Minetest logo

IRC log for #minetest-dev, 2013-04-14

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:18 Taoki joined #minetest-dev
00:37 Taoki joined #minetest-dev
00:54 dexter0 joined #minetest-dev
01:04 dexter0 joined #minetest-dev
01:36 Taoki joined #minetest-dev
02:08 proller__ beta version of math mapgen ready: https://github.com/proller/minetest/compare/master...math2
02:20 proller__ joined #minetest-dev
03:23 RealBadAngel joined #minetest-dev
04:33 hmmmm hey guys, if you return 0 in a lua api, it marks that the call failed, but how is that reflected in lua?
04:34 Exio hm
04:35 Exio i think you'll end faster by trying :P
04:35 hmmmm i wouldn't know what i'm looking for
04:36 Exio i mean, print (minetest.stuff) or w/e it is in lua for checking what it returns or so, no idea really how it  manages that stuff
04:36 Exio maybe it just returns nil?
04:36 hmmmm ah
04:36 ShadowNinja Do you mean returning 0 from a C++lua api function?
04:37 hmmmm yes...
04:37 ShadowNinja I would imagine you would get nil
04:37 hmmmm anyway, returning nil seems reasonable
04:37 hmmmm i just hope it's not the last argument that was pushed onto the stack
04:38 ShadowNinja you could just lua_pushnil of course.
04:39 hmmmm i could, but i'm not the one doing it
04:39 hmmmm there are lots of apis that, on failure, return 0 but never push a return value to the stack
04:39 hmmmm i could just as well read the official documentation and find out, but i thought asking here would be quicker....
04:42 hmmmm Taoki, there are a couple questionable things about ParticleSpawner that I am seeing while fixing up the HUD code that I'd like to ask you about...
04:42 hmmmm what is the purpose of Server::spawnParticleAll and Server::spawnParticle?  they are _never_ called
04:42 Exio jeija coded the ParticleSpawner
04:42 hmmmm oh, i thought that was taoki
04:42 hmmmm sorry
04:42 Exio hehe :P
04:42 hmmmm jeija is not here, which is inconvenient
04:42 hmmmm and of course not there either
04:43 ShadowNinja he was around today, pushed some commits to mesecons
04:45 hmmmm and now i see why blue42u wanted to use strings as internal IDs for the bars....  the way i envisioned it being done was quite different
04:47 hmmmm i'm gonna change that map to a vector like it probably should be
04:48 hmmmm technically speaking, there was nothing really *wrong* with his original code
04:48 hmmmm it's just that i am not half-assing this... i'd like the bar api to be done well and done right the first time
04:48 hmmmm the way i envisioned it
04:49 hmmmm i really wish i could say that minetest meets or exceeds commercial code quality, like many other FOSS projects, but i can't and it hurts
04:55 Exio https://github.com/minetest/minetest/pull/650
04:55 Exio shouldn't it be closed?
04:56 hmmmm it isn't?
04:56 Exio aw
04:56 Exio did you close it a second after i finished loading it?
04:56 hmmmm yes
04:56 Exio pressed F5 and it is closed
05:00 hmmmm hmmm
05:00 hmmmm i do find myself having to fix a lot of stuff with this feature
05:01 Exio what feature?
05:01 hmmmm the lua hud
05:01 Exio luahud/api?
05:01 Exio ah
05:01 hmmmm it's going to be very nicely done and well polished though
05:19 Exio ha, perfect
05:22 Exio https://github.com/EXio4/minetest/commit/0e79f85312aced31f0cf755e7a8d404eb07d2b0a
05:34 hmmmm i guess i can add that in
05:35 Exio i tried changing the colors for a white crosshair / alpha channel too, and works btw :P
05:47 ShadowNinja how can I check if a node has a certain group from C++?
05:49 Exio there is some code doing that https://github.com/Jeija/minetest/commit/6715fa43c0bed797fc5a5b7cd4a982d8b7b41f3d
05:50 ShadowNinja thanks
05:51 hmmmm there's code that does almost exactly that in itemgroup.h
05:51 hmmmm (the ONLY code in that file)
05:53 ShadowNinja big file, we should split that up a bit ;-)
05:56 hmmmm i guess celeron was expecting the whole concept of item groups to be more grand than it turned out to be
05:59 Exio i would not think he was excepting having a "real" game .. he started using trollfaces for the players
05:59 Exio :P
06:14 ShadowNinja Seriously, what I just wrote is compiling but I might have made ten new warnings, but I can't tell because it is buried under other warnings, is fixing those on your list?
06:16 ShadowNinja now saplings should grow on anything in the grower group, testing is the issue...
06:17 ShadowNinja Can anyone think of a better group name than "grower"?
06:23 Exio i can't
06:50 ShadowNinja hmmmm: issue 643 should be closed
06:54 hmmmm i don't know what issue 643 is and i don't really care about it
06:55 hmmmm just because i'm the only person around at the moment doesn't mean i need to take care of the closure of "issue 643"
06:55 hmmmm especially when i'm going to bed
06:58 Exio good night?
