| Time |
Nick |
Message |
| 01:25 |
|
v-rob joined #minetest |
| 01:27 |
|
jadzia joined #minetest |
| 02:01 |
|
YuGiOhJCJ joined #minetest |
| 02:26 |
|
queria joined #minetest |
| 02:27 |
|
Unit193 joined #minetest |
| 02:57 |
|
Gustavo6046 joined #minetest |
| 03:58 |
|
aldo joined #minetest |
| 05:42 |
|
Flabb joined #minetest |
| 06:08 |
|
CWz joined #minetest |
| 07:34 |
|
specing_ joined #minetest |
| 07:52 |
|
Kimapr7 joined #minetest |
| 07:57 |
|
beanzilla joined #minetest |
| 07:59 |
|
beanzilla joined #minetest |
| 07:59 |
|
SwissalpS joined #minetest |
| 08:52 |
|
absurb joined #minetest |
| 09:05 |
|
Fixer joined #minetest |
| 09:22 |
|
calcul0n_ joined #minetest |
| 09:41 |
|
Flabb_ joined #minetest |
| 10:00 |
|
Kimapr0 joined #minetest |
| 10:04 |
|
Kimapr5 joined #minetest |
| 10:09 |
|
delta23 joined #minetest |
| 10:11 |
|
Kimapr8 joined #minetest |
| 10:12 |
|
entuland joined #minetest |
| 11:23 |
|
Flabb joined #minetest |
| 12:02 |
|
Wuzzy joined #minetest |
| 12:14 |
|
specing joined #minetest |
| 12:32 |
|
entuland joined #minetest |
| 12:49 |
|
erlehmann joined #minetest |
| 12:51 |
|
Gustavo6046 joined #minetest |
| 13:06 |
|
Gustavo6046 joined #minetest |
| 13:14 |
|
calcul0n__ joined #minetest |
| 13:17 |
|
Gustavo6046 joined #minetest |
| 13:27 |
|
Gustavo6046 joined #minetest |
| 13:49 |
|
Yoda_ joined #minetest |
| 13:52 |
|
Gustavo6046 joined #minetest |
| 13:59 |
|
kamdard joined #minetest |
| 14:00 |
|
Gustavo6046 joined #minetest |
| 14:13 |
|
Gustavo6046_ joined #minetest |
| 15:11 |
|
j1233 joined #minetest |
| 15:34 |
MTDiscord |
<Lone_Wolf> Are particle velocities in nodes/s? |
| 15:39 |
|
Extex joined #minetest |
| 16:00 |
Krock |
@Lone_Wolf yes |
| 16:03 |
independent56 |
I have a 2 questions about advtrains. |
| 16:04 |
independent56 |
1 - why do routes refuse to set sometimes, forcing me to use the map to find the stopped trains andfind the front of the queue, and enable that one route? |
| 16:05 |
independent56 |
2 - Do railway maintainers on servers suffer this problem on a regular basis? |
| 16:05 |
independent56 |
3 - Does this happen in LF? |
| 16:05 |
independent56 |
4 - If not, why not? |
| 16:08 |
independent56 |
Forgive me, i lied about the quantity of questions |
| 16:15 |
MTDiscord |
<Lone_Wolf> Sweet, thanks |
| 16:15 |
|
Gustavo6046 joined #minetest |
| 16:17 |
Krock |
LF = linefeed |
| 16:17 |
Krock |
CR = carriage return ?? |
| 16:18 |
specing |
LF = linux forks |
| 16:23 |
|
sys4 joined #minetest |
| 16:24 |
independent56 |
Yeah, sorry. I am used to calling it "LF" because i wrote about the server a lot (in a degrading way) in a personal document. |
| 16:25 |
independent56 |
LF is entirely for trains. But my passion for trains caused me to overlook the quality of moderation there |
| 16:26 |
|
Talkless joined #minetest |
| 16:29 |
|
Gustavo6046 joined #minetest |
| 16:36 |
|
tuxifan joined #minetest |
| 16:40 |
|
Kimapr6 joined #minetest |
| 16:46 |
|
SwissalpS joined #minetest |
| 16:46 |
erlehmann |
i want to create an LBM that updates node meta with new content. is it a good idea to register the lbm name based on the new content i want to put in the node meta? |
| 16:47 |
SwissalpS |
a descriptive name is a good name :) |
| 16:49 |
Krock |
ignore comments and descriptive names. return to minimalism. use one-letter variables and compress your code! |
