Minetest logo

IRC log for #minetest-dev, 2015-02-20

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

All times shown according to UTC.

Time Nick Message
00:34 Bet5 joined #minetest-dev
00:35 Bet5 hello
00:38 Bet5 I have a problem with schematic decorations on Mapgen v7, because the schematic don't replace existing nodes
00:39 Bet5 so i have e.g. trees floating on grass
00:43 Bet5 I try to add "force_placement = true" line in decoration definition, but this not help
00:43 Bet5 I suppose this is a bug?
00:45 compunerd joined #minetest-dev
01:24 T4im joined #minetest-dev
01:38 paramat joined #minetest-dev
01:42 paramat Bet5, i noticed this too, force placement is not usable when defining schematic decorations despite being usable with 'minetest.place_schematic'. if i can find time and can figure out how to, i will fix this
01:44 shadowzone joined #minetest-dev
01:47 Bet5 paramat, thanks, it will be nice if you fix this ;)
01:48 paramat although it's not really a bug because schematics were originally meant to not force place, the option was added to 'place schematic' on request from modders
01:51 Bet5 I noticed also, that MG v7 always place schematics not exactly on surface, but one node under
01:56 paramat yes i noticed that, then realised this is useful for jungletrees with wide root systems that are placed on a slope, it stops a root node floating above air
02:03 paramat i originally wanted to add roots to the future trees in mgv5/v7 but was told that the loss of dirt nodes caused when farming trees would be a problem for 'skyblock' type gaming
02:05 Bet5 btw, is it possible to place decorations on water? I tried to set ' place_on = "default:water_source" ' and decorations will not generate
02:05 Zeno` joined #minetest-dev
02:07 Bet5 this would be useful for water plants like lilies etc.
02:11 paramat not sure, would indeed be good
02:15 ShadowNinja VanessaE: For this release, in that particular way, no.  But a merge-based solution for this release and something closer to what he described for the next would be good.
02:16 paramat so ... after my merging mistake yesterday i decided i will need to practice the proper merging method within one of my mod repos, in the meantime i will be asking others to merge my PRs
02:20 paramat so please could anyone review and possibly merge #2175 ? these changes were approved by hmmmm. the corresponding MTgame commit just needs a quick edit then i will contact the MTgame team asap to get that merged
02:20 ShadowBot https://github.com/minetest/minetest/issues/2175 -- Mgv5: Remove blobgen, add large caves. De-link terrain level from water level. by paramat
02:24 deltib joined #minetest-dev
02:34 paramat ok game#412 is ready but will need review by the team
02:34 ShadowBot https://github.com/minetest/minetest_game/issues/412 -- mapgen.lua: Add dirt/sand/gravel blobs to mgv5/mgv7. Use blob ore for clay. Tune flowers by paramat
02:39 ShadowNinja ~tell nrzkt Um, what happened to my comments on #2327?
02:39 ShadowBot ShadowNinja: O.K.
02:42 Miner_48er joined #minetest-dev
02:47 paramat left #minetest-dev
03:20 false_chicken joined #minetest-dev
03:36 kahrl about the version issue, why not tag the "bump version" commit on master with a new tag, say 0.4.12-dev
03:37 VanessaE kahrl: at this point, you may as well wait for nrzkt's 3/4 patch and then tag it as 0.5.0
03:37 VanessaE s/tag/bump/
03:37 kahrl heh
03:38 Zeno` tag < dev/random
03:38 VanessaE ha
04:05 Zeno`` joined #minetest-dev
04:09 sebastian joined #minetest-dev
04:10 sebastian joined #minetest-dev
04:42 Hunterz joined #minetest-dev
05:14 Zeno`` I don't understand master's history
05:14 Zeno`` oh well
05:31 decimalguy left #minetest-dev
06:09 Megaf joined #minetest-dev
06:35 chchjesus joined #minetest-dev
06:36 blaze joined #minetest-dev
06:41 chchjesus joined #minetest-dev
06:47 Megaf joined #minetest-dev
06:53 alexxs joined #minetest-dev
06:55 Hunterz joined #minetest-dev
07:15 Brains joined #minetest-dev
07:33 nore joined #minetest-dev
08:10 nrzkt joined #minetest-dev
08:12 nrzkt ShadowNinja: i added a commit with your suggestions, but i don't know why your comments disappear :s
08:20 ImQ009 joined #minetest-dev
08:26 Zeno` joined #minetest-dev
08:26 Zeno` nrzkt, hi :)
08:26 nrzkt Zeno`: hi !
