiscord |
<SX> good if you find some use for it, btw luarocks scm dev package is always tip of master branch. I've tried to keep it in working order with all things using it but behavior might change. |
04:24 |
erle_sleepy |
well for minecarts when we fixed it in mineclonia i actully made a rail testing world |
04:24 |
erle_sleepy |
actually |
04:24 |
erle_sleepy |
which made it very tedious to test |
04:25 |
erle_sleepy |
i mean players rely on stuff like minecart teleportation |
04:26 |
MTDiscord |
<SX> mineunit probably cannot replace play testing ever but for me it has been helpful when doing large modifications / additions / refactoring to ensure that API stays compatible and certain behavior wont change, tetsing all that in game would take hours every session and when you find typo you'll start over... |
04:26 |
erle_sleepy |
i.e. the mechanic where you can tp a minecart from one track to another using a curse mesecons machine |
04:26 |
erle_sleepy |
cursed |
04:27 |
erle_sleepy |
it can be used to switch tracks |
04:27 |
MTDiscord |
<SX> sounds kinda fun :) |
04:28 |
erle_sleepy |
“ensure that API stays compatible” is something that i wish to do too |
04:28 |
erle_sleepy |
and yes, tests enable fearless refactoring |
04:29 |
erle_sleepy |
SX the sad thing about stuff like minecart tp is that often devs are like “that was undocumented, we remove it” |
04:30 |
erle_sleepy |
even when it is a useful mechanic |
04:31 |
erle_sleepy |
mcl2 removed lavakelp for example in a refactoring |
04:34 |
MTDiscord |
<SX> well, that happens. then if it too many ppl think it is bad fork happens. but I guess I'm leaving now, good luck with testing o/ |
04:36 |
erle_sleepy |
i go sleep |
04:36 |
erle_sleepy |
SX thanks again mineunit looks awesome |
04:43 |
|
Lesha joined #minetest |
05:00 |
|
MTDiscord joined #minetest |
05:39 |
|
Hawk777 joined #minetest |
05:44 |
|
sagax joined #minetest |
05:48 |
|
clavi joined #minetest |
05:49 |
|
sy joined #minetest |
05:49 |
|
clavi joined #minetest |
05:49 |
|
jonadab joined #minetest |
05:49 |
|
MinetestBot joined #minetest |
05:49 |
|
MinetestBot joined #minetest |
05:55 |
|
Calinou joined #minetest |
05:55 |
|
frostsnow joined #minetest |
05:55 |
|
sofar joined #minetest |
05:55 |
|
Emerald2 joined #minetest |
05:55 |
|
meldrian joined #minetest |
05:56 |
|
meldrian joined #minetest |
06:00 |
|
Unit193 joined #minetest |
06:01 |
|
riff-IRC joined #minetest |
06:37 |
|
search_social joined #minetest |
06:40 |
|
Sven_vB joined #minetest |
07:29 |
|
CWz joined #minetest |
08:24 |
|
specing_ joined #minetest |
08:50 |
|
olliy joined #minetest |
09:23 |
|
absurb joined #minetest |
09:32 |
|
Swift110-mobile joined #minetest |
09:33 |
|
hook54321 joined #minetest |
09:43 |
|
TomTom joined #minetest |
10:22 |
|
calcul0n joined #minetest |
11:00 |
|
sagax joined #minetest |
11:02 |
|
Miniontoby1 joined #minetest |
11:02 |
Miniontoby1 |
Hey guys |
11:03 |
Miniontoby1 |
https://pastebin.com/QLvwGhCP |
11:03 |
Miniontoby1 |
Please help ^ |
11:06 |
|
AristotIe joined #minetest |
11:15 |
|
nac_ joined #minetest |
11:15 |
* Miniontoby1 |
found https://forum.minetest.net/viewtopic.php?t=19712 and will see if it fixes my problem |
11:16 |
Miniontoby1 |
(didnt fix my problem) |
11:18 |
Miniontoby1 |
is github.com also for you guys down? |
11:25 |
MTDiscord |
<Sublayer plank> have you appended minetest.get_modpath() to the dofile filepath string? |
11:27 |
Miniontoby1 |
nope |
11:27 |
Miniontoby1 |