06:59 Exio blue42u opened this issue 3 days ago
06:59 Exio Crash on shutdown (double free)
06:59 Exio issue 643 ^
07:11 ShadowNinja thexyz: https://github.com/minetest/minetest/pull/654
07:13 ShadowNinja or darkrose, or proller__, anyone with push
07:18 darkrose it's in
07:23 * VanessaE peeks in
08:07 emptty joined #minetest-dev
08:35 emptty joined #minetest-dev
09:04 dexter0 joined #minetest-dev
09:06 vidplace7 joined #minetest-dev
09:14 Calinou joined #minetest-dev
09:21 darkrose joined #minetest-dev
09:27 ImQ009 joined #minetest-dev
09:29 Zeg9 joined #minetest-dev
09:32 emptty1 joined #minetest-dev
09:43 PilzAdam joined #minetest-dev
09:46 PilzAdam darkrose, we dont want merge commits
09:47 * darkrose doesn't actually care
09:48 PilzAdam they screw up the history and look ugly
10:01 * Calinou doesn't care too
10:01 Calinou the end user WILL care though.
10:02 Calinou "look, the git log is full of merge requests! fugly!"
10:10 ImQ009 joined #minetest-dev
10:17 iqualfragile joined #minetest-dev
10:23 darkrose joined #minetest-dev
11:07 emptty joined #minetest-dev
11:29 proller__ math ready to commit - https://github.com/proller/minetest/commit/3f03fc63d5b57c0d2024ba8552e3473b7e5e6597
11:30 PilzAdam json format?
11:30 iqualfragile joined #minetest-dev
11:31 PilzAdam I dont think json is good for minetest.conf
11:32 PilzAdam and why do you have so many commented out code in the new file?
11:35 proller__ there is may be 10-20 params for one fractal
11:36 proller__ json is simplest way to specify
11:36 proller__ some comments for remove
11:36 proller__ some contain good params
11:40 proller__ some - mandelbulber prepare, if will be possible include it sources
12:13 RealBadAngel http://pastebin.com/V7nXhBZd
12:14 RealBadAngel for some wondering whats that, it is group "wood"
12:14 RealBadAngel and thus list of items with one can craft default:chest or pick, shovel etc
12:21 RealBadAngel PilzAdam, you there?
12:37 PilzAdam yes
12:40 RealBadAngel im polishin now crafting guide code and saw how "wood" group is used
12:41 RealBadAngel it become a mess
12:42 RealBadAngel folks cannot add wooden (or stone) stuff to logical group "wood" because of recipes...
12:43 PilzAdam proller__, it just goes against any standards that are in minetest.conf
12:43 jin_xi joined #minetest-dev
12:45 proller__ i can make not configurable fractals
12:45 proller__ with only one algo selector
12:45 RealBadAngel PilzAdam, any idea how to solve it?
12:46 PilzAdam RealBadAngel, most of the nodes you listed are stairs/slabs/not_cubic_nodes of moreblocks
12:46 PilzAdam these blocks go against the voxel idea
12:46 RealBadAngel yup, those things are made of wood
12:47 RealBadAngel if so then delete imediatelly stairs from default game
12:47 RealBadAngel also slabs ;)
12:47 PilzAdam the slabs and stairs are not in the group wood
12:48 PilzAdam because "group:wood" stands for "a full node that is wooden"
12:48 RealBadAngel no, it reads as thing made of wood
12:48 PilzAdam and wooden tools can be made out of "group:wood"
12:49 PilzAdam not things made out of wood, wooden tools are made out of wood and are not in the group wood
12:49 PilzAdam the group would be called "wooden" or "made_out_of_wood" or whatever
12:49 PilzAdam "wood" means it is woood
12:49 PilzAdam -o
12:49 RealBadAngel wouldnt it be better to use group:planks instead?
12:49 jin_xi "the voxel idea" idk, there are nodeboxes, i think that makes that argument invalid
12:50 PilzAdam planks are called "wood" in the default game
12:50 PilzAdam jin_xi, ehm
12:50 jin_xi +1 for plank group, use wood group for burning only
12:51 RealBadAngel PilzAdam, please try to understand, with such usage of groups it leads to stupid situations
12:52 RealBadAngel wooden things cannot be called wooden
12:52 RealBadAngel stones cannot be added to stone group
12:52 RealBadAngel for example i cant add marble or granite to the stone group because it will lead to marble furnace
12:53 PilzAdam groups are should be used based on how the game uses them
12:53 PilzAdam -are
12:53 jin_xi i think the change to allow crafting from material groups is not fully thought out
12:54 RealBadAngel indeed
12:55 PilzAdam it is meant to avoid copying the same crafting recipe over and over again
12:55 jin_xi sure
12:55 RealBadAngel but side effects are worse than that
12:55 PilzAdam and you cant add the wood group to things where you dont want these crafting recipes
12:55 PilzAdam add goups based on current usage
12:56 PilzAdam and not based on how you would call it IRL
12:56 RealBadAngel so you wont allow to call granite "stone" yes?
12:56 RealBadAngel because it is not a stone, right?
12:56 PilzAdam if I dont want to make the normal stone tools out of granit, then I wouldnt add it to the stone group
12:57 RealBadAngel cant you see your logic error in this? really?
