03:22 <+SirCmpwn> which is especially interesting when you consider that I'm not sending any of those 03:22 < TkTech> MCP *used* to copy the names from our wiki 03:23 <+SpaceManiac> I mostly used the wiki names in Glowstone 03:25 <+SirCmpwn> does that look right? https://gist.github.com/SirCmpwn/daefd641e28ad097ec38 03:25 <+SirCmpwn> <- means client to server 03:26 <+SpaceManiac> I think I would send the chunks before the PlayerPositionAndLook 03:26 <+SpaceManiac> because that's the packet that makes the world appear 03:26 <+SpaceManiac> but it looks alright 03:28 <+SirCmpwn> doesn't help 03:28 <+SirCmpwn> still this stupid exception 03:29 < TkTech> Updating craft.ne? 03:29 <+SirCmpwn> yeah 03:29 <+SirCmpwn> https://gist.github.com/SirCmpwn/a34c0f885d40cd80d0b4 03:30 <+SirCmpwn> C.N.Client works, I just have to get the server working again 03:32 <+SpaceManiac> Have you tried doing it unencrypted? 03:32 <+SirCmpwn> yeah, no luck 03:33 <+SirCmpwn> I have the packets C.N.Client got, in order, I'll try those 03:34 <+SirCmpwn> no good 03:34 <+SirCmpwn> C.N.Client log: https://gist.github.com/SirCmpwn/a1a0680f4dbe865cdcff 03:35 <+SirCmpwn> this is interesting: when I stop after the JoinGamePacket, the client gives me a JSON exception 03:36 < TkTech> You dun' got a packet error 03:36 <+SirCmpwn> hmm, but the client sends me ClientSettings and PluginMessage before crashing 03:36 <+SirCmpwn> join game is packet ID 1, yes? 03:42 <+SirCmpwn> I dunno, this looks right to me: https://gist.github.com/SirCmpwn/f9cca0aa70fd0ebfa406 03:42 <+SirCmpwn> plus the client can read that packet just fine 03:42 <+SirCmpwn> C.N.Client can, that is 03:45 < Not-002> [fNbt] fragmer pushed 1 commit to master [+2/-0/±4] http://git.io/3cUpog 03:45 < Not-002> [fNbt] fragmer 960e655 - Continuing work on fNbt.Serialization API 03:56 <+SirCmpwn> I can't seem to get the client to output networking info to the console, after following instructions on the wiki 03:56 <+SirCmpwn> is anyone else able to? 03:57 <+SirCmpwn> aforementioned instructions: http://wiki.vg/Debugging 03:58 <+SirCmpwn> ah, nevermind, I got it 04:01 <+SirCmpwn> I wonder why the client thinks I'm sending it a plugin message before I send join game 04:01 <+SirCmpwn> I especially wonder when you consider that it seems to have read the join game packet correctly 04:04 <+SirCmpwn> I am so confused :( https://gist.github.com/SirCmpwn/ec714e50a7cea24153d0 04:04 <+SirCmpwn> fuck it, I'll come back to it later 04:06 <+SpaceManiac> The client sends a plugin message on join with MC|Brand vanilla 04:12 <+fragmer> TkTech: who's in charge of wiki.vg these days? 04:15 <+SirCmpwn> SpaceManiac: not to me 04:16 <+SirCmpwn> SpaceManiac: or maybe it didn't even get that far 04:18 <+SpaceManiac> yeah, might not get that far 04:18 <+SirCmpwn> it does send me SOME info before blabbering on about JSON bullshit 04:19 <+SpaceManiac> if you're sending chat messages they're json now 04:19 <+SirCmpwn> I am not sending chat messages 04:19 <+SirCmpwn> and they've been JSON for a few versions no 04:19 <+SirCmpwn> now* 04:19 <+SpaceManiac> dunno then 04:23 < umby24> SirCmpwn: @ The gist you posted, are you sending the packet ID before WritePacket gets called? 04:23 <+SirCmpwn> yes, but the relevant code is somewhat confusing 04:23 <+SirCmpwn> fyi, the same network stack *works* from a client perspective 04:23 < umby24> ok 04:31 <+SirCmpwn> there was a new packet added at some point that is not documented 04:31 <+SpaceManiac> opcode? 04:31 <+SirCmpwn> no clue 04:31 <+SirCmpwn> I sent a disconnect packet and the client thought it was a plugin message 04:31 <+SirCmpwn> I think 04:31 <+SpaceManiac> they're pretty close, maybe you have your opcodes off? 04:32 <+SpaceManiac> don't forget the opcodes are different per direction 04:32 <+SirCmpwn> nevermind, I was wrong 04:34 <+SirCmpwn> god damn 04:34 <+SirCmpwn> I even try to just send the client a disconnect packet and the little punk gives me that bka cannot be cast to fs error 04:35 <+SpaceManiac> oh 04:35 <+SpaceManiac> the disconnect packet's json too 04:35 <+SirCmpwn> I know 04:35 <+SirCmpwn> I copy+pasted one from a vanilla connection 04:35 <+SirCmpwn> I can disconnect before I send LoginSuccess, but not after 04:35 <+SpaceManiac> Hmm 04:35 <+SpaceManiac> Maybe it's your encryption 04:35 <+SirCmpwn> this connection is cleartext 04:36 <+SpaceManiac> Hmm. 04:36 <+SpaceManiac> Well the opcode for disconnect changes 04:36 <+SirCmpwn> I guess I can turn encryption back on 04:36 <+SirCmpwn> yeah, I know 04:36 <+SpaceManiac> Maybe you're sending the wrong one? 04:36 <+SirCmpwn> nope 04:36 <+SirCmpwn> I'm sending the right one 04:36 <+SpaceManiac> well what do the bytes look like? 04:37 <+SirCmpwn> no change, crypto or