Time Nick Message 00:03 PilzAdam paramat, https://github.com/minetest/minetest_game/pull/374 I guess you missed these fields? 00:06 * paramat looks 00:08 paramat i made pine needles non-flammable because i thought they would be too waxy to burn easily. sorry i forgot the other stuff 00:12 paramat i guess since leaves decay and are lost they could be collected and composted/crafted back into dirt to fill root holes 02:40 paramat back later 04:01 Zeno` #1916 is nice! 04:01 ShadowBot https://github.com/minetest/minetest/issues/1916 -- Restore finalColorBlend implementation in shaders. by RealBadAngel 05:10 hmmmm will anybody cry if I get rid of noise2d_perlin() and friends? 05:10 hmmmm maybe i should just deprecate them and make new functions that take noiseparams, and remove the macros 05:28 paramat noise at point is needed, perlinmaps cant do everything 05:32 paramat btw i attempted to code player settable biome noise parameters, but had no idea what i was doing, so trashed that 05:40 kaeza hmmmm, the new commit seems to work perfectly. thank you 05:47 kaeza unrelated, but now mgv5 is crashing. here's the backtrace: https://gist.github.com/kaeza/8dc47768a5c6e6574f9c 05:54 paramat ! 06:04 paramat warning during build: src/mg_schematic.cpp:60:49: warning: comparison between signed and unsigned integer expressions 06:10 paramat yes mgv5 segfaults for me 06:11 paramat sometimes 06:12 hmmmm hmm 06:12 hmmmm the "sometimes" about this bug is what irks me 06:13 hmmmm did you notice in that backtrace what the value of the emergemanager pointer is? 06:13 hmmmm 0xbadf00d 06:13 kaeza lol 06:15 hmmmm it's funny, but it tells you that it was a placeholder value 06:15 paramat hehe 06:16 hmmmm nobody is going to return 0xbaadf00d on failure, if anything i bet it's the byte pattern that got written to the piece of memory the emergemanager pointer resided when it was freed 06:17 hmmmm EmergeManager clearly exists, it's 0x63cd6a0 06:18 hmmmm or an uninitialized value 06:20 hmmmm https://github.com/minetest/minetest/commit/2f134ce24fa3c9e32f0bbe063adc8114d70091de 06:21 paramat wow thanks 06:21 hmmmm paramat: you didn't need to rename the parameter to emerge_ - the scoping rules specify that local variables have precedence over class scope 06:22 hmmmm although having a name that shadows another name is generally bad 06:22 hmmmm i'll have to (or you, i don't know) fix that problem in all mapgens all at once 06:23 paramat weird i didnt write that 'emerge_' 06:24 hmmmm ahh could have been celeron 06:24 paramat ah i copied from his mgv5 branch 06:25 paramat when i should have checked the corresponding line in mgv7 06:25 hmmmm that's really my own bug 06:25 hmmmm i didn't realize the parameter was named differently 06:26 hmmmm but yet it compiled because there was, in fact, an 'emerge' name in scope 06:31 kaeza I tend to not use underscores at the beginning or end of an identifier for reasons like this. it takes more time to notice it is a different name than if it was something like `the_emerge` or something like that 06:31 kaeza but then, it's a matter of preference, I guess 06:31 hmmmm MapgenV5::emerge is a member that is in all uses, private, and should probably be prefixed with m_ 06:31 paramat works on 4 tries of mgv5, thank you 06:31 hmmmm sorry paramat 06:32 hmmmm i keep breaking your stuff 06:32 hmmmm i do tend to introduce a lot of bugs :-) 06:32 paramat no problem. i should warn you i have a very random head 07:24 paramat testing new biome api nodes: shore, underwater, water_top https://github.com/paramat/minetest/commit/ad63c81d962eb0a748ecd8f42ec90a048813390f 07:24 Calinou they should all be bucketable, beware 07:26 paramat sure, but these are not new nodes, nodes are assigned to these variables 07:55 hmmmm shoot 07:55 hmmmm that's a really stupidly simple idea 07:55 hmmmm a shoreline is the only actual case where we'd need precise biome heights 07:55 hmmmm feels hacky though... 08:02 Calinou Zeno`, you broke view range minimum changing… 08:02 Calinou I can't increase view distance with +/- anymore 08:03 Zeno` ahhh... ok 08:03 Zeno` will fix 08:08 Zeno` fixed. Will merge shortly. Thanks Calinou 08:08 Calinou no problem 08:22 Zeno` will merge #1918 soon after Travis tests (trivial bug/regression fix) 08:22 ShadowBot https://github.com/minetest/minetest/issues/1918 -- Fix regression (increase/decrease viewing range with +/- keys) by Zeno- 08:24 Calinou looks good (no code style mistakes) 08:24 Calinou I think PRs should require approval of 2 core devs instead of 3 08:25 Calinou would get them merged faster 08:26 Zeno` This is ok to merge without approval 08:27 Zeno` i.e. small patch that fixes a trivial mistake 08:28 sol_invictus Calinou: where is that coming from? >instead of 3 08:28 Calinou small PRs would require approval of only 1 08:28 Calinou sol_invictus, I've read/heard that somewhere 08:29 Zeno` pull requests only require approval from 1 other dev 08:29 Zeno` so if you make a PR and I agree and another dev agrees then it can be merged 08:29 Calinou ah 08:30 Calinou but they still get merged so slowly :( 08:30 Calinou especially the good, user-visible ones 08:30 Zeno` Calinou, RBA' 08:30 Zeno` Calinou, RBA's shader PR fixes this: #1705 08:30 ShadowBot https://github.com/minetest/minetest/issues/1705 -- Regression in smooth day-night transition 08:35 Calinou thanks 08:36 Zeno` and it works. I'd merge it right now but I think RBA will turn up in a minute anyway 08:37 Calinou https://github.com/minetest/minetest/issues/1919 ⇒ what is your opinion? 