12:58 PilzAdam as I said above, groups should be used based on the existing usage, instead of based on logic
12:58 RealBadAngel rotfl
12:58 RealBadAngel lets throw away logic! its just a game
12:59 RealBadAngel wooden things are not really wooden, stones are made of thin air, because they cannot be stones :P
13:00 RealBadAngel imho youre defending your idea with nonsense arguments and just cannot admit you have screwed it
13:00 PilzAdam who said that it was my idea?
13:01 Zeg9 joined #minetest-dev
13:01 RealBadAngel you have added those recipes
13:01 PilzAdam based on what lua-api.txt and c55 said about the groups
13:01 RealBadAngel im not against groups
13:02 RealBadAngel im against not well thought usage of it
13:02 RealBadAngel use damn planks for makin a chest
13:02 PilzAdam so you say stone pickaxe is not made out of stone?
13:03 RealBadAngel i say other stones cannot be put in logic group stone because you will force them to be usable in nonsense recipes
13:04 PilzAdam a granite stone pickaxe is the same as a cobblestone pickaxe; what is nonsense in this?
13:04 RealBadAngel marble pickaxe is also cool?
13:05 PilzAdam why not?
13:05 RealBadAngel what about obsidian?
13:05 PilzAdam we make steel out of iron by cooking it in a furnace
13:05 PilzAdam then we can also make marble, granite, cobblestone and stone pickaxe
13:05 PilzAdam wich all have the same specs
13:06 RealBadAngel i see you just dont want to understand. i will add function to unregister (wipe) unwanted recipes
13:06 RealBadAngel my mods will just remove this shit
13:07 PilzAdam make a game if you dont like it
13:07 PilzAdam I really think technic should be a game
13:08 RealBadAngel its not about technic at all
13:08 RealBadAngel its about logic
13:08 RealBadAngel but you already threw it away
13:08 PilzAdam what is not logical in making stone tools out of stone?
13:09 RealBadAngel http://pastebin.com/V7nXhBZd
13:09 RealBadAngel read this group
13:09 RealBadAngel and find there any technic item
13:09 PilzAdam that are nodes that go against the voxel idea
13:09 RealBadAngel wtf is this voxel idea?
13:10 PilzAdam so they of course dont fit the game mechanics for a voxel game
13:10 RealBadAngel where have you dug it?
13:10 RealBadAngel since when we are limited to such nonsense?
13:10 PilzAdam you play minetest and dont know what a voxel game is?
13:11 RealBadAngel im suprised that you are talkin such bullshits
13:11 RealBadAngel like stairs being against something
13:11 RealBadAngel for christ sake, are you on drugs or what?
13:12 RealBadAngel according to your "voxel idea" all similar games are using invalid items
13:13 RealBadAngel mt, mc, terasology, whatever else is out there
13:13 PilzAdam Im not against adding a small amount of stairs/slabs
13:13 PilzAdam but the massive usage of morestairs is clearly too much
13:14 PilzAdam and then you expect it to fit perfeclty into the voxel world?
13:14 RealBadAngel its a mod and its not yours, so hands off
13:14 PilzAdam but you are asking to support such mods
13:15 RealBadAngel im asking to support logic
13:15 jin_xi shouldnt morestairs be able to drop some of its nodes now that 6d facedir is in?
13:15 PilzAdam and changing stuff that is designed for a voxel game to support non-voxel things
13:15 RealBadAngel if not i will allow folks to bring logic back
13:15 RealBadAngel PilzAdam, as i said before, delete then all non-cubic items from game
13:16 RealBadAngel signs, fences, doors, plants, rails.....
13:16 PilzAdam do you actually listen to what I say?
13:16 RealBadAngel theyre raping VOXEL idea
13:16 RealBadAngel yes i do
13:16 PilzAdam you clearly dont
13:16 RealBadAngel but it looks like you dont listen to others
13:16 PilzAdam because you are repeating yourself
13:17 RealBadAngel because im talkin bout the same thing all over?
13:17 RealBadAngel fix the damn recipes
13:17 RealBadAngel because theyre screwed
13:17 PilzAdam no, mods use them wrong
13:17 RealBadAngel see?
13:18 RealBadAngel all over the same
13:18 RealBadAngel next round? :)
13:24 PilzAdam RealBadAngel, do you have an idea what needs to be changed?
13:28 RealBadAngel already told you. add planks=1 group
13:28 RealBadAngel this way all the kinds of planks could be used in all recipes
13:29 PilzAdam then people will add 2D planks with raillike drawtype and say: "You shouldnt be able to make wooden tools out of this plank; the recipes are fucked up"
13:30 RealBadAngel dont think so
13:30 PilzAdam this is exactly what you are doing currently?
13:30 PilzAdam -?
13:30 RealBadAngel wood group shall be used for all the wooden things, planks are a subset
13:31 PilzAdam the default:tree node is not in the wood group
13:31 PilzAdam "wood" is used differently in Minetest
13:32 RealBadAngel yeah, im talkin bout this right now. it is used wrong way
13:36 RealBadAngel using planks in recipes is far more logical
13:38 RealBadAngel and regarding furnaces, picks etc with stone group, use of group shall be removed from those recipes at all
13:39 RealBadAngel here any explanation and change wont help. it is just wrong
13:40 PilzAdam so stone tools out of stone are illogical?
