Minetest logo

IRC log for #minetest-dev, 2014-09-28

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

All times shown according to UTC.

Time Nick Message
00:08 Strikingwolf left #minetest-dev
00:14 doomtron joined #minetest-dev
00:46 Taoki joined #minetest-dev
01:26 Miner_48er joined #minetest-dev
01:27 mberends joined #minetest-dev
01:30 mberends_ joined #minetest-dev
01:49 GrimKriegor joined #minetest-dev
02:00 Zeno` joined #minetest-dev
02:07 jin_xi progress: http://youtu.be/kTL8IKk4YlM
02:12 petrusd987 joined #minetest-dev
03:22 AnotherBrick joined #minetest-dev
03:22 chchjesus joined #minetest-dev
03:53 NakedFury joined #minetest-dev
04:32 mberends_ joined #minetest-dev
05:08 AnotherBrick joined #minetest-dev
05:35 nore joined #minetest-dev
06:36 Krock joined #minetest-dev
06:43 Miner_48er joined #minetest-dev
06:44 BlockMen joined #minetest-dev
06:45 BlockMen going to push this in 30 minutes if noone claims anything: https://github.com/BlockMen/minetest/commi​t/b75e7148c2c6ad9a602d78d7ec6356273e3a4214
06:48 nore BlockMen: if that works, then I'm ok with it
06:48 Hunterz joined #minetest-dev
06:50 BlockMen nore, yes it does :)
06:51 nore so merge it then :)
06:53 BlockMen done
06:56 VanessaE I didn't even know you could DO that..
07:13 RealBadAngel BlockMen, why you think multiline tooltips were broken at all?
07:13 BlockMen RealBadAngel, click on the issue mentioned in commit msg
07:15 RealBadAngel i can see the image, but multiline tooltips were working perfectly for me
07:16 RealBadAngel http://i.imgur.com/r2TpDP6.png
07:18 mos_basik_ joined #minetest-dev
07:19 BlockMen RealBadAngel, the tooltips of items (inventory) have their own tooltip rendering
07:19 BlockMen try it out with a multilined item decription instead, you will be surprised ;)
07:20 RealBadAngel ah, indeed they do have own
07:20 RealBadAngel lemme check tooltip element then
07:23 RealBadAngel i did that similar way, your seems to be a bit easier
07:25 RealBadAngel well, finding number of lines there is easier but creates a copy of the tooltip
07:26 BlockMen how did you that?
07:26 RealBadAngel http://pastebin.com/u3gfg2fz
07:26 RealBadAngel that also gets number of lines but without copying the string
07:28 BlockMen but another int, size_t and while loop
07:29 BlockMen i think this is both similar in terms of performance
07:30 RealBadAngel propably yes so i think its ok
07:31 kahrl BlockMen: split() handles escape, you probably don't want that
07:31 kahrl it is really meant for parsing the formspec language
07:33 RealBadAngel in that case my way is correct
07:34 kahrl I wonder why m_tooltip_element->getTextHeight() doesn't account for the number of lines
07:36 RealBadAngel i dont remember know why, but it was broken somehow, i kept getting weird effects using that
07:37 kahrl ok I looked at the implementation, it is indeed broken
07:38 kahrl it only looks at the number of lines in BrokenText (pun not intended) when word wrap is enabled
07:38 kahrl in 1.8 at least
07:40 kahrl irrlicht trunk has a comment there: // TODO: Text can have multiple lines which are not in BrokenText  // This is likely not correct. But as I have no time for further  // investigation I just fix it for now by return the true height here.
08:02 ImQ009 joined #minetest-dev
08:12 BlockMen since split() is not useable and a the row detection should be consistent we should change it to this https://github.com/BlockMen/minetest/commi​t/61ed56f916f71a708b56260d7cba2160e8166529
08:12 BlockMen kahrl, RealBadAngel ^
08:14 RealBadAngel im not sure, imho iterating the string is a bit better than creating unnecesary copy of it. what do you think kahrl?
