Time Nick Message 00:17 Jordach kilbith, consider allowing ffmpeg to decode audio streams 00:19 Jordach things that openal cannot do at runtime, like decoding live streamed ogg vorbis 10:57 sfan5 merging #9339 in 10 minutes 10:57 ShadowBot https://github.com/minetest/minetest/issues/9339 -- Fix pathfinder bugs: returning nil frequently, broken A*, jump through solid nodes by Wuzzy2 11:08 sfan5 ok what 11:08 sfan5 it appears github has marked the commit as being mine 11:08 sfan5 I will force-push this to give Wuzzy proper attribution 11:09 sfan5 done 12:15 kilbith I'm not very satisfied with the arm inertia I made 3 years ago 12:16 kilbith this is too "lazy" moving 13:27 rubenwardy Breaking news: programmer doesn't like code he wrote years ago 13:52 erlehmann always write code like the person who has to maintain it knows where you live 13:53 erlehmann because if it's you, they will know 13:53 erlehmann (and it's likely you) 13:53 rubenwardy Today's new code is tomorrow's legacy project 14:55 jonadab Nah, it's fine, future you may know where you sleep, but he probably doesn't have a time machine, so he can't actually come back and do anything to you. 14:55 jonadab Which is good, because programmers would be extinct by now otherwise. 15:25 erlehmann jonadab, future you can kill you in your future 15:25 erlehmann but probably not kill you in your sleep (unless you are aiming for a darwin award anyway) 15:28 jonadab Future me can kill _himself_, but that is not my problem now :-) 15:38 erlehmann i see 15:38 erlehmann you don't have object permanence 15:47 jonadab Heh. 15:48 jonadab Anyway, that would be a Pyrrhic victory for him. 18:56 sfan5 Krock: do you think the set_sky PR is ready for merge? 18:56 sfan5 I can fix the three remaining things at merge http://sprunge.us/FJ66ZW?iff 18:56 sfan5 eh ?diff to get the syntax highlighting 18:57 sfan5 (the third being that is_color_table does not pop the read values from the stack) 18:57 Krock yes sure 18:58 sfan5 alright 18:58 sfan5 but first I'll actually test the fabsf thing with clang once 19:01 sfan5 yup worked 19:02 sfan5 merging #8964 with aforementioned changes in 10 minutes 19:02 ShadowBot https://github.com/minetest/minetest/issues/8964 -- set_sky improvements, set_sun, set_moon and set_stars by Jordach 19:08 sfan5 hm I need to be careful not to merge it incorrectly 19:11 sfan5 ..and also going to fix the indentation of the copyright notice in skyparams.h 19:11 Jordach yes how can i help 19:12 Jordach sfan5, Krock? 19:12 sfan5 nothing you need to do 19:13 Jordach ah 19:31 Krock thanks for merging, sfan5 19:31 sfan5 sure 19:38 sfan5 Krock: if you have some time leftover, please take a look at #9459 and #9465 19:38 ShadowBot https://github.com/minetest/minetest/issues/9459 -- Fixes around ServerActiveObject on_punch handling by sfan5 19:38 ShadowBot https://github.com/minetest/minetest/issues/9465 -- Fix linking failures when compiled with Clang 9 by sfan5 19:40 Krock how about clang for windows? 19:40 sfan5 unaffected I think 19:45 Krock I'd prefer to apply it for cross-compiled clang versions as well - just to be safe 19:46 sfan5 in the linked clang bug they diagnose it to be due to a glibc change, which definitely doesn't exist on Windows 19:48 Krock > Since it's not really an llvm regression, but a glibc change 19:48 Krock well then. go ahead. 19:49 sfan5 also, I've never heard of anyone using clang for windows cross-compiling 19:49 sfan5 even though it's supposedly easier than setting up mingw 19:51 Krock me neither. barely anyone would even compile it on Windows 19:51 Krock or also for cross-compiling 19:51 Krock both is supported according to clang's page 19:52 sfan5 ah great clang-format is complaining 19:55 sfan5 pushing trivial code style fix 20:00 Krock sfan5: (loot) warning once on startup is a saner check, hence my suggestion. lazy checking may also spam the logs 20:01 Krock not spam too much, but still every time it's spawned in a chest 20:01 sfan5 higher incentive for modders to fix their stuff ;) 20:01 Krock that's another POV.. yeah. 20:02 sfan5 I need the check inside the populate_chest() anyway, so checking it on register would just be additional code 20:02 sfan5 hm 20:03 Krock rubenwardy: opinions? 