otherwise 04:37 <+SirCmpwn> I wish I had an easy way to dump that 04:37 <+SirCmpwn> encryption or no, the JSON error on JoinGame is also always the same 04:37 <+SirCmpwn> so my encryption is presumably being set up right 04:38 <+SpaceManiac> http://sockspy.sourceforge.net/sockspy.html 04:38 <+SpaceManiac> my preferred network debugging tool 04:45 <+SirCmpwn> wow, I'm a moron 04:45 <+SirCmpwn> the server was disconnecting the client after handling an exception 04:45 <+SirCmpwn> and the disconnect message from the exception handler was not valid JSON 04:45 < umby24> lol 04:47 <+SirCmpwn> woo logged into server 04:58 <+SirCmpwn> incoming 04:58 < Not-002> [Craft.Net] SirCmpwn pushed 41 commits to master [+6/-1/±136] http://git.io/hih_fw 04:58 < Not-002> [Craft.Net] SirCmpwn 71603df - Update Craft.Net.Server to 1.7.5 04:58 < Not-002> [Craft.Net] SirCmpwn 3a58391 - Fix online mode on C.N.Server 04:58 < Not-002> [Craft.Net] SirCmpwn 2e77d72 - Merge branch 'master' into 1.7.x 04:58 < Not-002> [Craft.Net] SirCmpwn pushed 1 commit to master [+0/-0/±1] http://git.io/WE1IGw 04:58 < Not-002> [Craft.Net] SirCmpwn 36f1618 - Remove some test code 04:59 < Not-002> [Craft.Net] SirCmpwn deleted branch 1.7.x 05:00 < Not-002> [Craft.Net] SirCmpwn pushed 1 commit to master [+0/-0/±1] http://git.io/bSd7rg 05:00 < Not-002> [Craft.Net] SirCmpwn 3572685 - Update latest supported version 07:26 < Not-002> [Glowstone] SpaceManiac pushed 3 commits [+2/-0/±10] http://git.io/51xnBA 07:26 < Not-002> [Glowstone] SpaceManiac 643595c - Vague start on shift-click stuff 07:26 < Not-002> [Glowstone] SpaceManiac 177d4fd - Implemented chat tab completion. 07:26 < Not-002> [Glowstone] SpaceManiac 2bbfbaf - Updated ANSI color support for new color codes. 08:01 < TkTech> fragmer: "in charge" would probably be me 08:01 < TkTech> Not that I ever do anything 08:11 <+fragmer> TkTech: I've had a couple reports of people being unable to register at all 08:12 <+fragmer> "Login error Incorrect or missing confirmation code." 08:12 <+fragmer> perhaps the captcha broke? Could you check please? 08:34 < TkTech> fragmer: No errors in the log and I just registered fine 08:34 < TkTech> I'm assuming they're somehow overlooking the first field 08:35 <+fragmer> alright, I'll pass the info on. Thanks for looking into it 10:51 < Not-002> [mc-erl] clonejo pushed 7 commits to master [+2/-3/±53] http://git.io/vK8hBQ 10:51 < Not-002> [mc-erl] clonejo 2b0d6e0 - extended .gitignore 10:51 < Not-002> [mc-erl] clonejo 69a8806 - License is now GPLv3, nbt.erl moved out as a dependency 10:51 < Not-002> [mc-erl] clonejo 4268684 - removed proxy 10:51 < Not-002> [mc-erl] ... and 4 more commits. 17:07 <+sadimusi> TkTech: most of the people getting the error on the wiki tried "Alexej" and "No" 17:07 <+sadimusi> TkTech: the questions might be a bit misleading 17:18 < TkTech> sadimusi: You're almost certainly right 17:18 < TkTech> sadimusi: Any recommendations? 17:18 < TkTech> (I'm not going to respond for a bit, about to head out) 19:10 <+AndrewPH> TkTech: "Markus Persson, also known as '______'" 19:22 < KnownUnown> :/ 19:22 < KnownUnown> dangit, why did I have IRCCloud set to highlight "known"? 20:36 < MrARM> TkTech cluck about MinecraftCoalition in the footer 20:37 < MrARM> *Click 20:37 < MrARM> says that there is no such page 20:38 < MrARM> No, says there is no content in this page 20:38 < MrARM> And this gives for all the pages here 20:39 < TkTech> Yup 20:39 < TkTech> No one ever wrote an About page. 20:41 < MrARM> And there is no privacy policy, right? 20:42 < TkTech> There is no Privacy on the internet. 20:42 < TkTech> (No, again, no one ever wrote one) 20:43 < MrARM> xD 20:44 < MrARM> A way to get all user accounts from the wiki: http://wiki.vg/index.php?title=Special%3ALog&type=newusers&user=&page=&year=&month=-1 20:44 < MrARM> names of couse 20:44 < MrARM> *course 20:44 < shoghicp> ¬¬ 20:44 < shoghicp> not a bug ;) 20:44 < shoghicp> oh 20:44 < shoghicp> most of those are spam accounts 20:44 < shoghicp> xD 20:46 < TkTech> MrARM: Sorry, are you trying to point something out? That is a normal page. 20:46 < MrARM> no 20:46 < MrARM> just were browsing it 20:48 < MrARM> I think, does it exist in wikipedia? 20:54 < TkTech> Yup 20:54 < TkTech> It's horrific on wikipedia 22:18 <+SpaceManiac> Hmm, I should consider redoing my world I/O, still using the original RegionFile class... 22:28 < shoghicp> well, I've to implement Regions in PocketMine :S --- Day changed mer. mars 05 2014 00:17 < Ellis> Hmm... Anyone know of an efficient way of handling server-sided lighting for a custom server I am writting in C++. I've implimented server sided lighting that works the same way as the client does.. But am struggling to think of a way to deal with dynamic light sources (e.g. torches) 00:19 < Ellis> I'm using a single 16*16*128 byte buffer for light values and have written code that handles all the skylight and even placing of torches. But I can't think how to remove the light values when a torch is broken. 