Minetest logo

IRC log for #minetest-dev, 2013-04-08

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

All times shown according to UTC.

Time Nick Message
01:19 Exio irrlicht should support an easy way to do a "multi-pass shaders"
01:22 hmmmm proller, https://github.com/minetest/minetest/commit/7c0e707cf1d0031788567bfec66f6c13ec901cdc
01:23 Exio wb hmmmm :P
01:23 hmmmm hello
01:27 RealBadAngel param2 for ores? whats that?
01:29 ShadowNinja RealBadAngel: it is for infinite springs with finite liquid
01:29 hmmmm it's needed for infinite water springs with finite water activated, but you can use it for other things as well
01:30 hmmmm when the colorlike node type comes around, you'll be able to specify what color the ore is you're placing, for example
01:30 hmmmm or if you're placing things with facedir
01:30 RealBadAngel may be funny
01:31 RealBadAngel i will play with it
01:31 hmmmm hmm
01:32 hmmmm for our nether, instead of just doing exactly what minetest does, it'd be neat if we had glowstone occuring as an ore inside the netherrack which was a light source
01:33 hmmmm it might be fun in general to play around with ore that emits light occuring in air
01:34 ShadowNinja hmmmm: should commits to the games be counted as contributors?
01:34 hmmmm no
01:34 hmmmm there would be way too many people
01:34 RealBadAngel hmmmm, have you tried to build with jit?
01:34 ShadowNinja well, maybe a seperate section?
01:34 hmmmm i haven't
01:34 hmmmm i need to do that
01:34 RealBadAngel so what are you waitin for?
01:34 RealBadAngel need feedback :)
01:34 hmmmm shadowninja, can you imagine the wars that would start?
01:35 ShadowNinja ok, guess it is best left alone
01:36 hmmmm i don't mean to act like we're above people who only contribute to the game, but it's much easier to contribute a texture or some small fix or whatever.. when something is easier, there is more incentive to do it
01:36 VanessaE joined #minetest-dev
01:37 hmmmm they'd make any small bullshit tweak or whatever and bug those with commit access until they add it, and then demand to get added to the credits tab for efame
01:38 ShadowNinja yea, I can see that
01:38 hmmmm alllright
01:39 * hmmmm cracks fingers
01:39 hmmmm time for sapiers' pull flood
01:39 RealBadAngel lol, good luck :)
01:39 RealBadAngel may the force be with you ;)
01:39 hmmmm it's something that needed to be done eventually
01:40 hmmmm then after i have all of his stuff squared away, i'll check out luajit
01:40 RealBadAngel btw i checked some of the commits by blue42 or something like that
01:40 RealBadAngel that pull with blocking is nice
01:41 hmmmm he's the Metology guy if my memory serves me correctly
01:41 hmmmm yeah, it's awesome that he split it up
01:41 hmmmm i am definitely going to take a look at those things
01:41 RealBadAngel block pull will need protocol bump, hes adding new key
01:42 hmmmm the lua hud will also bump it
01:42 RealBadAngel lctrl by default
01:45 RealBadAngel effect is nice when youre holding sword
01:45 RealBadAngel it does look like blocking
01:46 hmmmm just looked at it, i'm going to need to clean up some pieces of code, but overall it looks good
01:48 RealBadAngel btw, regarding jit. link to static or dynamic lib?
01:48 RealBadAngel im linkin now to .so
01:48 hmmmm hm
01:48 hmmmm no idea if it matters
01:49 hmmmm let's do a static link so things don't explode if the user updates
01:49 RealBadAngel ok, i will change it then
01:52 hmmmm https://github.com/sapier/minetest/commit/6b256e8ca5029ccd30327bdb3a1e4a44b64c3296#L0R682  lolwhat
01:52 ShadowNinja more static linking :-(
01:53 hmmmm if it's a very very very dirty hack, are you not finished with this one?  would you rather me wait for this leak fix?
01:53 hmmmm @sapier
01:55 hmmmm and why is that outside of the class definition again?  *blinks*
02:05 hmmmm christ, i can't believe that SourceImageCache didn't even bother to attempt to clean itself up
02:05 hmmmm who wrote this crap to begin with
02:05 hmmmm :)
02:08 RealBadAngel hehehe
02:54 RealBadAngel we have found a bug that limits number of nodes
02:55 RealBadAngel http://pastebin.com/01tmh64A
02:55 RealBadAngel place this somwhere in default
02:56 RealBadAngel and increase loop until it starts to segfault
02:56 RealBadAngel 1600 is the value for my set of mods
02:56 RealBadAngel for clean mt it will be higher
02:56 hmmmm who's 'we'
02:56 VanessaE celeron55 told me once of the cause and the likely limit, but I don't remember what the cause was, nor whether it is fixable.  Only that it imposed a limit somewhere near 4096
02:56 RealBadAngel me and VanessaE
02:56 VanessaE hmmmm: RealBadAngel and I have been working in the background
02:56 hmmmm 4096 is CONTENT_MAX
02:57 hmmmm in any case, 1600 is not 4096
02:57 VanessaE hmmmm: it was supposed to be 65534.
