Minetest logo

IRC log for #minetest-dev, 2016-05-10

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

All times shown according to UTC.

Time Nick Message
00:01 est31 joined #minetest-dev
00:04 est31 ~tell PilzAdam I see thanks for notifying me. I will merge the updated translations shortly, and reset the repo. This will solve the error.
00:04 ShadowBot est31: O.K.
00:06 est31 I'll merge translations one last time before release
00:06 est31 not now
00:07 est31 I'll lock the translation repo at 18:00 CEST, may 10
00:07 est31 translators should do their edits by then.
00:07 est31 then I'll push all the commits to a github branch, where translators can inspect them
00:07 est31 at 20:00 CEST I'll merge the translations into minetest master
00:08 est31 18:00 CEST, may 10 is approx. 16 hours from now
00:08 est31 I think merging translations was an open issue?
00:09 est31 nore, I think its better left non-translated.
00:22 STHGOM joined #minetest-dev
00:28 Wayward_One paramat, on a hunch, I tried a direct clone of MoNTE48's androidbuttons branch instead of a patch, and the volume buttons indeed work now
00:29 paramat good. thanks
00:29 Wayward_One np
00:34 paramat #4086 now looks potentially good on testing by others, anyone else want to review/approve? kahrl ShadowNinja
00:34 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
01:18 paramat will merge #4103 in a moment
01:18 ShadowBot https://github.com/minetest/minetest/issues/4103 -- Lua_api.txt: Fix documentation for facedir rotation by paramat
01:19 paramat merging
01:21 paramat cookie coming! =]
01:23 paramat merged
01:51 asl97 joined #minetest-dev
02:35 paramat left #minetest-dev
03:02 Zeno` joined #minetest-dev
05:04 asl97 left #minetest-dev
05:13 Hunterz joined #minetest-dev
05:53 bugzapper joined #minetest-dev
06:41 jin_xi joined #minetest-dev
06:54 nrzkt joined #minetest-dev
07:18 Megal_ joined #minetest-dev
08:17 Amaz joined #minetest-dev
08:27 bugzapper joined #minetest-dev
09:04 nrzkt joined #minetest-dev
09:55 rubenwardy joined #minetest-dev
10:22 crazyR joined #minetest-dev
11:08 betterthanyou710 joined #minetest-dev
11:10 betterthanyou710 void you mean adding in a new material
11:10 betterthanyou710 like dark mese
11:10 betterthanyou710 i think just purple
11:10 betterthanyou710 the yellow is a little bit weird for me
12:19 proller joined #minetest-dev
12:29 Fixer joined #minetest-dev
12:39 STHGOM joined #minetest-dev
12:40 PilzAdam joined #minetest-dev
12:46 naltamur joined #minetest-dev
12:47 naltamur hello
12:51 damiel joined #minetest-dev
13:22 naltamur1 joined #minetest-dev
13:28 naltamur1 are you ready for icon on website?
13:28 naltamur1 http://imgur.com/17lz9VM
13:31 naltamur1 left #minetest-dev
13:32 naltamur joined #minetest-dev
13:39 rubenwardy naltamur, heart doesn't match very well
13:39 naltamur rubenwardy: I know :)
13:40 naltamur It was a old screen
13:40 naltamur I dont know what icon I can put
13:41 rubenwardy maybe http://fortawesome.github.io/Font-Awesome/icon/globe/
13:43 naltamur okay I will add it
13:43 rubenwardy I'm not convinced on icons though, will have to see final product ;)
13:53 Amaz left #minetest-dev
14:31 iqualfragile joined #minetest-dev
14:35 KaadmY joined #minetest-dev
14:35 Void7 joined #minetest-dev
15:07 hmmmm joined #minetest-dev
15:08 Taoki joined #minetest-dev
15:31 DI3HARD139 joined #minetest-dev
15:33 RichardTheTurd joined #minetest-dev
15:39 Krock joined #minetest-dev
16:04 jin_xi joined #minetest-dev
16:12 rubenwardy #2704 and #4105 are duplicates of #1367
16:12 ShadowBot https://github.com/minetest/minetest/issues/2704 -- [feature request] text overlay on nodes
16:12 ShadowBot https://github.com/minetest/minetest/issues/4105 -- Add an API for in game text?
