Time Nick Message 00:05 ShadowNinja sapier: I was looking at your security v2 and noticed the top of two files says Minetest-c55 00:07 sapier thx I'll fix it 00:11 ShadowNinja troller: you forgot to add constant scale here: https://github.com/minetest/minetest/pull/543/files#L0R319 00:14 hmmmm shadowninja, i am looking at the cloud menu thing and there's a lot to discuss 00:15 hmmmm right off the bat, it doesn't compile, you're missing a semicolon. then the clouds are way too fast, reducing the dtime multiplier to makes it look better 00:16 hmmmm there are quite a few lines in your code that not only go over 80 lines, but more like over 100 lines 00:16 VanessaE I didn't notice a problem with the speed of the clouds? 00:16 hmmmm you didn't? 00:17 VanessaE no, they seemed fine on my end as of that little patch I added. 00:17 VanessaE (a comment in the pull request) 00:17 hmmmm was that patch added into that pull request 00:18 VanessaE dunno 00:19 hmmmm and the footer takes up the entire window length 00:20 VanessaE that's what my little 2-line (effective) change does. 00:20 hmmmm yeah... 00:20 hmmmm you see, now when you go to use the original footer, it doesn't look right at all 00:21 VanessaE well that's easily solved by adding some blank space to it 00:21 VanessaE but I supplied a new foote rimage too 00:21 VanessaE s/\ r/r\ / 00:21 hmmmm yeah.. how do other people feel about that? 00:21 VanessaE dunno, not many have commented on it so far 00:22 VanessaE *checks pull request* 00:22 hmmmm i'd like the non-cloud menu to be pixel-perfect to what it was before tihs patch 00:22 hmmmm which means we'd have to add the border around the footer 00:22 hmmmm would you volunteer to do that? 00:22 VanessaE in the context of a image that's scaled, can you define "pixel perfect"? 00:23 hmmmm well 00:23 VanessaE oh you mean size relative? 00:23 hmmmm yes 00:23 VanessaE hm, I should be able to manage something 00:23 VanessaE want the dark grey background too? 00:23 hmmmm okay 00:23 hmmmm hmm 00:23 hmmmm yes 00:23 hmmmm just exactly the way it was 00:23 VanessaE ok, give me a few and I'll upload one. 00:23 hmmmm (except without the -c55) 00:28 ShadowNinja hmmmm: oops, I thought I compiled it before pushing 00:30 ShadowNinja fixed and pushed 00:30 hmmmm it's okay, you didn't need to push it 00:30 VanessaE http://digitalaudioconcepts.com/vanessa/hobbies/minetest/images/menulogo.png 00:31 VanessaE as long as you center it on the window, it should be exactly as was before. 00:31 VanessaE (center horizontally I mean) 00:32 VanessaE might be off by one pixel vertically, I will fix that if needed. 00:32 hmmmm yep, looks good 00:32 hmmmm thanks 00:32 VanessaE np 00:32 ShadowNinja should I use that? 00:33 hmmmm shadowninja, i'm taking care of it from here on out 00:33 hmmmm you don't have to do anything more 00:52 hmmmm hmmm 00:53 hmmmm you know, i'm looking at the clouds and i'm wondering if it'd look better coming at you 00:53 hmmmm what do you guys think? 00:55 ShadowNinja I would have to try it but I think sideways would look better 00:56 hmmmm nevermind that the cloud object is initialized wrong.. 00:56 VanessaE hmmmm: actually yeah 00:56 VanessaE let's see it coming towards the player 00:58 hmmmm hang on, let me fix Clouds::Clouds() 00:59 ShadowNinja VanessaE: change v3f(-100,60,0) to v3f(0,60,100) in my branch line 1628 01:02 ShadowNinja hmmmm: I notice Clouds takes smgr twice, shouldn't that be changed? 01:02 hmmmm hmm? 01:02 RealBadAngel talkin to yourself? 01:02 hmmmm it does? 01:02 RealBadAngel hi all 01:02 hmmmm hello 01:03 ShadowNinja smgr and smgr->getRootSceneNode() 01:03 ShadowNinja can't it get it itself? 