Time Nick Message 00:18 red-001 exio4, reviewed the pr 00:18 red-001 g_settings is a global so you can just access it directly 00:19 exio4 yeah, I am too used to pass "globals" as parameters due to lack of them in my main language, going to fix this 00:21 red-001 well fear not minetest doesn't use extern much 00:24 exio4 I pushed the changes without testing them so far, I am compiling minetest on my laptop now (but it will take a while) 00:28 exio4 it seems to work! is the setting name fine or would be there a better name for it? 00:31 red-001 ask #minetest-doc 00:46 * red-001 awaits a code review t-shirt 00:52 exio4 heh :p 08:03 sfan5 merging #6532 in a few mins 08:03 ShadowBot https://github.com/minetest/minetest/issues/6532 -- Don't try to craft a non-existent item by EXio4 08:13 sfan5 force pushing because i forgot to squash (sorry) 13:21 jas_ #6536 13:21 ShadowBot https://github.com/minetest/minetest/issues/6536 -- Inventory image no longer draws overlays? 15:19 rubenwardy could be worth adding the "hacktoberfest" to some simple issues 15:19 rubenwardy especially lua related ones 15:20 rubenwardy then when this month is over, rename the label to "Beginner-friendly 15:20 rubenwardy +" 15:20 rubenwardy *hacktoberfest label 15:28 sfan5 nice idea 15:39 rubenwardy heh, most easy issues are only documentation 15:40 rubenwardy the rest are C++ changes 15:40 exio4 sfan5: what is the review about? 15:40 exio4 rubenwardy: I have stolen two of the easy changes by now 15:43 exio4 what is SmallJoker's IRC nick? 15:43 rubenwardy Krock 15:44 exio4 oh, he's not here :( 15:46 exio4 when he talks about range limit in #6535, do I just do a MAX/MIN combo or do I check and write a message into verbosestream or infostream? 15:46 ShadowBot https://github.com/minetest/minetest/issues/6535 -- Configurable recent chat by EXio4 15:56 red-001 exio4, just do a max/min combo 15:57 exio4 red-001: then it should be fine now :) 16:01 exio4 nevermind 16:06 red-001 exio4, wouldn't it be better to move this to the ChatBackend constructor? 16:08 exio4 I was thinking of that at first, but ChatBackend is also used for the console, and it has a scrollback of 500 lines by default 16:08 red-001 thats ChatBuffer 16:09 red-001 ChatBackend creates two chat buffers 16:09 red-001 s/two/three 16:09 exio4 oh, sorry 16:10 red-001 so just create the recent messages one without a size and set it 16:16 exio4 red-001: ah, I had issues following where the constructor was actually being called so I went with the place where I know it would work (and had initialization routines) 16:22 exio4 I am not 100% sure moving it to the constructor would end up with the setting being refreshed if you change the settings in-game and start a new game? 16:26 red-001 in-game setting changes are a hack anyway 16:26 red-001 and the chatbuffer should be destroyed when the client is 16:26 red-001 otherwise we got a memory leak 16:27 exio4 OK, moving to the constructor then, and removing applySettings 16:29 red-001 wait 16:29 red-001 I think this might be some werid static thing 16:29 red-001 exio4, 16:30 exio4 red-001: git stash then? :P 16:30 exio4 // Remove stale "recent" chat messages from previous connections 16:30 exio4 chat_backend->clearRecentChat(); 16:30 red-001 yeah 16:30 exio4 this gave me the hint the chat_backend persists (and just checked, and the console persists connections) 16:30 red-001 dunno why it's designed like that 16:31 exio4 designed?! 16:31 red-001 looks like it's created once in clientlauncher 16:31 exio4 MT has changed a lot since I checked it last time apparently :p 16:31 red-001 and then just passed around 16:31 red-001 exio4, lol 16:38 exio4 rubenwardy: re: Hacktoberfest tag, C++ changes might be relatively easy/small to get the label 16:39 rubenwardy yeah, I've got a few 16:39 rubenwardy feel free to suggest some 16:40 exio4 I am hunting for them after I get this pull request relatively pulished : 16:40 exio4 :P 18:06 Krock Git blame pointed at me again. Working on a patch to fix the misbehaviour of strict_protocol_version_checking 18:09 * red-001 tells the admin of localhost that a fix is being worked on 18:19 Krock Will push https://pastebin.com/raw/kbHWmcCi in 10 minutes 18:23 paramat hang on 18:23 sfan5 looks correct to me 18:24 paramat but it's missing brckets around the ORs to make the logic clear 18:25 Krock there should be no brackets around the ORs 18:25 paramat oh wait, saw it wrong 18:25 paramat fine 18:25 Krock could even remove the new parentheses but for the sake of readability I kept them 18:29 Krock pushing... 18:36 exio4 thinking about the collision checking algorithm, of course an octree etc would be much better, but couldn't it be easily improved by doing a line instead of a cube? 18:37 exio4 (checking collisionMoveSimple) 18:41 exio4 there's right now a check to avoid looping more than 100 times but there's a vector with all the possible blocks every time it is called 18:42 exio4 it sure needs to be profiled, but I cannot do it until wednesday 18:43 Krock Having that function optimised or even rewritten would be great. However, also needs careful testing 18:48 celeron55 19:31:03 < exio4> designed?! 18:49 celeron55 lol! 18:51 Krock bug by design 22:06 exio4 paramat: just saw my email and there's a message on changing it from lines to messages, but it's not in the PR, do you think it is reasonable to have _lines or..? 22:24 paramat my comment might be in one of the 'outdated line comments' 22:25 paramat yes click 'show outdated' 22:26 paramat my second from last 'review' in the thread 22:27 paramat it's not critical, but strictly 'messages' is more correct than 'lines', let's see what other devs say 22:28 exio4 ah, sorry 22:29 paramat added my comment to the thread as a comment to be visible 22:29 paramat no problem 22:30 exio4 I'll change it to _messages 23:42 exio4 are there any other changes missing? 23:45 exio4 BTW, I have noticed you need to run the update script twice when there are changes to settings_translations_file.cpp 23:45 exio4 once which is going to delete things, and another time for adding the new strings 23:52 paramat rubenwardy is it possible to use 'position[,]' when defining a sfinv formspec? i'm trying but no luck so far 23:52 rubenwardy position? 23:52 rubenwardy do you mean size? 23:52 paramat no 23:53 paramat i want to offset a sfinv formspec to be on one side of the screen instead of centred 23:53 rubenwardy ah 23:53 rubenwardy just put it in the formspec 23:53 paramat hm ok i'll gist what i have 23:55 paramat https://gist.github.com/paramat/0720b1e885b4d9d32bd97d81ee90096f 23:56 paramat this is an edit of simple skins mod 23:56 paramat see line 3 23:56 paramat has no effect when i try it, still centred 23:59 paramat i know 0.4 is close to centre i was just trying that value out in case it worked