08:15 kahrl I think it doesn't really matter in this case
08:17 RealBadAngel propably not
08:18 RealBadAngel BlockMen, so ok
08:18 RealBadAngel until some1 decides to put a book into tooltip ;)
08:28 BlockMen RealBadAngel, then better limit tooltip length :P
08:28 BlockMen and ok, pushing it then
08:32 Krock <3 BlockMen for fixing "low prior" bugs :)
08:41 BlockMen :)
08:41 PenguinDad joined #minetest-dev
09:10 jin_xi joined #minetest-dev
09:33 rubenwardy joined #minetest-dev
09:34 edlothiol joined #minetest-dev
09:38 ImQ009 joined #minetest-dev
09:43 Megaf_ hm, the fullscreen option in minetest.conf doesnt seem to be working
09:44 kahrl have you tried restarting it?
09:45 Megaf_ several times
09:47 Calinou joined #minetest-dev
09:49 sfan5 Megaf_: os x?
09:54 Megaf_ sfan5: Debian
09:55 Megaf_ it works on OS X, but the OS X is a much older builder
09:55 sfan5 there are reports that it doesn't work on OS X at all
09:56 Megaf_ it works very fine on OS X
09:58 sfan5 ¯\_(ツ)_/¯
09:58 PenguinDad Megaf_: which irrlicht version?
10:10 Megaf_ that' s a very good question
10:10 Megaf_ Irrlicht log: Irrlicht Engine version 1.7.3
10:11 Megaf_ I will updated to the latest one
10:12 Megaf_ It seems like there are some bugs with Minetest on Debian Wheezy
10:13 Megaf_ client seg faults when you send a long message on 32 bit with irrlicht 1.7.3
10:13 Megaf_ client doesnt go fullscreen in some hardware
10:28 alexxs joined #minetest-dev
10:49 Krock I guess pull #1515 is ready to merge? Sometimes it actually gives the situation, where this fix is needed
10:50 Krock (playing with falling nodes)
10:58 werwerwer joined #minetest-dev
11:14 rubenwardy joined #minetest-dev
11:22 BlockMen Krock, why is that rounding needed?
11:25 Krock BlockMen, because a misplaced falling node can get stuck
11:25 Krock example: if they have a X and/or Z movement speed
11:31 BlockMen ok. i think it can be merged then (since sapier already said it looks fine to him)
11:31 Krock it's not important - just to complete that one
11:49 GrimKriegor joined #minetest-dev
12:16 Amaz joined #minetest-dev
12:27 sol_invictus joined #minetest-dev
12:29 Animetrom joined #minetest-dev
12:32 Animetrom joined #minetest-dev
12:38 Phroh|orca joined #minetest-dev
13:09 rubenwardy joined #minetest-dev
13:22 Zeno` BlockMen, I don't get it
13:22 Zeno` Why split up the string like you do in b75e714 ?
13:24 rubenwardy joined #minetest-dev
13:27 Krock Zeno`, exactly. counting all '\n' would be much more efficient
13:27 Zeno` There should be a + 1 in parenthesis but the idea is right
13:27 Zeno` I don't get it :(
13:28 Zeno` https://github.com/minetest/minetest/commi​t/b75e7148c2c6ad9a602d78d7ec6356273e3a4214
13:28 CraigyDavi Urm "minetest.exe --worldname server" seems to be broken...
13:29 Zeno` surely (after adding the +1) that would be more sane. But maybe I'm going insane
13:29 rubenwardy joined #minetest-dev
13:31 Zeno` CraigyDavi, it seems to work for me
13:32 Zeno` oh wait... I'm not using Windows, so I dunno
13:33 CraigyDavi I do: "minetest-server.exe --server --worldname server" and get "World '--worldname' not available. Available worlds:
13:33 CraigyDavi server"
13:34 CraigyDavi Yeah, this is happening on windows
13:41 PenguinDad CraigyDavi: which version/commit hash?
13:41 CraigyDavi 9a685a4f2e
13:43 BlockMen CraigyDavi, it got fixed already: https://github.com/minetest/minetest/commi​t/5ca5630a116458910fdffa097417bb211b89ba4e
13:43 PenguinDad CraigyDavi: then your build is 2 commits too old to contain the fix
13:43 CraigyDavi Ah thanks didn't see that
13:43 BlockMen Zeno`, idk. havent thought off that and the others did not aswell
13:44 BlockMen also, does that work for "normal" std::string as for any array?