01:03 hmmmm oh, *shrug*, i don't know what the design intention was 01:04 hmmmm oh i know why, because you can't execute code in an initializer list 01:05 hmmmm i'd rather not fool around with something that doesn't need fixing 01:06 hmmmm anyway that clouds code, it was passing what was thought to be the cloud height as the seed, so the clouds would be the same exact thing every time. 01:07 hmmmm and what's more is that the cloud height would depend on cloud_height in the config, so i had to fix that as well 01:07 ShadowNinja ah, that's why changing it did nothing 01:07 hmmmm so i hardcoded the cloud height for that scene as 100 01:08 hmmmm it looks pretty nice. anyway, what's your final verdict on clouds coming toward you? i think it's neater that way, personally 01:08 VanessaE works for me 01:09 ShadowNinja looks fine either way 01:10 ShadowNinja will the clouds look higher than before, I was trying to change that 01:10 ShadowNinja +? 01:10 hmmmm they'll look a tiny bit lower 01:11 hmmmm alright, i'm really happy with the tweaked camera angle, the randomized seed, and the dtime multiplier at 3 01:11 hmmmm when i go to play minetest, this is what i'd want to see 01:12 hmmmm we'll see the consensus when people update and make tweaks if necesssary along the way to 0.4.6 01:12 ShadowNinja ok, just attribute me please 01:14 hmmmm take a look at how it is 01:14 VanessaE pulling it now. 01:15 hmmmm i'll personally be using the cloud menu from now on 01:15 VanessaE uh oh 01:15 VanessaE I think you used the wrong menulogo 01:15 hmmmm why's that? 01:15 hmmmm oh great hold on 01:15 VanessaE well, or rather, you deleted it. 01:15 hmmmm yeah 01:15 hmmmm it's menufooter.png now 01:15 hmmmm remember? 01:16 VanessaE right 01:16 VanessaE it's not in your commit, plus it doesn't build 01:16 VanessaE /home/vanessa/Minetest-Textures-and-mods/minetest/src/main.cpp:605:2: error: expected ‘,’ or ‘;’ before ‘if’ 01:16 ShadowNinja he didn't pull latest, I assume he will add a fixer commit now 01:17 hmmmm ? 01:17 hmmmm uh oh 01:17 ShadowNinja you pushed my old version that didn't compile 01:17 hmmmm what on earth 01:17 hmmmm i amended it 01:17 hmmmm OH SHIT 01:17 VanessaE you bwoke it! :'( 01:17 hmmmm i did not add the files 01:18 hmmmm okay, it's not bad 01:18 hmmmm it's like a couple minutes 01:18 ShadowNinja ah, forgit git add? 01:18 ShadowNinja *forgot 01:19 hmmmm gulp 01:19 hmmmm okay here we go 01:19 VanessaE merge conflict. 01:19 VanessaE I'll re-clone 01:19 hmmmm force it 01:19 hmmmm yeah that 01:20 VanessaE here goes nothin' 01:20 VanessaE (building against irrlicht 1.7.x) 01:20 hmmmm it'll work fine. 01:22 VanessaE ok.. 01:22 VanessaE interesting 01:22 hmmmm next up, get rid of that "TAPE AND GLUE" thing on the side of the singleplayer menu, it makes minetest seem like it's hacked together 01:22 hmmmm ugh 01:23 hmmmm it's definitely not, we put effort into this 01:23 VanessaE hmmmm: seems to work for me; clouds are too low to see with my HDX splash but no biggie 01:23 VanessaE +1 on getting rid of the glue. 01:23 VanessaE oh wait, here come some clouds now 01:23 Exio tape and glue? 01:23 hmmmm aside from cloud height, is everything else okay? 01:24 VanessaE Exio, "Single player" menu. 01:24 VanessaE hmmmm: yep, lemme check one thing 01:24 VanessaE hmmmm: yep, it's good. 01:25 ShadowNinja looks good here 01:26 hmmmm aaaahhh! i found a nyancat in the air for the first time ever 01:26 VanessaE wait 01:26 VanessaE hm 01:27 ShadowNinja I have found one of those too 01:27 VanessaE hmmmm: I don't see the default menufooter 01:27 VanessaE (with clouds turned on I mean) 01:27 hmmmm yeah that's intentional, remember? 01:28 VanessaE oh right 01:28 hmmmm with clouds turned on it uses menufooter_clouds.png 01:28 VanessaE right, I forgot. 01:29 VanessaE hrm 01:29 VanessaE menu_clouds = false 01:29 VanessaE should disable them in favor of the default old background, right? 01:29 VanessaE oh wait 01:30 VanessaE wrong file 01:30 VanessaE yep that works 01:30 VanessaE however, that new footer I drew isn't being used anywhere 01:31 ShadowNinja just symlink/copy it 01:31 VanessaE just making sure I'm following things clearly. 