16:12 ShadowBot https://github.com/minetest/minetest/issues/1367 -- Proper display of text on the surface of a node(box)
16:12 rubenwardy #3528 is a partial duplicate
16:12 ShadowBot https://github.com/minetest/minetest/issues/3528 -- Runtime dynamic textures via mod API
16:19 naltamur tomorrow new version :) I'm excited
16:28 kaeza joined #minetest-dev
16:29 kaeza awww heck
16:29 kaeza is weblate locked again?
16:30 proller joined #minetest-dev
16:33 Hunterz joined #minetest-dev
16:33 naltamur How Can I have account for wiki?
16:34 kaeza Calinou, ^
16:34 est31 joined #minetest-dev
16:35 est31 okay please review: https://github.com/est31/minetest/commits/master
16:35 est31 (if there are any translators)
16:35 est31 i will shorten the commit count as far as i can by merging commits from same authors
16:36 est31 I'll push it in 1 hour 30 mins, perhaps a bit later
16:38 est31 I have left out *one* commit about the addition of canadian french translation
16:39 est31 I didnt see anything translated tehre
16:39 est31 there*
16:39 sofar +1
16:40 sofar I looked a bit at the Dutch translation... in general it looks great but the mapgen options are just walls of text
16:40 * sofar too lazy for that :)
16:41 est31 :)
16:42 DI3HARD139 joined #minetest-dev
16:43 kaeza can give +1 for spanish. that said, it sucks that it's at ~50% :/
16:51 hmmmm wow my quality of life has significantly improved
16:51 hmmmm i discovered git add -p
16:53 est31 that subcommand is really cool yeah
17:02 nrzkt joined #minetest-dev
17:05 est31 Okay pushed now a shortened version of the commit log https://github.com/est31/minetest/commits/master
17:05 est31 ten commits less!
17:05 est31 will push it in one hour
17:17 KaadmY joined #minetest-dev
17:22 Void7 joined #minetest-dev
17:27 proller joined #minetest-dev
17:38 Player_2 joined #minetest-dev
17:42 Void7 joined #minetest-dev
17:52 naltamur1 joined #minetest-dev
17:52 naltamur1 hello
17:53 naltamur1 have you got any suggestions for icons on website?
17:54 Krock what website?
17:55 naltamur1 minetest website
17:58 paramat joined #minetest-dev
18:06 loggingbot_ joined #minetest-dev
18:06 Topic for #minetest-dev is now Minetest core development and maintenance. Feature freeze since 2016-05-01, release on 11th. Last release: 0.4.13, Aug 20 2015. Chit-chat goes to #minetest. Consider this instead of /msg celeron55. http://irc.minetest.net/minetest-dev/ http://dev.minetest.net/
18:07 naltamur1 http://imgur.com/RVf7oK7
18:09 Darcidride naltamur1 => Beautiful :)
18:10 naltamur1 Darcidride: I'm ready to make a pull request
18:16 paramat we should drop the use of 'near-infinite' though
18:17 Fixer multiple subgames icon and modding api icon
18:18 Fixer modding api probably should have that puzzle icon
18:18 Fixer or not
18:18 Fixer btw, there are more than 3 generators now :)
18:18 Fixer 5/6/7/flat/mgvalleys
18:19 paramat mgfractal singlenode
18:19 paramat most have 10 biomes
18:20 paramat so 6 mapgens, singlenode is hidden and is empty for mod use
18:21 hmmmm i'd rather not advertise multiple mapgens in those words though
18:21 hmmmm maybe something more subtle and less technical like "explore different kinds of worlds"
18:22 hmmmm "choose from one of many terrain generation algorithms"
18:22 naltamur1 https://github.com/minetest/minetest.github.io/pull/61
18:24 Fixer agree with hmmmm
18:25 hmmmm i also lol @ "near-infinity"
18:25 hmmmm this is an oxymoron
18:25 hmmmm you can never be near infinity
18:25 Fixer and 65000 not impressive anymore it seems
18:25 Fixer people trevel even more far
18:26 sofar some on a daily basis :)
18:26 hmmmm the number isn't as impressive as minecraft's
18:26 naltamur1 okay
18:26 Fixer yeah
18:26 hmmmm but people don't ever travel that far, they teleport to the edge of the map and then whine about how there isn't enough space
18:26 naltamur1 I add only icons ;)
18:26 naltamur1 and not numbers lol
18:26 Fixer lol, some people actually travel
18:26 Fixer i treveled myself in 2011
18:27 paramat thoughts on #4086 ? this is beyond me but it seems to test well according to Wayward_One
18:27 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
18:27 hmmmm i think it takes a full day running at full speed in order to get from one edge of the map to another
18:27 hmmmm or something like that
18:27 Fixer it is faster
18:27 Fixer at least from spawn to 32000 it took me few hours for sure
18:27 Fixer there was convinient highway so I just put a rock on a key
18:28 paramat i once drove to the edge in one of my cars on my auto-generated roads
18:29 Fixer and of course at the end of the map we got server crash
18:29 Fixer should be fixed by now
18:32 paramat hmmmm dungeonparams is an excellent idea. not sure about seperating out caveparams, cavegen is increasingly mapgen-specific and caves seem more a part of base terrain
18:32 hmmmm ??!?
