| Time |
Nick |
Message |
| 00:58 |
|
illwieckz joined #minetest |
| 01:21 |
|
ShadowBot joined #minetest |
| 01:48 |
|
Bombo joined #minetest |
| 01:48 |
|
Bombo joined #minetest |
| 02:16 |
|
sy joined #minetest |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:38 |
|
Verticen joined #minetest |
| 04:43 |
|
est31 joined #minetest |
| 04:47 |
|
TheSilentLink joined #minetest |
| 05:14 |
|
Talkless joined #minetest |
| 06:12 |
|
gregon joined #minetest |
| 06:12 |
|
gregon1 joined #minetest |
| 07:07 |
|
definitelya joined #minetest |
| 07:44 |
|
gregon joined #minetest |
| 07:52 |
|
TomTom_ joined #minetest |
| 08:12 |
|
lissobone joined #minetest |
| 08:38 |
|
jaca122 joined #minetest |
| 08:39 |
|
mrkubax10 joined #minetest |
| 08:54 |
|
s20 joined #minetest |
| 09:09 |
|
Verticen joined #minetest |
| 09:47 |
|
s20 joined #minetest |
| 10:32 |
|
Hanicef joined #minetest |
| 10:42 |
|
s20_ joined #minetest |
| 10:42 |
|
Verticen joined #minetest |
| 11:00 |
|
s20 joined #minetest |
| 11:02 |
|
proller joined #minetest |
| 11:33 |
|
cranez joined #minetest |
| 11:37 |
|
Desour joined #minetest |
| 11:45 |
|
qqq joined #minetest |
| 11:46 |
|
Glaedr joined #minetest |
| 11:46 |
mtvisitor |
!up wunderwelt.one:30000 |
| 11:46 |
MinetestBot |
wunderwelt.one:30000 is up (12ms) (IPv4) |
| 11:46 |
mtvisitor |
!up joesworld.zapto.org:30001 |
| 11:46 |
MinetestBot |
joesworld.zapto.org:30001 seems to be down (IPv6) |
| 11:47 |
|
sy joined #minetest |
| 11:48 |
|
zm joined #minetest |
| 11:49 |
|
sy joined #minetest |
| 12:23 |
|
ireallyhateirc joined #minetest |
| 12:31 |
|
lissobone joined #minetest |
| 12:31 |
lissobone |
Thank goodness, I can now join... |
| 12:33 |
|
dumpster_lord joined #minetest |
| 12:36 |
dumpster_lord |
Hi, I'm writing a city generator for my game using sfan5's new `minetest.register_mapgen_script` thing and I'd like to know if there's a way to pass data into that mapgen environment |
| 12:37 |
dumpster_lord |
the problem is I want my city mapgen to be moddable (users could define their own buildings, etc.) and the environment's isolation makes it quite hard because I can't access globals or read files |
| 12:38 |
dumpster_lord |
one solution I considered was passing data in VoxelManip data but sounds like a hack |
| 12:38 |
lissobone |
I am now, too, interested in this new procedure |
| 12:38 |
lissobone |
I want to know what does it accept as arguments |
| 12:39 |
dumpster_lord |
minetest.register_mapgen_script(path) |
| 12:39 |
|
gregon1 joined #minetest |
| 12:39 |
|
gregon2 joined #minetest |
| 12:40 |
dumpster_lord |
example: minetest.register_mapgen_script(mod_path.."/mapgen.lua") |
| 12:40 |
lissobone |
And "path" refers to a file containing some script? |
| 12:41 |
lissobone |
Something is sparking in my head, but cables do not connect, I need to know more in order to devise a solution |
| 12:41 |
lissobone |
I will test it |
| 12:41 |
dumpster_lord |
here's the original example: https://gist.github.com/sfan5/1a124eb496a39734f53fb4daffc47bee |
| 12:42 |
dumpster_lord |
I have a kind of working example in my game: |
| 12:42 |
dumpster_lord |
the init.lua file (mapgen script registered at the bottom): https://codeberg.org/lord_of_the_dumpster/perfect_city/src/branch/wip-roadgen/mods/pcity_mapgen/init.lua |
| 12:42 |
dumpster_lord |
and here actual mapgen: https://codeberg.org/lord_of_the_dumpster/perfect_city/src/branch/wip-roadgen/mods/pcity_mapgen/mapgen.lua |
| 12:43 |
dumpster_lord |
Steps to follow are: 1. register a mapgen script with minetest.register_mapgen_script(mod_path.."/foo.lua") |
| 12:44 |
dumpster_lord |
2. in the foo.lua make a mapgen function that accepts args: local function mapgen(vm, pos_min, pos_max, blockseed) |
| 12:44 |