08:37 Calinou I could do it 08:38 Zeno` Is there a tool to convert md to groff? 08:39 Zeno` I guess a man page might be overkill though for an API 08:40 Zeno` Personally I think MD is a good step anyway 08:42 Calinou man page isn't as readable as MD 08:44 Zeno` I agree 08:44 Zeno` Not for this anyway 08:44 Calinou https://github.com/minetest/minetest/pull/1441 08:45 Calinou this has been dormant, why wasn't it merged? 08:45 Calinou this helps modders know which ores are registered 08:46 Zeno` no idea, I think it's a good idea 08:50 Krock Calinou, it's a good idea but ShadowNinja's comment should be read 08:50 Krock the codes are not optimal yet 08:51 Zeno` it optimal but perhaps not as readable as it could be ;) 09:54 RealBadAngel hi 09:55 Krock moin, RealBadAngel 09:58 Zeno` RBA, merge! 09:58 Zeno` heh :) 09:59 RealBadAngel hehe, ok 10:00 Krock does #1920 happen on linux too? 10:00 ShadowBot https://github.com/minetest/minetest/issues/1920 -- Ignore .git directories in modpacks 10:06 RealBadAngel Zeno`, meged 10:06 RealBadAngel +r 10:11 Zeno` nice 10:19 Krock #1921 linux test please 10:19 ShadowBot https://github.com/minetest/minetest/issues/1921 -- Ignore .name directories and files by SmallJoker 10:27 Zeno` Krock where does it show up? (before the patch) 10:27 Krock Zeno`, in the mod enable/disable menu 10:28 Zeno` checking 10:28 Zeno` I probably never see it because I used worldmods/ ?? 10:28 Krock example: clone mesecons and you'll see a .git directory in the modpack 10:29 Zeno` yeah I do it a bit different. Setting up a test 10:34 Zeno` weird 10:34 Zeno` doesn't show them for me 10:34 Zeno` looking into it 10:34 Krock it must be a modpack 10:35 Zeno` plantlife is a modpack? 10:35 Zeno` certainly looks like it 10:35 Zeno` I'll try mesecons as well 10:40 Zeno` Krock, you cannot remove the check for ".." 10:40 Zeno` still checking the other stuff 10:40 Krock ".."'s first letter is '.' 10:41 Krock will be skipped aswell 10:41 Calinou also check for ~ at the end of the name 10:41 Krock why ~? 10:41 Calinou file managers usually hide these 10:41 Calinou backups 10:41 Krock ah 10:42 Zeno` yes, that's true 10:42 Zeno` this is all wrapped in #ifdef _WIN32 so I cannot test 10:42 Krock Zeno`, I tested the win32 part 10:42 Krock but not the linux one 10:43 Zeno` sorry, that's what I meant. 10:43 Zeno` first impression is that it looks fine, but I don't see the original issue for some reason. Give me 5 to work out why :) 10:44 Krock ok. 10:45 Krock eww. I forgot again how the function was called to get the length of a string 10:46 Zeno` lines 259 and 260 do not appear necessary 10:47 Zeno` because they're not even reached if line 256 is false 10:49 Krock oh yes. 10:50 PilzAdam good morning 10:50 Zeno` (which is not something you added I know) 10:51 Krock moin, PilzAdam 10:51 PilzAdam sfan5, https://github.com/minetest/minetest_game/pull/373 is tested now 10:51 Zeno` But these linux lines explain why I don't see the issue. In that case I approve. May as well fix that superfluous check while you're at it though 10:51 sfan5 nore already agreed, right? 10:51 sfan5 if so then you can merge it 10:52 Calinou hi PilzAdam 10:52 nore sfan5, I did 10:52 PilzAdam yes, nore agreed 10:52 nore hi btw 10:52 sfan5 hi nore 10:53 PilzAdam merged 10:53 PilzAdam what about https://github.com/minetest/minetest_game/pull/374 ? 10:55 Krock Zeno`, updated. 10:56 Krock (also replaced spaces by tabs) 10:57 sfan5 PilzAdam: looks good; thats a small fix; I'd suggest naming the commit 'Correct node definitons for pine saplings and pine needles' 10:57 PilzAdam ok, I'll merge it then with the update commit message 11:00 Zeno` Krock, that looks good 11:02 Zeno` #1922 11:02 ShadowBot https://github.com/minetest/minetest/issues/1922 -- Fix f6 debug/profiler display by Zeno- 11:16 Zeno` Do any devs agree with #1917 ? 11:16 ShadowBot https://github.com/minetest/minetest/issues/1917 -- Add space by srifqi 11:18 Zeno` Any other devs* 11:20 PilzAdam Zeno`, yes, 1917 seems good 11:22 Zeno` Ok, I'll merge in 10 minutes (roughly) 11:23 Zeno` Also, there seems to be a lack of comments on #1772 11:23 ShadowBot https://github.com/minetest/minetest/issues/1772 -- Increase third person view distance by Calinou 11:24 Zeno` I was wary about it at first but after testing I think it's better than the current *2 11:28 PilzAdam sfan5, nore, https://github.com/minetest/minetest_game/pull/375 11:28 Zeno` merging 1917 now 11:28 PilzAdam also, RealBadAngel, do you want to look at ^, since you originally wrote it (IIRC) 11:37 Zeno` I'd like to merge #1772. Any comments? Any agreement? 