Minetest logo

IRC log for #minetest-dev, 2016-05-27

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

All times shown according to UTC.

Time Nick Message
00:13 Tmanyo joined #minetest-dev
00:24 torgdor joined #minetest-dev
00:28 torgdor joined #minetest-dev
00:31 kaeza joined #minetest-dev
00:38 stormchaser3000 joined #minetest-dev
00:38 Taoki joined #minetest-dev
00:39 Tmanyo joined #minetest-dev
00:40 torgdor joined #minetest-dev
00:50 torgdor joined #minetest-dev
01:00 est31 joined #minetest-dev
01:16 torgdor joined #minetest-dev
01:17 torgdor_ joined #minetest-dev
02:04 paramat joined #minetest-dev
02:13 paramat working on improving night sky https://github.com/minetest/minetest/issues/4146#issuecomment-221745654
02:33 * est31 is still working on the joystick thing
02:35 Zeno` paramat #3919
02:35 ShadowBot https://github.com/minetest/minetest/issues/3919 -- Add base64 encoding and decoding to the lua api. by red-001
02:36 Zeno` it's an "official" +1 if ShadowNinja's comment regarding newlines was addressed
02:36 paramat ok
02:36 Zeno` I think it has been
02:37 Zeno` hmmmmm's comment has been addressed as well
02:38 Zeno` it's been done in a strange way though (maybe they had a squash problem I dunno)
02:38 Zeno` yeah it's official, I'll update my comment
02:39 paramat nice
02:41 paramat labelled #4135 as high priority
02:41 ShadowBot https://github.com/minetest/minetest/issues/4135 -- Biomegen by kwolekr
02:42 est31 I guess it blocks lots of mapgen stuff
02:42 paramat hmmmm would like extra review for that due to size
02:43 est31 okay adding action needed label
02:43 est31 to track the required review
02:44 paramat yeah much more mapgen re-organization is coming
02:45 Zeno` can you change the name of biomegen?
02:45 Zeno` I stare at it for 2 minutes everytime I see it wondering wtf "bio megen" is
02:46 paramat lol
02:46 Zeno` seriously :) heheh
02:46 Zeno` bio is the much more common prefix in real life
02:46 est31 heh
02:46 Zeno` biome is never used as a prefix so I always read it as bio megen and can't shake the habit
02:47 Zeno` normally I don't mind contractions but this one makes a fool out of me everytime I see it!
02:48 Zeno` maybe it's because my background is in biology/botany/ecology I dunno
02:49 paramat as a mapgen bitch i always see biome gen
02:50 paramat anyway could you review that more?
02:50 est31 you are all english speaking peolpe
02:50 est31 your brains are just not wired for composite words
02:51 est31 you like the words to be two syllables or less xP
02:51 paramat i feel it almost has enough review now actually
02:51 est31 Rindfleischetikettierungsüberwachungsaufgabenübertragungsgesetz
02:51 est31 thats a nice word isnt it
02:52 Zeno` no, no... brain is wired to interpret a common prefix (bio) as a common prefix!
02:53 Zeno` geomegen would be just as confusing
02:53 Zeno` (for geomes)
02:53 Zeno` if they were such a thing
02:53 Zeno` sofar would know
02:55 Void7 joined #minetest-dev
02:55 hmmmmm Zeno`:  is that a code review +1 or an approval of the concept for that PR?
02:55 Zeno` hmmmmm, I have no objection to the PR
02:56 Zeno` we already have the base64 stuff, why not let Lua use it?
02:56 hmmmmm I was sorta hoping to hear what the use case was from red-001
02:56 Zeno` there's a link in the first comment isn't there?
02:56 Zeno` (giving a use case)
02:56 hmmmmm "It would be cool to have this builtin"
02:56 hmmmmm "would be cool"
02:56 paramat there'a use case in the linked issue
02:57 hmmmmm oh, I see now
02:57 hmmmmm is our implementation of base64 the version that works with json params?
02:58 Zeno` I didn't know there was a specific implementation that JSON relied upon...
02:58 hmmmmm i'm not sure either
02:58 hmmmmm just being extra careful
02:58 Zeno` I know there are different implementations. Hmm
02:59 hmmmmm Zeno`:  I just created that PR so that somebody was able to see all the individual commits merged together
02:59 est31 hmmmmm, go to the feature request issue
02:59 hmmmmm Biomegen is the name of the branch
02:59 est31 it has one use case outlined
02:59 est31 with the http api, they apparently want to send over binary data
02:59 hmmmmm est:  yes I am looking at 3804
03:00 hmmmmm alright