13:40 RealBadAngel marble pick, marble shovel....
13:41 RealBadAngel obsidian furnace, granite axe
13:41 RealBadAngel is it logical?
13:41 PilzAdam obsidian isnt in the stone group
13:41 RealBadAngel should be
13:41 PilzAdam and the rest: yes
13:41 RealBadAngel because it is stone
13:41 PilzAdam obsidian is more glass than stone
13:42 RealBadAngel but the recipes shall use cobble only
13:42 PilzAdam why?
13:42 PilzAdam stone furnace cant be made out of stone but only out of cobble?
13:43 RealBadAngel because for example some1 would have obsidian tools
13:43 PilzAdam wheres the logic in that?
13:43 RealBadAngel different than basic ones
13:43 RealBadAngel just an example
13:43 RealBadAngel with this you limit usage of all the stones in the game
13:44 RealBadAngel and force everythin to serve just basic items
13:44 PilzAdam the mod authors limit themselves if they add the node to the stone group
13:44 PilzAdam noone forces them to do so
13:44 jin_xi hey guys i think we're complecting the situation... groups should be a convenience thing only to sort out materials like stone or wood and types like block, slab or so
13:46 RealBadAngel by now even with only 3 stones in stone group its is nonsense that processed item
13:46 RealBadAngel like smooth stone serves the same as non processed one
13:47 RealBadAngel also nonsense is not calling rest of stonelike nodes in game stones
13:47 PilzAdam you are differentiate the different kinds of stones too much
13:48 rubenwardy joined #minetest-dev
13:48 ecube joined #minetest-dev
13:48 ecube joined #minetest-dev
13:50 RealBadAngel and about forcing it is you forcing the modders to do illogical things like not calling stonelike nodes stones
13:50 RealBadAngel not vice versa
13:50 RealBadAngel you made limitations
13:51 RealBadAngel and thx to it, when modders would like to put wooden things in "wood" group it ends with 115 possible recipes to craft a chest
13:52 RealBadAngel out of anything wooden
13:52 RealBadAngel instead of just planks
13:52 RealBadAngel please re-think the situation
13:54 PilzAdam the moreblocks argument is invalid
13:54 PilzAdam this mod uses the group wrong
13:55 rubenwardy why is the donations page locked?
13:56 rubenwardy *readonly
13:56 PilzAdam because celeron55_ logs the money he gains in there
13:57 rubenwardy ah ha
13:58 rubenwardy oh, I see
13:58 rubenwardy should use a table rather than a list
13:59 hmmmm joined #minetest-dev
15:08 iqualfragile joined #minetest-dev
15:09 iqualfragile PilzAdam: i think yaml would be better
15:13 sfan5 yaml?
15:13 sfan5 if we add some sort of format to minetest, why not use json as there's already a json library in minetest?
15:14 ShadowNinja comments? https://github.com/minetest/minetest/pull/656
15:15 PilzAdam "grower"?
15:15 rubenwardy "soil"?
15:15 rubenwardy "fertile"?
15:15 RealBadAngel i was about to suggest soil too
15:16 Exio he said that this morning (here and night in some places)
15:20 Calinou joined #minetest-dev
15:20 celeron55_ 22:18:55 < BlockMen> celeron55, would you be ok with a little change at layout? ... <- why? it's not really better or worse or anything, just different
15:21 celeron55_ you could try putting some hills in there or something, if you want to try to make it fancier
15:22 ShadowNinja pull request updated, one more vim fix now...
15:23 ShadowNinja there, it should be ready now
15:24 PilzAdam ehm, soil isnt good
15:24 PilzAdam I use it for farming
15:24 ShadowNinja do you have a better idea?
15:25 ShadowNinja node_that_saplings_grow_on :-)
15:26 celeron55_ it might make sense to change the one farming uses to "farmable_soil" and use "soil" for more regular ground
15:27 celeron55_ (unless there are too bad compatibility issues)
15:27 PilzAdam I could use soil=2 for dry and soil=3 for wet soil
15:28 ShadowNinja Saplings grow on any level of soil
15:29 celeron55_ having the value of the soil group represent the grade of the soil sounds good
15:31 ShadowNinja hmm, yes, trees can grow on any level but special plants require higher levels
15:32 PilzAdam to fit the other groups level system we should make 3 = normal dirt; 2 = dry plowed; 1 = wet plowed
15:32 PilzAdam (and maybe 4 = grass?)
15:35 RealBadAngel growable=1 ?
15:35 ShadowNinja no, that would be backwards
15:36 ShadowNinja growable would also work
15:36 PilzAdam I would use growable for the plant itself
15:36 salamanderrake joined #minetest-dev
15:36 RealBadAngel node doesnt have to be soil for things be able to grow on it
15:37 RealBadAngel cacti can grow on sand
15:37 ShadowNinja should trees be able to grow on sand?