11:37 ShadowBot https://github.com/minetest/minetest/issues/1772 -- Increase third person view distance by Calinou 11:38 PilzAdam Zeno`, testing... 11:41 Zeno` At least it's easy to test with different values :) 11:41 PilzAdam Zeno`, it's better 11:41 Zeno` ok, so merge? 11:41 PilzAdam I'd say yes 11:41 Zeno` merging ;) 11:41 PilzAdam another thing I noticed: if you fly with noclip through stone the camera is always "pushed" to the player 11:42 PilzAdam maybe don't move the camera closer to the player if noclip is enabled? 11:42 * Zeno` looks 11:45 Zeno` is that an issue? 11:46 PilzAdam yes, at least for me 11:48 PenguinDad Zeno`: it's an issue for me too 11:50 Zeno` where should it be re-positioned to if noclip is active? 11:50 Zeno` (this seems to be a current issue rather than something introduced by the patch) 11:50 PilzAdam yeah, its unrelated to 1772 11:51 Zeno` I just made a quick test and disabled repositioning if player can have noclip and noclip is also enabled for them 11:52 Zeno` the problem is that it makes outside viewing "incorrect" just because noclip is enabled 11:52 Zeno` I think that possibly I should merge 1772 and a new issue be opened by somebody? 11:53 Krock ^looks okay 11:53 Zeno` PenguinDad, can you confirm this is an existing issue and not something related to #1772? 11:53 ShadowBot https://github.com/minetest/minetest/issues/1772 -- Increase third person view distance by Calinou 11:53 PilzAdam yes, merge 1772 first 11:53 Zeno` ok 11:54 PenguinDad Zeno`: it's of course unrelated to #1772 11:54 ShadowBot https://github.com/minetest/minetest/issues/1772 -- Increase third person view distance by Calinou 11:54 Zeno` ok, just making sure ;) 11:55 Krock Zeno`, does changing the font size not help to solve #1922 ? 11:55 ShadowBot https://github.com/minetest/minetest/issues/1922 -- Fix f6 debug/profiler display by Zeno- 11:56 Zeno` Krock, doesn't seem to. Try it before and after the patch (feedback is of course most welcome) 11:57 Krock Zeno`, I see. the height of that window is broken before the patch 11:58 Zeno` yeah 11:58 Krock why is getTextHeight() not doing its job well anymore? 11:58 Krock looks like it returns 0 all the time 11:59 Zeno` nope 11:59 Zeno` it returns 16 11:59 Krock hmm okay 12:02 PilzAdam Zeno`, https://gist.github.com/PilzAdam/5386e4e1b38dbd6bff87 12:03 Krock Zeno`, it's not possible to change fov in-game 12:04 Zeno` Krock? It wasn't before either (not with my tests anyway) 12:04 Krock Zeno`, it was possible before those settings by variables came 12:05 PilzAdam Zeno`, in singleplayer /set modifies the client's settings 12:06 Zeno` Krock, there is a big TODO there. That will be address. I think the most important thing now is the speed improvement. Unless changing fov while playing is a deal-breaker of course... 12:07 Zeno` PilzAdam, ah, I see now. Because that would affect the server instance "simple single player" launches 12:07 Krock good, I hope it's fixed before release of 0.4.11 12:07 Zeno` Krock, you really change fov commonly when playing? 12:08 Krock Zeno`, no but when I do this, I don't want to rejoin to see how it looks like 12:09 Zeno` hmm, ok 12:10 Zeno` there is a much larger (and more correct) solution planned for after 0.4.11 (because I didn't want to introduce major changes at this late stage in the dev/release cycle) but I can make fov changeable in singleplayer games again 12:18 Zeno` https://github.com/Zeno-/minetest/commit/d7d1ee82deeab3256cea6e1a2df334601390e928 12:19 Zeno` will merge in 10 12:20 Krock wait for an other agree 12:21 Zeno` well, it's a simple "revert" of one line (basically) but ok 12:22 Zeno` PilzAdam, should I merge that? 12:22 Zeno` :D 12:22 PilzAdam one could ask why you only do it for fov, since all client side settings can be changed in singleplayer... 12:23 Krock better modify the /set command somehow 12:23 PilzAdam Krock, no, definetly not 12:23 Zeno` one could ask why using the non_cached versions increases the time spent in the main client loop by 7-10% :) 12:23 PilzAdam don't fix glue with more glue 12:24 PilzAdam I don't know.... ask someone else 12:24 Zeno` The real solution is a whole new class but adding that is something a relatively large change at this point (with release around the corner) is not.. good :) 12:25 Zeno` The downside is some setting cannot be changed when playing simple single player. Most commercial games have lots of settings that require re-launching the game 12:27 Zeno` e.g. changing antialiasing options in all games I can think of require a game restart 12:27 Zeno` the few that don't close the device and re-open it 12:28 Zeno` Maybe some settings are more important than others :) 12:31 kilbith Zeno: why merged Calinou PR so that you were testing the movable camera earlier? 