01:33 VanessaE (I mean the wide, cutaway view of the landscape, not the "==MINETEST==" one) 01:33 hmmmm hm 01:33 ShadowNinja oh, hmmmm wanted it to be pixel perfect 01:33 hmmmm i just noticed we have a credits page 01:34 hmmmm yeah nevermind, it's good enough 01:35 hmmmm irrlicht isn't correctly rendering the umlaut a in that one guy's name, and i'm not even existing :( 01:36 VanessaE you should fix the latter, for sure 01:36 VanessaE your name belongs near the head of the list given the work you've done on this 01:36 hmmmm d'aw, thanks :) 01:36 ShadowNinja not sure what you mean by the former but is matttpt spelled correctly? 01:36 hmmmm it isn't 01:37 hmmmm what a shame that half of these people aren't even around anymore 01:37 ShadowNinja I have seen about 7 out of 15 01:38 VanessaE kahrl keeps dropping by every so often, of all people 01:38 VanessaE erlehmann too. 01:38 hmmmm does he use a different nick? haven't seen him drop by at all 01:38 VanessaE nope, same old nicks for both 01:39 VanessaE [03-07 02:05] hmmm: many distro maintainers despise bundled libraries and want to use system libraries if at all possible 01:39 hmmmm yeah that's the last thing he said 01:39 hmmmm and we still need to fix the json thing 01:39 hmmmm so like, rename it back to json so system libraries can take over, i guess? and rename jthread completely 01:40 VanessaE (I last saw erlehmann quitting from #minetest back on Jan. 20) 01:40 hmmmm yeah exactly 01:40 VanessaE I dunno, I just hate json or really anything that isn't plain "key = value", one line per key. 01:41 VanessaE it's like trying to write a document in XML when all you need is the occasional tag. 01:42 hmmmm json fits in with lots of web technologies 01:42 hmmmm like celeron said, with cURL, it could prove useful later on 01:42 VanessaE fair enough. 01:42 VanessaE let's just not use it where it isn't really needed though 01:43 hmmmm right, like the whole noiseparams thing using json was ridic 01:44 hmmmm alright, now let's decide how to update the credits 01:44 hmmmm what I think we should do is something like Xfce has, a scrollable textbox with each person's real name and then email 01:44 VanessaE top 10 or so people, in order by number of contributions? 01:45 hmmmm top 10 is pretty arbitrary and people would understandably feel left out 01:45 VanessaE true 01:45 hmmmm if we use a scrollable textbox we can add everybody 01:45 VanessaE but would anyone use it? Maybe just an automatic scroller? 01:46 VanessaE you know, a movie-style closing credits sort of thing 01:46 hmmmm well i'd use it 01:46 hmmmm it's undoubtably better than having some number of people all on one line 01:55 ShadowNinja hmmmm: can dungeons be enabled/disabled with a config option? 01:55 hmmmm yes, you must add 'dungeons' to the map generation flags 01:55 hmmmm mg_flags = what you had before, dungeons 01:56 ShadowNinja good, I didn't notice that change in the commit, was the option there but unused before? 01:56 hmmmm yes 01:57 ShadowNinja are they enabled by default? 01:57 hmmmm no 02:00 ShadowNinja is there a reason #minetest has +C but not -dev? 02:11 hmmmm no idea... 02:11 hmmmm hrmm so we have to decide once and for all how we're going to classify contributors 02:13 ShadowNinja Just make sure I qualify :-) 02:14 hmmmm sure 02:14 ShadowNinja made a commit within x months? 02:14 hmmmm we still need to acknowledge people who did a lot but don't contribute anymore 02:15 ShadowNinja or have made over x commits? 02:15 hmmmm the way Xfce divvys up people is: Core Developers, Active Contributors, Previous Contributors, Translators 02:16 hmmmm i know i quote from xfce a lot but i feel like it's a tidy, organized operation 02:16 hmmmm and i have the credits for it right in front of my face 02:16 ShadowNinja do you know where a page describing those is? 02:17 Exio hmmmm works for the Xfce Spam Team! 02:17 Exio :P 02:17 VanessaE hmmmm: can I suggest my "bigger, sharper fonts ( http://forum.minetest.net/viewtopic.php?id=4436 ) be added as the defaults for when MT isn't running in freetype mode? 02:17 VanessaE (some formspecs will need tweaked to accommodate the larger size) 02:17 hmmmm http://www.xfce.org/about/credits 02:18 hmmmm i'll investigate// 02:18 VanessaE "goodies supervision".. interesting title (I know what it means) 02:18 hmmmm the people who guard and distribute the heroin? 