Time Nick Message 02:15 diemartin would be useful if callback priority was documented somewhere 02:18 diemartin e.g. If I define on_place for an item, and on_rightclick for a node, what gets called first when using that item on that node? 02:23 sofar messing around with entities, I can now no longer connect to my server: it crashes hard on login (segfault) 02:24 sofar I may have an image with a too long texture string 02:24 sofar that's a bad client side bug :/ 02:24 sofar https://gist.github.com/sofar/8be79e03c3aa3aa1cfee7a7ec24ff581 02:24 sofar I really need a fix for this, I can't just nuke my world to get past this problem 02:26 sofar sigh 02:45 paramat please can anyone review #5157 ? simple, tested, fixes a crash 02:45 ShadowBot https://github.com/minetest/minetest/issues/5157 -- Objectpos over limit: Avoid error caused by sector over limit by paramat 02:52 VanessaE paramat: merge #5172 plz ;) 02:52 ShadowBot https://github.com/minetest/minetest/issues/5172 -- Improve mesh shading by numberZero 02:53 paramat ah that was what i was working on, will review 05:43 Zeno` +1 for 5172 05:43 Zeno` there is no need atm for that to be optimised by having early return IMO 05:44 paramat yeah just wondering which would be faster 05:44 paramat i'll test that soon 05:48 Zeno` by all means :) 06:15 Zeno` paramat, did MTG have any updates related to detached_inventory? 06:17 Zeno` nvm, found it 06:22 paramat ok good 06:22 Zeno` heh :) 06:28 Zeno` thanks for the help =D 09:27 nerzhul merging #5172 09:27 ShadowBot https://github.com/minetest/minetest/issues/5172 -- Improve mesh shading by numberZero 09:29 paramat nerzhul are you able to review #5157 ? fairly simple 09:29 ShadowBot https://github.com/minetest/minetest/issues/5157 -- Objectpos over limit: Avoid error caused by sector over limit by paramat 09:30 nerzhul paramat, can you add unittests for this function, permitting to verify when it should work or not ? 09:30 nerzhul this can help for future usage 09:31 nerzhul this function is a very good candidate to be tested with UT 09:33 nerzhul please test 0,0,0 X limitation, Y limitation, Z limitation (both negative and positive, just at world limits) and a random position between -LIMIT,+LIMIT 09:37 nerzhul !tell rubenwardy when you merge your PR please squash commits to not polute the log 09:37 ShadowBot nerzhul: O.K. 09:47 paramat hm maybe, never coded UT before. bbl 10:34 Thomas-S Minetest technic mod crashes in this line when metadata is nil: https://github.com/minetest-technic/technic/blob/master/technic/tools/mining_drill.lua#L265 10:34 Thomas-S Is this related to the latest changes to metadata? 10:37 sfan5 does it happen before the meta changes? 10:38 Thomas-S If I remeber correctly, it didn't. I will test it with an old build 10:40 sfan5 as of current master nothing should break 10:40 Thomas-S It works fine in old clients. 10:40 sfan5 there's compat code in the to_table method of itemstacks 10:42 Thomas-S Sorry, there was a typo in my last message: Instead of clients, I meant servers. 10:44 Thomas-S After updating minetest (nothing else changed) the crash occurs again. There must be some backwards compatibility problem with the latest changes. 10:49 sfan5 it's quite obvious that the lua crash happens on the server 10:49 sfan5 i'd open an issue to have rubenwardy take a look at it 10:53 paramat nerzhul unit tests for object placement will have to be a separate PR, it will take me a long time to learn how to code those and this is a bugfix for a crash. so if possible could you review #5157 ? 10:53 ShadowBot https://github.com/minetest/minetest/issues/5157 -- Objectpos over limit: Avoid error caused by sector over limit by paramat 10:53 paramat anyone else? 10:54 Thomas-S sfan5: Issue is opened: #5181 10:54 ShadowBot https://github.com/minetest/minetest/issues/5181 -- Backwards incompatibility of the latest metadata changes 11:02 nerzhul paramat, it's not unit test for object placement, the function only tets position, it's an easy test to do 11:02 nerzhul you must do it with the function as we have problems with this function, TDD permit to ensure it works properly 11:08 paramat it's not easy for me, it's unreasonable to delay this PR for UT when it will take me over a week to learn how. UT are not at all essential for this PR to go ahead 11:08 nerzhul paramat no 11:08 nerzhul unit tests are easy to use 11:09 paramat i'm not even sure i want to code a UT for this, if it's easy you can do it :] 11:09 nerzhul it's easy for you too 11:09 nerzhul open test_utilities.cpp 11:09 nerzhul add a void testObjectPosOverLimit() 11:10 nerzhul and do like other funcitons, UASSERT(object_pos_overlimit(param) == requiredvalue) 11:10 nerzhul it's about 10 lines of code and you know what the function awaits 11:10 paramat est31 nore sfan5 Zeno` please can anyone review #5157 ? simple, tested and fixes a crash. it's actually semi-trivial 11:10 ShadowBot https://github.com/minetest/minetest/issues/5157 -- Objectpos over limit: Avoid error caused by sector over limit by paramat 11:11 paramat ok i might do that, but in a separate PR 11:12 paramat thanks for the explanation 11:13 nerzhul no no paramat include it in your PR 11:13 nerzhul it's easy to do, i don't want it to be forgotten and it permit to ensure you PR works properly and enhance the review process 11:14 paramat open an issue then it won't be forgotten. UT are never absolutely essential 11:15 nerzhul here it's essentiel as this code has been modified many times and each time there are fails. 