18:32 hmmmm paramat, did you remove large_caves from CaveV5 and CaveV7?
18:33 paramat nope
18:33 hmmmm the boolean large_caves
18:33 hmmmm all caves generated with these versions are large
18:33 hmmmm there are no small tunnels now
18:33 paramat pseudorandom yes
18:33 paramat correct
18:33 hmmmm right
18:33 paramat except in mgv6
18:34 hmmmm please don't do that
18:34 hmmmm there is a convenient boolean option that you can set to false
18:34 paramat good idea
18:34 hmmmm now i have to spend time re-adding it back...
18:34 paramat lol
18:34 hmmmm paramat, i would highly appreciate it if you started thinking about mapgen changes in a more abstract sense
18:35 hmmmm more generalization
18:35 paramat sure no problem
18:35 hmmmm just because some feature isn't being used in a mapgen at a certain point of time doesn't mean it needs to be completely removed or permanently welded to a specific mapgen
18:35 paramat i will have to after your changes
18:36 sfan5 do we still intend to release tomorrow
18:36 hmmmm and for your upcoming mapgen, i hope you'll try to put some effort forward into thinking about how you can maybe adapt that version of caves to apply to all versions of caves with an option maybe
18:37 naltamur1 so tomorrow?
18:37 paramat i will
18:37 hmmmm instead of copying a huge block of code and then making a couple mapgen-specific changes that aren't really specific to the mapgen
18:37 hmmmm thanks.
18:37 paramat i have an idea for improving dungeon density noise which will enable super-complex structures
18:38 hmmmm try holding off on it until i'm done with all this stuff
18:38 proller joined #minetest-dev
18:39 hmmmm see paramat, the idea behind having a separate CaveV6 and CaveV7 was never to make a specific piece of cave code for each mapgen, and then each cave can be purpose made, it was to isolate V6's caves so that you can freely develop caves using the same basic algorithm without harming v6's reverse compatibility
18:39 paramat when you add dungeonparams this will be relevant, so please let me know before you start on that
18:40 hmmmm reusable code is a good thing
18:40 paramat i see
18:41 hmmmm having 6 copies of code that is almost exactly the same is generally frowned upon because it makes it difficult to maintain and track down bugs
18:41 hmmmm also it's much more prone to mistakes
18:41 hmmmm and bloatier
18:41 est31 joined #minetest-dev
18:41 paramat agreed
18:41 KaadmY joined #minetest-dev
18:42 est31 yeah its a good code practice to have reusable code
18:42 Fixer hmmmm, are you just refactoring it or there are also some new features?
