Time Nick Message 00:14 luk3yx Sorry Josiah (but to be fair minetest.conf calls it `serverlist` and has done since before the rename) 00:28 celeron55 i think it's sometimes appropriate to connect words like that to form effectively a proper noun (i.e. a unique name) 00:28 celeron55 i think server_list would actually be a bit weird, being quite a specific thing 00:29 MTDiscord Names can be multiple words though, to be fair 00:29 celeron55 well also none if it actually matters in the end 00:31 MTDiscord This is true 00:32 MTDiscord anyways, how long did you want to wait before creating the repo? 00:32 MTDiscord c55 ^ 00:47 v-rob I assume core devs will automatically have access to minetest_docs by merit of being part of the Minetest organization, right? 00:48 MTDiscord Ofc, though they aren't necessarily a part of the main team 00:49 MTDiscord The idea is to separate the workload 00:49 v-rob To be sure 00:52 MTDiscord Do we have any plans for assigning certain people the areas that they know well or is it all gonna be free for all? 00:54 MTDiscord I think that forced specialization would be our downfall, people will naturally work on what they're most comfortable with. 01:18 MTDiscord Agreed, for now I'd expect it to be a free for all for filling out the actual text, but organized as to what needs to be filled out and the template to review on 02:05 rubenwardy bug report https://twitter.com/Zweieck2/status/1476528073361264640 02:07 MTDiscord "String too long" makes me think its likely something on their end, as thats a Lua error 02:07 MTDiscord Usually is, anyways 02:07 MTDiscord yep, that's a windows issue I've experienced 02:07 MTDiscord we have much shorter max foldername/filename limits 02:07 MTDiscord linux is 4000, I think windows is like <500 02:09 MTDiscord Oof, yeah looking at that thread I smell Windows issues 02:38 ShadowNinja Looks like that's a packet/serialization error, likely due to a string over 65546 characters. Although we'd kind of need a backtrace to know which field it is. 02:40 ShadowNinja s/4/3/ 02:47 erlehmann i have recently compiled minetest on aarch64 (reform2) and there are several cases in the code where it seems that at least irrlicht has the ability to copy strings past buffer ends by default? at least the compiler warnings told me that 02:48 erlehmann i don't have a log rn, am on x86 11:38 sfan5 if you want/need logging for irc just ping me 14:51 ROllerozxa https://github.com/minetest/minetest/pull/11747#issuecomment-1003567032 I guess this is why making IrrlichtMt a Git submodule would be useful... One known working commit of IrrlichtMt would be matched with each commit of Minetest, instead of having deal with it manually for PRs that are forked behind current master like this one. 14:53 sfan5 useful fact: the last tagged version of Irrlicht is assured to build with MT master 20:49 lhofhansl Planning to merge #11914 in a few. 20:49 ShadowBot https://github.com/minetest/minetest/issues/11914 -- Recalculate normals before adding mesh to the scene. by x2048 20:50 lhofhansl Or hmm... After minetest/irrlicht#90 it's not needed. 20:51 lhofhansl I can confirm that the irrlicht change fixes the problem I have observed. 21:33 lhofhansl If someone would have a look at these two. I'd vote for merging the Irrlicht fix and not the Minetest fix. 21:34 Krock will merge #11892 and #11878 in 10 minutes 21:34 ShadowBot https://github.com/minetest/minetest/issues/11892 -- Better document sky_color scope by Zughy 21:34 ShadowBot https://github.com/minetest/minetest/issues/11878 -- Print announce error response by ShadowNinja 21:35 Krock I don't know about the irrlicht one... 21:36 Krock having it fixed in the irrlicht engine would make more sense 21:44 Krock merging 21:46 Krock done