08:26 nrzkt you want to discuss about UT ? :p
08:27 Zeno` I can only stay a minute because I must fill water and stuff (big storm coming my way)
08:27 Zeno` yeah
08:27 nrzkt i see that on french news this morning
08:27 Zeno` I plan to fix the assert issue soon... I can do stage 1 tonight if you like
08:27 nrzkt which ASSERT issue ?
08:27 nrzkt you mean replacing ASSERT in unit tests ?
08:28 Zeno` assert() does not, currently, get replaced with no-op when NDEBUG is defined
08:28 Zeno` because it's used incorrectly throughout the minetest source code
08:28 Zeno` and therefore assert() must remain in release builds
08:29 Zeno` as for the unit tests... yes I think they can return a value in Release build but abort in Debug
08:29 Zeno` test.cpp is yucky
08:30 nrzkt ofc. I can patch my PR
08:30 Zeno` can you hold off for a few hours? I'll be back later. I just need to fill water containers and stuff in case I lose power
08:30 Zeno` when I get back I'll have time to discuss leisurely
08:31 nrzkt okay, but for re-enable unit tests in release, i fix my patch to re-enable them, okay ?
08:31 Zeno` yep sure
08:31 Zeno` and then we talk in a few hours :)
08:31 nrzkt no problem, good luck
08:31 Zeno` I need to prepare for the storm
08:31 Zeno` thanks :)
08:59 kilbith joined #minetest-dev
09:14 psedlak joined #minetest-dev
09:23 MattJ joined #minetest-dev
09:47 FR^2 joined #minetest-dev
10:01 gravgun joined #minetest-dev
10:29 nrzkt ~tell celeron55 openhub link on http://minetest.net doesn't work
10:29 ShadowBot nrzkt: O.K.
10:34 selat joined #minetest-dev
10:35 twoelk joined #minetest-dev
10:41 proller joined #minetest-dev
11:06 roniz joined #minetest-dev
11:13 Zeno` joined #minetest-dev
11:17 Zeno` nrzkt, save me some time... where are connections closed?
11:17 Zeno` sockets released and all that stuff
11:18 proller joined #minetest-dev
11:26 Zeno` what a $*&@($(8 mess
11:26 proller joined #minetest-dev
11:27 Zeno` https://github.com/minetest/minetest/blob/master/src/debug.h#L75
11:28 Zeno` who wrote that crap? :(
11:30 kilbith Zeno`, c6e3797c
11:34 proller joined #minetest-dev
11:44 Zeno` what is that commit
11:44 Zeno` anyway, it will be tedious to fix :(
11:53 crazyR joined #minetest-dev
11:53 crazyR joined #minetest-dev
11:54 MinetestForFun joined #minetest-dev
11:55 roniz_ joined #minetest-dev
12:45 SopaXorzTaker joined #minetest-dev
12:50 SopaXT joined #minetest-dev
13:06 Zeno` @2362
13:06 Zeno` #2362
13:06 ShadowBot https://github.com/minetest/minetest/issues/2362 -- Begin assert fix by Zeno-
13:11 Zeno` ^^ kahrl
13:11 Zeno` you should never have added that assert hack :p
13:40 leat joined #minetest-dev
13:40 shadowzone joined #minetest-dev
13:46 gravgun joined #minetest-dev
14:03 Taoki Woah, so many performance improvements in the latest GIT logs. I shall check them out when I can. Thanks Loic Blot :)
14:03 Taoki Also, I'm seeing a commit about a heavy network rewrite! Does this improve networking performance at last?
14:04 Taoki ed04e8e9e407f0dd57fa83a9732b3a3968cb80e0
14:05 celeron55 no; it's more about organization
14:05 celeron55 (oh also, i'm actually not here; don't even look at me)
14:06 Taoki ok
14:06 Taoki Anyway, my Creatures mod seems to be broken after the update. Mobs no longer move and find any targets
14:06 Taoki Maybe the find_near function broke? I saw a comment about that
14:07 Zeno` celeron55, where was the 0.4.12 release discussed? (it's not in the #minetest-dev logs)
14:08 Zeno` There was suddenly "celeron55 says it must be released today"
14:08 Zeno` that's the only mention of the sudden rush of activity
14:08 celeron55 i just wanted it to be out of the way from development standpoint
14:09 Zeno` I agree... but nobody seemed to know about it?