15:38 RealBadAngel some of them, maybe
15:38 RealBadAngel palms for example
15:39 ShadowNinja should default trees be able to?
15:39 PilzAdam no
15:48 iqualfragile i think that a high number should represent a fertile soil because that will make it easier to extend
15:48 iqualfragile nobody realy wants to create plants that have less needs
15:48 iqualfragile but a plant wich needs fertilized soil is more likely
15:49 Exio joined #minetest-dev
16:12 OWNSyouAll joined #minetest-dev
16:34 OWNSyouAll joined #minetest-dev
16:45 ShadowNinja anyone with push access, this can be closed: https://github.com/minetest/minetest/issues/625
16:49 RealBadAngel closed
16:52 OWNSyouAll joined #minetest-dev
16:58 OWNSyouAll joined #minetest-dev
17:06 Exio what do you think about https://github.com/minetest/minetest/pull/653 ?
17:06 Exio </spam> :P
17:12 OWNSyouAll joined #minetest-dev
17:16 ShadowNinja annother issue to close: https://github.com/minetest/minetest/issues/510
17:17 ShadowNinja and this: https://github.com/minetest/minetest/issues/509
17:18 Calinou https://github.com/minetest/minetest/pull/412#issuecomment-16351091 <= comments?
17:18 Calinou PilzAdam: merge this, kthx
17:18 Calinou it causes zero problem
17:20 ShadowNinja I agree, there are a lot of pulls that have no reason not to be merged but are ignored
17:25 OWNSyouAll joined #minetest-dev
17:26 PilzAdam RealBadAngel, add a comment with the commit SHA1 if you close issues/pull requests
17:28 PilzAdam ShadowNinja, thx for keeping an eye on the bug tracker
17:29 RealBadAngel PilzAdam, ShadowNinja commented it already enough with pointing to commit
17:30 PilzAdam oh, didnt see that
17:30 PilzAdam sorry
17:38 ShadowNinja sfan5: Do you still have this issue? https://github.com/minetest/minetest/issues/452
17:38 sfan5 lemme test
17:39 sfan5 *clicks on one-click minetest installer+starter*
17:39 OWNSyouAll joined #minetest-dev
17:39 sfan5 ShadowNinja: nope
17:42 ShadowNinja OK, then someone please close it.
17:42 sfan5 i did
17:43 ShadowNinja If "won't add" then it should probably be closed: https://github.com/minetest/minetest/issues/394
17:44 ShadowNinja this too: https://github.com/minetest/minetest/issues/398
17:45 PilzAdam ShadowNinja, its open to prevent other people to create a new issue for that
17:45 PilzAdam (I guess)
17:45 rarkenin_ joined #minetest-dev
17:45 rarkenin_ joined #minetest-dev
17:46 ShadowNinja well it will still be there, just closed, it will show up if you search for it (like you should)
17:46 Exio i wonder, why not implement this https://github.com/minetest/minetest/issues/398?
17:46 Exio (as, why won't add?)
17:47 PilzAdam Exio, because people can set it in their monitors
17:47 Exio ehm..
17:48 PilzAdam at least thats what c55 said about it; and he flagged it as "wont add"
17:49 ShadowNinja I will add it if someone knows how and where it should be done
17:49 ShadowNinja seems like this should be closed: https://github.com/minetest/minetest/issues/374
17:50 Exio ShadowNinja: I'd implement it but I'm not sure if it may affect the performance or not or w/e
17:50 Exio (i think a "multiplier" between 0 to 2 in the light values would be 'perfect', but no idea really
17:50 Exio (in light.h)
17:51 ShadowNinja Probably not, I would think Irrlicht has a option for this
17:51 ShadowNinja Changing light.h is probably a bad idea
17:51 Exio yep
17:51 Exio probably, but i have no idea :P
17:52 ShadowNinja This looks like it can be closed: https://github.com/minetest/minetest/issues/355
17:53 Exio https://github.com/EXio4/minetest/commit/46a4a4b253cf2b02aecaea642ac327af419d4d1f
17:53 Exio - https://github.com/minetest/minetest/issues/517
17:56 ShadowNinja Great, pull request and have the issue closed
17:57 Exio https://github.com/minetest/minetest/pull/657 :D
17:58 Exio i'm the fan of configuration
17:58 Exio :P you can change... EVERYTHING!
18:00 ShadowNinja This should be pulled https://github.com/minetest/minetest/pull/463 and this closed https://github.com/minetest/minetest/issues/336
18:00 Exio i was thinking about that
18:00 Exio but doserj had other way to fix it
18:01 Calinou Exio: too bad, PilzAdam is against it
18:01 ShadowNinja well pull either, as long as it works
18:01 Calinou but yes, mouse sens is needed
18:02 Calinou mouse accel would be appreciated too
18:02 Calinou :p
18:02 ShadowNinja And when is anyone going to get around to merging db_backends_2?
18:03 PilzAdam when thexyz rebases it
18:04 ShadowNinja IPv6 has almost no merge conflicts, just removing markers and changing about 5 characters to get the menu setting to fit properly
18:04 ShadowNinja At least last I tried, a few weeks ago I think
18:04 Calinou localhost fix could be merged too
18:05 Calinou you know, some people use windows in the world
18:05 Exio localhost fix?