lissobone |
I am reading, don't worry, I did not vanish' |
| 12:44 |
lissobone |
Without the trailing apostrophie! |
| 12:45 |
dumpster_lord |
3. at the bottom of foo.lua start the function minetest.register_on_generated(mapgen) |
| 12:45 |
dumpster_lord |
And that's it |
| 12:45 |
dumpster_lord |
you can use the vm object now |
| 12:46 |
sfan5 |
dumpster_lord: the current best way to do one-way data transfer is to serialize data, write it a file in the world folder and do the reverse in the mapgen script |
| 12:47 |
sfan5 |
(https://github.com/minetest/minetest/issues/14451) |
| 12:47 |
dumpster_lord |
sounds illegal but solves my problem, thanks :) |
| 13:00 |
|
sparky4 joined #minetest |
| 13:01 |
|
Lupercus joined #minetest |
| 13:23 |
|
Lupercus joined #minetest |
| 13:50 |
|
shaft joined #minetest |
| 13:51 |
shaft |
Can someone quickly review and merge my ts_furniture pull request? https://github.com/minetest-mods/ts_furniture/pull/20 Really trivil. Just fixes the camera and stops people from sitting on air. |
| 13:52 |
shaft |
It's been annoying me for a looong time now. |
| 14:10 |
|
flux__ joined #minetest |
| 14:22 |
|
flux__ joined #minetest |
| 14:46 |
|
silverwolf73828 joined #minetest |
| 15:01 |
|
proller joined #minetest |
| 15:02 |
sfan5 |
just did |
| 15:19 |
|
Lupercus joined #minetest |
| 15:25 |
|
bdju joined #minetest |
| 15:32 |
|
gregon joined #minetest |
| 15:55 |
shaft |
Okay I applied the suggestions |
| 15:55 |
shaft |
pls merge |
| 15:55 |
|
mrkubax10 joined #minetest |
| 16:11 |
|
gregon joined #minetest |
| 16:23 |
|
definitelya joined #minetest |
| 16:50 |
|
sys4 joined #minetest |
| 16:51 |
|
est31 joined #minetest |
| 16:51 |
|
fling joined #minetest |
| 16:52 |
|
f_ joined #minetest |
| 16:55 |
|
TomTom joined #minetest |
| 17:36 |
|
f_ joined #minetest |
| 17:39 |
|
Lupercus joined #minetest |
| 17:45 |
|
Lupercus joined #minetest |
| 18:02 |
BuckarooBanzai |
regarding #14451: would something like the mod-storage as common ground for async- and sync environment be "easy" to implement? |
| 18:02 |
ShadowBot |
https://github.com/minetest/minetest/issues/14451 -- Allow passing arbitrary data to mapgen Lua env (also at runtime) |
| 18:07 |
sfan5 |
the mod storage is 1. not thread safe currently 2. forces all writes and reads to go to disk/database |
| 18:14 |
|
Lupercus joined #minetest |
| 18:29 |
|
f_ joined #minetest |
| 18:52 |
MTDiscord |
<mistere_123> how hard would it be to implement a mod sql interface, where each mod gets its own database file, and the interface is asynchronous (so you can submit requests and get responses later without blocking the main thread? Would it be enough to include luasqlite on the async environment and restrict its access to the world and moddata folders? It would help mods like sban not need to be unsandboxed |
| 18:54 |
MTDiscord |
<mistere_123> Should I make an issue? |
| 18:54 |
sfan5 |
it's just lua bindings for sqlite, not very hard |
| 18:55 |
sfan5 |
however aside from permanently marrying the engine to sqlite this request sounds like "we use this external library in a mod and it's convenient, can you please copy it into the engine?" |
| 18:55 |
MTDiscord |
<wsor4035> dont make an issue |
| 18:55 |
MTDiscord |
<wsor4035> there is on already iirc |
| 18:56 |
MTDiscord |
<wsor4035> (assuming i recall correctly anyways) |
| 18:56 |
sfan5 |
the hard part might be sandboxing sqlite, in case it has support for e.g. 'INTO OUTFILE' |
| 18:56 |
sfan5 |
also I don't think sqlite is written to be secure against an attacker who can execute arbitrary SQL(?) |
| 18:57 |
MTDiscord |
<wsor4035> see https://github.com/minetest/minetest/issues/3374 |
| 18:57 |
|
mrkubax10 joined #minetest |
| 18:57 |
MTDiscord |