02:18 ShadowNinja I mean describing what "core developer" or "active developer" means etc 02:19 hmmmm ah you're right, it doesn't have it on that page 02:22 hmmmm http://www.freebsd.org/doc/en/articles/contributors/article.html 02:22 ShadowNinja Can progress text be added to the loading screen for the freetype build? 02:23 hmmmm core developers, i guess, are the people with commit privileges, active developers are people who made contributions that were committed within the past say, 3 months? and previous developers would be people who made contributions before but not any time recently 02:24 hmmmm i guess... i have no idea what's going on with the freetype stuff 02:25 VanessaE the timing/categories sounds fair to me 02:25 ShadowNinja that sounds good 02:26 ShadowNinja *whew* that is a lot of contributors 02:26 hmmmm it's not going to really take that long to gather 02:26 hmmmm doesn't xyz have a minetest.net email address? 02:26 VanessaE hmmmm: I guess c55 likes to use that ohloh page or whatever it was 02:27 hmmmm oh, lol! 02:27 VanessaE https://www.ohloh.net/p/minetest-c55/contributors?query=&sort=kudo_position 02:27 VanessaE that one 02:27 hmmmm it strikes me as unprofessional for some reason 02:28 ShadowNinja I mean on the freebsd page 02:28 ShadowNinja I imagine the list of all contributors can be found with git log and maybe some pipes 02:29 VanessaE that might work too 02:31 RealBadAngel btw, any volunteers to implement player:set_yaw ? 02:33 hmmmm ahh! the problem is that we don't know the email addresses for many people 02:34 RealBadAngel ask them? 02:35 RealBadAngel and meanwhile leave field blank 02:36 hmmmm that's what i'm doing 02:36 RealBadAngel <-- mk@realbadangel.pl 02:36 hmmmm ya i have yours 02:38 RealBadAngel so, no volunteers for set_yaw? ;) 02:38 hmmmm nope ;\ 02:38 RealBadAngel damn :) 02:38 RealBadAngel shouldnt be that hard anyways 02:39 RealBadAngel theres already set_yaw for entity 02:39 VanessaE can someone at least make enabling freetype not disable the use of bitmap fonts? 02:39 VanessaE (wow, I talk good Engrish ;) ) 02:40 RealBadAngel as for a hobbit ;) 03:01 RealBadAngel ShadowNinja, maybe you will want te be a volunteer? 03:02 ShadowNinja hmm, maybe, if it wouldn't be to hard 03:02 ShadowNinja would it require aditions to the protocol? 03:05 RealBadAngel i think not 03:05 RealBadAngel look at lua entity set_yaw 03:06 RealBadAngel this will provide you scriptapi ways 03:06 RealBadAngel apply the same to player object just 03:07 RealBadAngel yaw have to be updated also by pressing keys so it is called there somwhere 03:08 RealBadAngel treatin use of this method as a response for a key pressed (and yaw changed then) shall do the trick 03:08 VanessaE ok, weird bug. If the server is compiled with -O3, it doesn't not send adequate map updates to the client with things like worldedit (even if you just //set say a 50 by 50-node platform) 03:09 hmmmm sounds like a gcc bug 03:09 VanessaE might be 03:10 ShadowNinja is there some sort of scriptapi_player.cpp? 03:11 RealBadAngel ah those splits 03:11 RealBadAngel grep for get_player_name 03:12 RealBadAngel and put it there 03:13 RealBadAngel http://www.youtube.com/watch?v=TGJHySGnMps 03:14 RealBadAngel i like those Crack animations :) 03:27 VanessaE hmmmm: also happens with -O2 03:28 hmmmm but not O1? 03:28 VanessaE I'm recompiling to -O1 now. 03:28 RealBadAngel i got the gates workin properly 03:29 VanessaE when I say "adequate" I mean only the few nodes closest to me, say a 5 node radius, were being updated to the client. 03:29 RealBadAngel only thing left is set_yaw for teleporting players 03:29 hmmmm >< 03:29 RealBadAngel also updated sounds 03:30 RealBadAngel and included iris open and close for future upgrades 03:30 VanessaE hmmmm: same at -O1 03:30 VanessaE this is a problem. 03:31 VanessaE re-trying without a -O setting. 