04:24 <+SpaceManiac> I noticed the favicon on wiki.vg is broken 04:53 < Not-002> [Glowstone] SpaceManiac pushed 3 commits [+4/-1/±7] http://git.io/3elLDA 04:53 < Not-002> [Glowstone] SpaceManiac 142c3e8 - Added biome id conversion and I/O support. 04:53 < Not-002> [Glowstone] SpaceManiac 960eea7 - Added and load from recipes.yml file. 04:53 < Not-002> [Glowstone] SpaceManiac 9d28276 - Re-added tab list display. 06:18 < Not-002> [mc-erl] clonejo pushed 2 commits to master [+0/-0/±3] http://git.io/4SHqbg 06:18 < Not-002> [mc-erl] clonejo ee4c453 - reworked README 06:18 < Not-002> [mc-erl] clonejo fe6130d - eliminate compiler/dialyzer warnings 06:20 < Not-002> [Glowstone] SpaceManiac pushed 1 commit [+0/-0/±2] http://git.io/7wu09g 06:20 < Not-002> [Glowstone] SpaceManiac eaf20ef - Added help, version, configDir, and configFile command-line options (adds #88). 07:19 < Zack_> Hey, does anyone here know of a timber-like mod for 1.7.2? 07:19 < Zack_> I know there is Tree Capacitator, but it is not quite up-to-expectations. 07:19 < Zack_> Was hoping for a mod that if you break any wood log on the tree, it destroys any other logs on that tree. 07:20 <+clonejo> Zack_: This channel isn't about mods, but about standalone custom software. 07:21 < Zack_> I meant to say plugin, and was directly sent here from #minecraft :c 07:22 <+clonejo> o_O 07:22 <+clonejo> Well, we have nothing to do with Bukkit, MCP or Forge. 07:25 < Zack_> Thanks anyways- and thanks for the info c: 07:26 <+clonejo> no problem 15:24 < Ellis> Question about the new position data encoding.. How can negative coordinates be sent? 15:25 < Ellis> For example, take the coordinates (-63, 0, 0) 15:30 < Ellis> As a long, -63 = 0xFFFFFFFFFFFFFFC1. And that with 0x3FFFFFF: 0xFFFFFFFFFFFFFFC1 & 0x3FFFFFF = 0x3FFFFC1. Then shift it 38: 0x3FFFFC1 << 38 = 0xFFFFF04000000000. When we add the y and z, it still is 0xFFFFF04000000000. 15:31 < Ellis> But then decoding that: 0xFFFFF04000000000 >> 38 = 0x3FFFFC1 15:32 < Ellis> Whcih is 67108801 15:37 < Thinkofdeath> You get 0x3FFFFC1 because you are doing an unsigned shift instead of a signed shift http://ideone.com/5lUBez 15:37 < Thinkofdeath> Ellis: ^ 15:42 <+sadimusi> Ellis: if you're trying to encode the complete coordinate, shouldn't it be 12 bytes long? 15:46 < Ellis> Thinkofdeath: Thank you! Just need to try and get it wroking in C++ now, where right shift on signed values is undefined ;D x.x 15:46 < Ellis> sadimusi: The prerelease coordinate system squeezes all three values into a long 15:46 < Ellis> (8 bytes) 15:46 <+sadimusi> sounds fun :D 15:49 < Ellis> x: 3.25 bytes, y: 1.5 bytes, z: 3.25 bytes ... oh yeah... great "fun" ;P 15:49 <+sadimusi> only for block coordinates I hope 15:50 < Ellis> Yeah, any integer positions 15:51 <+sadimusi> if your languages shifting is quirky you can always di a &0x2FFFFFF afterwards 15:52 <+sadimusi> *do 15:52 <+sadimusi> and *3 16:00 < Ellis> Ugh... C++, why are do you do this to me... http://i.ellisvlad.com/y7bLU.png 16:00 <+Fador> ..but in real life it actually works just fine 16:02 < dx> "undefined behavior" is the spec directly telling you "YMMV, the compiler *might* do what you expect" 16:46 < Ellis> .... all this time... I thought it was my coding skill failing me... http://i.ellisvlad.com/Cm4y9.png FML 16:47 < dx> hah, yep, that's the kind of bug that can take you months to get fixed 17:12 <+clonejo> it's actually one the compiler could have told you about 17:12 < Flemmard`> unused parameter .. ? 17:12 <+clonejo> yup 17:13 < Flemmard`> yea but all doesnt do that 17:13 <+clonejo> ?? 17:15 < Flemmard`> all compilers, by default .. sure you can always enable all that :> 17:16 <+clonejo> it's default in Erlang :3 17:16 < Flemmard`> cool 17:16 < Flemmard`> lol 17:20 < Ellis> Idk where it is in Eclipse CDT :P 17:20 < Ellis> Fixed the tiny position mistype bug... now I have broken the whole lighting system :D http://i.ellisvlad.com/Hwgn2.png 17:24 < Flemmard`> if you based all your results on that failing position .. eh .. :P 17:38 < Ellis> Fixed lighting... http://i.ellisvlad.com/ywx3r.png I'm fixing so many huge bugs today :D --- Day changed jeu. mars 06 2014 03:31 < cathode> hi there, whats the best way to profile a running minecraft server to diagnose the cause of "can't keep up!" messages? 03:44 <+clonejo> cathode: Which server software are you using? 