02:58 hmmmm the content identifier can hold values up to 65536, sure, but right now the maximum is 4096
02:58 RealBadAngel 1600 + that what i has already defined with my set of mods
02:58 hmmmm it can be increased
02:58 hmmmm but nevermind any of that
02:58 hmmmm it shouldn't segfault if it hits the limit
02:58 hmmmm i'll check it out some other time
02:58 VanessaE hmmmm: if you wish to test, I can arrange a test set.
02:58 hmmmm no, thanks, that's really not necessary...
02:59 RealBadAngel loop above its most simple one
02:59 VanessaE anyone know why was it fixed at 4096?  what's the logic there?
02:59 RealBadAngel propably some atlas remains
02:59 RealBadAngel dunno
02:59 hmmmm has nothing to do with the atlas
03:00 hmmmm and no idea, really
03:00 RealBadAngel so get rid of it
03:00 RealBadAngel i will fix it when i got back from work today
03:00 hmmmm for things that iterate through the entire range of values, that would increase slowness by a factor of 16
03:00 hmmmm i wouldn't change CONTENT_MAX unless we know exactly why it's 4096 right now.
03:02 hmmmm also, some code in the mapblock serialization would need to be modified, namely the size of the node id mapping table, which should be min(MAP_BLOCKSIZE^3, CONTENT_MAX)
03:02 VanessaE Sep 08 2012 11:42:03 <celeron55>        actually it's 4096 due to a leftover constant 8)
03:02 VanessaE Sep 08 2012 11:42:12 <celeron55>        it should be changed some day
03:02 VanessaE (minetest-delta)
03:03 VanessaE Sep 08 2012 11:52:38 <celeron55>        yes but there is a static table of node definitions, the size of which is 4096
03:03 VanessaE Sep 08 2012 11:52:45 <celeron55>        go over that and get a segfault
03:03 hmmmm it'd be nice to know what the constant is there
03:03 VanessaE Sep 08 2012 11:53:04 <celeron55>        i saw that only when i started to experiemnt with that branch
03:03 VanessaE Sep 08 2012 11:53:08 <celeron55>        ment*
03:03 VanessaE Sep 08 2012 11:53:49 <celeron55>        by the way, if the size of that static table is set to 65536, it will always consume at least 50MB of RAM
03:03 VanessaE maybe that's why.
03:04 hmmmm don't really get why it's not a vector
03:05 VanessaE Sep 08 2012 11:54:01 <celeron55>        maybe it should be dynamic 8)
03:05 hmmmm hmm.  maybe.
03:07 VanessaE I say make it dynamic, max 32 bits, and be done with it.  :-)
03:08 VanessaE (anything worth doing is worth overdoing)
03:08 VanessaE really though, it's the Unix concept - either limit it to 1, or make it (effectively) infinite.
03:10 hmmmm max 32 bits
03:11 hmmmm lol^
03:11 hmmmm okay, you fork minetest and have fun with that
03:15 VanessaE note the smiley :)
03:18 hmmmm but on a serious note, minecraft has 8 byte MapNode structures
03:18 hmmmm just think about that for a bit
03:18 hmmmm they use twice as much space
03:18 VanessaE all I know is I can't run moretrees' "newstuff" branch on my server.
03:19 VanessaE (because I'm already pretty close to that 4096 limit)
03:19 hmmmm could it be possible that you have too much stuff?
03:19 hmmmm instead of adding more blocks, try making better use of that stuff
03:19 VanessaE entirely possible, sure.  thing is, everything on my server is being used somewhere
03:20 hmmmm anyway back to pushing memory leak fixes
03:20 VanessaE *nod* that's more important right now
03:21 hmmmm 8 more
03:21 hmmmm aww man, i go to work on my own thing (mapgen v7) for less than a week and this is what happens
03:22 VanessaE could be worse - you coulda been trying to do this during the big merge a couple months ago.
03:22 VanessaE (well a few months ago)
03:23 hmmmm is few > couple?
03:23 hmmmm i often have this discussion with my classmates
03:23 VanessaE to me, few = 3 to 4.
03:23 VanessaE couple is exclusively 2.