20:03 rubenwardy err 20:04 Krock :3 involved into a random discussion 20:04 sfan5 context: game#2613 20:04 ShadowBot https://github.com/minetest/minetest_game/issues/2613 -- Dungeon loot: Warn on unknown registered loot items by sfan5 20:05 Krock also: I'd recommend to begin a feature freeze by this Saturday 20:05 sfan5 yeah 20:06 Krock there are a few points in the "meetings" wiki article - if you have more stuff to discuss feel free to add it 20:08 sfan5 I have no idea how #9352 should be fixed for example 20:08 ShadowBot https://github.com/minetest/minetest/issues/9352 -- minetest.get_player_information sometimes return nil (running at mod load, in singleplayer) 20:09 sfan5 because I looked into it briefly and the only reason the rtt values should fail to return is when the peer doesn't exist 20:09 sfan5 but that doesn't make sense, because a peer definitely exists 20:10 Krock > getClientConInfo(player->getPeerId(), con::AVG_RTT, &avg_rtt)) 20:10 Krock well, according to paramat's post, the average rtt is broken, and that's a known things 20:10 Krock -s 20:11 Krock and the average rtt defaults to -1 at start IIRC 20:12 Krock connection.h L618 20:12 Krock which results in the return value -1 which is interpreted as error 20:13 sfan5 oh it makes sense now 20:13 sfan5 guess I can take it off of the meeting notes then, I can make a PR to fix that stuff tomorrow 20:14 sfan5 btw I fixed the comment in #9459 20:14 ShadowBot https://github.com/minetest/minetest/issues/9459 -- Fixes around ServerActiveObject on_punch handling by sfan5 20:29 kilbith so the next step is to merge Jordach's atmos mod into minimal or MTG 20:41 sfan5 merging it into minimal would be useless 20:42 kilbith or builtin idk 20:43 kilbith the engine should expose *nice* defaults 20:43 kilbith and the current defaults are no better than atmos 20:44 Krock the engine is a modding base, not a modding template 20:45 Krock modders will always come up with better solutions than what can be found in minimal or MTG 20:45 kilbith I'm not your average modder dud 20:45 kilbith e 20:53 sfan5 merging #9369, #9459, #9465 in 5 minutes 20:53 ShadowBot https://github.com/minetest/minetest/issues/9369 -- Fix potential security issue(s), documentation on minetest.deserialize() by luk3yx 20:53 ShadowBot https://github.com/minetest/minetest/issues/9459 -- Fixes around ServerActiveObject on_punch handling by sfan5 20:53 ShadowBot https://github.com/minetest/minetest/issues/9465 -- Fix linking failures when compiled with Clang 9 by sfan5 20:58 kilbith what interesting features MTG has to bring up for 5.2? 20:58 kilbith I am almost hesitant to clone it anymore by default 20:59 kilbith I mean, there are bigger subgames now in terms of popularity 20:59 kilbith MineClone or ITB 21:00 kilbith if only MTG could add atmos and craftguide, that'd save the day 21:02 sfan5 oh, 5 mins over already 21:04 sfan5 kilbith: this one is actually marked for inclusion with 5.2 https://github.com/minetest/minetest_game/pull/2396 21:06 sfan5 ok merging done 21:07 kilbith btw the BAW team is in hiring campaign 21:07 kilbith if anyone's interested 21:08 sfan5 hm looks like MTG has nothing exciting at all for 5.2 21:09 * kilbith misses BlockMen and PilzAdam... 21:10 kilbith at least they knew how to make good game features each release 21:11 kilbith the furnace could benefit of animated_image[] for example 21:12 kilbith why this isn't done already 21:12 kilbith the book could benefit of hypertext[] definitely 21:13 sfan5 do you want to let users write html? or include a markdown renderer? 21:14 kilbith no the title could have a bigger size and the author could be italic 21:14 kilbith that's all 21:15 sfan5 oh 21:15 kilbith hypertext wraps better the text than textarea moreover 21:15 Jordach a *simplified* markdown renderer wouldn't be that bad 21:16 * rubenwardy compiles MT and installs atmos 21:18 sfan5 minetest/src/network/clientpackethandler.cpp:1247:29: warning: unused variable 'params' [-Wunused-variable] 21:18 sfan5 std::vector* params = new std::vector; 21:18 sfan5 ^ memory leak 21:23 Jordach sfan5, the old networking handler used that code for strange reasons 21:23 Jordach it's there to ensure it doesn't fly off the wall 21:23 sfan5 no that's actually unused, otherwise clang wouldn't be reporting it 21:25 Jordach want me to fix that 21:26 sfan5 sure 21:26 sfan5 the "bool clouds" a few lines down is also unused 21:26 Jordach that should be skybox.clouds = true 21:27 Jordach want me to push to set_sky_3 or construct a new branch 21:28 sfan5 make a new branch and PR 21:29 kilbith that should count as trivial 21:29 kilbith so push ahead 21:29 Jordach it probably breaks the merge used 21:34 Jordach #9466 21:34 ShadowBot https://github.com/minetest/minetest/issues/9466 -- Fix memory leak in protocol 38 set_sky by Jordach 21:36 sfan5 thanks 21:36 sfan5 WHY THE FUCK DOES GITHUB CHANGE THE COMMIT OWNERSHIP TO MY NAME 21:36 sfan5 FUCKING SOY HIPSTERS IN SILICONVALLEY CANT WRITE SHITTY SOFTWARE THAT WORKS 21:37 Jordach i mean you would've fixed it 21:37 sfan5 ok but that doesn't matter 21:37 sfan5 you made the commit so why does that ownership disappear????????