18:05 ShadowNinja That is a ugly hack, not a fix
18:05 Calinou no u
18:05 Calinou stop about "hacks"
18:05 Exio if it is what i recall
18:05 ShadowNinja connecting to localhost doesn't work
18:05 Exio if (host = L"localhost") ...
18:06 ShadowNinja yep, that
18:06 Calinou if you don't bind 127.0.0.1 to localhost your system will be insanely slow
18:06 Calinou it causes zero downsides
18:06 Calinou so stop arguing about the possible uses of gentoo and merge it
18:06 ShadowNinja what?
18:07 Exio the same what ShadowNinja said
18:07 ShadowNinja Can someone pull some pull requests now?
18:08 Calinou see
18:09 sapier joined #minetest-dev
18:14 ShadowNinja maybe this should be closed: https://github.com/minetest/minetest/issues/47
18:15 PilzAdam no
18:18 Exio PilzAdam: https://github.com/minetest/minetest/issues/137
18:19 ShadowNinja Added in common: https://github.com/minetest/minetest/issues/137
18:19 ShadowNinja oh lol, too late
18:19 Exio :D
18:20 Exio and.. https://github.com/minetest/minetest/issues/182
18:24 PilzAdam Exio, ShadowNinja, could you please also the commit SHA1 so I dont have to search it myself?
18:25 Exio ok
18:27 Jordach joined #minetest-dev
18:27 Exio https://github.com/minetest/minetest_game/commit/e82c12282233113774e6f55f459fb89a457a981c ?
18:27 Exio searching the other
18:27 PilzAdam already found it
18:28 ShadowNinja got lavacooling
18:28 PilzAdam just for future close requests
18:30 celeron55 joined #minetest-dev
18:30 emptty joined #minetest-dev
18:31 Exio 6823ce99a7deabe410dd8b143b688cd364490cec
18:32 Exio it is the commit for the apple trees
18:32 sapier I'm looking for a solution to client entity movement prediction sometimes running havoc. What do you think about limiting predicition time ?
18:49 hmmmm merge pull request
18:49 hmmmm mmmmmmmmm.....
18:49 hmmmm darkrose, i'm glad that you don't actually care, but the rest of us DO care.
18:50 hmmmm minetest isn't some "do what you feel like" project
18:50 hmmmm we have rules and regulations that are there for a purpose, mostly to not shit up the codebase
18:50 hmmmm so if you don't care about shitting up the codebase, then i don't know if we care to have you on the list of committers
18:50 hmmmm i'm just saying that about the "i don't care" in general
18:50 hmmmm obviously a merge pull request isn't too bad.  i just don't like that attitude at all
18:51 ecube joined #minetest-dev
18:51 ecube joined #minetest-dev
20:03 emptty Just for the record, I just uploaded a clean package of version 0.4.6 to debian servers
20:04 emptty Since I introduced a new binary package while repackaging it, it needs to be reviewed by the ftp masters, but it will land into the distro within the week.
20:06 * Jordach wonders about the influx of new users
20:06 emptty don't expect any flood of users, yeah
20:06 emptty there is already a package in debian, but it was still for 0.3.1 or something
20:07 emptty you might get some more people from ubuntu once it automatically migrate from debian to ubuntu
20:17 celeron55 Jordach: i find it odd too
20:30 jin_xi joined #minetest-dev
20:35 celeron55 <+PilzAdam> to fit the other groups level system we should make 3 = normal dirt; 2 = dry plowed; 1 = wet plowed
20:38 celeron55 ^ i don't think doing it that way for this group is wise; unless you want to make it extendable to the "more dead" direction rather than "more fertile"
20:38 celeron55 the logic for the digging time groups is that because 0 is "infinite time", then 1 is the longest time, 2 is shorter than that and so on
20:38 celeron55 it isn't "backwards"
20:39 celeron55 it's just the way that makes sense in the context of 0 being "can't dig"
20:39 celeron55 for soil, 0 is "can't grow", so 1 should be "some things can grow"
20:39 celeron55 and so on
20:46 sapier joined #minetest-dev
20:47 hmmmm sapier
20:47 hmmmm i've been meaning to ask you, which of your pull requests do i commit for the final memory leak fix
20:48 hmmmm at this point, i've fixed (on my own) a really dumb double free on emergemanager's dtor, and a texture-related attempt to drop a NULL texture
20:49 sapier yes I've fixed one of those in sky memory leak fix too ... and written here at least 3 times ;-)
20:49 sapier I'll remove it and add a new sky fix
20:50 hmmmm there are like three different pull requests iirc
20:50 sapier btw I've just had this strange minetest grabs mouse and freezes it again ... not quite sure if this might be an irrlicht bug
20:50 hmmmm i've had that forever
20:51 sapier I'll check wich of the pull requests are still valid
20:51 hmmmm right after you close a formspec, minetest loses focus or something
20:51 sapier in this case no formspec was involved
20:51 hmmmm odd
20:51 sapier I had to reboot even killing x server didn't help
20:52 hmmmm sounds like an irrlicht bug then
20:52 hmmmm you need a really low-level screwup for something that bad to happen
20:53 sapier https://github.com/minetest/minetest/pull/635 is still valid as I accidentaly added patch to constructor (which is quite useless as there isn't anything in there)
20:53 hmmmm i don't think i should even try to merge that
20:54 hmmmm rather, i'll just take note of what you fixed and modify it against a current base
20:55 sapier no problem it's just moving those lines from constructor to destructor
