Time  Nick       Message
00:27 celeron55  http://c55.me/random/2012-07/screenshot_3059758902.png
00:36 celeron55  basically the same as the 11-color wool mod, but adds dark grey, pink, dark green and violet
00:50 celeron55  if i'd add groups basedye_magenta, basedye_red, basedye_yellow, basedye_green, basedye_cyan, basedye_blue, basedye_white and basedye_black to appropriate dye items (one to each), some extended dye mod could pick up from there by defining group-based crafting recipes... ehm... maybe
07:41 Nemo08     hi! anybody knows, in 0.4, 0.4 map format has changed? standart mapper make strange image, my mapper make some image..
07:42 Nemo08     *0.4.1
08:00 cy1        Nemo08: 0.4 was almost a rewrite. The map is sqlite there instead of sector directory flat file.
08:01 cy1        0.4.1 not real differentfrom 0.4, just use the latest git of your favorite gitter
08:05 darkrose   and don't ask questions like that in here
08:07 Nemo08     em? sector directory flat file was until 0.3. current map generator work with 0.3 sqlite maps fine, but not with 0.4. I think format of records wes changed, but still not documented in mapformat.txt
08:07 Nemo08     sorry
08:07 Nemo08     where i may ask it?
08:08 cy1        #minetest-delta?
08:10 Nemo08     ok ,thanks
08:58 EdB        vorbis doesn't seems to come with a cmake files
09:20 EdB        sorry about that, cmake was distrub by the fact taht I move src folder
10:57 celeron55  VanessaE: http://paste.dy.fi/aAa
10:58 celeron55  http://paste.dy.fi/axk
10:59 jordach    celeron55, this looks really good. i cant wait to update my mods.
11:09 Nemo08     celeron55: may i ask? how changed node data format in sqlite?
11:10 celeron55  Nemo08: that question does not make any sense
11:10 Nemo08     this part:
11:11 Nemo08     zlib-compressed node data: - content:   u8[4096]: param0 fields   u8[4096]: param1 fields   u8[4096]: param2 fields - The location of a node in each of those arrays is (z*16*16 + y*16 + x).
11:11 celeron55  basically, if someone wants to make a colorable block or item, it will be colorable with any dye as long as it can be colored with any of basecolor_*, excolor_* or unicolor_*, the colored variations needed being up to the one making it; to fully support all dyes, either 9, 17 or 89 colors
11:12 celeron55  Nemo08: what version are you talking about?
11:12 Nemo08     24
11:13 celeron55  param0s are 2 bytes wide, compared to 23 where they are 1
11:13 celeron55  and param2 does not ever contain any part of the node id
11:13 celeron55  if somebody updates minetestmapper.py, that'd be useful to have
11:14 Nemo08     thanks :)
11:14 celeron55  hmm... also there can now be node timers... eh, now i remembered something
11:15 celeron55  hell
11:15 celeron55  the node timer format sucks
11:15 celeron55  you need to know the length of the data of a node timer to be able to read it
11:15 celeron55  it isn't written in the data itself
11:15 celeron55  s/read it/skip it
11:16 Nemo08     it compressed?
11:16 celeron55  i'll just modify it to something i like
12:27 celeron55  https://github.com/celeron55/minetest/commits/master
12:33 celeron55  oh shi-
12:34 celeron55  i think doors break with some version now
12:34 celeron55  i think not with 0.4.0/1 but something after them
12:37 celeron55  hmm, they don't seem to break server-side
12:37 celeron55  the client just shows them wrong if you click them
12:38 celeron55  ehm... wtf
12:39 jordach    what is exactly wrong with them?
12:40 celeron55  they are very buggy
12:46 celeron55  ...bisect time
13:15 celeron55  what the fuck?!
13:15 celeron55  https://github.com/celeron55/minetest/commit/ea62ee4b61371107ef3d693bda4c410ba02ca7e6#L3R4019
13:17 celeron55  darkrose: ^
13:19 darkrose   heh, oops
13:22 celeron55  that didn't fix anything though
13:31 celeron55  HA
13:31 celeron55  u32 MapNode::serializedLength(u8 version)
13:31 celeron55  applies to single nodes serialized
13:37 celeron55  https://github.com/celeron55/minetest/commit/0cf1ed544c91a01b9c2098cd3e11a49c8f0c1ddb
13:38 darkrose   love the commit message :D
15:14 celeron55  hmm, dye and wool mod ready
15:15 jordach    is the node prefix wool: ?