13:44 Zeno` It works on std::string... it would not work on a c style string
13:45 Zeno` but splitting the string seems overkill to me
13:46 Zeno` plus, using a standard function is very likely to be faster than anything you and I could code (not that it matters much in this case)
13:46 RealBadAngel Zeno`, we were discussing that already
13:47 Zeno` RealBadAngel, so why was the decision made to not use the simple way?
13:48 RealBadAngel http://irc.minetest.ru/minet​est-dev/2014-09-28#i_3948960
13:49 Zeno` I've read all that and I still don't understand
13:50 Zeno` a one liner is turned into making a new function and splitting the string up?
13:51 rubenwardy joined #minetest-dev
13:51 RealBadAngel what one liner?
13:51 Zeno` std::count() with a multiply
13:52 Zeno` counting the '\n' characters
13:55 RealBadAngel does count works witch chars or strings?
13:55 RealBadAngel from what i can see it searches the string for a char
13:56 Krock joined #minetest-dev
13:56 RealBadAngel http://rosettacode.org/wiki/Cou​nt_occurrences_of_a_substring#C.2B.2B
13:58 Zeno` one sec
13:59 Zeno` we're not looking for a sub-string though, but a character
13:59 RealBadAngel no, we are looking for a string
13:59 Zeno` why's that?
13:59 RealBadAngel "\n" are two chars
14:00 Zeno` that's incorrect
14:00 Zeno` the environment will handle that
14:00 Zeno` http://codepad.org/jYX2lmo0
14:01 Zeno` it doesn't matter that '\n' might be 2 characters
14:01 Zeno` if it's 2 characters it will match the 2 characters
14:01 RealBadAngel if it works then indeed thats better way than mine or blockmens string slpitting
14:02 RealBadAngel kahrl, what do you think?
14:06 RealBadAngel but on the other hand propably that count is something like my code
14:06 Zeno` well it is but the libc probably does byte alignment and stuff
14:07 Zeno` which is kind of irrelevant
14:07 Zeno` for such small strings heh
14:09 Zeno` http://codepad.org/8M2LZw2W
14:09 Zeno` I wish I had a Windows machine but it's in the Standard somewhere
14:12 RealBadAngel my code was basically the same, just checked
14:13 RealBadAngel just BlockMen and kahrl decided it doesnt really matters and chosen split way
14:13 Zeno` :/
14:14 BlockMen RealBadAngel, you agreed aswell :P http://irc.minetest.ru/minet​est-dev/2014-09-28#i_3949089
14:14 RealBadAngel yup
14:15 RealBadAngel just pointed out originally i wanted iterating the string rather than copyin it
14:16 BlockMen for what i found now on web is that everyone prefers a different way for counting things like that and noone could prove his way was the best
14:17 BlockMen but if you guys think it needs to be changed for that case, do it ;)
14:23 RealBadAngel BlockMen, if theres a method (count) to do that, why not using it?
14:23 rubenwardy \n is one character
14:23 rubenwardy You may also have to look for carriage returns, though
14:43 Guest28228 joined #minetest-dev
14:43 Miner_48er joined #minetest-dev
14:45 Zeno` seems \r is handled by irrlicht whatever the case may be
14:56 Guest39002 joined #minetest-dev
15:03 AnotherBrick joined #minetest-dev
15:20 petrusd987 joined #minetest-dev
15:29 zat joined #minetest-dev
15:29 hmmmm joined #minetest-dev
15:36 AnotherBrick joined #minetest-dev
16:00 alexxs joined #minetest-dev
16:16 werwerwer joined #minetest-dev
16:52 AnotherBrick joined #minetest-dev
16:55 EvergreenTree joined #minetest-dev
17:05 NakedFury joined #minetest-dev
17:05 kaeza joined #minetest-dev
17:11 diemartin joined #minetest-dev
17:42 Megaf_ folks, about my fullscreen mode not working
17:42 Megaf_ Irrlicht log: Could not find specified video mode, running windowed.