03:32 hmmmm eh nevermind 03:32 hmmmm sounds like a bug... 03:32 RealBadAngel and i think i will get friday off at work 03:32 RealBadAngel so propably i will get looooonger weekend 03:33 hmmmm alright how does this look guys http://pastebin.com/iLGEVeYb 03:33 hmmmm can't seem to find jeija's email, and shadowninja has a fake email 03:34 VanessaE it looks good, as for jeija, try jeija@mesecons.net 03:34 hmmmm ok 03:34 VanessaE I think that's the one he prefers 03:34 RealBadAngel looks ok 03:35 VanessaE hmmmm: without a -O setting it still bugs out. 03:36 VanessaE I'm only getting updates in roughly 5 a node radius. 03:36 hmmmm actually that sounds about right now that i'm thinking about it 03:36 hmmmm 'cause look: 03:37 VanessaE no, this is wrong. 03:37 VanessaE it used to update everything out to about 5 mapblocks away 03:38 VanessaE I fly around the area that was changed and receive no updates. 03:38 hmmmm hold on i can't find it 03:38 VanessaE I have to restart my client to force it 03:39 VanessaE *sigh* please just go test it 03:39 hmmmm in server.cpp line 1557 it sends bulk change events to players within 5 blocks 03:39 hmmmm far_d_nodes 03:40 hmmmm so clearly, disable_single_change_sending must be enabled 03:40 VanessaE worldedit. Find some mostly-flat ground. do //pos1 and then walk out about 40 to 50 nodes diagonally. try to stay at the same Y elevation. //pos2. //set cobble 03:40 hmmmm which means that more than 3 nodes must have been added to the change queue 03:40 VanessaE what you'll see is a few nodes around you will update, and maybe a couple of whole mapblocks in the distance, but nothing in between 03:40 hmmmm well, are you sure this doesn't happen with previous versions? 03:41 VanessaE it didn't beofre. 03:41 VanessaE no. 03:41 VanessaE I'm very familiar with the normal map update pattern when using worldedit. 03:41 hmmmm well from what i'm reading in the code, this is how it *should* work.. 03:41 hmmmm can you find some version where it didn't happen so i can compare? 03:42 VanessaE try going back to commit e204bedf1d. 03:42 VanessaE I know it worked "right" there. 03:42 VanessaE I'll try, I mean 03:42 hmmmm at the menulogo update? 03:42 hmmmm that's really not too far back, eek 03:42 VanessaE yeah, that's a good spot to start 03:43 hmmmm in fact the only things to change since then are the cloud menu, the STL fix, my mapgen cleaning, moving some code to somewhere else... 03:43 hmmmm okay, i can say with 90% certainty that this was caused by the new STL migration 03:43 VanessaE I figured that might be the case 03:44 VanessaE lemme build from this older commit just to be sure of how far back it goes 03:44 hmmmm i'm looking at what was changed in server.cpp by that right now 03:44 VanessaE pk 03:44 VanessaE ok 03:44 hmmmm oh god... what if it was something in connection.cpp... that'd be even worse.... 03:45 VanessaE oh shit 03:45 VanessaE you know something? 03:45 hmmmm ? 03:45 VanessaE A LOT of people have been complaining of being unable to connect to most servers 03:45 VanessaE mine seems to be an exception to that rule, people have no trouble connecting to me 03:46 hmmmm the people who are having trouble connecting are using the most recent version from git? 03:46 hmmmm and you're sure of this? 03:46 VanessaE in some cases yes, in other cases no. 03:46 VanessaE but yeah, I've seen the complaints, and somehow my server's been the only one that's working reliably for those who complained. 03:46 VanessaE (go figure that one out) 03:47 VanessaE BAM! 03:47 VanessaE as I predicted. 03:47 hmmmm huh 03:47 VanessaE it works properly at e204bedf1d 03:47 hmmmm great -.- 03:48 VanessaE shall I work my way forward? 03:48 hmmmm no 03:48 VanessaE ok 03:48 hmmmm logically, the only thing that can possibly be the problem is the STL thing 03:49 VanessaE how hard would it be to revert the STL commit? 03:50 hmmmm hrmm 03:50 hmmmm i'd have to rewrite history in order to revert it 03:50 hmmmm since it's way too late for that, we'd have to re-patch everything that commit changed 03:50 VanessaE shit 03:52 VanessaE well, may as well go back to -O3 at least :D 03:53 hmmmm you can live with it for a little while, right? 