| 16:50 |
|
riff-IRC joined #minetest |
| 16:52 |
erlehmann |
SwissalpS, Krock i mean this so that if the content changes the node meta is always updated |
| 16:57 |
|
jonadab joined #minetest |
| 17:00 |
|
Hawk777 joined #minetest |
| 17:00 |
|
Gustavo6046 joined #minetest |
| 17:01 |
SwissalpS |
something tells me we are responding to a question you did not ask. In other words: I probably didn't quite understand exactly what you are asking |
| 17:04 |
MTDiscord |
<Warr1024> An LBM is triggered when a node is contained in a mapblock which is just loaded, i.e. from disk or mapgen. If what you want is to set meta every time a node is actually placed into the world, then that can be more or less complex, depending on the ways that node could be placed. |
| 17:05 |
MTDiscord |
<Warr1024> You can use an on_construct call to catch most ways a node can be placed; you miss only vmanips and fluid transform IIRC. An LBM is useful to update nodes that already existed in the map before you add the construction hook. |
| 17:06 |
|
j1233 left #minetest |
| 17:06 |
erlehmann |
yeah so |
| 17:07 |
erlehmann |
in this case mineclone2 has removed formspecs from node metas |
| 17:07 |
erlehmann |
but mineclonia has them there |
| 17:07 |
erlehmann |
you know, it's the lag friendly solution |
| 17:07 |
erlehmann |
so if you migrate a server from mineclone2 past commit 819dbc6224c3b96ad4094cccf3d9150f3ef61d45 to mineclonia, ender chests will not work |
| 17:08 |
erlehmann |
i want to introduce an LBM that sets the formspec for existing ender chests |
| 17:08 |
erlehmann |
it is, as Warr1024 correctly assumed, usually set when an ender chest is built |
| 17:09 |
SwissalpS |
yeah, sounds like lbm can help there. one time only needed, most likely. |
| 17:09 |
erlehmann |
well, the trick is: by making the name of the lbm depend on the formspec using minetest.sha1(formspec) (is that correct?), i want to ensure that every time the formspec is updated, a new lbm is created. |
| 17:10 |
erlehmann |
now i am looking for reasons to not do that |
| 17:10 |
MTDiscord |
<Warr1024> LBMs have to have a modname: prefix, but other than that, I don't see much problem with that idea |
| 17:10 |
MTDiscord |
<Warr1024> except in the case where they change the formspec to A, then to B, then back to A again |
| 17:11 |
SwissalpS |
one would need to carefully test that with every update |
| 17:11 |
MTDiscord |
<Warr1024> yeah, it feels like either manual versioning, or an external script that's a bit more complex than just the formspec contents itself, would be needed. |
| 17:11 |
SwissalpS |
you can also leave a trail in the meta of the node too. |
| 17:12 |
erlehmann |
wdym trail |
| 17:13 |
MTDiscord |
<Warr1024> Keeping track of past values in the node won't help you inform the engine of the need to run an LBM to find those nodes in the first place though. |
| 17:13 |
SwissalpS |
most likely the lbm that deletes meta does not delete all entries, only certain known ones. so your own meta would stay untouched. Alowing you to check it and other meta entries to determine what needs to be added. you could cache data in your own keys for example |
| 17:13 |
MTDiscord |
<Warr1024> So if they change the formspec from A to B to A2 then you'll need to retire the A and the B LBMs and have an A2 one to make sure that all the ones that have run both A and B go back to A again. |