14:09 Taoki Can anyone please check if minetest.find_nodes_in_area and minetest.env:get_objects_inside_radius are still working properly?
14:09 celeron55 now it is; the actual publication of it is still TBD
14:09 celeron55 (i won't do it)
14:11 Zeno` well, it's released now. No point in crying over spilt milk I guess
14:11 Taoki Ok, they are actually. Something else in the recent comments broke my mod. Will have to see what
14:13 LazyJ joined #minetest-dev
14:13 shadowzone joined #minetest-dev
14:14 LazyJ I'd like to increase the size of the playername's above their avatars. What line of code is it and where?
14:16 Taoki Ok, the bug seems to happen when existing mobs are restored after the server restarts. Could there be a problem with persistence?
14:28 Taoki Found the bug. Apparently nested tables are no longer persisted properly on entities between server restarts.
14:29 Taoki For example: If a mob has the table property self.property.value = 1, and you persist self.property in staticdata, the next time the server starts it will be 0 and not remembered as 1.
14:29 Taoki Can someone please take a quick look? This is likely to break several mods, and already has mine.
14:31 roniz joined #minetest-dev
14:32 rubenwardy joined #minetest-dev
14:46 Taoki Yeah... the problem appears to be in set_staticdata. If in there you try to save self.table and it contains a self.table.something = 5 for instance, that self.table.something will not be saved. A recent change in GIT appears to have broken this.
14:47 Taoki Anyone around to take a look please?
14:52 fz72 joined #minetest-dev
14:58 fz72 maybe this bug was in 0.4.11, too. I had some problems with animals_modpack
14:58 Taoki It didn't happen for me over some 2 weeks ago
15:04 Taoki Ok, the problem is actually weirder. The table is seen as empty to begin with. Something is really wrong here...
15:10 LazyJ left #minetest-dev
15:14 Taoki Ok, the breakage seems to be in Lua itself.
15:14 Taoki abc = {one = {0, 1}, two = {0, 1}}
15:14 Taoki print("Count: "..#abc)
15:14 Taoki The above returns 0... which is not normal
15:16 kilbith joined #minetest-dev
15:25 Zeno` that seems normal
15:26 Zeno` because you don't use numbers it's not an array in the normal sense
15:27 Zeno` > abc = {one = {0, 1}, two = {0, 1, 3, 4}}
15:27 Zeno` > print("Count: "..#abc["two"])
15:27 Zeno` Count: 4
15:27 * VanessaE peeks in
15:27 Zeno` VanessaE can explain it better
15:27 VanessaE actually you did quite well :)
15:29 Zeno` not really :/ I expect that abc when the "indices" are not numbers are implemented as a hash array, or sparse array, so #abc would make no sense
15:29 Zeno` I dunno how to explain though
15:29 VanessaE use ipairs() if you need to iterate over a table using non-numeric keys
15:30 Zeno` yeah, it's a table... not an "array"
15:30 VanessaE # only returns the highest numeric key in the table, not actually the number of elements in it
15:31 VanessaE that's why it fails on non-numeric keys, as far as I know
15:31 Zeno` kind of :P
15:42 fz72 Taoki I think when you use tables as static data und have to serialize them. For me normal tables doesn't work
15:42 Taoki fz72: I believe I found the issue. It's to do with how my code converts certain tables. Fixing it now.
15:43 Taoki A change in Minetest has indeed introduced the breakage, but my code was doing some things the wrong way too.
15:48 fz72 joined #minetest-dev
15:50 SudoAptGetPlay joined #minetest-dev
15:55 alexxss joined #minetest-dev
16:15 SopaXT joined #minetest-dev
16:17 Miner_48er joined #minetest-dev
16:20 celeron55 Taoki: if there is an engine issue, make sure to submit an issue on github
16:39 ImQ009 joined #minetest-dev
16:57 Calinou joined #minetest-dev
17:22 Hunterz joined #minetest-dev
17:25 guest365 joined #minetest-dev
17:26 proller joined #minetest-dev
17:29 Calinou joined #minetest-dev
17:45 selat joined #minetest-dev
17:48 T4im joined #minetest-dev
18:08 MinetestForFun joined #minetest-dev
18:13 proller joined #minetest-dev
18:15 Krock joined #minetest-dev
18:42 Krock2 joined #minetest-dev
18:43 Calinou joined #minetest-dev
19:30 CraigyDavi joined #minetest-dev
19:34 Taoki Alright, I spent several hours fixing the related issues in my code. Sadly there is an issue in Minetest as well, and I shall submit an issue on Github.