03:23 hmmmm the consensus is that "a few" means 2, "a couple" means 3, 4, or 5
03:23 hmmmm oh
03:23 hmmmm so it's the opposite for you
03:23 VanessaE *nod*
03:24 hmmmm your way makes more sense (the word couple does refer to one and another)
03:24 VanessaE I read "couple" as in the definition for a two-person unit - i.e. "don't they make a nice couple?" etc.
03:26 VanessaE now, you ask them to define "several" and you can expect about a dozen different answers :)
03:26 hmmmm but all answers will be less than a dozen
03:26 VanessaE in theory yeah
03:53 hmmmm hah... look at how much cleaner it is to not use iterators https://github.com/sapier/minetest/commit/c23bcd8ca8d91691fb671fd83cab5dcf37c0df96
03:55 hmmmm hmm.. i need to modify that one a bit before i merge it, because now, i in fact _do_ create biomedef inside of the emergemanager
03:57 hmmmm how does that work even?  apparently he did rebase it on top of the mgv7 commit
04:03 hmmmm it doesn't work.. hm
04:04 hmmmm and no, the map doesn't delete mgparams, although it certainly should
04:36 troller joined #minetest-dev
05:07 celeron55 hmmmm: i don't know why it's 4096
05:07 celeron55 someone has probably just forgotten to raise it when the id field was increased to full 16 bits
05:09 celeron55 before that the maximum was 4096, and the real maximum count was something like 2048+126 (some trickery to allow some amount of nodes to have full param1+param2)
05:09 celeron55 so the raise to 16 bit effectively doubled the amount of nodes, so i guess it made everyone happy for this long
05:11 emptty joined #minetest-dev
05:13 VanessaE celeron55: and then I came along ;)
05:13 VanessaE (and threw a monkey wrench into the whole damn works)
05:27 jin_xi joined #minetest-dev
05:41 * VanessaE wanders off to bed, leaving said monkey wrench behind ;)
05:41 VanessaE night.
07:23 emptty joined #minetest-dev
07:53 serengeor joined #minetest-dev
07:55 jin_xi joined #minetest-dev
08:43 jin_xi joined #minetest-dev
09:04 troller joined #minetest-dev
09:28 emptty joined #minetest-dev
09:32 ImQ009 joined #minetest-dev
10:18 PilzAdam joined #minetest-dev
11:08 rubenwardy joined #minetest-dev
11:17 Jordach joined #minetest-dev
11:17 Jordach joined #minetest-dev
11:34 celeron55 hmm... as "-" isn't allowed in modnames, maybe we could add direct compatibility for mods extracted from github packages by ignoring the rest of the directory name if it contains a "-"
11:34 celeron55 i don't see githug going away nor github's behavior changing in respect to this, so we could just adjust to it
11:34 celeron55 githug :D
11:36 * Exio githugs celeron55
11:40 celeron55 the name of that hypothetical behavior is "githug" from now on
11:41 celeron55 so who would like to have githug or does someone see a problem in it?
11:41 Exio hehe
11:41 PilzAdam we alread have this for "_game" at the end of game IDs
11:41 PilzAdam it should be documented somewere
11:43 Jordach celeron55, yup, githug ftw
11:43 Exio well, i hope i can start looking at the code what applies the shaders, the multi-pass shader is needed :P
11:44 Exio did anyone work or anything for/about that?
11:53 celeron55 PilzAdam: i think _game is documented somewhere (is it?)
11:54 PilzAdam it isnt
11:56 celeron55 hmm, it seems so
12:11 Taoki PilzAdam: Submitted a bug fix commit to my fall bobbing, it was not frame rate dependent. If we can look at it again today, please check the latest commit in the pull request
12:28 Taoki celeron55: If you have a minute maybe you can merge this upstream. Ready and tested on my end. https://github.com/minetest/minetest/pull/609
12:30 ImQ009 joined #minetest-dev
12:41 rubenwardy left #minetest-dev
12:41 rubenwardy joined #minetest-dev
12:48 PilzAdam line_of_sight() is broken, it returns the "step" param
12:49 celeron55 i'd imagine it will return that parameter if you try to check it in something you can't see through
12:50 celeron55 umm
12:50 celeron55 well obviously no 8)
12:50 PilzAdam no, its definetly broken
12:56 RealBadAngel joined #minetest-dev
13:05 Taoki I'm getting the feeling minetest.show_formspec is broken too...
13:06 Taoki It does work, but on_receive_fields doesn't function
13:08 Taoki In minetest.register_node I have both: minetest.show_formspec(clicker:get_player_name(), "x:y", formspec_string) ---- and ---- on_receive_fields = function(pos, formname, fields, sender) ---- But when clicking any button in the formspec, the on_receive_fields function is completely ignored.