12:31 Zeno` kilbith, because I have a huge new class for the movable camera 12:32 Zeno` But I don't feel that adding a PR for it before 0.4.11 is wise 12:32 Zeno` I'd rather concentrate on existing issues this late in the cycle 12:32 Zeno` Besides it needs a lot of testing 12:32 kilbith gtg, will read logs sorry 12:39 Zeno` one could also argue that it's not a bug because the documentation says that /set is for server settings 12:39 Zeno` simple_single_player_mode is a bit of a spanner in the works though :( 13:13 PilzAdam Zeno`, have you looked at that gist? 13:17 Zeno` yeah 13:17 Zeno` I only tested for about 5 minutes but in that time it seemed ok 13:18 Zeno` It's similar to what I did (as a quick and dirty solution just to test) but higher up in the "hierarchy" 13:19 Zeno` I think the camera for 3rd person view could do with a lot of tweaks and improvements 13:20 Zeno` It's an area rich for experimentation and testing 13:27 PilzAdam yep 13:36 Zeno` Comments on #1922 ? 13:36 ShadowBot https://github.com/minetest/minetest/issues/1922 -- Fix f6 debug/profiler display by Zeno- 13:52 Zeno` who coulda thunk that a bug fix would be so hard to merge 13:53 PilzAdam Zeno`, merge it if it works 13:55 Krock Is this a MTG-wide issue with the non-switchable screwdriver? 13:56 PilzAdam did someone say screwdriver? https://github.com/minetest/minetest_game/pull/375 13:58 Krock no no. current screwdriver is good. 13:58 PilzAdam what? are you serious? 13:59 Krock yes. 13:59 Krock it contains rotation possibilites into all directions 14:00 PilzAdam can you explain how it works in one sentence? 14:01 Krock A screwdriver lets you rotate a node with a specified mode which is set by holding left mouse button and shift. 14:01 PilzAdam that sentence doesn't explain what a "mode" is 14:02 PilzAdam for the rewritten screwdriver: "lefclick rotates around axis and rightlick changes axis" 14:02 Krock which axis? 14:02 PilzAdam all of them 14:02 Krock aha 14:03 Krock might takes a while until I get that new system 14:03 PilzAdam I took me a minute or two; you can see the effect best with stairs 14:03 PilzAdam leftclick rotates the smaller part, while the bigger "bottom" stays where it is 14:04 PilzAdam and rightclick changes where the "bottom" part is 14:04 Krock ah, but I think it requires more clicks until the target is reached 14:05 PilzAdam with the old system you simply clicked randomly until you get your wanted result 14:05 PilzAdam the new system allows you to actually understand what you are doing 14:05 Zeno` that'd be nice :) 14:08 Krock I see stairs got a rewrite, I can not put a stair onto an other anymore 14:08 PilzAdam Krock, no? 14:09 Krock hmm then I must blame an other subgame 15:25 Tesseract Lock https://github.com/minetest/minetest_game/pull/360 ? Comments are off-topic or posted without reading the description. Or delete comments. 15:26 Tesseract Also cg72 is accusing of using her ideas without credit in another (closed) issue, which AFAIK isn't correct. 15:32 PilzAdam sfan5, https://github.com/minetest/minetest_game/pull/376 15:33 Krock PilzAdam, why default.LIGHT_MAX ? It's a global thing 15:34 PilzAdam Krock, everything global a mod defines should be in a namespace 15:34 PilzAdam it could be moved to builtin, though 15:34 RealBadAngel Zeno`, PilzAdam adding new style settings class wont hurt, because it will work mutal 15:34 Krock I agree. Please remove it on that place after adding it to the builtin mod 15:35 RealBadAngel new class will use old way on startup - thats first 15:35 RealBadAngel second, we can replace calls for settings one by one 15:36 RealBadAngel at the very end old way will remain there just for lua mods 15:37 RealBadAngel with 0,5k of calls in whole code, mainly in loops we can only gain speed 15:37 RealBadAngel and IF c55 doesnt understand that, lets buy him lower end machine to check the impact 15:39 Zeno` this is what I tried to do a week or so ago 15:39 Zeno` and it's what *needs* to be done (I agree with RBA) 15:40 Zeno` But, I am happy to wait until 0.4.11 is released to do this 15:40 RealBadAngel about screwdriver, PilzAdam, if somebody cant read, thats not my fault 15:40 Zeno` mainly because there is so much stuff to debug already 15:41 RealBadAngel modes of screwdriver are explained on use 15:41 Zeno` and more stuff to debug gets added every day 15:41 Zeno` there has to be a feature freeze 15:41 Zeno` not this willy nilly let's make a new fancy class do dah and merge it without asking other devs 15:42 rubenwardy PilzAdam, how many mods will that LIGHT_MAX change break? 