| 17:14 |
erlehmann |
nah in this case the meta is pretty easy |
| 17:14 |
SwissalpS |
Warr: true, unless you have the lbm run on every load and look for specific node. |
| 17:14 |
MTDiscord |
<Warr1024> You ... frankly could just run on every load I guess :-) |
| 17:14 |
MTDiscord |
<Warr1024> Checking a meta string is not THAT expensive. |
| 17:14 |
SwissalpS |
it sounds like the simpler solution yes. |
| 17:15 |
MTDiscord |
<Warr1024> The wasted effort of rechecking a handful of nodes unnecessarily may be well worth it to avoid the headaches of having to track down every weird corner cases that upstream changes may trigger. |
| 17:15 |
SwissalpS |
+1 |
| 17:15 |
MTDiscord |
<Warr1024> Just (1) try the run_on_every_load option or whatever it's called, and then (2) don't worry about it again unless you encounter a problem with it, and then come back and ask again :-) |
| 17:16 |
MTDiscord |
<Warr1024> Ender chests should be rare enough that you wouldn't notice a problem anyway. Somebody would have to build an entire fortress out of the things before e.g. slow loading times from this should become a problem. |
| 17:17 |
SwissalpS |
dynamically creating lbms sounds like asking for trouble |
| 17:18 |
MTDiscord |
<Warr1024> I mean, modding MT is asking for trouble to begin with ... and automatic dynamic LBMs sounds interesting :-) |
| 17:18 |
MTDiscord |
<Warr1024> If you don't like asking for trouble then you don't want to see what I've done with ABMs :-) |
| 17:19 |
erlehmann |
Warr1024 show and tell |
| 17:20 |
erlehmann |
pls |
| 17:21 |
MTDiscord |
<Warr1024> Exhibit A: the engine caches info about ABMs for speed, but the cache strategy is completely disabled if there are more than like 63 ABMs. I have a game that has 82 ABMs. I realize that many of them have similar signatures from the engine's perspective, so I create a system for automatically multiplexing them that allows me to squash them into about 36 and make caching work again. |
| 17:21 |
MTDiscord |
https://gitlab.com/sztest/nodecore/-/blob/master/mods/nc_api_active/abmmux.lua |
| 17:23 |
MTDiscord |
<Warr1024> Exhibit B: ABM neighbor checks are expensive, so if you want an ABM to run when common node A is next to rare node B, it's more efficient to check for B next to A, but I have a number of ABMs already written in the first form, and it would be a pain to invert them across the whole mod ecosystem. Instead, I'd rather just add a "neighbor_invert" flag to the ABM API that does this for me in the most reasonable way I can: |
| 17:23 |
MTDiscord |
https://gitlab.com/sztest/nodecore/-/blob/master/mods/nc_api_active/abminvert.lua |
| 17:24 |
MTDiscord |
<Warr1024> There's probably other crazy stuff in there too. I used to have an auto-rate-limiting ABM system, but that ended up being stubbed out as ABM time budgeting constraints have evolved to make it no longer make sense. |
| 17:25 |
erlehmann |
Warr1024 what is the reason for the caching being disabled? |
| 17:40 |
|
Rafi59 joined #minetest |
| 17:48 |
robyndrake |
what version of lua is used with minetest, and where can I download such an interpretor? |
| 17:50 |
robyndrake |
please? |
| 17:51 |
Krock |
Lua 5.1 |
| 17:51 |
Krock |
with with JIT or without |
| 17:51 |
Krock |
luarocks might use it, dunno. |
| 17:52 |
MTDiscord |