13:09 Taoki What could be broken?
13:22 Taoki Never mind, I found the problem. I had an abm registered to replace the node and it was re-creating it. My mistake
13:28 hmmmm joined #minetest-dev
13:35 rubenwardy1 joined #minetest-dev
13:38 rubenwardy joined #minetest-dev
13:38 RealBadAngel hmmmm, have you tried luajit?
14:01 ImQ009 joined #minetest-dev
14:05 proller__ joined #minetest-dev
14:10 proller joined #minetest-dev
14:41 serengeor joined #minetest-dev
14:45 proller__ joined #minetest-dev
14:59 jojoa1997|Tablet joined #minetest-dev
15:00 jojoa1997|Tablet left #minetest-dev
15:03 rubenwardy joined #minetest-dev
15:06 iqualfragile joined #minetest-dev
15:07 rubenwardy where is the new damage system documented?
15:22 Calinou joined #minetest-dev
15:22 celeron55 in lua_api.txt
15:24 iqualfragile which hash does the forum use for its passwords?
15:24 rubenwardy https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L658 ?
15:25 dimeshake joined #minetest-dev
15:28 iqualfragile celeron55: is it md5 or sha224 or ?
15:32 jojoa1997|Tablet joined #minetest-dev
15:38 celeron55 iqualfragile: as far as i know, sha1
15:38 celeron55 rubenwardy: yes
15:38 iqualfragile ok, thanks
15:38 celeron55 why does that interest you
15:39 * jojoa1997|Tablet cough hacker cough
15:49 rubenwardy1 joined #minetest-dev
15:55 rubenwardy1 joined #minetest-dev
16:16 dexter0 joined #minetest-dev
18:07 proller joined #minetest-dev
18:13 nyuszika7h_ joined #minetest-dev
18:27 emptty joined #minetest-dev
18:36 BlockMen joined #minetest-dev
18:37 BlockMen celeron55, is that ok to mark as external link? https://dl.dropbox.com/s/sox5e6e57imek2z/dokuwiki_prev_2.png
18:38 emptty joined #minetest-dev
18:42 proller joined #minetest-dev
18:58 iqualfragile celeron55: yes, its salted sha1
18:58 proller joined #minetest-dev
18:58 iqualfragile it interests me as im writing a database for the mods and i want to allow the users to login using their forum-accounts
19:03 emptty joined #minetest-dev
19:03 iqualfragile you stated that you would be willing to host one& the guy doing it at the moment does not use normalized databases
19:04 rubenwardy hmm
19:20 rarkenin joined #minetest-dev
19:20 rarkenin joined #minetest-dev
19:44 celeron55 iqualfragile: wut
19:44 iqualfragile celeron55: wut "wut"?
19:45 celeron55 i don't know what you are doing, but if you want to integrate something to the forum, you need to be in contact with thexyz
19:46 iqualfragile i dont realy need to integrate anything, i just add 3-4 tables to the database and access the table containing the users
19:46 iqualfragile i.e.: read-access
19:46 celeron55 that too
19:46 celeron55 BlockMen: i think they are ok
19:46 JesseH joined #minetest-dev
19:47 celeron55 BlockMen: the dokuwiki integration seems to be coming along much better than i actually thought
19:47 celeron55 it's like it all was made for that 8)
19:48 JesseH Is there anyway to assist the core-development with my Lua knowledge?
19:48 celeron55 i guess it does limit a bit how the page content can be formatted, but it's a small price to pay for editing abilities
19:49 BlockMen celeron55, yes it was a nice idea with the dokuwiki
19:49 BlockMen and the way the templates are made i can much personalize
19:49 BlockMen like the menubar
19:49 jojoa1997|Tablet joined #minetest-dev
19:50 jojoa1997|Tablet Has anyone thought of having an in browser minetest for chrome
19:50 celeron55 oh god
19:50 celeron55 jojoa1997|Tablet: this is not a suggestions forum
19:50 jojoa1997|Tablet Ok
19:50 JesseH My question goes unanswered :(((
19:50 celeron55 jojoa1997|Tablet: what you just said is considered spam here
19:50 JesseH lol
19:51 jojoa1997|Tablet Oh sorry :-[
19:51 celeron55 JesseH: probably nothing right at the moment, but hang around and see for a day or few
19:51 JesseH Alrighty!
