Time  Nick         Message
00:54 ShadowNinja  Zeno`: Re: Shader update patterns; I don't like the factory pattern any more than you, but it (or a variation of it) is necessary in that situation (unless you make a major API redesign or something I guess).  I removed the SkyBackgroundColorProvider, but I had to replace it with a pointer to a pointer to the Sky, which isn't mush (if any) better.  It seems like those things were actually done pretty well, despite the "superfluous
00:54 ShadowNinja  enterprise Javainess" that it look like at first.
01:48 Zeno`        ShadowNinja, I agree
01:50 Zeno`        ShadowNinja, I think (hope) I said when I mentioned that I hated that, that I also thought it was probably the best way
01:50 Zeno`        so I have no objection to the PR at all. It's just how it has to be
01:50 Zeno`        as much as I hate it :)
01:51 Zeno`        it's a good PR btw
01:51 Zeno`        I've tested it a few times now
05:46 Hijiri       we should be introducing monads instead
05:47 est          Hijiri: when you rewrite minetest in haskell, we can do that
05:48 est          xD
07:10 paramat      forum spambot please ban https://forum.minetest.net/viewtopic.php?p=234186#p234186
12:35 gregorycu    Thank you ShadowNinja
14:25 gregorycu    Fuck I'm good
14:27 gregorycu    So, I sped up mapblock_mesh_generate_special, from 1.1ms on average time, to 0.65ms
14:40 gregorycu    I know what you're thinking, "How did he do that. Save 0.4ms!"
15:06 paramat      i'll probably start on a PR for this soon https://github.com/minetest/minetest/issues/4390#issuecomment-247217212
15:09 paramat      part of #4390
15:09 ShadowBot    https://github.com/minetest/minetest/issues/4390 -- Settings / .conf: Mgv5 ground noise unsettable / missing. 2 problems with .conf auto-generation
15:46 gregorycu    #4544
15:46 ShadowBot    https://github.com/minetest/minetest/issues/4544 -- mapblock_mesh_generate_special improvements by gregorycu
15:47 gregorycu    Shouldn't be too controversial
16:00 paramat      thanks for that
16:14 paramat      #4545 might merge this later (in a few hours)
16:14 ShadowBot    https://github.com/minetest/minetest/issues/4545 -- Conf.example: Re-add deleted noise parameter documentation by paramat
16:24 paramat      as far as i know rubenwardy is currently busy starting university
16:25 gregorycu    University? Why is he doing that?
16:33 Calinou      gregorycu: I'm at university too :P
16:33 gregorycu    Go get a job ya bludger
16:34 gregorycu    http://www.urbandictionary.com/define.php?term=bludger
16:43 gregorycu    #4346
16:43 ShadowBot    https://github.com/minetest/minetest/issues/4346 -- Improved getPointedThing() by juhdanad
16:43 gregorycu    Looks good, we need to make sure we don't let this get stale
17:05 paramat      game#1293 which means game#709 can now be closed
17:05 ShadowBot    https://github.com/minetest/minetest_game/issues/1293 -- Add settingtypes.txt for Minetest Game by paramat
17:05 ShadowBot    https://github.com/minetest/minetest_game/issues/709 -- Add settingtypes.txt by PilzAdam
17:12 paramat      game 1293 has the required approval so i might merge it later with game#1292
17:12 ShadowBot    https://github.com/minetest/minetest_game/issues/1292 -- Doors: Trim open fencegate collison box again by paramat
17:20 paramat      will merge games 1292 1293 in a moment
17:25 paramat      now merging
17:28 paramat      complete
17:29 paramat      whoo 17 PRs
17:32 celeron55    the PR open/closed ratio of _game is kind of impressive
17:45 Fixer        gregorycu is on coding spree
17:46 Fixer        celeron55: in good way, or bad?
17:52 Krock        PR open/merge ratio must be huge, not the one of open/close
17:53 celeron55    i disagree
17:53 celeron55    if PRs are shit, they shouldn't be merged, and the shittyness of PRs is not dependent on the project itself
17:57 paramat      will merge #4545 in a moment
17:57 ShadowBot    https://github.com/minetest/minetest/issues/4545 -- Conf.example: Re-add deleted noise parameter documentation by paramat
18:03 paramat      now merging
18:04 celeron55    that will need a better solution in the long term
18:05 paramat      yeah
18:09 paramat      merged
18:45 est31        rejecting a change is much harder than ignoring it
18:45 est31        because you dont have to come up with an argument
18:45 est31        when you ignore a change
18:46 est31        tbh, most changes look good concept wise though, but reviewing them is just hard
18:46 est31        like that PR for getting the selected node that does the raycasting
18:47 est31        its nice concept wise, but the algorithm needs to be reviewed.
18:49 Krock        requires someone with the knowledge who takes the time to analyze it
21:15 SloanOnLinux hi