Minetest logo

IRC log for #minetest-dev, 2013-10-01

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

All times shown according to UTC.

Time Nick Message
00:10 cisoun joined #minetest-dev
00:40 cisoun left #minetest-dev
00:54 hmmmm aghh
00:55 hmmmm I promise i'm going to do minetest stuff on weekends at least
00:55 hmmmm or if i switch my schedule to 9/80, i'll devote my friday i have off to minetest
00:57 Kray joined #minetest-dev
01:11 smoke_fumus joined #minetest-dev
01:14 Naked joined #minetest-dev
02:09 smoke_fumus|2 joined #minetest-dev
02:15 n4x joined #minetest-dev
03:21 blaaaaargh joined #minetest-dev
04:39 ImQ009 joined #minetest-dev
04:59 BrandonReese_ joined #minetest-dev
05:59 Weedy joined #minetest-dev
06:00 Weedy_la1py joined #minetest-dev
06:02 Weedy_la2py joined #minetest-dev
06:02 Weedy_lappy joined #minetest-dev
06:16 smoke_fumus joined #minetest-dev
06:45 Calinou joined #minetest-dev
07:34 darkrose joined #minetest-dev
07:39 djdduty joined #minetest-dev
08:03 Akien joined #minetest-dev
08:19 kaeza joined #minetest-dev
08:24 IceCraft joined #minetest-dev
08:42 IceCraft joined #minetest-dev
09:39 proller joined #minetest-dev
10:12 diemartin joined #minetest-dev
11:26 ImQ009 joined #minetest-dev
11:38 Semilevel joined #minetest-dev
11:53 BlockMen joined #minetest-dev
12:03 Zeitgeist_ joined #minetest-dev
12:28 proller joined #minetest-dev
12:30 Ritchie_ joined #minetest-dev
12:53 kahrl I'll merge #933 in a few minutes (PilzAdam already said it was good)
12:54 kahrl (if I find remaining whitespace problems I'll fix them)
12:55 IceCraft joined #minetest-dev
13:18 Naked joined #minetest-dev
13:50 werwerwer joined #minetest-dev
13:59 proller joined #minetest-dev
14:22 ecube joined #minetest-dev
14:26 kahrl before I continue working on httpfetch, I'd like to get some pulls merged that might conflict with it
14:26 kahrl for example, #846
14:27 kahrl I'm not sure about some of it though
14:27 kahrl for example, the engine.start at mainmenu.lua:458 (when doubleclicking a serverlist entry) might need setting.set calls too
14:28 kahrl (I'm not sure that a CHG event is always guaranteed to be sent before a DCL event)
14:29 kahrl also, just from reading the code, I believe with this pull request the port is no longer saved when starting a server from the menu
14:32 kahrl ^ yeah, indeed that is the case
14:36 kahrl actually...
14:36 kahrl this is because tab_server() hardcodes 30000 as the initial server port
14:36 kahrl wtf?
14:37 Exio4 blame sapier
14:41 Semilevel joined #minetest-dev
14:43 PilzAdam joined #minetest-dev
14:43 ShadowNinja #925 needs a decision soon...
14:44 kahrl ShadowNinja: I don't like removing the metatable
14:44 PilzAdam I dont like that either
14:44 kahrl it could probably be made faster by building the metatable once instead of in every vector.new
14:45 PilzAdam #941 is definitely a "no" too, since it breaks the current API
14:45 ShadowNinja kahrl: It is nice, but inconsistant and slow. This likely slows down vectors more than assert..
14:45 kahrl what is inconsistent about it?
14:45 thexyz "likely" is a bad word, avoid
14:45 ShadowNinja PilzAdam: Shouldn't mods work find with it?
14:46 PilzAdam ShadowNinja, it doesnt work with ipairs(), and basically every mod I know uses ipairs() on it
14:46 PilzAdam so you can close that pull request
14:47 thexyz that's not how it works
14:47 ShadowNinja kahrl: local pos = minetest.function_returning_pos() pos + {x=1, y=2, z=3} doesn't work vector.add(pos, 4) + {x=1, y=2, z=3} works.
14:47 thexyz ShadowNinja should explain why this is better than what we have now
14:47 thexyz and then if it really is much more better then it could be merged some day
14:47 BrandonReese joined #minetest-dev
14:48 kahrl ShadowNinja: well a pos is not a vector, so it's not really inconsistent
14:48 ShadowNinja It is better because, as kahrl said, finding a specific player is faster, just get player_list[name] instead of looping through the table and checking player:get_player_name() (more us->C calls)
14:49 ShadowNinja kahrl: They are identical except for the metatable.