03:52 < SinZ> cant keep up is usually a vannila message 03:56 < cathode> Forge 03:56 < cathode> ~179 mods loaded 03:57 < cathode> i'm wondering if there's a way to tell like, which tile entity might be consuming lots of cpu time 03:57 < cathode> or mod 03:58 < cathode> there's either a bug, too many entities for the server to handle, or more likely there's an infinite loop of itemducts or something where things are going 'round in a circle 03:59 < cathode> i'm getting the message about once every 20 seconds in the server console 04:01 < SinZ> the message occurs when you go below 20 tps 04:02 < SinZ> try the mod Opis, its a really good profiler for mc forge instances 04:03 < SinZ> it can tell you what is causing the tickrate lag 04:03 < cathode> ok will do 04:03 < cathode> interestingly... 04:03 < cathode> the tickrate issue corresponds with disk writes... https://www.dropbox.com/s/kmdhzdk1sjyjbl4/Screenshot%202014-03-05%2019.03.16.png 04:04 < cathode> at least, the same frequency 04:06 < cathode> SinZ - dumb question but does Opis need to be installed on the client too? 04:07 < SinZ> doesn't need to, but has some advantages if you do, like visualising the info on the client, rather than dumping the info on a file in the server folder 04:08 < cathode> ok. 04:40 < cathode> ok now i gotta figure out how to use Opis... 04:40 < cathode> interestingly i found that it was already installed with the feed-the-beast monster modpack 04:46 < cathode> ok so while almost all other tileentities were using a handful of microseconds, the ME Interface tileentities were using 2800-3500 microseconds... 06:09 < Not-002> [mc-erl] clonejo pushed 3 commits to master [+1/-0/±6] http://git.io/o8KXCA 06:09 < Not-002> [mc-erl] clonejo ba92357 - moved inventory handling to separate module, fixing #17 (shift clicking) 06:09 < Not-002> [mc-erl] clonejo 2197bb5 - setup fixes 06:09 < Not-002> [mc-erl] clonejo a1d42ba - extended gitignore 06:20 <+clonejo> meh, I still have to port mc-erl to the 1.7 protocol 15:33 < Not-002> [mcprotocol] thinkofdeath pushed 1 commit to master [+0/-0/±46] http://git.io/44xA9Q 15:33 < Not-002> [mcprotocol] thinkofdeath 4a5e6ae - 14w10a --- Day changed ven. mars 07 2014 13:03 < shoghicp> MCMrARM: ping 13:43 < MCMrARM> shoghicp: Pong 13:43 < shoghicp> MCMrARM: the app is up again :D 13:43 < MCMrARM> :D 13:43 < MCMrARM> is the fake down? 13:43 < shoghicp> nope 13:44 < MCMrARM> :( 13:44 < MCMrARM> I'll put the sources today 13:44 < shoghicp> I'll send a DMCA takedown 13:44 < shoghicp> ok 13:48 < shoghicp> MCMrARM: ping brandon about the ARM PHP binary with the modified uclibc 13:49 < shoghicp> he is here xD 13:49 < MCMrARM> ik 13:49 < MCMrARM> no wait 14:36 < MCMrARM> Bradon can I have that arm php bin with modified uclib? 14:38 < MCMrARM> Brandon ↑ 17:11 < shoghicp> MrARM: "File extresion: php" 17:11 < shoghicp> ^ fix that 17:11 < MrARM> Typo or what? 17:11 < shoghicp> yes 17:12 < shoghicp> and show the version somewhere bigger 17:12 < MrARM> any more critic? xD 17:12 < shoghicp> not only on the extended description 17:12 < MrARM> for what? 17:12 < MrARM> It handles updating. .. 17:13 < MrARM> shoghicp how it should be? 17:13 < MrARM> ah, extension 17:13 < shoghicp> put the version at the right of the "Description" bar, in italics 17:15 < MrARM> But I can't. .. 17:15 < MrARM> Without breaking the whole layout 17:15 < shoghicp> xD 17:16 < MrARM> It's just a TextView, with margin and background in a linearlayout 17:16 < MrARM> it isnt like css 17:16 < shoghicp> ok 17:16 < MrARM> I may add version in the same field as date is 17:17 < shoghicp> and remember the field changes 17:17 < shoghicp> XX downloads 17:17 < shoghicp> instead of Downloads: XX 17:17 < MrARM> ... 17:17 < AlphaBlend> 1 downloads 17:17 < AlphaBlend> Downloads: 1 17:18 < MrARM> Yeah 17:18 < AlphaBlend> the latter looks better, so you don't even need to modify download to add an s to it 17:18 < shoghicp> AlphaBlend: well, on the design.. it looks strange to me 17:18 < AlphaBlend> it's the colon 17:18 < AlphaBlend> i'd think ;p 17:19 < MrARM> shoghicp it is like that in play store.... 17:19 < shoghicp> and all plugins get 2 downloads when published ;D 17:19 < MrARM> xD 17:19 < MrARM> I don't understand field changes... 17:20 < MrARM> What you meant 17:20 < MrARM> shoghicp ↑ 17:21 < shoghicp> ? 17:21 < MrARM> 1717 and remember the field changes 17:21 < MrARM> shoghicp ↑ 17:21 < shoghicp> downloads + current rating 17:22 < MrARM> Remember for what 17:22 < shoghicp> well, that is an small thing 17:22 < MrARM> For what I should remember it? And where? 17:22 < shoghicp> for nothing! 17:22 < shoghicp> just I told you a few weeks ago 17:22 < shoghicp> xD 17:23 < MrARM> I don't think I remember that xD 17:23 < MrARM> You mean it adds something or what? 17:24 < shoghicp> well, forget it 20:26 <+clonejo> hm, notifico didn't work 21:32 < TkTech> clonejo: I was restarting it due to a conversation in #notifico 21:32 <+clonejo> okay, np --- Log opened lun. mars 10 12:12:47 2014 12:12 -!