03:00 est31 i dont know though whether its a really that good use
03:00 hmmmmm it's a fine use
03:00 est31 after all http never has been meant to be used with base64
03:00 est31 its a hyper TEXT transfer protocol
03:00 est31 made in ASCII
03:01 hmmmmm you can transfer binary too
03:01 est31 ah yeah
03:01 est31 now I remember
03:01 hmmmmm but we shouldn't even bother discussing how correct the usage is
03:01 est31 you can download pdf files as well over http
03:01 est31 or gif or sth
03:01 hmmmmm web development is a cesspit with incredibly bad design choices cemented on top of other bad design choices
03:01 est31 ++1
03:02 hmmmmm if somebody needs base64 for some web thing, then i don't doubt them
03:02 est31 but most of it is open source, and ITS SANDBOXES
03:02 est31 SANDBOXED*
03:02 est31 thats more than posix could ever dream of
03:02 hmmmmm so
03:03 hmmmmm 1). a C++ base64 implementation will always be faster than a lua version
03:03 Zeno` I was thinking of using base64 for something only a month or so ago (before I saw this PR which may not even have been PR at the time) for something not even related to web design heh
03:03 est31 hmmmmm, not sure about that
03:03 hmmmmm 2). a solid implementation of base64 already exists in the core
03:03 hmmmmm 3). it would be wasteful to duplicate functionality
03:03 est31 there is an overhead calling lua <-> c++
03:03 est31 dont forget that its also copied twice
03:03 hmmmmm do you know how much overhead the call is?
03:04 est31 no
03:04 hmmmmm i actually tested it once
03:04 est31 do you know how much overhead doing it in lua is :)
03:04 est31 3) sounds convincing though, even though base64 is pretty simple
03:04 hmmmmm a lua-to-C call takes about 35 microseconds with -O0 on my modern processor
03:04 hmmmmm using luajit
03:04 Zeno` As far as overhead is concerned I'm not terribly bothered by it. BUT if people are going to start using base64 in mods (and one person already has) it would be better IMO that all mods use the same base64 implementation (the one we use)
03:05 Zeno` instead of every mod rolling their own base64 and us having one in C++ also
03:05 hmmmmm absolutely
03:05 est31 well okay then
03:05 est31 +1 concept approval
03:05 hmmmmm so i strongly approve of that PR after thinking about it
03:05 est31 no code review done though
03:05 hmmmmm i was already leaning toward it
03:06 hmmmmm well it already has more than enough code review +1s
03:06 hmmmmm just trust me.  it's solid.
03:06 hmmmmm it's sweet and simple and does exactly what it needs to
03:06 est31 Zeno`, have you seen PM?
03:07 Zeno` which PM?
03:07 Zeno` oh
03:07 Zeno` hang on :)
03:08 hmmmmm actually whoops
03:09 hmmmmm it takes 35 nanoseconds, not 35 microseconds - i measured the time with 1000 iterations
03:15 Void7 joined #minetest-dev
03:30 Hunterz joined #minetest-dev
04:28 thePalindrome joined #minetest-dev
04:33 thePalindrome joined #minetest-dev
05:13 Hunterz joined #minetest-dev
05:42 Etzos joined #minetest-dev
05:46 paramat #4163
05:46 ShadowBot https://github.com/minetest/minetest/issues/4163 -- Sky: Darker, bluer sky at night by paramat
06:09 paramat hmmmmm i have explained a mapgen setting issue here #4164
06:09 ShadowBot https://github.com/minetest/minetest/issues/4164 -- Setting mapgen in a mod can result in incorrect ore/biome/decoration/nyancat registrations
06:10 hmmmmm i am aware of that issue
06:11 hmmmmm it's not that auto-setting in a mod is bad, it's just that your mod can't predict the future
06:12 Etzos joined #minetest-dev
06:13 hmmmmm maybe there can be a second pass at registration time where mapgen params are read only
06:13 hmmmmm maybe setting the mapgen should be disallowed from inside of the mod execution body
06:14 hmmmmm the preferred mapgen should be in like a dependencies.txt-type file that's read
06:14 hmmmmm and then how would conflicts be resolved?  hmmm
06:15 paramat i'm not too worried due to the workaround explained at the end
06:15 Taoki joined #minetest-dev
06:15 paramat maybe a note in docs would be good
06:17 hmmmmm well that does work around the problem
06:17 hmmmmm while introducing new problems of its own
06:18 hmmmmm this is going to break everybody's mods