15:15 celeron55  wool.
15:15 jordach    thanks
15:17 celeron55  i don't use any longer-than-needed or more-obscure-than-needed mod names in upstream
15:17 celeron55  people should prefix their mods with something if they don't want collisions
15:17 jordach    mines cotton: so we are safe
15:17 celeron55  as everybody probably understands
15:18 jordach    i know
15:35 celeron55  that is why generally enterprisey packaging systems put a vendor name as the first namespace
15:35 celeron55  it isn't because "WE DID IT" 8)
15:39 * VanessaE wanders in
15:39 VanessaE   (saw the paste, reading it over now)
15:42 VanessaE   seems okay
15:42 VanessaE   not sure about the magenta entry (line 54) but the rest looks reasonable.
15:44 celeron55  how can you be not sure about it? you know where the colors are based, and based on it there is nothing unreasonable in anything it contains
15:45 VanessaE   only because I went with the HSV definition of magenta (300 degrees on the color wheel, exactly midway between blue and red)
15:45 VanessaE   just a different definition of the word, that's all.
15:46 celeron55  ehm... what?
15:46 VanessaE   you match it to "red violet" :-)
15:48 celeron55  it's just fine
15:48 VanessaE   ok
15:48 jordach    not this again. it was fine when i checked it.
15:49 celeron55  i am wondering if i should make user-added mods of the same name have a priority over in-game ones
15:50 celeron55  currently it goes exactly the other way, which also has it's reasons
15:50 celeron55  i think i'll not touch that
15:50 jordach    no, it would be better if the in-built were higher priority, because if one (mod version) over-writes another colour, then we have a problem
15:52 celeron55  it kind of is up to the mod if it wants to override things, but on the other hand if that would become a common practice, all inter-mod compatibility would be lost
15:53 celeron55  i'm trusting the order i and kahrl set up back then 8)
15:55 darkrose   user-mod crafts override game-mod crafts, but the inverse happens with nodes?
15:55 jordach    yes, id 50/50 this, suggest that you can change the defs, but can also use the default colour defs
15:56 celeron55  i am not talking about colors
15:56 celeron55  i am talking about things in general
15:56 jordach    ah
15:57 celeron55  can you override crafts with mods?
15:57 celeron55  i am in the belief that the first recipe always takes precedence
15:57 celeron55  but i don't know
15:57 darkrose   sort of, if 2 recipes exist with different outputs, the last recipe is used
15:58 celeron55  oh i see
15:58 darkrose   that's how it looked when I was working out how to do the get_craft_recipe anyway
15:58 celeron55  i just browsed to the code that iterates over recipes and it indeed does that
15:58 celeron55  it's made by the wise man, kahrl 8)
15:59 celeron55  aaaanyway, so that is well
21:49 celeron55  ugh... i need to modify the way the node metadata inventory callbacks work
21:49 celeron55  they don't allow resolving moves between different kinds of inventories
21:50 celeron55  if the engine is to check whether a move is allowed and what is the maximum item count allowd, it needs to get those limits from both, and then do the move, and then report to both inventories of what it decided to do based on the limits
21:50 celeron55  the node metadata inventory callbacks currently don't allow that; they work like "do this if you can, give back items that you reject"
21:50 celeron55  which just doesn't work
21:50 celeron55  because the source inventory may not want to take them back
21:52 VanessaE   would that be why chests sometimes destroy/delete an item when you try to swap one with another?
21:53 celeron55  ummmm
21:53 celeron55  dunno
21:53 VanessaE   *shrug* just curios.
21:53 celeron55  but this system sucks, i will replace it with a sane one
21:53 celeron55  happily it is just the callbacks, nothing persistent
21:59 VanessaE   curious*
22:00 VanessaE   well have fun :-)
23:40 celeron55  pushed