03:53 RealBadAngel im off to work, c ya later folks 03:54 VanessaE oh sure 03:54 VanessaE I can just stay at this earlier commit for now 03:55 RealBadAngel btw, i do have working trees compiled every 2-3 days, so i can track the exact point when it got worse 03:55 hmmmm i am looking and i am not finding 03:56 hmmmm all the core-to-stl changes in server.cpp seem okay 03:56 RealBadAngel and i can go back to mid of january with that 04:01 hmmmm vanessa 04:01 VanessaE hm? 04:01 hmmmm would you like to try something? 04:01 VanessaE sure, gladly. 04:01 VanessaE (this ought to be good ;) 04:03 hmmmm in server.cpp, replace lines 1608 and 1909 with this: modified_blocks2[i->first] = m_env->getMap().getBlockNoCreateNoEx(i->first); 04:03 VanessaE against current HEAD? 04:03 hmmmm yes 04:03 VanessaE sure, gimme a sec to finish something 04:05 VanessaE 1909? That's got an access-denied message? 04:06 hmmmm 1609 04:06 hmmmm i mean 04:06 VanessaE ah 04:06 VanessaE that makes more sense ;) 04:06 hmmmm wait hold on 04:06 VanessaE *waits* 04:07 hmmmm that can't be the problem 04:07 VanessaE ok, not making that change. 04:07 hmmmm i'm sorry 04:10 VanessaE no worried. 04:10 VanessaE s 04:12 * VanessaE argues with a persistent shadow on the map 04:20 VanessaE BAM 04:20 VanessaE fuck you, you stupid shadow :) 04:20 hmmmm i give up 04:21 VanessaE that can't be good.. 04:26 VanessaE It's a drastic step, but I think it would be a good idea to back out these last few commits, I'm a little concerned that this could lead to data loss for someone 05:22 ShadowNinja RealBadAngel: Adding setYaw is not so simple for me 06:06 thexyz hm, can somebody check connection.h/cpp? i might've fucked something there 06:07 hmmmm the packets are reliable, it would've cropped up some sort of error if that's the case 06:08 thexyz i also have stl commit splitted to several ones (to help bisecting) on my pc which i can't access at the moment 06:08 hmmmm perhaps change back server.cpp and whatever else is necessary to the old version and see if that fixes the problem 06:08 hmmmm well it's not time-critical 06:08 hmmmm just needs to be done soon 06:08 thexyz will be able to on 17th 06:09 hmmmm ouch 06:09 hmmmm well whatever 06:09 thexyz yeah 06:55 VanessaE I'm off for the night also. 07:15 celeron55 the credits tab should roughly correspond to http://www.ohloh.net/p/minetest-c55/contributors (with size of individual commits taken into account, which that page does not do; eg. kahrl has tended to have very large squashed-together commits) 07:16 celeron55 squashing is good and shouldn't be discouraged by lessening the credit of those who do it 08:54 thexyz some guys are complaining about being able to replace chest with other node 09:45 proller ShadowNinja|Away, ok, but i remake scale in new dev mapgen 10:07 celeron55 launchpad's daily builds are now tracking minetest/minetest 10:08 celeron55 so they'll be again up to date from now on 12:17 proller looking at many errors like ACTION[ServerThread]: Server: Player with an empty name tried to connect from 12:18 proller maybe use system username + rand(1000) like http://stackoverflow.com/questions/11587426/get-current-username-in-c-on-windows 12:19 proller autofill username field ? 12:45 celeron55 Administrator1, Administrato2, Administrator347 12:50 celeron55 proller: the problem with that is that if it's pre-filled into the field automatically, you have to erase it to input your actual nick which sucks, and on the other hand if a such a nick is silently sent to the server if the field is blank you don't remember it the next time and it'll properly conflict with some other person on an another server 12:50 celeron55 s/properly/propably/ 12:52 proller generate it only in first time and save for next time 12:54 proller or not system user, but like guest12345 13:47 proller or maybe generate name on server, using hash from client ip 14:29 celeron55 i guess it could work so that if the user leaves the name empty when connecting to a server, the client will input guest+randint(1,1000) to it and leave that name to it for next times too as if the name was input by the user 14:29 celeron55 no os dependency, no effect on correct usage, no anything 17:14 hmmmm http://ompldr.org/vaHJiNg/2013-03-14-131345_1154x890_scrot.png yes? no? 17:14 hmmmm i'm going to make the listbox a little bigger 17:15 Calinou looks nice 17:15 Calinou YELLOW TEXT 17:15 Calinou OMERAGAD 17:15 Calinou (can we also have color codes which can be used by mods, puleeez?) 