Time Nick Message 01:42 nolsen I think a new configuration should be added for servers that can let server admins or owners choose if they want to allow or ban android users. 01:43 nolsen Would be useful to keep the ping timeout noise down. 01:43 nolsen Since there's no point playing multiplayer. 01:43 nolsen on an android. 01:49 Wayward_Tab No point? 01:49 Wayward_Tab It works fine, depending on the mods 01:49 Wayward_Tab I play multiplayer on android almost daily 01:52 nolsen Well it's noisy on some servers like mine 01:52 nolsen I have more than 123 mods loaded. 01:54 Wayward_Tab ah, ok 03:00 paramat hmmmm noise defined biome blend seems ready, please could you review sometime tonight? https://github.com/minetest/minetest/pull/2764 03:06 paramat adds 0.6ms to chunk generation time 15:20 nolsen Maybe strict_protocol_version_checking should be changed. 15:20 sfan5 why 15:21 nolsen Because if a server is 0.4.12-dev, and a client is 0.4.12, it should still let the client connect. 15:21 nolsen It's not that outdated. 15:21 kilbith this is already the case 15:22 nolsen I tried enabling strict_protocol_version_checking on my "Always the latest commit" server so I can get rid of people who are too lazy to update, but it kicked me out. 15:23 nolsen It should let development and stable version connect as long it's the same number. 15:23 nolsen Because most people use Stable. 15:25 sfan5 umm 15:25 sfan5 nolsen: 0.4.12-dev has some possibly(?) incompatible changes like SRP for login 15:25 sfan5 this is why 0.4.12 clients can't connect 15:25 sfan5 or something like that 15:25 nolsen I can connect just fine. 15:25 nolsen No problems. 15:26 sfan5 ¯\_(ツ)_/¯ 15:27 nolsen Could I change the server version name saying it's stable to let stable people in? :P 15:27 nolsen wait it might not let dev clients in. 15:35 kahrl there was a pull request for making strict protocol version checking check for an interval instead of a single version 15:35 kahrl https://github.com/minetest/minetest/pull/1731 15:35 kahrl I don't think the reason is was closed applies anymore 15:35 kahrl it* 15:37 kahrl there, reopened it 15:37 kahrl (as per ShadowNinja's comment) 15:41 nolsen " Failed — 1 errored, 1 failing, 2 pending, and 1 successful checks" 15:42 nolsen You think everything fails, don't you jenkins and Travis CI 15:42 nolsen s/don't/do 17:02 est31 in 5 minutes pushing https://github.com/est31/minetest/commit/b7bb83209e5d2787783e6ec50549e8e750ad9edb 17:04 est31 updated https://github.com/est31/minetest/commit/6dcf549ba9d522ed991184401658e85d3554ad8e 17:27 Taoki I found a very weird bug in minetest.serialize_schematic: It will randomly detect the file at the correct path (eg: modpath/filepath) or instead try to find the schematic at ./ (current directory). 17:27 Taoki Basically, I run minetest.serialize_schematic() in a part of my code and it sees the schematic in modpath/ Then I run it in another and it looks for it in ./ 17:28 Taoki (posting here as well since this is certainly a bug) 17:32 est31 Taoki, when are the two parts of your code executed? 17:32 est31 at load vs while the game is running 17:33 Taoki est31: It's a longer code, can't post all of it easily. Basically there is a function in init.lua in which serialize_schematic is called. There are two other lua files, and each calls this function. For one of them, serialize_schematic sees the file at the correct path. For the other, it tries to take it from ./ 17:33 Taoki I cannot understand why on Earth they would be different 17:33 Taoki est31: After the game has started, and the world is running for several seconds 17:38 est31 Taoki, so you execute both functions while the game is running? 17:39 est31 or is one function called when the game is loading? 17:39 Taoki est31: Yes, one after the other. Both triggered at nearly the same time 17:39 Taoki No, both execute only after startup 17:39 Taoki Long after 17:40 est31 usually, it shouldnt work 17:40 est31 you should rather cache the mod's path 17:40 est31 and then use absolute paths 17:40 Taoki I want it to detect the schematic automatically really. I tried putting it in "schems" as adviced by someone, but that didn't work 17:41 est31 with caching I mean minetest.get_modpath(minetest.get_modnames()) 17:41 est31 and what do you mean with automatic detection 17:42 est31 also, minetest won't support access to the mod directory after it has been started 17:42 sfan5 also, minetest won't support access to the mod directory after it has been started 17:42 sfan5 wat 17:43 est31 blame ShadowNinja for it 17:43 est31 in fact its a bug that it works 17:44 est31 (mod security) 17:48 Taoki Fixed it by writing the path manually I guess 17:49 est31 what we need