20:56 hmmmm as for this https://github.com/sapier/minetest/commit/f1d971f1373c6e2a77e55a57ecc3f2b47266da45  hmm
20:56 hmmmm i was intending to do that on my own at some point, except using EmergeManager::getGroundLevelAtPoint()
20:57 hmmmm and then scanning upward until air and then scanning downward until is_walkable
20:57 sapier didn't know there was a function like that
20:57 celeron55 it's bad practice to implement logic in lua wrappers
20:57 celeron55 most of that EnvRef::l_get_surface should be in ServerEnvironment
20:57 sapier ok I'll move code there
20:57 celeron55 or elsewhere
20:58 celeron55 it does solely map access so i guess it could be in Map instead too but whatever
20:58 hmmmm ServerEnvironment
20:59 hmmmm if you wish to use getGroundLevelAtPoint
21:00 sapier1 joined #minetest-dev
21:00 hmmmm your way might be a better idea since you can specify around where to look for the ground level at
21:00 sapier1 environment is already available for lua ... I'd prefere to keep/reduce dependency of lua callbacks if possible
21:00 celeron55 the interface could be tuned somewhat
21:01 sapier1 yes it's just a 1:1 implementation of what I've done in mobf and found out to be very costly
21:01 sapier1 if you have ideas what could help to I gladly improve it
21:02 hmmmm well like i said, seeing if the node is walkable first off, not just if it isn't air or content ignore
21:03 sapier water isn't walkable too
21:03 sapier but it's surface without any doubt
21:03 hmmmm really?
21:03 hmmmm i wouldn't call water a surface, and none of the mapgen components do either
21:03 sapier ok ... true ... depends in your point of view
21:04 sapier for a ship or duck water is surface ;-)
21:04 hmmmm why is water the surface level if your mob can't walk past water anyway
21:05 sapier ducks may walk on water ... but for what I use this fct it wouldn't make a difference
21:05 celeron55 in some cases it could make more sense to change yoffset to range (which would be search up and down, or the equivalent done somehow), because often the caller knows the rough level but just wants to find the fine height but without causing huge performance issues if it constantly guesses wrong
21:05 VanessaE waterlilies always spawn on the water's surface.....
21:06 sapier the top and bottom values are result of this function used most of time in on_generate where you have them by callback
21:06 sapier so in this particular usecase range is not best way to go
21:07 sapier maybe supplying a list of nodes to be considered above surface would be an option
21:07 celeron55 i'd suggest adding a third optional boolean parameter only_air (or a parameter table that can contain it)
21:07 sapier if you supply nothing default to air, if you do use those nodes
21:07 hmmmm every time someone says "list of nodes"..
21:08 celeron55 it could then handle the water lily case, but default to a more generic one
21:08 VanessaE <offtopic>  Could someone please extend the register_alias() function to allow specifying a facedir?  This will be extremely handy for conversion of maps/inventories to use the '6d facedir' code.
21:08 VanessaE (well maps anyway)
21:09 celeron55 you should learn C++ and do it yourself
21:10 celeron55 it's once again one of those things that aren't hard, but will take so much time and focus that anyone working on anything else won't do it
21:10 sapier what do you mean hmmmm?
21:10 sapier syntax would change to get_surface(basepos,yoffset,node_not_solid) e.g.
21:10 sapier get_surface({0,0,0},5,{ "air","default:water_source","default:water_flowing"})
21:11 celeron55 eh, no lists
21:11 hmmmm exactly
21:11 celeron55 it's so utterly non-generic
21:11 celeron55 you can't possibly list everything in a compatible way ever
21:11 hmmmm and if you were doing that, i'd tell you to use groups
21:11 sapier so any better suggestions giving same functionality?
21:12 sapier i don't want to add a get_surface_air() get_surface_water() get_surface_lava() ;-)
21:13 sapier btw this function doesn't have to be added to core it can be done in lua too ... but it's a big performance issue there
21:13 celeron55 i think the only sane cases here are: 1) consider any non-walkable node as "not ground", 2) consider nothing else than air "not ground"
21:14 hmmmm the thing about lua api is that we don't want to blindly add things to the core to accelerate specific mobs, but instead give an interface that all mods would be able to use
21:14 sapier https://github.com/minetest/minetest/pull/627 is sky cleanup
21:14 celeron55 if somebody needs more than that, they can iterate the rest with lua
21:14 hmmmm right, what celeron said
21:14 sapier ok no problem
21:14 hmmmm so have a boolean to switch what sort of search you're using
21:14 sapier I'll change it that way
21:16 celeron55 also, make it immediately clearer in the documentation that it searches the surface upwards from the given position (or anywhere in the range basepos.y...basepos.y+yoffset, however it actually is supposed to do it)
21:17 celeron55 now it's kind of mentioned in there as the last words in a way 90% of readers won't notice
21:17 sapier ok
21:18 celeron55 interface is always the most important; it's the hardest to improve later
21:20 sapier true ... lua api is somehow inconsistent with _ usage ... getnode, get_player_by_name ... any chance to declare one of those notations deprecated?