15:42 Zeno` let's fix what is there now! *before* adding other stuff :( 15:43 rubenwardy some mods us light = light_max - constant 15:43 RealBadAngel making fancy new class with no use (yet) is the best way 15:43 rubenwardy *use 15:43 PilzAdam RealBadAngel, there are 2 problems with the current screwdriver: 1) some modes depend on the position of the player that uses the screwdriver 2) mode switching is too slow and unhandy 15:43 RealBadAngel then adding a few places where it will be used its the right way 15:43 PilzAdam rubenwardy, 0 15:43 Zeno` RBA, I'm not talking about the class we propose to merge 15:43 RealBadAngel yes 15:44 RealBadAngel first we need clean implementation 15:44 RealBadAngel without usage at all 15:44 Zeno` Although if I am against other classes being created, used, and merged at this point in time I can't argue that our new class should be different 15:44 Zeno` let's just get 0.4.11 out the door! 15:44 Zeno` right at this moment every day is a moving target 15:44 Zeno` :( 15:44 rubenwardy https://github.com/crazyginger72/mtlite_game/blob/24fa618adef3d6eecc0ec6997f41957250e65b7b/mods/farming/init.lua#L16 15:44 rubenwardy https://github.com/HybridDog/zegaton_game/blob/63e82a2c5993dd9070cb4b7d218b28562df8ec22/mods/homedecor/television.lua#L23 15:44 rubenwardy These aren't broken? 15:45 RealBadAngel PilzAdam, ofc theyre depending on players position 15:45 rubenwardy bad examples, those are games 15:45 RealBadAngel they were designed to work that way 15:45 kaeza rubenwardy, LIGHT_MAX global is still defined for backward compat 15:45 rubenwardy Okay 15:45 RealBadAngel look at face, bring top there, then rotate 15:45 RealBadAngel for example 15:45 PilzAdam rubenwardy, yes, because a) these are different games, so they don't even use minetest_game; and b) I added LIGHT_MAX under "Legacy" 15:46 PilzAdam RealBadAngel, that is bad design 15:47 kaeza PilzAdam, BTW, there are ways to make access to global LIGHT_MAX, etc, issue a warning, but not sure if that is desired 15:47 PilzAdam RealBadAngel, also some modes even depend on the face you point at _and_ the player position 15:48 PilzAdam when rotating stairs its easy to point at a face that you don't even want to point at 15:48 Zeno` Let me ask this. How are we going to release 0.4.11 if new features are added every day without peer review? 15:49 Zeno` There is no time to test and squash bugs :( (IMHO) 15:49 Zeno` I'm only saying this because 0.4.11 is supposed to be released soon (?) 15:50 Zeno` at the start of the cycle it's a different story 15:50 PilzAdam Zeno`, features are added until a feature freeze is declared 15:50 Zeno` PilzAdam, yes I understand, but ... yeah 15:51 RealBadAngel PilzAdam, if you show me better and working screwdriver i will agree. atm current system is good for me. but im open for innovations 15:51 PilzAdam RealBadAngel, https://github.com/minetest/minetest_game/pull/375 15:51 PilzAdam easy to understand 15:51 PilzAdam the whole explanation fits into the tooltip (!)! 15:52 RealBadAngel PilzAdam, i will try it 15:55 PilzAdam this also has a few other benefits: no 32x32 textures needed; you can use the screwdriver while sneaking; the chat is not spammed with text about modes 15:56 Zeno` I guess what I'm saying is that there is (probably) enough bugs and "blockers" to call a feature freeze *now* 15:56 RealBadAngel PilzAdam, i will test it. If i find it comfortable and easy to understand i will accept it 15:56 Zeno` if we don't then more will be added and 0.4.11 release will be delayed yet again 15:57 RealBadAngel Zeno`, we still have some time till Xmas edition 15:58 RealBadAngel ive closed one blocker today 15:58 Zeno` not a lot though 15:58 RealBadAngel will try to close another one 15:59 Zeno` fair enough 15:59 RealBadAngel with more changes, like settings class we can wait 15:59 RealBadAngel for next dev phase 15:59 PilzAdam sfan5, nore, will any of you (or better both) be available today for approving pull requests for minetest_game? 15:59 PilzAdam if not then I might as well do something else 16:00 RealBadAngel its kinda big change and shouldnt be applied in a hurry 16:00 sfan5 PilzAdam: maybe later 16:00 PilzAdam ping me then 16:00 RealBadAngel lets do .11 first (with closing as many issues we can) and then we can start adding new stuff 16:01 RealBadAngel theres enough of new things since .10 17:15 Krock RealBadAngel, could you give me a +1 for #1921 ? 17:15 ShadowBot https://github.com/minetest/minetest/issues/1921 -- Ignore .name directories and files by SmallJoker 17:18 Tesseract Feature freeze tomorrow? That way there's still some weekend time for development. The freeze should last at least a week. 17:22 Calinou Tesseract! I used you to preview my textures 17:22 Calinou anyway… thanks for merging my PR about third person distance 18:50 PilzAdam sfan5, https://github.com/minetest/minetest_game/pull/368 the texture are updated now 18:52 Krock PilzAdam, how about ^[brighten from wet soil -> dry soil? 18:52 PilzAdam (nore already agreed to that one) 18:52 Calinou ^[brighten may be too bright, its effect can't be controlled 18:52 PilzAdam Krock, some texture packs might not display dark soil as dark 19:02 hmmmm are we in feature freeze yet? 19:03 Krock /topic does not say so. 19:06 kahrl the target for that was around the 10th IIRC 19:07 kahrl ...which means I should fix the serverlist icons pull today 19:07 Krock Yes please. 