06:58 est31 https://github.com/est31/minetest/commits/gamepad
06:59 est31 What do you think: should I squash the two commits
06:59 est31 or work on keeping them separate
06:59 est31 the first commit doesnt touch any real functionality
06:59 est31 its just about moving stuff around
07:00 est31 but without the second commit, some of the changes of the first commit aren't really straightforwardly
07:00 est31 -ly
07:02 paramat i'm thinking keep seperate
07:03 jin_xi joined #minetest-dev
07:03 hmmmmm what's the point in making GenericKeyList a template though..?
07:05 est31 its a remainder from me thinking that I need it later on in JoystickController
07:05 est31 but then I chose to go with bitset
07:08 est31 hrmm what was again the reason for that???
07:09 est31 well it wont really make a difference I guess
07:09 est31 in speed
07:09 est31 because KeyType::INTERNAL_ENUM_COUNT is so small
07:10 est31 its probably faster to iterate the bitset than to bother with all the indirectness of a GenericKeyList
07:17 est31 hmmmmm, do you think I should make it a non-template again?
07:18 hmmmmm perhaps
07:18 Calinou joined #minetest-dev
07:18 hmmmmm you shouldn't make things templates just because
07:18 Anchakor__ joined #minetest-dev
07:19 est31 https://github.com/est31/minetest/commit/11f64f9c0a84d547dc4aeaf75647a9319fc85f9a
07:20 est31 okay removing it, and above link will help me to find the commit hash should I ever reconsider
07:21 hmmmmm when i write something using templates i try writing it again without templates
07:22 hmmmmm and then i compare the two
07:22 hmmmmm the latter version almost always comes out nicer
07:49 nrzkt joined #minetest-dev
08:49 Amaz joined #minetest-dev
09:04 paramat nore or mtgame devs, please can anyone review/approve game#1001 ? fairly simple and tested
09:04 ShadowBot https://github.com/minetest/minetest_game/issues/1001 -- Fire: move fire node removal out of ABM. by sofar
09:05 Megaf joined #minetest-dev
09:27 Thomas-S joined #minetest-dev
09:37 troller joined #minetest-dev
11:01 est31 joined #minetest-dev
11:05 Fixer joined #minetest-dev
11:26 Megaf joined #minetest-dev
11:41 Lunatrius joined #minetest-dev
11:43 damiel joined #minetest-dev
11:44 Lunatrius` joined #minetest-dev
12:02 proller joined #minetest-dev
12:07 paramat joined #minetest-dev
12:25 xunto joined #minetest-dev
12:37 proller joined #minetest-dev
12:43 STHGOM joined #minetest-dev
13:40 Darcidride joined #minetest-dev
13:45 Darcidride joined #minetest-dev
13:57 Megaf joined #minetest-dev
14:06 paramat left #minetest-dev
14:11 est31 joined #minetest-dev
14:14 lisac joined #minetest-dev
14:17 proller joined #minetest-dev
14:41 Lunatrius` joined #minetest-dev
14:43 Hunterz joined #minetest-dev
14:43 Megal joined #minetest-dev
14:44 Darcidride joined #minetest-dev
14:47 DI3HARD139 joined #minetest-dev
14:49 Lunatrius joined #minetest-dev
14:51 Darcidride joined #minetest-dev
15:01 Darcidride joined #minetest-dev
15:26 KaadmY joined #minetest-dev
15:32 hmmmm joined #minetest-dev
15:32 torgdor joined #minetest-dev
15:33 electrodude512 joined #minetest-dev
15:38 Puka joined #minetest-dev
15:49 Krock joined #minetest-dev
16:13 turtleman joined #minetest-dev
16:35 electrodude512 joined #minetest-dev
16:39 DFeniks joined #minetest-dev
16:41 STHGOM joined #minetest-dev
16:50 torgdor joined #minetest-dev
17:03 Amaz joined #minetest-dev
17:07 electrodude512 joined #minetest-dev
17:29 Megal joined #minetest-dev
17:41 rubenwardy joined #minetest-dev
17:43 bry_ joined #minetest-dev
18:08 proller joined #minetest-dev
18:18 torgdor joined #minetest-dev
18:21 Lunatrius joined #minetest-dev
18:37 proller joined #minetest-dev
18:39 Void7 joined #minetest-dev
18:51 Krock joined #minetest-dev
19:01 electrodude512 joined #minetest-dev
19:03 Void7 joined #minetest-dev
19:08 proller joined #minetest-dev
20:19 Amaz joined #minetest-dev
21:16 Void7 joined #minetest-dev
21:18 troller joined #minetest-dev
21:21 electrodude512 joined #minetest-dev
21:41 proller joined #minetest-dev
21:44 Calinou joined #minetest-dev
23:05 troller joined #minetest-dev
23:07 torgdor joined #minetest-dev
23:25 troller joined #minetest-dev
23:51 Void7 joined #minetest-dev

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