11:15 nerzhul Unit test will permit to ensure you are doing it right 11:17 paramat testing and use will show it's working. it's not essential. it's likely i will either not want to write this UT or will take weeks to get around to it 11:18 paramat no problem, i'll ask others for review 11:21 paramat i've tested it thoroughly, read the thread 11:24 paramat hmmmm was our strictest dev and even he didn't try to force a contributor to write a UT and delay a good simple PR 11:24 juhdanad paramat: does Minetest have a 'save for older version' option? 11:24 paramat erm don't know 11:25 juhdanad #4682 would be less controversial then... 11:25 ShadowBot https://github.com/minetest/minetest/issues/4682 -- Fix water flooding onto lava by juhdanad 11:25 paramat yeah i'm really sorry to be unsure about that PR 11:26 juhdanad No problem. 11:27 paramat please can devs consider the possible change in block serialisation in this https://github.com/minetest/minetest/pull/4682#issuecomment-276918751 11:28 paramat can your other PR be made to work without version change? 11:29 juhdanad It could, but not so well. 11:29 sfan5 paramat: increasing a serialization version is not uncommon or bad and it should not be tried to work around it 11:30 sfan5 the "this is why you can not play on your maps on other versions of Minetest after you opened them in this version." thing happened about 20 times in the past already and it wasn't a concern 11:30 sfan5 why would it be now? 11:32 juhdanad An example: imagine a TNT which explodes next to an unloaded block. Due to the explosion the light changes. Without that fix light would stay incorrect in the unloaded chunk. 11:34 juhdanad Also, the two PRs are ~1000 line long, and it would be a lot of work to swap their logical order, and re-test them. In their current state, my PRs are well tested by Fixer. 11:35 paramat sfan5 ok. seeing as version change seems essential for the 2nd PR i think it's worth it then 11:36 paramat i can see how the 1st PR will be useful for much more than just huge lavacooling mountains 11:37 paramat so, i'll go through the 1st one again then will probably +1 11:37 paramat then i'll start looking at the 2nd 11:37 juhdanad Thank you! 11:38 paramat btw devs these are very high priority 12:46 nerzhul #5131 is now rebased and use the common Metadata storage, it can be merged :) 12:46 ShadowBot https://github.com/minetest/minetest/issues/5131 -- Add ModStorage Lua API by nerzhul 12:59 red-001 travis seems to be broken 13:10 nerzhul red-001, no, why ? 13:11 red-001 nvm it's just really slow 14:23 red-001 updated #5142 14:23 ShadowBot https://github.com/minetest/minetest/issues/5142 -- [CSM] Add `on_punchnode` callback by red-001 14:23 red-001 rebased it so that it doesn't conflict 15:04 red-001 rebase #5152 15:04 ShadowBot https://github.com/minetest/minetest/issues/5152 -- [CSM] Add `get_wielded_item` by red-001 15:04 red-001 the WIP tag can be removed 15:31 celeron55 this method is kind of funny: https://github.com/minetest/minetest/blob/master/src/inventory.cpp#L88 15:37 celeron55 "give me a node item in a pre-0.3 format and i'll just convert it to a second version of static node ids and then from that to a textual name; no big deal" :P 15:51 Fixer i'm still waiting for that tsundere.fi world.... coming out from somewher 15:52 Fixer e 16:49 VanessaE 2017-02-05 16:18:52: ERROR[Main]: ServerError: AsyncErr: environment_Step: Runtime error from mod 'areas' in callback environment_Step(): Invalid float vector dimension range 'x' (expected -2.14748e+06 < x < 2.14748e+06 got -2.14748e+06). 16:49 VanessaE (the affect mod's author says it is the engine's job to make this check) 16:49 VanessaE affected* 16:51 Fixer VanessaE: error on your server? 16:52 VanessaE yeah 16:52 Fixer Liberty Land server also has this, error is quite rare 16:52 VanessaE happens every so often. almost always crashes the same mod (Areas) 18:06 red-001 can anything be done about apps that don't release the full source code? 18:07 red-001 (and are based of minetest of course) 18:09 Calinou red-001: DMCA requests, which GOogle and Apple are supposed to obey 18:09 Calinou however, Minetest core developers have other things to do than send those DMCAs all day 18:10 red-001 I see 18:19 rubenwardy #5182 18:19 ShadowBot https://github.com/minetest/minetest/issues/5182 -- Fix incompatibility of ItemStack.to_table() introduced by stack meta by rubenwardy 18:19 rubenwardy Thomas-S, 18:19 Thomas-S Thanks for the fast bugfix! 18:24 nerzhul rubenwardy, please look at #5131 it now uses Metadata interface 18:24 ShadowBot https://github.com/minetest/minetest/issues/5131 -- Add ModStorage Lua API by nerzhul 18:29 diemartin VanessaE: could it be that "player teleported to the middle of nowhere" bug? 18:29 VanessaE I don't know, maybe 18:31 diemartin tye chwck is done. the engine just throws an error 18:31 diemartin check* 18:32 diemartin bah I blame the phone 18:58 VanessaE ShadowNinja: ^^^ 18:59 diemartin mod error? 19:04 VanessaE idk, he says it's the engine's fault. 19:06 diemartin uh probably