19:08 Krock And I need two +1 for my last pull request 19:11 kahrl Krock: the GetDirListing one? 19:12 Krock kahrl, exactly. 19:12 kahrl let me check 19:12 Krock I tested it on Windows. Unknown for Linux 19:12 kahrl I've always wanted to rewrite GetDirListing completely 19:13 kahrl the windows version I mean 19:13 kahrl to get rid of the C style mallocs and sprintfs 19:15 Krock You'll surely have time somewhen to do that :) 19:15 kahrl which aren't bad in themselves but I think there are some off by one errors 19:15 Krock But I think the serverlist has a higher priority 19:15 kahrl check my windirlisting branch 19:16 kahrl I don't have a windows compiler setup and at the time I wrote it, nobody wanted to test it, so it was forgotten 19:17 kahrl https://github.com/kahrl/minetest/commit/5420a3a7ed3280a466a457c6e1814b7fbfc9ccd4 19:18 Krock kahrl, looks good but could you please make it up-to-date, so I don't need to recompile the complete minetest source? 19:20 kahrl Krock: https://github.com/kahrl/minetest/commit/eca2fc046478627db6df8cd09d046a3cf013e47e 19:21 kahrl another thing (more related to your PR) is: some caller actually want the dotfiles, e.g. RecursiveDelete 19:21 kahrl callers* 19:22 kahrl so perhaps the best strategy is to add an optional parameter include_dot_files = false 19:23 Krock kahrl, I think it's better to do it like linux and simply hide it 19:23 kahrl why? 19:23 Krock Well, I could make it also lua-sided on the mainmenu 19:24 kahrl if RecursiveDelete isn't guaranteed to work then it should simply be deleted 19:24 kahrl recursively :P 19:24 Krock kahrl, https://github.com/minetest/minetest/blob/master/src/filesys.cpp#L256 Linux already ignores it 19:25 kahrl yeah, and I don't like how this is OS dependent 19:25 kahrl RecursiveDelete isn't affected on linux because it runs rm -rf (the actual command) 19:26 Krock so, you would suggest to remove that '.' limitation from linux codes too and add a check on the mainmenu codes? 19:26 Krock s/on/in/ 19:26 kahrl which mainmenu codes? 19:27 kahrl oh you mean because of #1920 19:27 ShadowBot https://github.com/minetest/minetest/issues/1920 -- Ignore .git directories in modpacks 19:27 Krock yes 19:28 kahrl well you could add a parameter ignore_dot_files to core.get_dirlist 19:28 kahrl *include_dot_files 19:28 Krock https://github.com/minetest/minetest/blob/master/builtin/mainmenu/modmgr.lua#L21 19:28 Krock I think I'll add the limitation there :) 19:30 kahrl how about the texture packs tab? 19:30 kahrl it also uses core.get_dirlist 19:31 Krock ah yes. 19:36 hmmmm ugh crap 19:36 hmmmm I accidentally de-optimized noise 19:37 hmmmm couple of commits ago i switched back to celeron's original trilinear interpolation that uses 31 operations as opposed to 21 19:39 Krock :/ 19:39 kahrl \: 19:43 Krock kahrl, merged your commit and mine together. compiling... 19:44 PenguinDad hmmmm: re-optimize it then! ;) 19:50 Krock kahrl, 20:52:28: ERROR[main]: GetDirListing: FindFirstFile error. Error is 3 19:51 kahrl ok, I'm glad I didn't simply merge it without testing then ;) 19:51 kahrl 3 is ERROR_PATH_NOT_FOUND 19:53 PilzAdam hmmmm, earlier today I tried to create a world with mgv5; when joining for the first time it said something like "could not create/read map_meta.txt" and then a segfault 19:53 PilzAdam I can't reproduce it anymore 19:53 hmmmm hrmm 19:53 hmmmm interesting 19:53 hmmmm would be nice if you can get a backtrace though 19:54 PilzAdam when I restarted Minetest and joined the world it filled map_meta.txt with params for v6 19:54 hmmmm v6 is the default_mapgen 19:54 Krock kahrl, it shows up the mods correctly 19:54 kahrl Krock: can you make it print what dirSpec is when the error occurs? 19:55 hmmmm it failed to read the params for the map currently so it figured "welp, let's just use the v6 params" 19:55 Krock kahrl, ye sure if you want to wait 10 minutes again 19:55 kahrl are you on an 80386? 19:56 Krock kahrl, Athlon 4000+ 19:56 kahrl what does it do in those 10 minutes then? :P 19:56 Krock It compiles the .cpp files in the first 10s and links the stuff for 9min 50 19:57 PilzAdam also, kahrl, sometimes when clicking on "New world" the header imagine turns just white 19:57 Krock because incremental linking is disabled 19:57 PilzAdam *image 19:57 Krock actually, it only takes 4s for filesys.cpp 19:58 kahrl I don't use incremental linking either as far as I'm aware and it takes 2s 19:58 kahrl but that's on linux 19:58 Krock LNK4075 warning: /INCREMENTAL is disabled due the parameter /LTCG 19:58 kahrl oh 19:58 kahrl so it actually compiles the whole minetest each time you link 19:58 kahrl minus the parsing stage 19:58 Krock I've tried to remove that parameter but it causes troubles in an other library 19:59 kahrl because with /LTCG the object files only contained the parsed source code in some tree format or something like that 19:59 kahrl contain* 19:59 Krock Using incremental linking would speed up my developing speed a lot :3 19:59 exio4 get a new computer, Krock 20:00 exio4 that speeds up everything, even minetest! 