Time  Nick         Message
01:12 paramat      lhofhansl #6483 tested, see my comment
01:12 ShadowBot    https://github.com/minetest/minetest/issues/6483 -- Retrieve extra speed-directed, narrow view cone to load blocks the player will encounter soon. by lhofhansl
01:18 lhofhansl    Thanks. Let's leave some of the layout fixes (in unchanged code) to another PR (caching all those setting soon in a different PR anyway).
01:26 paramat      ok
01:26 paramat      go ahead then
01:34 lhofhansl    shortening the if-statement really harms readability. I do want the short-circuit of the if, though. I can break it up no problem at all, but are you sure that's what you're asking for.
01:34 paramat      don't change the logic, obviously
01:35 paramat      but split the long lines, unfortunately it will split in the list of arguements, but 90 is a hard rule
01:36 paramat      *arguments
01:36 lhofhansl    Will. Just making sure it's a hard rule :)
01:38 paramat      "Try to keep lines under 80 characters. It's okay if it goes over by a few, but 90 character lines or larger are definitely unacceptable. (note: this column count assumes 4-space indents)"
01:38 paramat      however you will notice much code goes over 90, but this is stuff we are slowly fixing
01:38 lhofhansl    I should change my tab width to 4.
01:48 paramat      looking
01:50 paramat      ok nice but because 'continue' is indented by 1 tab the condition above must be indented by 2 tabs to distinguish
01:50 paramat      but the splitting is done very well
01:51 paramat      lines 274 275 one more indent
01:52 paramat      oh that was something i missed in earlier commits
02:10 paramat      i have to go eat, back in an hour
03:11 CalebDavis   where can i fins the builtin code for the pause menu?
03:12 CalebDavis   find^
03:15 CalebDavis   nvmd found it
03:23 paramat      lhofhansl looks fine now
03:31 lhofhansl    Thanks :)
03:43 lhofhansl    First push. Yeah.
03:44 paramat      good, thanks
03:47 paramat      lhofhansl when you merge a PR, copy the commit SHA by clicking the clipborad icon in the commit history, and paste it into a new comment in the PR thread
03:47 paramat      *clipboard
03:49 paramat      and you can close issue 1442 by commenting '6483 merged' or something
05:12 lhofhansl    should we reduce some of the unnecessarily wide u32/s32 in the network protocol for 0.5? (see client.cpp:writePlayerPos, where we send s32 for pitch and yaw and u32 for keyPressed, just a single bit). Won't save much, but it came up in some discussion on PRs.
05:15 paramat      nerzhul would be the dev to talk to about this, he *might* be around soon as it's 7am in france
14:55 deltasquared Krock: right, so I'm not sure if there exists another PR as suggested in the one you linked for documentation, but would it be safe to consider that function stable yet?
14:56 deltasquared me and VanessaE were considering using it in pipeworks.
14:56 Krock        so stable like getpos() which is deprecated for a long time but still available
14:57 deltasquared Krock: we just didn't want it to vanish because it's not documented yet
14:58 deltasquared granted, if something newer does come along pipeworks can be transferred to that, we just wanted to be sure it wouldn't vanish quitely without there being a replacement
14:58 Krock        just by the amount of attention it got till now may indicate that changes happen rarely to never there
14:58 deltasquared hmm, so I guess we'd just have to come along and whine loudly if someone decides to take it out in future because it's "obscure" :P
15:00 deltasquared Krock: would it help to open an issue on minetest on this then? to indicate we're at least intending to use it, possibly also request the relevant documentation being added
15:00 deltasquared if there isn't already, going to have a look
15:07 deltasquared Krock: right, there is a PR involving it, but it's quietly tucked into https://github.com/minetest/minetest/issues/6372 . would it be worth putting a comment there perhaps
15:08 Krock        yes, with a patch or PR it would be even better
15:09 deltasquared Krock: I thought a PR would sweeten the deal ;) I'll have to catch w_laenger later, if he's the one who came up with it
15:09 deltasquared patch though, can I file an issue with a patch attached? I'd prefer doing that to say forking the repository
15:11 Krock        why not fork it? That would open the possibility to help minetest more in the future.
15:11 deltasquared Krock: I dunno, it feels like a commitment (I know it's not *actually*)
15:11 deltasquared baby steps.