is a directory handle API 17:49 est31 mods can get a directory handle at loading time 17:49 est31 so that they still have access after loading time has ended 19:06 crazyR_ it would be really usefull if the world dir structure created mod_conf and mod_data. then encourage modders to put the settings conf files in the mod_conf folder and any data files int eh mod_data folder.... would make the world folder much cleaner and easer to look at 19:09 est31 I think we should split up settings altogether 19:10 est31 settings for worlds, and the client 19:10 est31 and perhaps more 20:08 crazyR_ thats sounds good est31. my request above was more todo with keeping the world dir tidy. it gets pretty messy when all the diffrent mods place random files in there. theres no order 20:19 est31 pushing fix in 5 minutes: https://github.com/est31/minetest/commit/0dc915421b252c33173e5f893dd529b2bdcccda2 20:20 sfan5 est31: looks good 20:21 est31 update https://github.com/minetest/minetest/commit/45a9bb310ee0584cacb9993ed9805da5006c07dc 20:21 sfan5 (except that IS_UTF8_MULTB_START is not used anywhere) 20:21 est31 yes it isnt 20:21 est31 might become handy one day dunno 20:21 est31 should I remove it? 20:21 sfan5 yes 20:21 sfan5 also I'd prefer character_idx to be named char_idx 20:22 est31 char != character 20:22 est31 char is a c type 20:22 est31 characters can span multiple chars 20:23 est31 (char == byte) 20:23 sfan5 it's pretty clear that char_idx is not the same as i 20:24 sfan5 nobody is going to assume that you're defining a duplicate of the variable that is counted by for 20:24 est31 yea, but I think it makes understanding easier 20:24 sfan5 hm 20:24 sfan5 maybe codepoint_idx would be better 20:29 * est31 wonders what character_idx counts, inside figure 2-8: 20:29 est31 http://www.unicode.org/versions/Unicode7.0.0/ch02.pdf#G25564 20:30 est31 I guess its codepoints 20:30 est31 or "encoded characters" 20:30 est31 err no 20:30 est31 only codepoints 20:37 VanessaE celeron55: [06-17 16:36] the development wiki appears to be down for now http://minetest.net/wiki/doku.php 20:37 VanessaE (confirmed) 20:38 VanessaE nevermind. outdated link. 20:42 est31 Wrapping inside multi-codepoint "abstract characters" isnt that bad I guess 22:45 paramat ShadowNinja, does https://github.com/minetest/minetest_game/pull/539 'Fix is_ground_content settings for nodes' look good? it just needs one more approval, i can push it for you 23:11 est31 paramat, you already have one +1 23:11 est31 so you can merge it, no? 23:12 est31 (as you are yourself member of the core team) 23:12 paramat i thought game needs 2 game devs, so i don't count? 23:12 paramat not sure 23:12 est31 hmm 23:13 paramat however it is mapgen stuff 23:15 paramat mtgame is a bit neglected, i am willing to join the mtgame team. or perhaps my vote should count for mapgen matters 23:15 est31 this page only speaks about "new features" http://dev.minetest.net/minetest_game_Development 23:16 est31 You can push something to upstream [1] only if two members of the core team [2] or the subsystem maintainer agrees on it. 23:16 est31 so from how I see it, mtgame has "maintainers" but no extra 2 approve rule 23:16 est31 on non feature changes at least 23:17 est31 this would be a feature change I guess https://github.com/minetest/minetest_game/pull/542 23:17 est31 so I need 2 people to approve 23:17 est31 or how many are present 23:18 est31 what "present" means 23:18 paramat for mapgen stuff i think i should only need one +1 23:18 est31 sfan5, ShadowNinja ^ 23:18 est31 yup 23:18 est31 as long as its no mapgen addition 23:18 est31 e.g. new feature 23:19 paramat yes i'll have to judge carefully 23:19 est31 yea bureocracy 23:19 paramat so i'll push this, otherwise dev is too slow 23:19 est31 yup 23:20 est31 thats what I think too 23:20 est31 one can have bureocracy if people live it 23:20 est31 and if people actually review PRs 23:22 est31 paramat, I +1 to you being in the mtgame maintainer team 23:23 est31 (if I have a say in this) 23:23 paramat hehe they may not want me =) 23:23 paramat they might trust me on mapgen stuff though 23:26 paramat 'To prevent problems with the development like before minetest_game is maintained by only five people' 23:27 paramat well i'll consider myself one vote for mapgen stuff.. 23:34 paramat hi hmmmm https://github.com/minetest/minetest/pull/2764 'Biome API: Noise defined biome blend' is ready for review. adds 0.6ms 23:34 est31 at least sfan5 while not commiting often (these days) still approves patches 23:34 est31 as in I say "pushing in 5 minutes" and sfan5 sais "ok" 23:35 est31 or "not ok, fix this" 23:35 est31 thats very good 23:38 * est31 wonders what has become of RBA's minimap