20:00 Krock exio4, it is not always a good soulation to speed up things with better hardware. 20:00 exio4 for compiling, if you can't switch compilers (something like, gcc -> clang), then it is the best you can do 20:01 Krock kahrl, 21:03:11: ERROR[main]: GetDirListing: FindFirstFile error. Error is 3, path: E\* 20:01 kahrl aha 20:01 Krock 21:03:34: ERROR[main]: GetDirListing: FindFirstFile error. Error is 3, path: E:\Programme\minetest\bin\..\textures\server\* 20:02 Krock 21:03:34: ERROR[main]: GetDirListing: FindFirstFile error. Error is 3, path: \Programme\minetest\games\base_game\mods\dye\media\* 20:02 Krock and many more 20:02 kahrl so... I guess those problems were there before but the old code simply didn't tell you about them 20:02 Krock yes 20:03 kahrl which makes sense if a lot of code calls this without first checking if the dir exists 20:03 Krock Well, it should ignore those things because it is not expectable that the 'models' directory exists in every mod 20:04 kahrl so replace if (dwError != ERROR_FILE_NOT_FOUND) with if (dwError != ERROR_FILE_NOT_FOUND && dwError != ERROR_PATH_NOT_FOUND) 20:04 Krock ok 20:05 Krock wait. that's not good. 20:05 Krock better return the empty listing if it hasn't been found 20:05 kahrl isn't that what it does? 20:05 Krock yes but it it already returns a not_found error, then why checking if there are filesin the directory? 20:06 kahrl I'm not sure what you mean 20:06 Krock Well, it returns an error message when it throws an other error then file_not_found or folder_not_found 20:07 kahrl yes 20:07 Krock Otherwise it tried tog et the files of that folder 20:07 kahrl I think that's reasonable 20:07 Krock *tries to get 20:07 kahrl no 20:07 Krock oh sorry 20:07 kahrl dir doesn't exist -> FindFirstFile returns INVALID_HANDLE_VALUE -> "if" branch is executed -> dwError is ERROR_PATH_NOT_FOUND -> inner if branch is not executed -> return listing -> empty listing is returned 20:08 Krock I messed up the stuff because I replaced the wrong line /headdesk 20:08 Krock sorry for the trouble 20:10 Krock kahrl, added your changes with the corrected line: https://github.com/SmallJoker/minetest/commit/c0a68b33 Is that okay? 20:10 kahrl looks good 20:11 Krock and since there's no other error code than 3, it must work 20:26 kahrl so, about #1891 20:26 ShadowBot https://github.com/minetest/minetest/issues/1891 -- Display serverlist flags as icons by kahrl 20:26 kahrl I agree that there is too little horizontal space for the server name in its current form 20:27 kahrl putting the icons closer together only helps very little 20:27 kahrl so how about something more radical 20:27 kahrl why not get rid of all the vertlabels on the left? 20:28 kahrl they're redundant with the tab header and taking up space for no reason 20:28 kahrl also very ugly since c86b78c9b 20:31 kahrl I also thought about reorganizing the whole menu (moving the server description below the serverlist and extend the serverlist to the right) but since I can't increase the menu height atm that's out of the question 21:32 paramat nore sfan5 roots are removed from pines, i will rebase in an hour then it will be ready for merge https://github.com/minetest/minetest_game/pull/366 21:42 kahrl updated #1891 21:42 ShadowBot https://github.com/minetest/minetest/issues/1891 -- Display serverlist flags as icons by kahrl 21:43 kahrl I removed the vertlabels and re-layouted, I think it looks pretty good 21:43 kahrl most importantly, the serverlist is now wider 21:44 kahrl http://imgur.com/a/PjfAY#1 21:45 kahrl btw, why does it always show "Favorites:" even when not showing the favorites list but the public serverlist? 21:45 kahrl I never really noticed that 21:56 Taoki Made a little pull request, #1924 21:56 ShadowBot https://github.com/minetest/minetest/issues/1924 -- Liquid node footsteps by MirceaKitsune 21:57 Taoki If it's accepted, I might add a swimming sound to minetest_game myself, and make a pull for that too. 22:00 paramat i have some watery swimming sounds in my 'trail' mod, you're welcome to use them 22:01 Taoki paramat: What's the link? Anyway doing them in Lua isn't a good idea. My pull fixes a small issue which prevented the footstep sounds of luqid nosed from being accounted, they will however work now 22:01 Taoki **liquid nodes 22:01 paramat link coming 22:03 paramat https://forum.minetest.net/viewtopic.php?f=11&t=6773 22:04 Taoki Thanks 22:04 paramat i think it should be done without view bobbing though for simplicity, also i can't stand that effect 22:05 Taoki View bobbing not working in water was probably a bug in itself, this fixed both it and footstep sounds. 