- Irssi: #mcdevs: Total of 135 nicks [1 ops, 0 halfops, 11 voices, 123 normal] 12:16 -!- Irssi: Join to #mcdevs was synced in 257 secs --- Day changed mar. mars 11 2014 01:26 < erai> what is the new 1.7.5 login sequence --- Log closed mar. mars 11 01:43:39 2014 --- Log opened mar. mars 11 01:44:01 2014 01:44 -!- Irssi: #mcdevs: Total of 140 nicks [1 ops, 0 halfops, 11 voices, 128 normal] 01:48 -!- Irssi: Join to #mcdevs was synced in 263 secs 01:49 <+SpaceManiac> erai: I don't think it's changed 01:50 < erai> the information in the faq page on the wiki is out dated 01:50 < erai> I mean after I enable encryption, How do I get the client to spawn? 01:52 < erai> or has that not changed as well? 01:52 <+SpaceManiac> that's also mostly not changed 01:52 < erai> ok thanks 01:52 < erai> Should I update the wiki with the new sequence after I finish? 01:53 <+SpaceManiac> it'd probably be helpful 05:23 < nickelpro> The wiki already has the correct login sequence http://wiki.vg/Protocol_Encryption 05:33 < nickelpro> Updated the wiki with roughly the correct sequence, not sure exactly when the vanilla client confirms its position. 13:36 < TkTech> Thinkofdeath: thinkmap +1 13:36 < Thinkofdeath> :) Thanks 13:37 <+Prf_Jakob> thinkmap? 13:38 < Thinkofdeath> http://thinkofdeath.co.uk/mapviewer/ 13:38 < Thinkofdeath> Really needs a better name 13:38 <+Prf_Jakob> Thinkofdeath: impressive 13:39 < Thinkofdeath> Thanks :) 13:41 <+Prf_Jakob> Thinkofdeath: html5 and webgl? 13:41 < Thinkofdeath> yep 13:42 <+Prf_Jakob> nice really smooth 13:43 < Thinkofdeath> Yeah that was the hardest part, ended up going with a view distance of 4 to keep the fps stable 13:44 < Thinkofdeath> Hopefully I can improve on that in the future 13:45 <+Prf_Jakob> yeah the problem with ^3, going just one extra distance explodes in the number of chunks needed to render. 13:45 < Thinkofdeath> Yeah, some people tried it with a view distance of 10 and apparently didn't have issues 13:46 < Thinkofdeath> I'm on a intel card so that might explain my issues :P 13:51 <+Prf_Jakob> ugh intel windows drivers... 14:11 < TkTech> Prf_Jakob: You actually had to implement them didn't you? 14:12 < TkTech> I forget what the VMWare display pretends to be 14:12 <+Prf_Jakob> TkTech: We have our own driver so we don't pretend to be anything else. 14:13 <+Prf_Jakob> Just in general Intel drivers suck 14:16 < shoghicp> Thinkofdeath: is there a way to fly in your viewer? 14:17 < Thinkofdeath> shoghicp: Hold down the right mouse button 14:17 < shoghicp> :D 14:17 < shoghicp> oh, and found the non-textured stairs 14:17 < Thinkofdeath> :P Yeah still not finished yet 14:18 < shoghicp> but it's nice :D 14:18 < Thinkofdeath> Thanks :) 17:32 < Jeebiss> I know I am a little late, but holy wow Thinkofdeath 17:36 < TkTech> Thinkofdeath: What issues? 17:36 < TkTech> Thinkofdeath: Did you post a problem somewhere? I found your project when it scrolled by on notifico. 19:46 < Thinkofdeath> Jeebiss: Thanks 19:48 < Thinkofdeath> TkTech: It used to have framerate issues on my intel card, using my nvidia one I could get solid 60 fps 19:48 < Thinkofdeath> Runs fine now though 20:08 <+clonejo> Thinkofdeath: your map is great, but you were too lazy asking for the correct keycodes 20:09 <+clonejo> wasd controls only work on qwerty keyboards 20:09 <+clonejo> *works 20:09 < Thinkofdeath> clonejo: https://github.com/thinkofdeath/ThinkMap/issues :) 20:10 < Thinkofdeath> Feel free to add to my todo list 20:10 <+clonejo> :) 20:12 <+clonejo> Thinkofdeath: your code looks correct :/ 20:12 <+clonejo> are javascript keycodes broken?? 20:15 < Thinkofdeath> clonejo: try pressing wasd on this http://jsfiddle.net/D5xMw/ 20:16 < Thinkofdeath> Maybe your browser is giving them different keycodes for your layout 20:19 <+clonejo> Thinkofdeath: when pressing the keys that map to wasd, I get the codes that are in your viewer's compiled javascript 20:19 < Thinkofdeath> odd 20:19 <+clonejo> testing with chromium … 20:21 <+clonejo> same issue 20:23 < Thinkofdeath> clonejo: See if the controls work on this, http://media.tojicode.com/q3bsp/ 20:23 < Thinkofdeath> if they do then I've done something wrong somewhere 20:24 <+clonejo> chars printed on my keyboard with the returned keycodes using my (non-qwerty) layout: W->86, A->85, S->73, D->65 20:26 <+clonejo> when I switch the layout to qwertz, I get: different values :/ 20:27 < TkTech> Thinkofdeath: Ahhh, yeah I'm on nvidia as well. 20:30 < Thinkofdeath> clonejo: Short of giving changable keybindings I don't think there is much I can do to fix it my end 20:30 < Thinkofdeath> Seems like a browser bug 20:32 <+clonejo> Thinkofdeath: e.keyCode is supposed to give you the character, not the keycode 20:32 <+clonejo> *not the "scancode" 20:33 < Thinkofdeath> I thought that was e.charCode? 