19:51 thexyz celeron55: we need more ops
19:51 PilzAdam JesseH, most question will stay unanswered for some hours
19:51 celeron55 or figure out and suggest yourself
19:51 jojoa1997|Tablet Jesse what was your Q
19:51 PilzAdam thexyz, take me
19:51 * JesseH couldn't be trusted with op.
19:51 JesseH Just saying...not now anyways :P
19:51 celeron55 thexyz: who even are ops now here?
19:52 thexyz celeron55: me, you
19:52 celeron55 who should be?
19:52 thexyz giving me +f would be another option
19:52 thexyz all core devs
19:52 JesseH celeron55, I forked minetest, gonna check out some things.
19:52 JesseH i don't know how much, if any game logic and what not are handled with Lua yet
19:52 celeron55 thexyz: well, i'll throw you +f so you can sort this out
19:53 celeron55 thexyz: you now have it
19:54 jojoa1997|Tablet left #minetest-dev
19:54 PilzAdam I suggest voice core devs in this channel
19:54 celeron55 BlockMen: is there a lot of work to do still?
19:55 celeron55 PilzAdam: maybe
19:55 jojoa1997|Tablet joined #minetest-dev
19:55 BlockMen celeron55, nope. just one or two boxes, maybe the link color
19:55 BlockMen if you want i could upload template later
19:55 RealBadAngel PilzAdam, any success building with luajit for windows?
19:56 PilzAdam I still dont have a libary
19:56 JesseH *.* where is the issues on minetests github page
19:56 celeron55 BlockMen: if it's usable, i think i could try to put it in use already - then if you have some updates, those can be taken into use later
19:58 BlockMen celeron55, i would say it is useable
20:00 RealBadAngel PilzAdam, https://github.com/malkia/ufo/tree/master/bin/Windows/x86
20:00 RealBadAngel theres jit dll
20:01 PilzAdam I already have a dll
20:01 RealBadAngel lib too
20:02 RealBadAngel https://github.com/malkia/ufo/blob/master/bin/Windows/x86/luajit.lib
20:04 thexyz would that be a good idea to have devs voiced by default?
20:04 PilzAdam yes
20:04 JesseH How does minetest keep up with issues and such?
20:05 celeron55 JesseH: https://github.com/minetest/minetest/issues https://github.com/minetest/minetest/pulls
20:06 JesseH O.o
20:06 * JesseH checks his vision./
20:06 thexyz proller: consider identifying
20:06 thexyz RealBadAngel: consider registering
20:06 JesseH I didn't see those, anyhow thanks celeron55
20:06 JesseH oh darn I was on your github
20:07 jojoa1997|Tablet Is this what I think it is? https://github.com/minetest/minetest/pull/612
20:08 BlockMen celeron55, if you want i could make the setup for the dokuwiki
20:10 celeron55 BlockMen: just throw me the template somehow
20:11 RealBadAngel thexyz, considered
20:11 JesseH Anyone started any other games with the minetest engine?
20:11 JesseH yet*
20:11 BlockMen celeron55, ok. i will just finish and upload in a few minutes
20:11 JesseH or should I ask this here
20:12 jojoa1997|Tablet What is the treavia build on github
20:12 sfan5 *travis
20:12 celeron55 JesseH: that's not really #minetest-dev stuff - this channel is mostly for developing the stuff that is in the official upstream repositories
20:12 sfan5 LEARN SPELLING!
20:12 jojoa1997|Tablet Brb getting my keyboard
20:13 JesseH alrighty celeron55
20:13 PilzAdam RealBadAngel, got it
20:13 celeron55 JesseH: and other officially supported things (like websites)
20:13 sfan5 JesseH: maybe... #minetest-mods
20:13 PilzAdam works :-)
20:13 JesseH alrighty!
20:13 celeron55 JesseH: but anyhow, check the forum's games section
20:13 JesseH forums .___.
20:13 * JesseH dies
20:14 RealBadAngel so if it works for both and linux i think we can merge https://github.com/minetest/minetest/pull/623
20:14 RealBadAngel *windows and linux
20:14 celeron55 JesseH: why die :P
20:14 PilzAdam testing on Linux with your latest changes
20:14 JesseH forums make me...cringe
20:15 ShadowNinja RealBadAngel, PilzAdam: Why are you voiced?