<wsor4035> also https://github.com/minetest/minetest/issues/10371 which is duplicate but has a few other thoughts |
| 18:59 |
MTDiscord |
<wsor4035> the tldr from the is the main two issues are security and if it should be db agnostic |
| 19:00 |
|
mrkubax10 joined #minetest |
| 19:00 |
MTDiscord |
<wsor4035> *....tldr of blockers from.... |
| 19:01 |
|
Talkless joined #minetest |
| 19:04 |
MTDiscord |
<wsor4035> personally commenting on the db agnostic thing, we already do have an issue with mods hard depending on luajit over stock lua in some cases, most notably relating to goto. there is common sql that you can write that would work across different backends, but it seems much easier that lua<>luajit to get yourself into a situation of hard depending on one sql backend |
| 19:06 |
MTDiscord |
<wsor4035> as for the security thing, could make it similar to the http api, where a mod has to be given permission to it? however in general that permission could probably be made much more ui/ux friendly |
| 19:10 |
|
mrkubax10 joined #minetest |
| 19:14 |
MTDiscord |
<mistere_123> Well, I dont care about whether it is sqlite, sfan5. Its not just a request to copy a library into the engine. It's that I have a usecase that I am developing where I need to use a full database solution. Currently the only option I am aware of is using luasqlite as a secure.trusted_mods mod. Aside from my present usecase, having full access to a full database seems like a generally useful feature. An imaginary usecase would be a lua |
| 19:14 |
MTDiscord |
mapgen mod that uses the async environment and the dummy backend to implement its own map storage system, so it could run water simulations in the background on the entire map, for example. |
| 19:16 |
MTDiscord |
<mistere_123> I guess I could use HTTP |
| 19:18 |
|
Thermoriax joined #minetest |
| 19:18 |
|
___nick___ joined #minetest |
| 19:22 |
Mantar |
no so imaginary: our mapchunk shepherd basically runs water simulation on the entire map, but unfortunately it's not in the background |
| 19:24 |
|
Verticen joined #minetest |
| 19:50 |
|
sys4 joined #minetest |
| 19:50 |
|
fling joined #minetest |
| 19:51 |
|
fling joined #minetest |
| 20:49 |
|
Lupercus joined #minetest |
| 22:03 |
|
jaca122 joined #minetest |
| 22:07 |
MinetestBot |
[git] OgelGames -> minetest/minetest: Hide internal detached inventory functions (#14594) a7bde8e https://github.com/minetest/minetest/commit/a7bde8e5237b28ff478e5585090620c670a8268e (2024-04-27T22:07:45Z) |
| 22:12 |
MinetestBot |
[git] appgurueu -> minetest/minetest: Fix set_fov-induced grayscreen 0837d67 https://github.com/minetest/minetest/commit/0837d674ebb098cb28e2573f36fd991e14bf2b3c (2024-04-27T22:12:15Z) |
| 22:14 |
MinetestBot |
[git] srifqi -> minetest/minetest: Translation updater: Add comment propagation (#14416) 815b5cb https://github.com/minetest/minetest/commit/815b5cb0867c2011be7031b74331b99de94326e4 (2024-04-27T22:12:30Z) |
| 22:14 |
MinetestBot |
[git] OgelGames -> minetest/minetest: Fix `InvRef` bugs and add unit tests (#14591) 05d5dc4 https://github.com/minetest/minetest/commit/05d5dc4cec37a64d01f1535c8e494a2a4a7e7036 (2024-04-27T22:13:44Z) |
| 22:14 |
MinetestBot |
[git] grorp -> minetest/minetest: Smooth scrolling (#14562) 73dbd2f https://github.com/minetest/minetest/commit/73dbd2f0abcc6528073af2790373878cfbaad756 (2024-04-27T22:14:31Z) |
| 22:25 |
MinetestBot |
[git] sfan5 -> minetest/minetest: Revert "Clean up some getServer() unsafety" df4c9e2 https://github.com/minetest/minetest/commit/df4c9e29034ed77c991d25af1891cd63e7518365 (2024-04-27T22:24:22Z) |
| 22:32 |
|
panwolfram joined #minetest |
| 22:58 |
|
Beauregard42 joined #minetest |
| 23:42 |
|
Lupercus joined #minetest |
| 23:52 |
|
Lupercus joined #minetest |