14:49 kahrl ShadowNinja: minetest.get_player_by_name
14:50 ShadowNinja ipairs use should probably be stoped, it is slower than pairs and doesn't give you indexes.
14:50 PilzAdam it doesnt matter if its better
14:50 PilzAdam if it breaks the API then it cant be merged
14:50 thexyz ooh
14:50 thexyz that's a poor world we live in
14:54 kahrl where did you get the info that ipairs is slower?
14:54 kahrl I found this: http://facepunch.com/showthread.php?t=875909
14:57 ShadowNinja Exio posted a link, if I can find it...
14:58 ShadowNinja http://forums.civfanatics.c​om/showthread.php?t=463488
14:59 thexyz PilzAdam: you mean, like, we never break the api? even in new major versions?
15:00 proller only PilzAdam can break api and make BUGS
15:00 PilzAdam thexyz, we could discuss to break it between major versions, but it should stay compatible mostly
15:01 thexyz proller: got it.
15:01 thexyz PilzAdam: this contradicts with what you said earlier
15:01 thexyz 18:45  <+   PilzAdam> | so you can close that pull request
15:02 thexyz but okay
15:02 PilzAdam there is no real benefit for that, so its probably not worth to break between major versions
15:03 thexyz API says "minetest.get_connected_players() -> list of ObjectRefs" without specifying what list is; Lua only has tables
15:03 ShadowNinja list implies table indexed by integers.
15:03 proller want to commit https://github.com/minetest/minetest/pull/897  and if many peoples will against - disable flying part
15:04 thexyz alright
15:04 Jordach joined #minetest-dev
15:04 PilzAdam thexyz, in lua-api.txt list means "Lua table with integers as index"
15:04 thexyz we ain't breaking api
15:04 jin_xi joined #minetest-dev
15:05 kahrl by the way, I know what sapier would have to say about using os.clock for profiling ;)
15:05 kahrl http://irc.minetest.ru/minet​est-dev/2013-08-19#i_3273180
15:06 thexyz PilzAdam: alirght
15:07 thexyz I won't be anal and won't complain about the fact you didn't mention that they are 1 ≤ (number of elements) and unique
15:21 celeron55 hmm... it seems the official lua terminology is "array"
15:22 celeron55 for list-like tables
15:23 celeron55 reference: the PIL book, eg. http://www.lua.org/pil/19.3.html http://www.lua.org/pil/7.3.html
15:24 celeron55 better: http://www.lua.org/pil/11.1.html
15:26 kahrl the lua source code calls them array too
15:27 celeron55 maybe lua_api.txt should define that list equals array in what it says
15:32 Semilevel joined #minetest-dev
15:37 kahrl https://gist.github.com/kahrl/6780389
15:38 kahrl perhaps without the explicit indices?
15:40 kahrl https://gist.github.com/kahrl/6780442
15:50 Semilevel joined #minetest-dev
15:55 kahrl https://github.com/minetest/minetest/pull/846 <-- posted a bug I noticed
15:56 kahrl what should we do about that pull request? I'm not sure I should go on with httpfetch before these issues are resolved
15:57 RealBadAngel joined #minetest-dev
15:58 celeron55 i hope those who are more involved with it than me can resolve it
16:15 n4x joined #minetest-dev
16:28 Calinou joined #minetest-dev
16:33 kaeza joined #minetest-dev
17:34 iqualfragile joined #minetest-dev
18:01 Taoki joined #minetest-dev
18:16 ShadowNinja IMO image button labels should show in the tooltip, not over the image. Or a aditional tooltip field could be added.
18:33 Akien joined #minetest-dev
18:35 ritchie joined #minetest-dev
18:35 ritchie joined #minetest-dev
18:43 Akien joined #minetest-dev
19:07 proller joined #minetest-dev
20:00 loggingbot_ joined #minetest-dev
20:00 Topic for #minetest-dev is now Minetest core development and maintenance. Chit-chat goes to #minetest. Consider this instead of /msg celeron55. http://irc.minetest.ru/minetest-dev/ http://dev.minetest.net/
20:00 lanxu joined #minetest-dev
20:28 Naked joined #minetest-dev
21:16 ShadowNinja https://github.com/minetest/minetest/pull/942
21:34 hmmmm joined #minetest-dev
21:40 iqualfragile_ joined #minetest-dev
22:04 werwerwer_ joined #minetest-dev
22:41 eeew joined #minetest-dev

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