<Warr1024> erlehmann: A single 64-bit integer is used as a bitmap to track the applicable/not-applicable state of an ABM within a given mapblock, and if there are too many ABMs to fit in that bitmap then it just gives up on the whole thing. I don't know exactly why it was done this way or what would have had to be done differently to allow it to expand dynamically. |
| 17:54 |
robyndrake |
will lua 5.1.5 work Krock, and thank you btw |
| 17:54 |
Krock |
robyndrake: yes |
| 17:54 |
robyndrake |
Thank you again |
| 17:55 |
Krock |
!next |
| 17:55 |
MinetestBot |
Another satisfied customer. Next! |
| 18:32 |
|
Gustavo6046 joined #minetest |
| 18:41 |
|
Fixer joined #minetest |
| 18:48 |
|
Boingo joined #minetest |
| 18:55 |
|
Gustavo6046 joined #minetest |
| 19:05 |
|
AristotIe left #minetest |
| 19:35 |
|
specing_ joined #minetest |
| 19:47 |
|
Gustavo6046 joined #minetest |
| 19:52 |
|
y5nw joined #minetest |
| 19:53 |
|
Gustavo6046 joined #minetest |
| 20:28 |
|
v-rob joined #minetest |
| 20:39 |
|
Extex joined #minetest |
| 20:48 |
|
riff-IRC joined #minetest |
| 20:52 |
|
Gustavo6046 joined #minetest |
| 20:52 |
|
Gustavo6046 joined #minetest |
| 20:59 |
|
Gustavo6046 joined #minetest |
| 21:01 |
|
Lone_Wolf joined #minetest |
| 21:18 |
|
Gustavo6046 joined #minetest |
| 21:20 |
|
Gustavo6046 joined #minetest |
| 21:25 |
erlehmann |
https://content.minetest.net/help/content_flags/ |
| 21:25 |
erlehmann |
the content flags are all for disgusting stuff |
| 21:26 |
erlehmann |
i guess buttplug.io interfacing would be banned :( |
| 21:26 |
erlehmann |
oh |
| 21:26 |
erlehmann |
i can just say it is for the xbox controller ^^ |
| 21:26 |
erlehmann |
of my … xbox |
| 21:26 |
erlehmann |
hehe |
| 21:31 |
|
Gustavo6046 joined #minetest |
| 21:35 |
MTDiscord |
<Warr1024> Not necessarily "disgusting" but "potentially objectionable". Tags are things you'd be looking to find, content flags are things you'd be looking to avoid. |
| 21:36 |
MTDiscord |
<Warr1024> I believe the de facto standard is that content with no content flags should be PEGI7 and content flags allow for specific exceptions to that standard where set. |
| 21:37 |
MTDiscord |
<Warr1024> Of course there are some packages that use them incorrectly, like a few that use the "horror" flag somewhat tongue-in-cheek. |
| 21:39 |
|
calcul0n__ joined #minetest |
| 21:45 |
|
Gustavo6046 joined #minetest |
| 21:47 |
|
Fixer_ joined #minetest |
| 21:49 |
|
Gustavo6046 joined #minetest |
| 21:52 |
|
Gustavo6046_ joined #minetest |
| 21:55 |
|
Extex joined #minetest |
| 22:00 |
|
Sven_vB joined #minetest |
| 22:04 |
|
Gustavo6046 joined #minetest |
| 22:25 |
|
Gustavo6046 joined #minetest |
| 22:35 |
Calinou |
Map Tools 2.2.0 released. Coin crafting recipes are now disabled by default: https://github.com/minetest-mods/maptools/releases/tag/v2.2.0 |
| 22:36 |
|
Gustavo6046 joined #minetest |
| 22:44 |
Calinou |
also More Blocks 2.2.0: https://github.com/minetest-mods/moreblocks/releases/tag/v2.2.0 |
| 22:44 |
Calinou |
and More Ores 2.1.0: https://github.com/minetest-mods/moreores/releases/tag/v2.1.0 |
| 23:01 |
|
Gustavo6046_ joined #minetest |
| 23:20 |
|
olliy joined #minetest |
| 23:22 |
|
olliy joined #minetest |
| 23:32 |
|
hecks joined #minetest |
| 23:33 |
|
Gustavo6046 joined #minetest |
| 23:39 |
MTDiscord |
<wwar> Nice updates :+1: |
| 23:46 |
|
Gustavo6046 joined #minetest |