20:35 <+clonejo> it seems like it doesn't 20:35 < Thinkofdeath> Does e.which give different results? 20:37 <+clonejo> e.which and e.keycode are always the same 20:37 < Thinkofdeath> Well thats all the event gives me 20:38 <+clonejo> then the js api is broken 20:38 < Thinkofdeath> Pretty much 20:38 < Thinkofdeath> Not much I can do about that, can't even detect what keyboard layout users are using to try and work around it --- Log closed mar. mars 11 22:11:22 2014 --- Log opened mar. mars 11 22:11:30 2014 22:11 -!- Irssi: #mcdevs: Total of 140 nicks [1 ops, 0 halfops, 12 voices, 127 normal] 22:15 -!- Irssi: Join to #mcdevs was synced in 257 secs --- Day changed mer. mars 12 2014 22:27 < electromatter> hello. What is a reasonable maximum packet size to accept from the client? 22:34 < TkTech> electromatter: No real concrete answer to that question. I used to give clients a 2MB heap. 22:34 < TkTech> electromatter: The more important thing is ensure that you timeout properly on reads, way too many 3rd party libraries and apps open to trickle attacks. 22:35 < electromatter> Ok thanks 22:35 < electromatter> I was worried about limiting the buffer size too small and making it hard to use the notchian client 22:42 < TkTech> electromatter: Best thing to do is to try it :) It's pretty simple to make a buffer that starts at MIN_SIZE and can realloc as needed up to MAX_SIZE 22:43 < Jeebiss> derp, i read that as tickle attacks, what an adorable attack 22:48 < TkTech> Jeebiss: That's also actually a thing. 22:48 < TkTech> https://groups.google.com/forum/#!original/sci.crypt/VfZ6GE35C9E/WsxykzJK0VwJ --- Day changed jeu. mars 13 2014 01:09 < nickelpro> Do Minecraft usernames have a max length (in bytes)? 01:14 < SinZ> 16 characters unless it changed 01:15 < SinZ> so 32 bytes 01:31 < nickelpro> SinZ: 32 bytes even after the switch to UTF-8? 01:31 < SinZ> nickelpro: maybe not 01:32 < nickelpro> SinZ: Ya that's my problem, "16 characters" is parroted everywhere but what does that mean with a variable byte encoding? 01:33 < SinZ> all strings in the protocol have a length header of a short 01:33 < SinZ> so you should know how long it actually is 01:34 < nickelpro> SinZ: It's a varint now, but ya I was hoping to avoid having to figure out how to find memory for each name and just get away with a char[16] or something 01:35 <+sadimusi> the names can't contain non-ascii letter anyway iirc 01:35 <+sadimusi> *letters 01:36 < nickelpro> sadimusi: So max length is probably 16 bytes then, thanks 01:36 < dividuum> how can i log the https request authlib does? 16:48 < roblabla|RealOne> ohi 16:50 < roblabla> http://wiki.vg/Pre-release_protocol#Animation < Is this a mistake ? I guess it probably is 16:52 <+sadimusi> looks to me like 0x0a and 0x0b switched places 16:53 <+sadimusi> but there are only changes in the new 0x0b 22:41 < Thinkofdeath> This is kinda useful https://sessionserver.mojang.com/session/minecraft/profile/4566e69fc90748ee8d71d7ba5aa00d20 (UUID -> Username + Properties) Any suggestions on which page it should go on? 22:42 < shoghicp> Session? 22:44 < Thinkofdeath> The session server doesn't have its own page, its currently mixed in with the encryption 22:44 < shoghicp> hmm 22:45 < nickelpro> Thinkofdeath: It sorta does http://wiki.vg/Authentication 22:45 < nickelpro> nvm that's auth server 22:45 < shoghicp> Session redirects there ;) 22:53 < nickelpro> so ya then it does have a page --- Day changed ven. mars 14 2014 08:54 < Payton> How do I update a github repo with the newest files saved locally? 08:54 < Payton> On git shell btw' 09:01 <+clonejo> Payton: the basic procedure is: git add .; git commit; git push origin master 09:02 <+clonejo> you might have to leave out the "git" part in the git shell 09:02 < Payton> I think I'm looking for git push 09:02 < Payton> What should I replace origin with? 09:03 <+clonejo> origin is the name of the remote repository 09:03 < Payton> and master 09:03 <+clonejo> that's the branch name 09:03 < Payton> so I cd to the dir I want to upload? 09:03 <+clonejo> origin is the default name for the repository you cloned from, if you did that 09:03 <+clonejo> you always push the whole commit 09:04 <+clonejo> but you can choose which files to include in the commit, so you might create a commit with just the changes in one directory 09:04 < Payton> How do I upload a specific file? 09:04 < Payton> And replace the existing one 09:05 < nickelpro> git add /path/to/file 09:05 <+clonejo> change the file, git add , git push 09:05 < nickelpro> git commit -am "Git commit message here" 09:05 < nickelpro> git push 09:05 < nickelpro> Assuming everything is already set up 09:05 < Payton> so git add MineTurtle? 