ß 21:39 rubenwardy how did you merge it? 21:39 rubenwardy did you push any commits to it? 21:39 rubenwardy usually it uses the first commit or the PR author 21:40 rubenwardy smells like bugs 21:59 sfan5 rubenwardy: no and I use the "squash and merge" thing 21:59 sfan5 https://github.com/minetest/minetest/commit/8d6a0b917ce1e7f4f1017835af0ca76e79c98c38 the same happened here.... 21:59 sfan5 I kinda feel like this isn't a bug but another poorly though out """"""""""feature"""""""""""""" by github 22:03 rubenwardy this is a bug then 22:03 rubenwardy yikes 22:06 sfan5 umm what https://github.com/isaacs/github/issues/1368 22:06 sfan5 this bug has existed since 2018? that can't be right 22:19 sfan5 https://github.com/isaacs/github/issues/1303#issuecomment-595380191 22:19 sfan5 oh 22:20 sfan5 idiots 23:21 kilbith hmm the clouds are glitchy now 23:22 kilbith with new set_sky + atmos 23:22 kilbith they disappear for like 10 ms. and reappear 23:25 kilbith lol wth: https://i.imgur.com/O3b9iks.png 23:25 kilbith the sun hasn't reached the horizon yet but there is still the sunset 23:25 kilbith how does review PRs? 23:26 kilbith how do you* 23:27 kilbith clouds are glitchy AF 23:29 sfan5 this either isn't a new issue or only happens with atomis 23:29 sfan5 atmos* 23:30 kilbith this is a new issue, clouds were behaving without glitches before 23:30 kilbith https://i.imgur.com/UHGQ4SL.png 23:30 sfan5 I'm sure paramat extensively tested that the default behaviour remains the same, because he very much cares about it 23:30 kilbith why isn't the sun not even clipped at this stage?? 23:31 sfan5 it never was? at least as far as I can remember 23:32 kilbith ok still you have this shocking gap between the sun and the sunset 23:32 kilbith try out 18:00 23:34 sfan5 no sunset texture visible (without atmos), it just barely creeps in at 18:30 when the sun actually hits the horizon 23:35 sfan5 https://a.uguu.se/fvCO2XoA5BGY_FZlMZGo.jpeg 18:00 https://a.uguu.se/MCQLOcBUW6vC_tTNGSzv.jpeg 18:30 23:36 sfan5 what I'm trying to say is, perhaps what you are seeing are bugs in atmos not in the new engine feature 23:38 kilbith ok maybe, the clouds glitches cannot be mod a bug 23:39 kilbith *a mod bug 23:39 kilbith brainfart 23:42 paramat the lack of good game features each release in MTG recently is simply due to lack of core dev time, as we are prioritising working on the engine, it is not due to lack of pilzadam and blockmen. very many good MTG features were added after they left 23:43 sfan5 eh, I'm sure we'd have more features if PilzAdam and/or BlockMen were still here 23:43 kilbith ? 23:43 kilbith why do you say that 23:44 sfan5 are you asking me? 23:44 kilbith oh, gotcha 23:44 sfan5 ... 23:44 kilbith because i'm ignoring paramat 23:45 kilbith uploadina a video to show the cloud glitch 23:46 kilbith uploading* 23:49 kilbith https://www.youtube.com/watch?v=GELiY2sV39I 23:49 paramat sfan, if they had time for MTG yes, but if they did not, which is the current situation, it would make no difference =) 23:49 kilbith glitch at 0:08 23:50 paramat set sky worked fine inmy tests, i did not test atmos because that is quite complex and the chances of bugs caused by it are high. so, probably an atmos bug 23:51 sfan5 you mean the split second in which they disappear? 23:51 sfan5 my guess would be that set_sky is called exactly then 23:51 kilbith yes 23:51 sfan5 ...by atmos of course, which would explain why it doesn't show in mtg 23:52 kilbith why it would be called? 23:52 kilbith the clouds doesn't change 23:53 kilbith this does glitch quite randomly 23:54 sfan5 dunno, I assume atmos periodically refreshes the clouds 23:54 sfan5 Jordach will know what exactly happens better in any casee 23:55 kilbith ok so yes it's refreshed every 0.1s 23:55 paramat might be interference with the weather mod that updates clouds every 8s 23:56 sfan5 might be interference with the weather mod that updates clouds every 8s 23:56 sfan5 fyi 23:56 kilbith but that's still a odd behavior that the clouds have to disappear when you call out set_clouds 23:57 kilbith weather not installed here 23:58 Jordach set_clouds wasn't touched 23:58 Jordach https://metacrane.com/ff/6881/atmos.zip 23:58 Jordach try that version 23:59 Jordach rubenwardy had a similar issue with it being *slightly* outdated