17:15 hmmmm huh 17:15 hmmmm well I did that manually by setItemColorOverride 17:16 hmmmm not all gui elements are guaranteed to have that capability 17:16 Calinou having color codes like ^f0f would be a nice thing, optionally, could be used by people talking too 17:16 hmmmm i had to make it yellow in lieu of being able to make it bold or change the font size 17:16 Calinou yellow looks good enough, no need for bold 17:16 hmmmm oh god, colored text, no we don't want a repeat of starcraft 17:17 hmmmm (well, colored text in chat anyway) 17:17 Calinou I said optionally ;) would probably be off by default 17:17 hmmmm colored text in chat adds so many loopholes, bugs, and what not 17:17 Calinou would be useful for mods 17:17 hmmmm in the end it's not worth it and games that do have it, end up removing it 17:24 hmmmm alright, so people, please do comment on the new credits box, i did adjust the y position of the listbox, so don't complain about that 17:27 thexyz hmmmm: change my email to xyz@minetest.net 17:27 hmmmm ok 17:35 rubenwardy hmmmm: very nice 17:36 rubenwardy I like the menu layout 17:42 hmmmm thanks 17:43 hmmmm that, along with removing "tape-and-glue", were just two steps to modernizing the main menu. next up is making a single, unified configuration menu that pops up in both the menu tab and in game (instead of having a single button in the game menu for sound volume, for instance - i find that tacky) 17:44 hmmmm also the layout should be revised too, instead of being just a bunch of checkboxes. if mutually exclusive options are present, for example, they should be so-called 'radio buttons', or option circles 17:46 rubenwardy yeah, definitely 17:52 jin_xi as much as i liked the tape and glue thing, the new menu looks fresh and nice! very cool 18:08 hmmmm oh by the way, i realized the exact reason why the dark spots happen on the map 18:09 hmmmm it's because when it's generating, the overtop does have a valid node that isn't content ignore, namely cave air. so it copies the light value of that node (which is initialized to 0) all the way down for that chunk 18:10 hmmmm i also think i know the reason why my lighting algorithm doesn't work right, and sadly i might have to do a lot more computations than i wanted 18:11 hmmmm but the reason that i suspect doesn't explain the unlit tree bases 18:48 PilzAdam hmmmm, +1 for the menu 18:53 rubenwardy nice clouds in menu 19:05 rubenwardy celeron55: http://forum.minetest.net/viewtopic.php?id=5106 19:06 rubenwardy and thexyz, PilzAdma 19:07 PilzAdam I like the background 19:07 PilzAdam thats all 19:08 rubenwardy It looks to similar to minecraft.net 19:08 rubenwardy *too 19:37 proller https://github.com/minetest/minetest/pull/512 20:11 hmmmm oh, proller, you need to use minetest.setting_getbool otherwise it'll execute if true or false 20:11 hmmmm i'll take care of that, though. 20:23 hmmmm also i'm gonna have to break up the lines a little 20:24 hmmmm realbadangel has this habit of making really long lines, like 120 characters, which is well over the 80 character limit, but you outdo everybody else with 235 column lines 20:34 PilzAdam hmmmm, https://github.com/minetest/minetest/commit/51ab87585a0e0db8ce67c79dfeeda6bc15d5d59e#L0R50 20:35 hmmmm hmmmm, 20:35 hmmmm oh crap i'm sorry 20:35 PilzAdam :-) 20:35 hmmmm it's... i can't help it 20:35 hmmmm my fingers just do pliz 20:36 Exio nyan cat floating in the air? 20:37 PilzAdam I guess you can --amend here 20:37 hmmmm yea 20:37 Exio i throught they were only underground 20:37 hmmmm they're not 20:38 hmmmm hrmmmm how am i going to do this.... perhaps