18:42 est31 question is when to apply it in practice
18:42 est31 sometimes it makes no sense, as the differences are too large
18:42 est31 or because of super-optimisation
18:42 hmmmm fixer, i need to refactor this to make any meaningful changes that aren't minor and incremental
18:42 est31 but idk about caves so i cant say anything about it
18:42 hmmmm otherwise i would get bogged down in a mess of copying the same modifications to 6 different files
18:43 Fixer ah ok
18:43 hmmmm but yes the original intention was to add new major features
18:44 paramat hmmmm not sure there is advantage to adding biomes after cavegen, the current order is simpler
18:44 hmmmm well the advantage is that there wouldn't be any requirement for cavegen to know about biomes
18:44 paramat decorations after cavegen yes though
18:45 hmmmm so if biomes are generated after caves, you'd have an issue where biomes generate inside the caves
18:46 paramat biome material in tunnel entrances is an excellent feature though
18:46 hmmmm but this wouldn't be a problem if we start using voxelmanip flags to mark the portion carved by tunnels as being "non-biomeable" areas
18:46 paramat so cavegen needs the biomemap
18:47 hmmmm and that can be solved by only adding the non-biomeable flag to nodes where the caves actually carved air/liquid from stone
18:47 hmmmm well at least that solves the case where you have /____
18:47 hmmmm but not \____
18:47 paramat i see
18:48 hmmmm but i don't think the current cave carving code handles the \____ case either
18:51 paramat "the advantage is that there wouldn't be any requirement for cavegen to know about biomes" i don't understand this
18:51 hmmmm why it's an advantage?
18:51 hmmmm it's called separation of concerns
18:52 paramat oh to help stuff in future
18:52 hmmmm from an architecture standpoint, it's best if a piece of code has as little dependencies as possible
18:52 paramat i would say it's better to avoid needing vm flags
18:52 hmmmm well that's your opinion
18:54 hmmmm here, i offer a generic way to tell the biome generation to "back off" from a certain area that could be applied to anything
18:54 hmmmm having to check one thing is simpler than having to check a node for any one that could have been placed in a biome
18:55 hmmmm and then, what if the mapgen that wants to use that code doesn't support biomemanager biomes?
18:55 hmmmm you're creating a new dependency that doesn't technically need to be there
18:55 paramat hmm good points
18:56 paramat seems good then
18:56 hmmmm maybe it would be easier for you to see if you were forced to re-use the same code in a different context
18:56 hmmmm such as a unit test
18:57 bugzapper joined #minetest-dev
18:58 paramat no need i can see :]
18:59 hmmmm i'm also adding a unit test for V6 cave generation
18:59 paramat anyway i really like what you're doing and looking forward to more, this organisation is overdue
19:00 hmmmm in general when we have a guarantee that a piece of map generation will have predictable, reverse-compatible output, it's a good idea to create a schematic of a chunk of map where the results are exactly 100% what is expected
19:00 hmmmm and then the unit test would be to run that piece of generation on some terrain, and then compare the voxelmanip contents with that of the schematic
19:05 est31 joined #minetest-dev
19:09 paramat so perhaps mgv6 cavegen should stay seperated to ensure it's stability? most of the others can be combined though, as i mentioned in my line comment
19:09 Void7 joined #minetest-dev
19:09 est31 I think thats what hmmmm is saying
19:09 paramat mgv5 and mgv7 only differ (i think) in which heightmap is used
19:10 hmmmm paramat, maybe you should look at this
19:10 hmmmm https://github.com/kwolekr/minetest/commit/a6b8b686f248a04a36a211437c7c28906ccb91ae#diff-b3b3932e4dba81e4e399d61603c86d8aR75
19:11 paramat however yes small prand caves should be re-added to be usable, i have plans to re-use those in future
19:11 paramat .. because they're fast
19:11 hmmmm if i have my way, YOU won't need to have plans to use them in the future, the modders will
19:12 hmmmm i want the mapgen api to be like a set of legos
19:12 hmmmm here is *a* way to make these types of caves, here is *a* way to make these ores, piece them together to make the kind of mapgen you want with minimal effort
19:12 paramat nice, i like LEGO
19:13 paramat it's always funny how americans call them legos
19:16 * paramat plays with his gravels
19:17 est31 there is something far more fun than LEGOs
19:17 hmmmm to show where i'm coming from with all this, you need to understand one of my original dissatisfactions with minetest was that i couldn't build stuff on a flat surface
19:17 est31 its called minecraft
19:17 est31 ermmm minetest
19:17 hmmmm but more generally i didn't have the ability to make mapgen features the way i want
19:18 hmmmm now part of the game is to write the lua you need (you're able to as an end-user) to make a map that has the qualities you want
19:21 paramat commit looks good btw
19:21 hmmmm ?