22:05 Taoki If it's disabled in the settings then of course it doesn't happen in either liquid or land :) 22:05 paramat those sounds are bubbly so good for underwater too 22:06 Taoki Yeah 22:06 paramat hm but they are perhaps too long to be triggered every footstep 22:07 Taoki I can make them shorter in Audacity, not a problem 22:15 kilbith kahrl: you can place players online/max in a tooltip and put a coloured "loading bar" icon for that 22:16 kahrl hmm, interesting idea 22:16 Taoki Updated it to also work for climbing. It should be possible to now give the ladder footstep sounds, and have climb sounds as you go up or down 22:17 kahrl kilbith: although that makes it hard to differentiate between 0/... and 1/... 22:17 kahrl which is often a crucial difference 22:18 kahrl also I'm not sure how to implement it with table[] 22:18 kilbith that why this bar should adapt to the percentage of online/max 22:19 kahrl but it would be too small to have any visual difference between, say, 0/50 and 1/50 22:19 Taoki Ok, actually I need to fix a few things... 22:19 kahrl because it would simply be shorter than 50 pixels 22:20 kilbith hmm, is it really important to differenciate 0/... and 1/... ? 22:21 kahrl I'm pretty sure for a lot of people it is 22:21 kilbith 0/... = empty bar, thats all 22:22 kahrl they don't want to play on a completely empty server and want to chat with other people, for others it's vice versa 22:22 kahrl yeah I guess 0/... could have a special icon 22:22 kilbith yes 22:23 kahrl but imo the absolute number of players is important too 22:23 kahrl for example if it's high, the chat will be very busy and might be hard to follow 22:24 kilbith would be specified in tooltip 22:24 kahrl right. 22:26 kahrl so the only question would be how to implement it 22:26 kahrl right now a tooltip can't be specified per row, only per column 22:26 RealBadAngel #1925 any objections? 22:26 ShadowBot https://github.com/minetest/minetest/issues/1925 -- Disable loading .mtl files. Theyre not used anyway. by RealBadAngel 22:27 kahrl and you can't draw arbitrary things into a table, but that could be solved by just having say six icons for 0%, 1%-25%, 25%-50%, 50%-75%, 76%-99% and 100% 22:31 kilbith and for the lag icon ? any plan for it ? 22:33 kahrl I'm not sure if the lag is accurate 22:33 kahrl if it is, then why not 22:35 kilbith also, i thought that server admins could place a small image under the description 22:35 kilbith limited in size ofc 22:37 kilbith that way servers should be more attractive 22:40 kahrl sounds good but is again way out of scope ;) 22:41 kahrl this would have to be somehow encoded and sent to the masterserver, then pulled from there by the client 22:42 kahrl ideally not within the usual json file that's downloaded, otherwise it will become completely impossible to debug this json 22:45 kahrl right now the purpose of this PR is to get done in time before feature freeze 22:45 kahrl any icing on the cake can be done next year :) 22:46 kilbith you can simply define a local location for that image - minetest.conf for enabling 22:48 kilbith hmm, nvm 22:48 kahrl I'm not saying it's hard to do 22:48 kahrl just would require masterserver and core changes (since the serverlist announcement is in C++) which I don't want in this PR 22:49 kilbith yes, do not bother with that 22:55 paramat sfan5, nore has left but this is now rebased https://github.com/minetest/minetest_game/pull/366 i removed the roots as requested by nore 23:02 Taoki Cool... climbing sounds on the ladder work out of the box with my patch :) 23:11 RealBadAngel can i merge #1925? its a one liner 23:11 ShadowBot https://github.com/minetest/minetest/issues/1925 -- Disable loading .mtl files. Theyre not used anyway. by RealBadAngel 23:27 Taoki Since I was commenting on a few pull requests, I was wondering about something: There's a rule that in order to accept a pull, at least 3 (if I'm correct) developers need to agree with the change. Do I... count? 23:28 Taoki Since I don't have direct pull rights on Github. But I'm hoping I am in the dev team as well :) 23:32 jin_xi +1 from me for Taoki on dev team, even tho my vote wont count... 23:34 Taoki Thanks. Not really sure how much I'm in for now, to be honest... 23:35 Taoki Truth be told, I wanted to discuss about that on the forum. I'm already familiar with the code, and would like to be able to contribute even more and better. 23:47 RealBadAngel can someone explain meaning of this? https://github.com/minetest/minetest/blob/master/src/client.cpp#L2654 23:47 RealBadAngel https://github.com/minetest/minetest/blob/master/src/client.cpp#L2624