now added it |
11:30 |
Miniontoby1 |
thanks |
11:32 |
Miniontoby1 |
It works |
11:34 |
MTDiscord |
<Sublayer plank> yay |
12:05 |
|
proller joined #minetest |
12:27 |
|
Peppy joined #minetest |
13:03 |
|
tech_exorcist joined #minetest |
13:07 |
|
Verticen joined #minetest |
13:18 |
|
ggITA52_ joined #minetest |
13:21 |
|
peterz joined #minetest |
13:22 |
|
appguru joined #minetest |
13:23 |
|
proller joined #minetest |
14:15 |
|
Big_Caballito joined #minetest |
14:17 |
|
proller joined #minetest |
14:19 |
Big_Caballito |
howdy! |
14:19 |
Big_Caballito |
howre things here? |
14:19 |
Big_Caballito |
ive been super busy, so im not keeping up on anything lol |
14:21 |
MTDiscord |
<Jonathon> decent |
14:22 |
Big_Caballito |
whens the next version coming out? |
14:22 |
Big_Caballito |
and when did we move away from freenode? |
14:22 |
MTDiscord |
<Sublayer plank> nobody knows when 5.5.0 is coming out |
14:22 |
MTDiscord |
<Sublayer plank> soon, hopefully |
14:23 |
Big_Caballito |
hopefully lol |
14:27 |
sfan5 |
when did we move away from freenode? <- in May, like everyone else |
14:29 |
Big_Caballito |
idk why i didnt hear about that lmao |
14:48 |
|
GNUHacker joined #minetest |
15:00 |
|
definitelya joined #minetest |
15:05 |
|
Noisytoot joined #minetest |
15:08 |
|
kamdard_ joined #minetest |
15:23 |
|
nuala joined #minetest |
15:40 |
|
Big_Caballito joined #minetest |
15:47 |
|
fluxionary joined #minetest |
15:50 |
|
Mc joined #minetest |
15:56 |
|
Extex joined #minetest |
16:03 |
|
Fixer joined #minetest |
16:26 |
|
Hawk777 joined #minetest |
16:55 |
|
proller joined #minetest |
17:27 |
MinetestBot |
[git] ExeVirus -> minetest/minetest: Allow for Game-Specific Menu Music (#11241) 52bfbf6 https://git.io/J1bZH (2021-11-22T17:26:46Z) |
17:29 |
MinetestBot |
[git] sfan5 -> minetest/minetest: Add backwards-compatible behaviour if too few CAO textures specified 206e131 https://git.io/J1bnZ (2021-11-22T17:27:49Z) |
17:38 |
|
garywhite joined #minetest |
17:38 |
|
garywhite joined #minetest |
17:41 |
|
Talkless joined #minetest |
18:07 |
|
grouinos joined #minetest |
18:16 |
|
Verticen joined #minetest |
18:27 |
|
erlehmann_ joined #minetest |
18:44 |
|
erlehmann_ joined #minetest |
19:20 |
|
proller joined #minetest |
19:50 |
|
LazyJ joined #minetest |
19:53 |
|
LazyJ joined #minetest |
19:55 |
|
LazyJ joined #minetest |
19:56 |
|
LazyJ joined #minetest |
20:07 |
|
proller joined #minetest |
20:07 |
|
LazyJ joined #minetest |
20:24 |
|
specing_ joined #minetest |
20:36 |
|
reumeth joined #minetest |
20:44 |
|
proller joined #minetest |
20:46 |
definitelya |
oh nice, game jam! |
20:50 |
|
sys4 joined #minetest |
20:54 |
|
GreenXenith joined #minetest |
20:55 |
GreenXenith |
definitelya beat me to the punch, but Im still gonna announce it myself anyway :P |
20:55 |
definitelya |
:o go for it! |
20:56 |
GreenXenith |
Hey all! We are holding a 3-week Minetest GAME JAM starting December 1st! The idea is to get more small, self-contained experiences into the Minetest game collection. You might even win some money. More details at https://forum.minetest.net/viewtopic.php?t=27512 or https://www.reddit.com/r/Minetest/comments/qzvd3o! |
21:05 |
|
Hawk777 joined #minetest |
21:24 |
|
exe_virus[m] joined #minetest |
21:28 |
|
appguru joined #minetest |
22:03 |
|
AristotIe joined #minetest |
22:36 |
|
proller joined #minetest |
23:00 |
|
Sven_vB joined #minetest |
23:15 |
|
illwieckz joined #minetest |
23:46 |
|
illwieckz joined #minetest |