09:05 < Payton> Okay 09:06 < nickelpro> Actually, not -am 09:06 < nickelpro> just git commit -m 09:06 < Payton> git add ? 09:07 < nickelpro> no 09:07 < nickelpro> git add 09:07 < nickelpro> git commit -m "Message" 09:07 < nickelpro> git push 09:08 < nickelpro> Assuming your repo already has Github set up as origin 09:08 < Payton> How would I replace files in a repository with the newly updated ones? 09:08 < nickelpro> That's what push does 09:08 <+clonejo> git isn't just a better ftp 09:09 <+clonejo> it tracks your changes, grouped in commits 09:09 < Payton> How would I push to a specific repository, more specifically a orginization repo 09:09 <+clonejo> you add another remote repository 09:09 < nickelpro> You have to set up that repo as origin in your local git repo 09:09 <+clonejo> nickelpro: nope 09:10 <+clonejo> git remote add 09:10 < Payton> clonejo, How do I do that? 09:10 < Payton> okay 09:10 < nickelpro> clonejo: yes, that's what I said, was typing the command 09:10 <+clonejo> then git push 09:10 < Payton> branch = master? 09:11 <+clonejo> yup 09:11 <+clonejo> nickelpro: but the remote repo can have any name, not just origin 09:11 < Payton> Repository address has to be the url to the repo? 09:11 < nickelpro> clonejo: Semantics, origin is the traditional name for a singular remote repo 09:12 < nickelpro> and if you do git push -u the remote will automatically track that branch, you wont have to specify every time you push 09:14 <+clonejo> Payton: http://git-scm.com/book is a nice read 09:14 < nickelpro> Payton: yes, for github its usually git@github.com:/.git though there are other protocols 09:14 < nickelpro> ^ agreed 09:16 < Payton> Saying it's not a git repo 09:16 < nickelpro> Payton: did you create the repo on github? 09:16 < Payton> yup 09:16 < nickelpro> private or public? If public, link 09:17 < Payton> https://github.com/MCPEBukkit/MineTurtle 09:17 < nickelpro> You already have files here, so you've set up this repo before. What's the problem? 09:18 < Payton> I'm trying to update the files with the new locally saved files.. 09:18 < Payton> As I said before 09:18 < nickelpro> Repo says you've done that before to, 8 times already 09:18 < nickelpro> You likely already have everything set up, unless this is a new repo 09:18 < nickelpro> you can just git push 09:19 < Payton> How do I push a specific file though 09:19 < nickelpro> git add 09:19 < nickelpro> git commit -m "Message" 09:19 < nickelpro> git push 09:19 < Payton> that will make a empty file though right? 09:20 < nickelpro> If the file you added is empty, it will make an empty file 09:20 < nickelpro> if you add a file with code in it, it will add that 09:20 < Payton> I want to replace the repo with this new file.. 09:20 < nickelpro> if you add a video file, it will add that 09:21 < nickelpro> You want to add MineTurtle.iml right? 09:21 < Payton> no 09:21 < nickelpro> You want to add some new file then? 09:21 < Payton> I want to add all of the files, I've updated everything 09:21 < Payton> I want to replace the whole repo. 09:21 < nickelpro> git add --all . 09:21 < nickelpro> git commit -am "Message" 09:21 < nickelpro> git push 09:22 < Payton> How will --all specify What I want to add.. 09:22 < nickelpro> git add --all . 09:22 < nickelpro> the period is importantr 09:22 < nickelpro> and it will add everything in your current repo, update any files you updated, and remove any files you removed 09:23 < Payton> It still says not a git repo.. 09:23 < nickelpro> How did you get a copy of these files? By cloning the repo or by downloading the zip? 09:24 < Payton> No.. this is my project.. 09:24 < nickelpro> How have you made commits so far? How has the current code gotten into the repo? 09:25 < Payton> I don't even remember, I used github for windows though xD 09:26 < nickelpro> So use Github for Windows then, dont bother with git shell if you're not familar with it 09:26 < l4mRh4X0r> Don't bother using git if you're not familiar with git shell 09:26 * l4mRh4X0r flees 09:26 < Payton> I can't.. I'm not booted into windows 09:26 < nickelpro> Payton, then...? 09:26 < l4mRh4X0r> Payton: well, you're looking for a folder containing a .git folder at any rate. 09:26 < nickelpro> You said you were using git shell, that's a windows program 09:27 * l4mRh4X0r slaps nickelpro 09:28 < l4mRh4X0r> Git shell is an environment. Conveniently with a shortcut on windows (because windows sucks) 09:28 < nickelpro> Sorry I thought he meant Git Bash, the command line that comes bundled with Windows git 09:28 < Payton> she* 09:28 < l4mRh4X0r> that's just plain bash. 09:28 < nickelpro> My apologies 09:28 < l4mRh4X0r> :P 09:28 < l4mRh4X0r> Anyway, Payton, you *should* be looking for the folder that contains the MineTurtle.iml 09:29 < l4mRh4X0r> and that folder *should* have a .git folder in it (might have to do ls -A to be able to see that one) 09:29 < Payton> why would the folder that I just made have a .git folder in it.. 09:32 < l4mRh4X0r> oh 09:32 < l4mRh4X0r> wait what 09:32 < nickelpro> ^ 09:32 < l4mRh4X0r> Payton: okay, so you didn't git clone it? 09:40 < Payton> The cloned copy does not have a .git folder either. 