20:15 celeron55 forums are useful for publishing stuff
20:15 celeron55 our forum isn't half bad :P
20:15 JesseH true that
20:15 PilzAdam ShadowNinja, we decided to voice core devs
20:15 JesseH I would ask for a link,but might get a rude answer xD
20:16 celeron55 you can find everything by a) googling, b) by links at http://minetest.net
20:16 JesseH not rude, but something I could have thought of myself <_>
20:16 jojoa1997|Tablet Forum.minetest.net
20:17 jojoa1997|Tablet It is a really friendly one
20:17 jojoa1997|Tablet I thinl
20:18 jojoa1997|Tablet Sorry
20:19 jojoa1997|Tablet Scratching off stuff from screen
20:19 jojoa1997|Tablet It turned on by accident
20:19 celeron55 jojoa1997|Tablet: stop flooding this channel with useless crap
20:19 celeron55 and this now was a warning before +q
20:20 jojoa1997|Tablet I really didn't mean to send
20:20 jojoa1997|Tablet left #minetest-dev
20:22 jin_xi joined #minetest-dev
20:22 * celeron55 feels time stretching as BlockMen's minutes become multiple minutes each on their own
20:23 celeron55 i need to sleep ->
20:23 BlockMen celeron55, https://www.dropbox.com/s/bniu5ic30qlqgtu/minetest_template.zip
20:23 PilzAdam too late :-p
20:23 BlockMen just put the "minetest"-folder in "lib\tpl"
20:24 PilzAdam RealBadAngel, everything works fine and is good to go
20:24 celeron55 well, i guess i'm not allowed to sleep then
20:24 hmmmm realbadangel, i did not
20:24 hmmmm i've gotta do that
20:24 hmmmm was sapier around?
20:25 PilzAdam hmmmm, LuaJIT works fine for me on Linux (with and without LuaJIT) and my cross compile works with LuaJIT too
20:25 hmmmm alright
20:25 hmmmm so it's just me who has to try it
20:26 RealBadAngel indeed :)
20:26 ShadowNinja celeron55: shouldn't you drop #minetest-delta/mods/doc?
20:27 thexyz ShadowNinja: #minetest-doc is registered by marktraceur and #minetest-mods isn't even registered
20:29 thexyz and it seems marktraceur got pissed off a while ago and isn't going to appear here
20:31 PilzAdam Minetest tends to segfault when Lua errors appear on startup
20:32 PilzAdam it also often segfaults on shutdown
20:36 hmmmm in general, i feel like we're going to see more problems with luajit on Linux because the platform is so nonuniform
20:36 emptty joined #minetest-dev
20:37 PilzAdam what I said is not related to LuaJIT
20:40 RealBadAngel hmm, that error with openal i mentioned earlier wasnt also jit related
20:41 emptty1 joined #minetest-dev
20:46 BlockMen celeron55, does the template work for you?
20:49 hmmmm mm oh, i need to ask sapier about two things it seems
20:49 PilzAdam is one of it the broken line_of_sight()?
20:49 hmmmm the try clause in the sky code, and the hack in the extruded item patch
20:50 hmmmm broken line_of_sight() is your question for him, not mine :p
20:50 PilzAdam well, you merged it
20:50 PilzAdam Ill blame you :-p
20:50 hmmmm hey look
20:50 hmmmm it doesn't matter because that code hasn't been released even
20:51 hmmmm as long as everything works for the release, it's fine.  it's not like if there's some bug right now, it's horrible
20:52 RealBadAngel hmmmm, hows testing jit?
20:52 PilzAdam bbl
20:52 hmmmm gud
20:53 thexyz um, what the fuck
20:53 thexyz http://i.imgur.com/HsJXSGs.png http://i.imgur.com/y22n3Zs.png
20:54 thexyz that hand just dropped off one of our players
20:54 hmmmm hahah
20:54 RealBadAngel rotfl
20:55 Exio hey, i want my hand back!
20:55 hmmmm -- LuaJIT not found, using bundled Lua.  good :)
20:56 PilzAdam sudo apt-get install libluajit-5.1-dev
20:56 hmmmm alright now compiling
20:58 BlockMen celeron55, the folders "lay" and "nav" are useless. i forgot to delete
20:58 hmmmm oh my god i'm in heaven
20:59 hmmmm 10ms on_generate calls with minetest_game
20:59 Jordach joined #minetest-dev
20:59 Jordach joined #minetest-dev
21:00 iqualfragile can someone give me some specification on what a moddb should be able to do?
21:00 iqualfragile should the files be hosted by minetest or should they just be linked
21:00 RealBadAngel write mods on its  own ;)
21:00 iqualfragile RealBadAngel: yeah, give me a few minutes
21:00 iqualfragile done!
21:01 iqualfragile can you all agree on what features the moddb should support and write a pm on the forums to me?