21:21 hmmmm well, if your idea of solving that is spamming the end user with messages in actionstream about how it's deprecated...
21:22 sapier that's the "hard" way of telling ... a little bit less invasive is marking it in doc
21:23 hmmmm perhaps :)  i'm just taking a jab at your idea about how to tell people to not use minetest.env: anymore
21:24 sapier I know ;-)
21:24 sapier at least ppl would notice something changed ;-)
21:24 hmmmm where did formspec come from?  who wrote that?  i am seeing a lot of shoddier formspec code
21:24 sapier "shoddier"?
21:24 hmmmm 106 column lines
21:25 hmmmm useless write to a buffer before writing to an os to write to the packet
21:25 hmmmm weird use of const
21:25 sapier what exactly do you mean? I don't understand
21:26 hmmmm also, making a copy of the string instead of a reference
21:26 hmmmm amateur mistakes
21:28 sapier what are you talking about? I've done some formspec changes while ago and a pull request to fix a bug there but Its 8 lines only :-)
21:28 hmmmm right, no, it's nothing to do with your modification
21:28 hmmmm in particular i'm looking at server::sendshowformspecmessage
21:28 hmmmm and the surrounding code
21:29 hmmmm it's like sendspawnparticle, which makes me believe that it's jeija's
21:29 sapier ok I've used it but not written
21:29 sapier and not checked either
21:30 hmmmm oh, this is nice, see Server::SendAddParticleSpawnerAll
21:30 hmmmm 15 parameters
21:30 sapier wow :-)
21:31 hmmmm i wonder if i would ever see half this shit in the industry
21:33 proller__ math mapgen 0.1 without external ready: https://github.com/proller/minetest/commit/723fcf96a392d1501e5d3b3ebec4fdb914b3a9f9
21:33 proller__ but with json in config
21:33 hmmmm yeah, we need to really talk about json in the config
21:34 proller__ other variants?
21:34 hmmmm should we allow it?  should we not?
21:34 hmmmm cons:
21:34 hmmmm 1. need json as a dependency all the time, even without masterserver
21:35 proller__ its now enabled all time
21:35 hmmmm 2. uses a pretty high-level interface for something that's supposed to be low-level like settings
21:35 hmmmm 3. if json screws up, everything's screwed up
21:35 hmmmm hmm
21:35 hmmmm pros:
21:36 proller__ now it for one variable, not all config
21:36 hmmmm can single out a specific field in a whole set of them
21:36 hmmmm more readable (but isn't that what making more settings is supposed to do?)
21:37 proller__ "0.5, 1, (100, 100, 100), 6467, 4, 0.75" - its more human friendly in json form
21:38 hmmmm and you need a Json::Value class for the settings
21:38 hmmmm that's a con
21:39 hmmmm if json is a full-time dependency right now, then i don't see any problem in adding it
21:39 proller__ Json::Value is very flexible but slower than struct
21:40 hmmmm hmm
21:41 hmmmm hey proller, i see a lot of comments in your code starting with "ok", what does that mean?  is that your way of keeping track of what is and isn't completely finished?
21:41 hmmmm just curious
21:41 proller__ ok - its #if 0 block to use with mandelbulb files
21:42 proller__ tested various fractals
21:42 hmmmm ah
21:42 proller__ if we can include this files - i make configurable via json
21:42 proller__ if not - will delete
21:43 hmmmm we should get some other peoples' opinions
21:46 ShadowNinja hmmmm: Is the IPv6 pull request OK? It only has a few minor conflicts and seems to work.
21:47 hmmmm i don't know, but we should be able to disable ipv6 support
21:47 ShadowNinja You can
21:48 ShadowNinja It has two conf options and a build option
21:48 ShadowNinja enable_ipv6 and ipv6_server IIRC
21:49 Taoki joined #minetest-dev
21:57 iqualfragile joined #minetest-dev
21:59 sapier as checking against air is less overhead than walkable I consider air check as default behaviour for get_surface
21:59 sapier any objections against this?
22:08 jojoa1997|Tablet joined #minetest-dev
22:14 jojoa1997|Tablet is it true that minetest 0.4.7 will be released soon?
22:15 sapier first time to here that jojoa :-)
22:16 jojoa1997|Tablet what do you mean?
22:16 sapier minetest 0.4.6 was just released ;-)
22:18 jojoa1997|Tablet ok someone told me that 0.4.7 will be realeased soon. i didnt believe it but wanted to make sure. With school it has felt like a year since 0.4.6 as released
22:20 sapier don't get me wrong I don't have anything to do with releases but 3 weeks would be quite fast for a new version ;-)
22:23 Exio joined #minetest-dev
22:56 sapier left #minetest-dev
23:34 Taoki joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext