Minetest logo

IRC log for #minetest, 2019-06-06

| Channels | #minetest index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:22 Daisae paramat What seed had the papyrus?
00:59 paramat hang on ...
01:00 illwieckz joined #minetest
01:01 paramat seed 1468216727764965012 at x 275 z -314
01:48 Copenhagen_Bram joined #minetest
01:53 Cornelia joined #minetest
02:05 twoelk|2 joined #minetest
02:10 Jordach joined #minetest
02:13 ircSparky im guessing the luajit package in apt is source only? or should i just need that to compile minetest with LuaJIT?
02:29 Coder12 joined #minetest
02:32 cheapie ircSparky: For compiling MT, you want libluajit-5.1-dev
02:35 ircSparky ok
02:37 ircSparky but i still have to compile luaJIT, right?
02:38 cheapie No, libluajit-5.1-dev will pull in all the precompiled LuaJIT stuff you need.
02:38 ircSparky oh, cool, thanks
02:38 ircSparky so then I just compile minetest?
02:39 cheapie If you have the other dependencies installed, then yes, cmake and make for MT would be the next steps.
02:53 paramat joined #minetest
02:53 nri joined #minetest
03:10 YuGiOhJCJ joined #minetest
03:23 Coder12 joined #minetest
03:24 DI3HARD139 joined #minetest
03:25 turtleman joined #minetest
04:10 lllI1I joined #minetest
04:15 Fritigern joined #minetest
04:27 DI3HARD139 joined #minetest
04:51 FrostRanger joined #minetest
05:26 est31 joined #minetest
05:53 Copenhagen_Bram joined #minetest
06:14 aheinecke joined #minetest
06:28 jluc joined #minetest
06:31 CWz joined #minetest
07:01 andyo joined #minetest
07:33 YuGiOhJCJ joined #minetest
07:53 FrostRanger joined #minetest
07:57 orwell96_mt joined #minetest
08:12 Daisae I also find papyrus with Carpathian seed 146821672776496501. That map is lacking ice, but otherwise looks nice to me.
08:18 Daisae When I use minetest with fullscreen=true, I have not had the slowness happen.  clavi,Copenhagen_Bram
09:24 pyrollo joined #minetest
09:45 proller joined #minetest
10:05 Fixer joined #minetest
10:45 clavi strange...
10:50 Daisae I want placed leaves to somehow disappear.  They could decay or be forgotten as dropped objects are.  Is there a simple, easy way to have that happen?
10:53 Daisae I created a closed semisphere of cobbelstone and underneath it there was still light.  Is that because I have parameter mg_flags=light ?
11:08 proller joined #minetest
11:12 orwell96_mt joined #minetest
11:13 ensonic joined #minetest
11:28 lllI1I joined #minetest
11:42 Cornelia joined #minetest
12:32 Daisae No, removing mg_flags=light did not let an enclosure make it dark.
12:37 Daisae I find settings lightness sharpness and darkness sharpness , but no user docs.
12:50 Daisae There is a comment about 'gradient' I guess that would be the slope of the curve at that point.
12:59 micheal65536 joined #minetest
13:00 micheal65536 i'm looking through the crafting recipes in the default mod and i just noticed that there's no cooktime specified for smelting ores (e.g. iron lump into steel ingot)
13:00 micheal65536 does anyone know what the default cooktime is?
13:04 micheal65536 left #minetest
13:39 BuckarooBanzai Cool: https://gaming.stackexchange.com/questions/tagged/minetest :D 1 question... :P
13:42 cdde joined #minetest
14:12 jluc rubenwardy, minetest is invited to join kde :
14:12 jluc https://phabricator.kde.org/T10034
14:13 gameri joined #minetest
14:36 Coder12 joined #minetest
14:51 Daisae how does one sprint using mod hbsprint ?
15:08 LMD joined #minetest
15:08 Daisae clavi: What 'Rendering backend' have you selected for compositing?  I had OpenGL 3.1.  Since I switched to XRender, I have no big slowdown.
15:17 orwell96_mt joined #minetest
15:41 Daisae Do any bee mods work with current minetest?
16:05 bingfengfsx joined #minetest
16:22 puzzlecube joined #minetest
16:31 bingfengfsx joined #minetest
16:32 bingfengfsx joined #minetest
16:38 Krock joined #minetest
16:42 proller joined #minetest
16:48 ensonic joined #minetest
16:55 nowhere_man joined #minetest
16:56 jluc joined #minetest
17:05 Volgaar joined #minetest
17:14 MinetestBot [git] SmallJoker -> minetest/minetest: HTTP API: Allow binary downloads and headers (#8573) cb00632 https://git.io/fjzm8 (2019-06-06T17:13:29Z)
17:32 lisac joined #minetest
17:32 p_gimeno Daisae: forcing decay of placed leaves is a bit difficult, because normally you need to remove a trunk node to start the process, otherwise it doesn't happen
17:32 Daisae p_gimeno: what about making them be forgotten, as dropped objects are?
17:33 Krock sneak-place them
17:33 Daisae Is sneak-place a thing already?
17:33 Krock yes, they will decay
17:33 Krock sauce: https://github.com/minetest/minetest_game/blob/master/mods/default/functions.lua#L402
17:33 Daisae I always sneak place and never notice things disappear.
17:33 Daisae though it is usually not leaves
17:34 Krock oh wait. it should be param2=0
17:34 p_gimeno Krock: by my reading of the code, it needs param2 = 0 (which is the easy part) and something to start the node timer (which is the difficult part)
17:34 Krock https://github.com/minetest/minetest_game/commit/3a55e1905
17:34 Krock p_gimeno: yeah
17:35 Daisae I mainly place leaves underneath me, while quickly jumping up repeatedly.
17:36 p_gimeno I guess you could add code to start the timer on place
17:36 Krock 1) overwrite leave_node_def.after_place to start a timer
17:36 Krock 2) profit
17:37 Krock cache the old after_place and run it after your timer setup to ensure you don't overwrite anything by accident
17:45 p_gimeno Daisae: https://notabug.org/pgimeno/Gists/src/minetest--mod-force-leaf-decay
17:46 Krock missing_license.jpeg
17:46 Daisae Thanks! I'll give it a try.
17:46 p_gimeno it's untested, it may not work at all
17:47 Krock jk. looks good except that you should cache the old default.after_place_leaves
17:47 Krock nvm
17:47 Krock question is whether any other mod would change the after_place callback
17:48 Krock for speed reasons, swap_node could be used instead of set_node
17:48 p_gimeno that applies to minetest_game too
17:49 Krock I wonder whether this would lock up itself due to an after_place loop
17:49 Krock p_gimeno: how? default registers it
17:50 p_gimeno https://github.com/minetest/minetest_game/blob/master/mods/default/functions.lua#L402
17:50 p_gimeno it uses set_node as well
17:51 p_gimeno all that mod does is adding a set timer and removing the condition to set param2
17:52 Daisae Is a mod.conf necessary?
17:52 p_gimeno I used depends.txt, you can use mod.conf instead
17:53 p_gimeno either way, it's essential that it depends on default
17:54 Fritigern joined #minetest
17:56 p_gimeno where is mod.conf documented? I haven't looked into the format yet
17:56 rubenwardy lua_api.txt
17:57 rubenwardy you should use mod.conf instead of depends.txt and description.txt if you're targetting 5.0.0+ only
17:57 p_gimeno ah thanks
17:57 p_gimeno I hadn't read the format of mod.conf yet so I was still using depends.txt
18:02 p_gimeno Daisae: as for your lighting problems, it depends on what nodes you used for the enclosure. If you used all 1x1 cubic blocks, that should not happen. But if you used e.g. slabs, that's expected to happen.
18:04 Coder12 joined #minetest
18:06 Krock hmm okay. Maybe then it's place_node which would call item_node_place again
18:13 pauloue joined #minetest
18:14 pauloue joined #minetest
18:17 Daisae I used multiple layers of full cobble blocks.
18:17 Daisae also in rectangular prisms / cubic
18:18 Daisae p_gimeno: the leaves do not decay
18:19 p_gimeno Daisae: they don't decay for me either, I'm debugging
18:19 p_gimeno rectangular prisms?
18:20 p_gimeno if something is not a 1x1x1 cube, that may be the culprit.
18:20 p_gimeno even when it is, some nodes propagate light
18:21 pauloue_ joined #minetest
18:23 Daisae p_gimeno: I mean that I used worldedit to place cobblestone cube blocks using a function i think it called 'cubic'
18:23 Daisae 5 flat surfaces enclosing a space above ground.
18:24 p_gimeno ahh
18:24 p_gimeno try //fixlight  maybe
18:24 Daisae I will try that.
18:25 Daisae I think there are light maps. I wonder whether it is that map and not updating it.
18:25 Daisae I see there are lightness and darkness sharness also. Could those be so unsharp or diffuse that light seeps through?
18:26 FreeFull joined #minetest
18:26 p_gimeno yeah, I don't think WorldEdit does a good job updating light
18:27 p_gimeno I've found the reason for the leaves to not decay, and it's irritating
18:27 p_gimeno basically, the function gets hardcoded into the nodes, it's not called dynamically if it's overridden
18:31 p_gimeno let me work around that...
18:32 p_gimeno this would solve it in MTG: http://www.formauri.es/personal/pgimeno/pastes/call_after_place_leaves.patch
18:33 p_gimeno but I'm working on the mod to make it work with the unmodified MTG
18:35 puzzlecube joined #minetest
18:36 Daisae Is that patch something you would want to incorporate into the next version of MTG?  If so, I think it is OK to rely upon it.
18:36 BuckarooBanzai7 joined #minetest
18:40 nerzhul joined #minetest
18:40 sy hi, seems like flicking though ui takes about 2s/page, with the last page thats only 1/4 full taking much less
18:41 p_gimeno Daisae: not my call. I'm not on Microsoft GitHub any more so I can't submit it as a PR. Maybe someone else in here wants to do that.
18:43 Krock make it a patch so that I can give you the credit for that
18:43 p_gimeno Daisae: updated https://notabug.org/pgimeno/Gists/src/minetest--mod-force-leaf-decay/init.lua
18:44 p_gimeno (and tested)
18:44 Krock (file is .patch but that's a diff)
18:46 p_gimeno Krock: well, .patch vs .diff is a distinction introduced by git, before that they were the same thing and I'm old school :)
18:46 p_gimeno actually probably by github
18:46 Krock til. it's a nice addition, though.
18:47 p_gimeno ok, let me commit it and make it a git-am-style patch
18:47 Krock git format-patch HEAD~1  ?
18:51 jas_ joined #minetest
18:53 p_gimeno http://www.formauri.es/personal/pgimeno/pastes/call_after_place_leaves.patch (same link, just reload)
18:56 Krock s/Author/Name/
18:56 Krock s/Name/From/
18:59 Krock game#2388
18:59 ShadowBot https://github.com/minetest/minetest_game/issues/2388 -- Allow overriding default.after_place_leaves easily by SmallJoker
19:07 fwhcat joined #minetest
19:10 Nezrok joined #minetest
19:27 Tux[Qyou] joined #minetest
19:33 p_gimeno Krock: thanks, and sorry about the format
19:34 p_gimeno that was the output of git log, I thought it was compatible
19:34 p_gimeno I've never worked with git am
19:35 scr267 joined #minetest
19:38 Cornelia joined #minetest
19:39 p_gimeno I'm wondering whether it would be a good idea to do 'return default.after_place_leaves...' so it counts as a tail call
19:43 Krock I don't think that makes a noticeable
19:43 Nezrok Are there any mapgens that have a world created "upside down", so to speak? Like all the regular mountains/hills/trees/etc, but flipped over?
19:44 Nezrok Might be a silly question, but I think it might be interesting to see
19:46 Krock p_gimeno: applied the change, also a note so that nobody tries to optimize that back again
19:48 p_gimeno ah good idea
19:50 awell joined #minetest
19:52 p_gimeno hm, it looks like there are more instances of that problem (with a different callback)
19:55 Pie-jacker875 joined #minetest
20:02 p_gimeno argh, I left out several because I did it on 0.4.17 and then ported it to master
20:05 p_gimeno ok, after going through the file, the only other event is default.grow_sapling
20:12 Daisae For the minetestmapper colors, there are 5 columns. I can guess what 4 would be, but what is 5 ?
20:13 Daisae for example:  default:water_source 39 66 106 128 224
20:13 p_gimeno Krock: http://www.formauri.es/personal/pgimeno/pastes/call_after_place_leaves_and_grow_sapling.patch
20:15 proller joined #minetest
20:18 Copenhagen_Bram joined #minetest
20:19 Pie-jacker875 so my server isn't starting anymore
20:19 Pie-jacker875 pipeworks is giving me some issue
20:20 fwhcat hi
20:20 Kray joined #minetest
20:20 fwhcat what issue Pie-jacker875 ?
20:21 Pie-jacker875 what():  Runtime error from mod 'pipeworks' in callback on_shutdown(): /var/lib/minetest/.minetest/mods/pipeworks/luaentity.lua:49: bad argument #1 to 'pairs' (table expected, got nil)
20:21 ShadowBot https://github.com/minetest/minetest/issues/1 -- GlowStone code by anonymousAwesome
20:21 Pie-jacker875 sorry for long line
20:21 Pie-jacker875 ah cool
20:22 p_gimeno Daisae: the code suggests it's r,g,b, alpha and thickness, not sure what thickness is about though
20:22 fwhcat no, don't bother about it it's unrelated to your issue.
20:22 Pie-jacker875 well cool that it tries anyways lol
20:23 p_gimeno Pie-jacker875: that's on callback on_shutdown, meaning Minetest is already shutting down, so the previous log entries would be more interesting
20:23 Pie-jacker875 hang on
20:23 fwhcat it's the # followed by a number, the bot considers it as an issue number on github.
20:24 Krock patch applied, PR updated
20:24 p_gimeno Krock: thanks
20:24 Krock thank you too
20:24 fwhcat Krock & p_gimeno this patch will really have an impact on performances ?
20:25 Krock not really, no.
20:26 Pie-jacker875 https://p.teknik.io/WSyv2
20:27 Pie-jacker875 maybe something wrong with the world
20:29 Pie-jacker875 or mismatched versions of mods
20:32 nowhere_man joined #minetest
20:32 Nezrok Tried disabling the mod and then running /clearobjects ingame?
20:32 p_gimeno Pie-jacker875: hmm, looked into it
20:33 fwhcat or check the content of the file luaentities in your world folder
20:34 p_gimeno Pie-jacker875: could you please add a line before line 49 in /var/lib/minetest/.minetest/mods/pipeworks/luaentity.lua ?
20:34 Pie-jacker875 luaentities file has return {}%   in it
20:35 p_gimeno Pie-jacker875: please add the following line between lines 48 and 49: if not luaentity.entities then return end
20:35 p_gimeno that error may be masking the real error, we need to know which error it actually is
20:37 p_gimeno the % is suspicious though
20:42 Pie-jacker875 joined #minetest
20:42 Pie-jacker875 https://p.teknik.io/55xm3
20:44 p_gimeno ok, that's the error that was being masked, thanks
20:44 Daisae In a MapGen, does a thread get assigned a task to generate a map chunk or map block ?
20:47 p_gimeno 2019-06-06 16:36:05: ERROR[Main]: /builddir/build/BUILD/minetest-5.0.0/src/serverenvironment.cpp:e5: std::string LBMManager::createIntroductionTimesString(): A fatal error occurred: attempted to query on non fully set up LBMManager
20:47 p_gimeno Pie-jacker875: sanity check, how many num_emerge_threads do you have?
20:48 Pie-jacker875 I'm retarded, how do I check?
20:48 p_gimeno in minetest.conf
20:48 p_gimeno if you haven't defined that, then probably the default (1)
20:48 Pie-jacker875 yeah it's default
20:49 Pie-jacker875 should I change it
20:49 p_gimeno no, it's OK
20:49 Pie-jacker875 it's all on an i5-4690k, so quad core
20:51 p_gimeno Daisae: I believe map chunks are enqueued, but that's paramat's area of expertise
20:51 p_gimeno Pie-jacker875: don't worry, it's just not safe to have more than 1 emerge thread because there are several bugs in that area
20:51 Pie-jacker875 ok
20:53 p_gimeno Pie-jacker875: that error is beyond my head, I have no idea what can possibly cause it, I don't know that part of the code
20:54 p_gimeno it could be triggered by some mod, but not sure which. One that sets up an LBM.
20:54 Pie-jacker875 I tried running the world in singleplayer and it caused a crash
20:55 puzzlecube joined #minetest
20:55 Pie-jacker875 seems to be similar, with LBM stuff
20:58 p_gimeno suggestion: back up your world.mt file (located in the world's folder) and then start disabling mods to see if the problem goes away
21:00 paramat joined #minetest
21:01 Pie-jacker875 if I disable every mod it still gives the same sort of erro
21:02 Pie-jacker875 error
21:02 paramat Daisae core mapgen runs in it's own thread yes, lua mapgens do not and so cause server lag
21:03 Daisae paramat: I was wondering how the work was divided into separate threads.  I guess a thread gets either a chunk or mapblock, but which?
21:03 Daisae ... gets to generate..
21:03 p_gimeno paramat: what? I was convinced Lua mapgens ran in the emerge thread
21:05 p_gimeno Pie-jacker875: interesting. There are several auxiliary files in the world folder, one of them might have pending LBMs or something like that. It would be interesting if you could find out, and if so, paste the file.
21:05 p_gimeno Listing the content might help.
21:06 p_gimeno I mean, showing us a listing of the world files
21:06 p_gimeno does anyone know if there are permanent LBMs, and if so, where are they stored?
21:07 Pie-jacker875 https://p.teknik.io/dxM9M
21:08 Pie-jacker875 wait a minute
21:08 Pie-jacker875 I think it's a permission issue
21:08 p_gimeno oooh
21:08 Pie-jacker875 I have files owned by root in my mods folder
21:09 Pie-jacker875 nope still crashes
21:09 Pie-jacker875 wait no I forgot the recursive flag
21:09 Pie-jacker875 okay nevermind still crashes
21:11 Fritigern joined #minetest
21:14 p_gimeno it would be more worrying if these files were in your world folder
21:14 p_gimeno so my first guess on the LBMs is env_meta.txt
21:17 Pie-jacker875 https://p.teknik.io/gNk9O
21:17 paramat well, i'm not an expert on threading, but lua mapgens cause server lag and interrupt lua code, so i thought they run in the main server thread. there are requests to run lua mapgen in a separate thread but that is difficult
21:20 paramat Daisae mapgen works on a mapchunk at a time because that is optimal
21:27 p_gimeno paramat: I'd have to check in more detail, but I do believe that Lua mapgens do run in the corresponding emerge threads, and if they interrupt the main thread it's because of a GIL https://en.wikipedia.org/wiki/Global_interpreter_lock
21:28 p_gimeno Pie-jacker875: bingo, could you back up the file then remove the line that says "lbm_introduction_times = ..." and try again?
21:32 paramat ah locking of course
21:32 paramat yes very likely
21:32 Pie-jacker875 oh uhm
21:33 Pie-jacker875 that file was actually only on the copy of the world I made for singleplayer
21:33 Pie-jacker875 it's 0 bytes on the server
21:33 Pie-jacker875 there is a map_meta.txt though
21:34 p_gimeno paramat: https://github.com/minetest/minetest/issues/3182#issuecomment-148986974
21:34 paramat thanks
21:36 p_gimeno Pie-jacker875: map_meta.txt should not matter for this error. Try backing up and deleting forceload_blocks.txt?
21:36 Pie-jacker875 force_loaded.txt is it?
21:37 p_gimeno er, yes, sorry
21:37 Pie-jacker875 no luck there, still crashes
21:38 p_gimeno with all mods disabled, I don't see any other potentially problematic file. I'm at a loss at what can be causing the error. Looks like something in the map itself.
21:39 Pie-jacker875 want me to upload it
21:39 Pie-jacker875 I'll throw it on my website with the player auth files removed
21:40 p_gimeno how large is map.sqlite?
21:40 Pie-jacker875 800MiB
21:40 p_gimeno mmm ok, it's about acceptable for a test
21:41 Pie-jacker875 I'll zip it
21:53 Pie-jacker875 https://piejacker.net/garbage/world.7z
21:56 p_gimeno "11 minutes remaining"
21:57 Pie-jacker875 oh good I was worried my internet speed was like 10 times worse
21:58 Pie-jacker875 good thing I did compress it lol
21:59 marisag joined #minetest
22:11 Pie-jacker875 is it still going
22:12 Pie-jacker875 oh nevermind I see it
22:12 p_gimeno 2m :)
22:12 Pie-jacker875 I have to head out shortly, I'll be back in like 3 hours
22:12 p_gimeno I may not be around by that time. Remind me tomorrow, please.
22:13 Pie-jacker875 right
22:13 Pie-jacker875 thanks for offering to help
22:14 p_gimeno no problem, not that I have been able to help much anyway
22:15 p_gimeno and it's here
22:15 Pie-jacker875 you've offered your time though
22:15 Pie-jacker875 at least a bit of it
22:15 Pie-jacker875 k well I'
22:15 Pie-jacker875 I'm off then
22:15 p_gimeno have a nice RL :)
22:18 nri joined #minetest
22:18 cdde joined #minetest
22:25 paramat joined #minetest
22:40 Taoki joined #minetest
22:47 pauloue_ joined #minetest
23:04 proller joined #minetest
23:26 awell left #minetest
23:34 Coder12 joined #minetest
23:45 turtleman joined #minetest
23:49 MistahDarcy joined #minetest

| Channels | #minetest index | Today | | Google Search | Plaintext