17:43 kaeza joined #minetest-dev
17:43 sfan5 paste the output of "minetest --videomodes"
17:43 Megaf_ kaeza: Hi
17:43 kaeza o/
17:44 Megaf_ sfan5: ok, it worked now
17:45 Megaf_ problem is, only 24 bit color is supported on Linux
17:45 Megaf_ but in minetest.conf 32 bit is used
17:46 Guest39002 Any comments on this?: https://forum.minetest.net/v​iewtopic.php?f=9&amp;t=10192
17:47 VanessaE Guest39002:  that belongs on #minetest
17:48 Guest39002 Umm... It’s a tutorial for Minetest. And I wonder if there is any chance for it to go into core. So I guess it’s on-topic. :)
17:50 VanessaE in that case it might fit here. :P
17:53 DuDraig joined #minetest-dev
17:55 Calinou joined #minetest-dev
18:03 casimir joined #minetest-dev
18:10 RealBadAngel http://i.imgur.com/FDXtgXq.png
18:10 RealBadAngel mesh nodes with shaders applied :)
18:34 nore joined #minetest-dev
18:37 petrusd987 joined #minetest-dev
19:05 diemartin joined #minetest-dev
19:17 werwerwer joined #minetest-dev
19:37 AnotherBrick joined #minetest-dev
20:49 iqualfragile joined #minetest-dev
20:49 OldCoder joined #minetest-dev
20:52 kaeza joined #minetest-dev
20:56 iqualfragile left #minetest-dev
21:00 diemartin joined #minetest-dev
21:14 petrusd987 joined #minetest-dev
21:18 proller joined #minetest-dev
21:22 Wuzzy2 joined #minetest-dev
21:34 ShadowNinja hmmmm: /msg ChanServ ACCESS #minetest-dev LIST -- #minetest has quite a few more ops than -dev.
21:35 khonkhortisan joined #minetest-dev
21:36 ShadowNinja Me and sfan manage ops in #minetest.  sfan and c55 have that access in -dev.
21:37 ShadowNinja (c55 also has the privileges in #minetest, but he doesn't go there much)
21:39 RealBadAngel ShadowNinja, wanna see some mesh nodes? :)
21:39 RealBadAngel i just made nice screenshot: http://i.imgur.com/eKgfuns.png
21:39 proller joined #minetest-dev
21:42 ShadowNinja RealBadAngel: Awesome.  :-D
21:42 kahrl inb4 <PilzAdam> not voxely enough
21:43 ShadowNinja Give games the freedom to decide how voxely they want to be.
21:44 kahrl well, I agree.
21:45 RealBadAngel with meshes anything can be done
21:48 jin_xi so i got all the pieces needed to move mt particles to irrlicht ones
21:48 jin_xi but its a big change and im a bit overwhelmed
21:50 RealBadAngel jin_xi, great
21:50 RealBadAngel dont forget bout rain :)
21:55 blaaaaargh joined #minetest-dev
22:00 ShadowNinja jin_xi: Hmmm, do you have solitary particles (not spawners) implemented?
22:02 jin_xi no
22:03 kaeza joined #minetest-dev
22:03 jin_xi i am not proficient enough to do full replacement and think i have come as far as i can get
22:04 jin_xi my question is more of the type how to proceed, should i make a gist and add it to that pull request
22:05 ShadowNinja Make it usable and make a PR.  Make sure to post performance differences and things that still need to be done.
22:07 jin_xi usable as in build and tryable, but wip
22:11 proller joined #minetest-dev
22:14 kaeza joined #minetest-dev
22:28 terminator joined #minetest-dev
22:36 kaeza joined #minetest-dev
22:38 jin_xi got a nice round number! https://github.com/minetest/minetest/pull/1700
22:43 EvergreenTree joined #minetest-dev
22:46 VanessaE the next one had better have some Enterprise reference damn it :)
23:21 doomtron joined #minetest-dev
23:35 Miner_48er joined #minetest-dev

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