19:34 casimir joined #minetest-dev
19:34 Taoki Apparently minetest.serialize -> minetest.deserialize no longer persist stacked tables properly.
19:36 ElectronLibre joined #minetest-dev
19:52 Fritigern joined #minetest-dev
19:56 paramat joined #minetest-dev
19:57 ElectronLibre joined #minetest-dev
20:02 Fritigern joined #minetest-dev
20:03 casimir Are biomes broken for everyone or just me?
20:04 casimir There should be no sand at all. Also, the trees get placed on water and sand. https://i.imgur.com/DdkfFxs.jpg
20:05 paramat hi, please can someone merge #2175 along with game#412 ? i think it's best i practice the merge method before attempting it myself
20:05 ShadowBot https://github.com/minetest/minetest_game/issues/412 -- mapgen.lua: Add dirt/sand/gravel blobs to mgv5/mgv7. Use blob ore for clay. Tune flowers by paramat
20:05 ShadowBot https://github.com/minetest/minetest/issues/2175 -- Mgv5: Remove blobgen, add large caves. De-link terrain level from water level. by paramat
20:06 paramat there have been changes to the biome API
20:07 Taoki I reported the issue regarding minetest.serialize. If anyone can please take a look into it: https://github.com/minetest/minetest/issues/2365
20:09 paramat casimir is your problem with the mgv5/v7 biome API?
20:09 casimir I know paramat. This is what the biome definition looks like: http://pastebin.com/TD5jL4NG
20:10 casimir Yes the API. v6 is unchanged.
20:12 casimir e.g. in the first biome, there is no sand defined at all, but when I increase height_shore then all the shore is sand.
20:13 est31 joined #minetest-dev
20:13 paramat yep there is new stuff in the API which defaults to sand on the shore if left out
20:14 paramat link coming ..
20:14 est31 paramat, you know mapgen, can you give me an answer to my question in https://github.com/Uberi/Minetest-WorldEdit/pull/74 ?
20:14 paramat yep will do that later
20:14 T4im didn't hmmmm add that already to the engine, est31?
20:15 est31 T4im, yes but you have to specify the coords manually
20:15 T4im ah
20:15 T4im so you use the mapgen region, nice
20:15 T4im worledit region*
20:15 est31 this duplication already exists elsewhere too
20:16 paramat casimir, https://forum.minetest.net/viewtopic.php?f=3&amp;t=10860
20:17 casimir oh, it seems I missed the node_shore_top.
20:18 paramat also, the biome API will soon simplify a little, that's my next job. there will only be 'node underwater' below 'shore height', shore nodes are too complex
20:21 paramat 'maximum extent for cavegen and mud reflow' is the 80^3 mapchunk plus a shell of mapblocks, so 112^3
20:21 paramat ^ est31
20:22 est31 paramat, thanks :)
20:33 twoelk joined #minetest-dev
20:35 paramat left #minetest-dev
20:43 ImQ009 joined #minetest-dev
20:53 ImQ009 joined #minetest-dev
20:59 Fritigern joined #minetest-dev
21:04 ElectronLibre joined #minetest-dev
21:10 domtron joined #minetest-dev
21:10 chchjesus joined #minetest-dev
21:12 ElectronLibre_ joined #minetest-dev
21:14 Fritigern joined #minetest-dev
21:19 indriApollo joined #minetest-dev
21:24 Miner_48er joined #minetest-dev
21:26 ImQ009 joined #minetest-dev
22:06 Fritigern joined #minetest-dev
22:15 shadowzone joined #minetest-dev
22:21 Calinou http://dev.minetest.net/Mapgen_Parameters
22:21 Calinou isn't this outdated? kwolekr said he is no longer involved in Minetest
22:22 Calinou also indev is gone
22:22 Calinou and there is v5 now
22:32 roniz joined #minetest-dev
22:37 domtron joined #minetest-dev
22:53 paramat joined #minetest-dev
22:54 paramat it is yes, as usual
22:55 Player_2 joined #minetest-dev
23:33 MattJ joined #minetest-dev
23:34 domtron joined #minetest-dev
23:34 Fritigern joined #minetest-dev
23:57 paramat okay i seem to have got my head around how to merge a PR by the proper method, i will merge #2175 in a few hours
23:57 ShadowBot https://github.com/minetest/minetest/issues/2175 -- Mgv5: Remove blobgen, add large caves. De-link terrain level from water level. by paramat

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