17:15 rubenwardy   #6531
17:15 ShadowBot    https://github.com/minetest/minetest/issues/6531 -- Profiler: Fix multiple parameters not being passed to callback regist… by rubenwardy
17:25 rubenwardy   I'll merge in ~5 mins
17:30 rubenwardy   done
18:59 exio4        is there anybody working #6511? I might try to tackle it if nobody minds :)
18:59 ShadowBot    https://github.com/minetest/minetest/issues/6511 -- Unknown items may be crafted
19:01 red-001      feel free to
19:03 red-001      it shouldn't be too hard to implement
19:04 exio4        yeah, looking for small issues, on an intel atom n2600 right now, heh
19:04 paramat      i don't know of anyone working on that
19:15 sfan5        paramat: is it planned to add biome-specific grass colors to mtg at some point?
19:18 paramat      we can easily have a different surface node per biome, and almost do already, new surface nodes for temperate coniferous forest and deciduous forest are intended. but hardware colouring, no i'm strongly opposed, there's no advantage for just a few colours
19:19 paramat      so hopefully soon every biome will have a unique surface node
19:20 paramat      there's an issue somewhere for MC type smooth colouring
19:20 paramat      #6441
19:20 ShadowBot    https://github.com/minetest/minetest/issues/6441 -- Per-biome grass and foliage colors, using hardware coloring
19:21 paramat      that explains my thoughts
19:22 paramat      something for a separate new biome system written by someone else, i won't work on it or maintain it
19:23 paramat      and probably best used in a new fancier subgame, which we need anyway
19:30 sfan5        oh right I should have looked on the issue tracker first
19:30 lisac        hey paramat, what program is that in the screenshot?
19:30 sfan5        also i was asking about hardware coloring, not actual different nodes
19:32 red-001      yeah what program is that
19:33 sfan5        lisac: looks like GeoGebra to me (the geometry version)
19:40 paramat      correct
19:41 paramat      i inverted the colours to save my eyes
19:42 paramat      realtime voronoi cell adjustment, essential for choosing biome points
20:01 lisac        oh, nice
20:16 red-001      strings in settings are utf-8 encoded right?
21:14 lhofhansl    going to merge #6530 in a few.
21:14 ShadowBot    https://github.com/minetest/minetest/issues/6530 -- Cache server config settings. by lhofhansl
21:17 red-001      lhofhansl, could you rename the commit/pr to indicate it also includes unrelated code style fixes?
21:26 nerzhul      lhofhansl, i don't like the approach where settings are dup in each remoteclient object :(
21:28 Krock        ^ but it's already an improvement
21:58 lhofhansl    I will not merge then.
22:01 lhofhansl    Well... on the other hand I cannot think of anything better
22:03 lhofhansl    So I'd like to merge anyway.
22:06 sfan5        maybe nrz does have a better idea since he's the one not liking the approach
22:38 paramat      thanks for the floats stuff
22:41 paramat      is there any advantage in caching in a constructor as opposed to using 'const static setting = g_settings->get..' ?
22:41 paramat      sometimes i wonder which to use
22:42 red-001      not really just looks cleaner
22:42 sfan5        the latter will cache when the function is first called AFAIK
22:42 red-001      there is an advantage for more complex data
22:42 red-001      of being faster
22:45 lhofhansl    can't do statics. These settings can be changed without restarting the executable (Singleplay or Local Game). That static would not be reloaded.
22:58 paramat      oops of course
22:59 paramat      actually no
23:00 paramat      these settings are ones that don't need to be changed during a game session
23:00 paramat      we have many settings that need a program restart to alter, they are made static for performance reasons
23:01 paramat      and we can always use 'thread_local' to be able to change them in main menu
23:01 paramat      but still, no problem as either method is ok
23:08 Hijiri       what format are floating points transmitted in?
23:09 Hijiri       it's no longer number / 1000 integers right?
23:09 sfan5        it still is
23:11 lhofhansl    yes, they do not need to change. I can start MT, play a SinglePlayer game, now these are all cached. Then I go to settings (not leaving MT) change the setting and do another SinglePlayer game, that would still see the old settings. (same for Local Game)
23:12 lhofhansl    Somebody who has since left MT (forgot the name) has "yelled" about not doing this. And I agree.
23:15 rubenwardy   hmmmm
23:38 lhofhansl    yes ;)