21:01 Exio minutes == 3 seconds? k
21:02 Exio hmmmm: 10ms? and before? :P
21:02 * iqualfragile feels time collapsing as Iqualfragiles's minutes become fractions fn their own
21:03 celeron55 BlockMen: got it up; http://c55.me/minetest4/doku.php?id=start
21:03 iqualfragile im gona check my inbox tomorow at about 18:00 would be nice to find some specifications there, but now i have to go sleeping
21:03 celeron55 seems that there is still work to do
21:05 celeron55 eg. the "page trace" is kind of screwed up, the page where you upload images is borked, and this what you get by clicking an image is broken: http://c55.me/minetest4/lib/exe/detail.php?id=start&amp;media=screenshot_terrain_2012-12-03.jpg
21:05 * VanessaE peeks in.
21:05 BlockMen celeron55, what do you mean?
21:05 BlockMen *sry, had not seen the next msg
21:05 ShadowNinja BlockMen: ohloh isn't linked near the bottom of the page
21:05 celeron55 it's a solid start though
21:06 * iqualfragile likes the old design better
21:06 hmmmm everything seems at least twice as fast with luajit
21:06 BlockMen ShadowNinja, that is just content thing, that made celeron55
21:06 hmmmm also when i a node like gravel drops, it falls smoothly and instantaneously
21:06 iqualfragile hmmmm: great! when will it be merged?
21:06 hmmmm it's not like there's a lag in between
21:06 VanessaE hmmmm:  nice, isn't it?
21:06 celeron55 well, people are free to try to make better templates for the site
21:07 hmmmm vanessae, it's not your modification if that's what you're thinking
21:07 hmmmm i'm using a very outdated minetest_game
21:07 BlockMen celeron55, you have to set "Number of "trace" to 0
21:07 VanessaE hmmmm: no, I mean using jit in general.
21:07 hmmmm yeah
21:07 BlockMen we dont need the breadcrubs
21:07 hmmmm realbadangel, it looks great, push it
21:08 iqualfragile the moddb should support texturepacks, shouldnt it?
21:08 RealBadAngel you know i have just merge button ;) you push it
21:08 celeron55 BlockMen: done
21:09 RealBadAngel btw, when i finished this system wide option i figured out how to properly bundle jit
21:09 hmmmm no no
21:09 hmmmm realbadangel, just do it yourself
21:09 hmmmm it's less work
21:09 celeron55 iqualfragile: this is about two things: the layout, and the changing of the main site to be dokuwiki-based so that the content is easier to update
21:09 hmmmm git fetch upstream && git rebase upstream/master git push upstream master
21:10 RealBadAngel shit, and you say its easier? ;)
21:10 hmmmm three commands
21:10 BlockMen celeron55, i know how i have to fix that with the image view. hold on
21:10 iqualfragile celeron55: you mean the transition to dokuwiki?
21:11 RealBadAngel im not used yet to git cli, still learning it with technic repo
21:11 PilzAdam hmmmm, RealBadAngel, I can merge it if you want
21:12 RealBadAngel dont wanna break something before im not sure im doin it properly
21:12 hmmmm it looks ugly when you see a "commited by blah" underneath
21:12 hmmmm for the majority of cases it's unavoidable
21:12 PilzAdam It looks even more ugly if RealBadAngel clicks the "Merge" button
21:12 RealBadAngel lol, indeed ;)
21:13 VanessaE hmmmm: what looks 10x worse is "authored by xxxx 6 months ago" when the associated code was last edited only seconds ago
21:13 PilzAdam on, click and I push it
21:13 hmmmm ugh fine
21:13 hmmmm i'll do it then
21:13 PilzAdam done
21:13 hmmmm well
21:14 hmmmm guess not
21:15 RealBadAngel i think we can close now 2 another jit related pulls
21:15 PilzAdam yea, go ahead, Ill do a win build
21:15 RealBadAngel ok
21:15 VanessaE I'll get a linux build out.
21:16 iqualfragile yaay
21:16 iqualfragile btw: i think its great that minetests development has gained this momentum
21:18 RealBadAngel those were 3 pull requests, also there was one with just fix to serialize.lua
21:18 RealBadAngel closed them all
21:18 PilzAdam probably lack of comunication
21:19 Exio what about https://github.com/minetest/minetest/pull/609?
21:20 PilzAdam it has still the setting...
21:20 BlockMen celeron55, how bout remove "table of contents"? it looks strange with it  http://c55.me/minetest4/doku.php
21:21 Exio PilzAdam: what "setting"?
21:21 Exio #fall_bobbing_amount?