i stash, reset soft, amend, push, reset, apply 20:41 PilzAdam Exio, http://forum.minetest.net/viewtopic.php?pid=40202#p40202 20:42 sapier looks like an earhquake happened ;-) 20:45 hmmmm crap 20:46 hmmmm >mfw i just realize i'm doing all this in branch liquid 20:55 celeron55 normal git usage 20:55 celeron55 pull upstream master - oh it was stable-0.3 20:58 sapier celeron what do you think about this, is there any (major) feature missing? https://github.com/minetest/minetest/pull/495 21:05 celeron55 i see some discussion about metatables 21:05 sapier yes metatables are a open issue with security 21:05 celeron55 manipulating them can't be enabled for sandboxed code 21:05 celeron55 there is no way of doing that securely 21:05 sapier yes but it can't be disabled for current minetest implementation 21:06 celeron55 how so? 21:06 sapier builtin uses raw access therefore it requires to access metatables too 21:07 celeron55 you can just grab setmetatable and getmetatable to local variables in there and hide them after loading builtin 21:07 celeron55 simple stuff 21:08 sapier I'll have a look at it again but I think it's not exactly that easy i think it would break serialize/desirialize too 21:08 celeron55 serialization doesn't care at all about any metatables 21:11 celeron55 stop thinking so complicatedly 8) 21:11 sapier I'll have a look at it it's some time since I authored this I'm not exactly sure why this wasn't as easy to replace as other things 21:12 sapier I tried to change as less as possible as there are strong objections against security ... at least this is my impression 21:12 celeron55 all objection against 1) security with obvious holes and 2) unnecessary restrictions in the name of "security" are valid 21:13 celeron55 if you allow metatables for mods, it's like you had no restrictions whatsoever 21:13 sapier there where objections against prohibiting calling arbitrary host applications .... if this is unnecessary ... ;-) 21:13 sapier all other objections are valid yes 21:14 sapier ok not all but some 21:14 hmmmm christ.... i just encountered two xorg problems back to back 21:15 sapier I know if there is any hole it can be exploited it's enough but that doesn't mean to make it as easy as io.execute("rm ~/*") 21:15 sapier the metatable problem needs to be fixed too but in between the other improovements are better than nothing 21:16 proller hmmmm, now better? https://github.com/minetest/minetest/pull/512/files 21:17 hmmmm t.t 21:17 hmmmm i did all of that stuff already 21:20 proller ups. thank you 8-) 21:21 hmmmm hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm 21:21 hmmmm to everybody: when i point out something with your pull request, if it's minor, i'll fix it on my own before committing, so don't bother 21:30 sapier celeron setmetatabe is used for all register functions therefore it doesn't seem to be easy to disable 21:30 proller hmmmm, about json for noise params: it allows to make unlimited number of params, without defining them with default values 21:32 proller like celeron55's idea about noise of noise - no problem to add 3-5 params for noise of scale, etc 21:32 sapier so unless register functions aren't replaced by minetest functions it can't be disabled 22:08 rark Hi, I'm having trouble with the buildscript. Can someone try to give me a little guidance? 22:09 rark I have a vanilla Ubuntu server installation with mingw, cmake, and assorted archiving utilities(such as p7zip-full) to extract the downloaded files. 22:10 rark I'm told by cmake that my compiler(installed through APT) failed to compile a simple test program. 22:11 rark When told it fails, it fails with Cmake reporting the compiler gave blank output. 22:12 rark The compiler in question is /usr/bin/i586-mingw32msvc-gcc 22:13 sapier apt-get install build-essential 22:13 sapier mingw is windows compiler you don't need it 22:14 sapier did you (re)run cmake prior calling make? 22:15 sapier btw this rool is for developers plz move to #minetest for questions like that 22:16 rark Sorry, I'll head right over.