19:22 paramat the commit you linked me to
19:22 hmmmm there are 3 others
19:22 cheapie Which side of a nodebox is the front?
19:23 paramat nice i will look
19:23 paramat erm south side i think
19:24 cheapie ...I suppose maybe I should re-word my question.
19:26 cheapie I'm attempting to work with a node that uses a nodebox and facedir. When I place the thing, would I be looking at the most +X side, the most -X side, the most +Z side, or the most -Z side?
19:27 paramat i see
19:30 naltamur1 left #minetest-dev
19:30 paramat -Z
19:31 paramat but this is discussion for minetest channel
19:32 cheapie OK, thanks.
19:34 elia joined #minetest-dev
19:48 Void7 joined #minetest-dev
19:51 nore est31: do you think that it is that important that the server strips the escape sequences sent to pre-0.4.14 clients?
19:51 nore old clients will but see a few extra characters
19:52 nore and 0.4.14 will be a stable release, with escape sequence stripping added but nothing using them
19:52 nore so there will always be a stable release between code that stripped escape sequences (client-side) and code that uses them
19:53 est31 yes but it would effectively mean to drop support for clients older than 0.4.14
19:53 est31 and > 70% of minetest clients are still on pre 0.4.13
19:54 est31 or at least i think
19:54 est31 forgot again what sfan5's last stat said
19:55 sfan5 so do we release tomorrow or now
19:55 sfan5 s/now/not/
19:55 est31 paramat? ^
19:56 nore est31: what I mean is, it is not an incompatible change
19:56 est31 hrmmm dunno what happens with \x1b
19:56 nore old clients with new servers already have some changes that make them not quite compatible (think connected nodeboxes)
19:57 est31 what does irrlicht do with that character
19:57 nore IIRC, \v was rendered as a space
19:57 nore I think it will be the same with \x1b
19:57 est31 it should be tested
19:57 * nore tests
19:59 paramat erm hmmmmm said 11th so yes if possible
19:59 paramat although there's #4086 left
19:59 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
19:59 paramat needs a decision
20:00 paramat but of course we can delay release a day or two if absolutely necessary
20:00 nore est31: it is rendered as either a space or not at all (I don't know which one)
20:01 nore but at least, it doesn't cause any problems
20:08 paramat left #minetest-dev
20:08 DI3HARD139 joined #minetest-dev
20:31 Amaz joined #minetest-dev
20:37 STHGOM joined #minetest-dev
20:48 asl97 joined #minetest-dev
21:35 asl97 #4106
21:35 ShadowBot https://github.com/minetest/minetest/issues/4106 -- Update Fedora dependencies list by lonniganseaweed
21:38 Void7 joined #minetest-dev
21:46 Darcidride joined #minetest-dev
21:56 troller joined #minetest-dev
22:01 DI3HARD139 joined #minetest-dev
22:25 paramat joined #minetest-dev
22:27 paramat can anyone confirm #4106 is correct? perhaps it should just be merged as trivial
22:27 ShadowBot https://github.com/minetest/minetest/issues/4106 -- Update Fedora dependencies list by lonniganseaweed
22:27 asl97 paramat: me
22:27 paramat ok :]
22:29 paramat i assume you use fedora and have tested this?
22:30 asl97 i just check the two package and it is indeed endswith -devel
22:31 paramat ok +1
22:31 asl97 like all other fedora package
22:32 paramat will merge soon as trivial then
22:34 paramat can #4086 be merged anyone? it may not be perfect but is it ok and harmless enough to merge in it's current form?
22:34 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
22:35 Fixer oh, my translation is somewhat horrific in some places
22:37 asl97 paramat: I would avoid using * since those three with it drag in other stuff like ClanLib, libvorbis-devel-docs, openal-soft-qt, openalchemist.  but that is another different problem
22:38 paramat ok
22:39 sofar paramat: 4106 is good, links added to show the patch is correct
22:39 paramat yes will merge soon
22:46 STHGOM joined #minetest-dev
22:50 paramat 4086 has been tested well and seems safe, so unless there are objections i will merge it later tonight
22:51 paramat with 4106
23:08 yang2003 joined #minetest-dev
23:25 electrodude512 joined #minetest-dev
23:54 yang2003 joined #minetest-dev

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