21:21 PilzAdam I dont like to add a setting for it
21:21 Exio err, remove the # :P
21:21 Exio i like that setting (as it is only one)
21:22 celeron55 BlockMen: done 8)
21:25 BlockMen celeron55, looks better. and here come the update: https://www.dropbox.com/s/grs59jhy16neg8u/minetest_template_v2.zip
21:25 celeron55 whoever wants to get to editing the site: create an account and note to me and i'll give the privileges
21:27 celeron55 lol now it's completely broken
21:28 celeron55 well, it's consistently broken now, rather than some page working and some not 8)))
21:28 celeron55 now i'll go to sleep ->
21:28 thexyz celeron55: the page looks broken right now
21:29 thexyz oh
21:29 thexyz i'm too slow
21:29 sapier joined #minetest-dev
21:30 thexyz ok, i fixed it
21:30 BlockMen celeron55, have you replaced the whole folder? if not you need to
21:31 BlockMen oh, too late
21:31 thexyz he forgot to purge the cache
21:31 thexyz but anyone can do that just by appending &purge=true to the end of css.php
21:31 BlockMen thexyz, yep. works now
21:34 sapier @hmmm this very very dirty hack is required as this class is constructed read only. Thats reason for making it outside of class too ;-) I'll have an additional look how much of minetest needs to be changed to do this in a clean way
21:34 hmmmm hey
21:35 hmmmm i had a merge conflcit with the sky cleanup order patch
21:35 hmmmm in your diff, there was a try { block around something
21:35 hmmmm which isn't there in the current master
21:36 hmmmm how did this try block get removed if it was there in the first place, or the other way around
21:36 sapier no it's there ... i had to move the try up in order to catch all exceptions within this fct
21:36 sapier otherwise minetest will just crash on e.g. lua errors
21:36 hmmmm for some reason that modification isn't reflected in any of the patches i added so far
21:37 BlockMen i'm also off now
21:37 BlockMen bye
21:37 sapier strange maybe some merge artifact
21:37 BlockMen left #minetest-dev
21:37 jojoa1997|Tablet joined #minetest-dev
21:37 hmmmm and yes, i did add the smartpointer fix first
21:38 hmmmm so the try{ is really supposed to be at line 1028
21:39 sapier wait I'll have alook I'm on laptop atm need to view code on github :-)
21:39 sapier its game.cpp ?
21:40 hmmmm yes
21:40 hmmmm i'll have to take a better look at it later
21:40 hmmmm bbl
21:40 sapier no try is supposed to be on 998 as it's in current latest
21:41 sapier I think eventmgr may throw one of the exceptions I've moved from main.cpp to game.cpp
21:41 jojoa1997|Tablet left #minetest-dev
21:42 VanessaE [04-08 17:36] * Menche says add a ENABLE_LUAJIT cmake option that is on by default
21:42 VanessaE I think this would be a good idea, actually.
21:43 RealBadAngel implement it?
21:44 sapier but you need to declare the sky pointer outside this try block in order do be able to delete it after it
21:44 sapier this information may have been lost in spliting this patch up
21:45 RealBadAngel also another thing: searching for libs takes dynamic ones first when both are aviable
21:45 RealBadAngel to use static one i shall let it search for exact name of library
21:45 sapier I assume this is distribution dependent?
22:00 proller 02:00:11: ERROR[main]: /usr/local/share/minetest/builtin/serialize.lua:200: invalid escape sequence near '"
22:00 proller дгфоше
22:00 proller luajit
22:00 sapier fix is already in pull requests
22:01 proller oh. its old system dir
22:01 proller from 0.4.4
22:02 RealBadAngel hehe
22:02 proller but runned from ~/minesrc/
22:04 sapier sometimes I wonder what some manufacturers think ... "intended use" for a electronic radiator controler: "The device ... must be protected from ... sources of radiant heat"
22:06 VanessaE *facepalm*
22:06 proller hmmmm, #0  0x00000000004a85e4 in MapgenV6::generateCaves (this=0x805486500,  -  /0 still exist
22:10 iqualfragile gn8
22:15 hmmmm proller, ?
22:15 hmmmm are you sure that's a division by 0
22:15 proller no
22:15 hmmmm show me the full error
22:16 proller Floating point exception: 8 (core dumped)
22:16 hmmmm well if you knew it happened in generateCaves(), you must have it on debug
22:16 hmmmm what line did it happen?
22:28 sapier left #minetest-dev
22:49 PilzAdam :-( hes gone again
22:49 PilzAdam anyway, sapier, if you read the logs: line_of_sight() is broken and it always returns the step parameter
22:54 PilzAdam hmmmm, https://github.com/minetest/minetest/pull/597 isnt that the "active objects only" setting you wanted?
23:02 proller hmmmm, how to see line? bt doesnt show, prints only func
23:02 proller CMAKE_BUILD_TYPE:STRING=Debug

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