09:41 < l4mRh4X0r> Payton: it does, but *nix hides folders and files starting with a dot by default 09:43 < Payton> Okay, it's there. 09:43 < Payton> Replace everything with new files? 09:43 < Payton> and then push? 09:44 < l4mRh4X0r> Well, make the changes you want to make (or copy them over) 09:44 < l4mRh4X0r> then git add -A; git commit; git push origin master 09:48 < Payton> saying "MineTurtle" isn't a repo. 09:48 < Payton> "could not read from remote repo" 09:49 < l4mRh4X0r> That's interesting. 09:49 < l4mRh4X0r> what's the output of git remove -v 09:49 < l4mRh4X0r> err 09:49 < l4mRh4X0r> git remote -v 09:51 < Payton> origin https://github.com/MCPEBukkit/MineTurtle.git fetch and push 09:52 < Payton> fixed, nevermind. 09:52 < Payton> wait hmm 09:53 < Payton> It said "Everything up to date" but nothing changed. 09:53 < Flemmard`> issues with git ? 09:54 < Payton> yup 09:54 < Flemmard`> what is the issue? 09:55 < Payton> When I push, it says "Everything up to date." but nothing changes. 09:55 < Flemmard`> have you commited your changes? 09:55 < Flemmard`> git log shows you your changes? 09:56 < Payton> no, it just gives me the message "Everything up to date." 09:56 < Flemmard`> git log 09:56 < Flemmard`> what shows it ? 09:56 < Flemmard`> git log doesnt show something like everything up to date 09:56 < Flemmard`> it should show you the commit log 09:57 < Payton> Wtf, it just now updates.. 09:57 < Payton> hmm 09:57 < Flemmard`> well if you have a real issue 09:57 < Flemmard`> clone in another folder 09:58 < Flemmard`> remove the .git folder from your actual folder 09:58 < Flemmard`> and cp -r * ../otherfolder/ 09:58 < Flemmard`> and redo the git add, git commit, git push 09:59 < Payton> okay 09:59 < Payton> Thanks 09:59 < Flemmard`> you're using git on windows ? 13:27 < natinusala> Hello, I came here because I saw this channel on the Protocol part of the wiki, the fact is that I would like to understand how the login process works in Minecraft 1.6.4. On the wiki, it's detailled for the 1.7.4 but I would like to have the same for 1.6.4, could someone help me ? Thanks :) 13:31 <+clonejo> natinusala: you can look at the edit history 13:32 < natinusala> i did this, but there is only the list of the packets, and not the details of the login process 13:32 < natinusala> http://wiki.vg/index.php?title=Protocol&oldid=4830 here is the latest revision of the 1.6.4 page 13:42 <+sadimusi> natinusala: are you looking for this? http://wiki.vg/index.php?title=Session&oldid=4157 13:43 < natinusala> the last part of the page will be very helpful, thank you :) 22:56 <+SirCmpwn> hey sadimusi, b.wiki.vg --- Day changed sam. mars 15 2014 03:05 < KnownUnown> logs speak wonders. 03:05 < KnownUnown> So, apparently, if a server isn't given permission to a world file on hand.... 03:05 < KnownUnown> the server refuses to load chunks correctly 03:05 < KnownUnown> you would think the server would actually show them 03:06 < KnownUnown> read access, but no wrtie 03:06 < KnownUnown> *write 09:30 < Not-003> [fCraft] fragmer * r2333 5 files : Added Schematic map exporter. 09:32 < Not-003> [fCraft] fragmer * r2334 4 files : Tweaked format settings, got rid of some leftover debug code, fixed a missing assembly reference in UpdateInstaller. 18:25 < winny> or you'd think the server would blow up 18:25 < winny> fail loudly and clearly 20:09 < KnownUnown> winny: that too --- Day changed dim. mars 16 2014 00:09 < Brott> dav1d: get tox 01:10 < dav1d> Brottweiler: ? 01:34 < Brottweiler> dav1d: you dont know tox.im? :( 01:34 < Brottweiler> dav1d: http://tox.im/ 06:22 <+SirCmpwn> tox was a joke last time I saw it 06:22 <+SirCmpwn> is it any better now 06:22 <+SirCmpwn> oh wow, that conversation was five hours ago 06:22 <+SirCmpwn> nevermind 06:22 <+SirCmpwn> you never notice until the timestamps are right next to each other, eh 07:47 <+clonejo> people do read backlog 11:42 < dav1d> Brottweiler: installing 11:51 < dav1d> o/join tox 11:51 < dav1d> whoops 12:03 < AndrewPH|Alt> http://i.imgur.com/RplC20w.png 12:03 < AndrewPH|Alt> wait wrong channel but you guys should enjoy that anyway 12:39 < Paprikachu> how does the json for a disconnect look like? 13:25 < Thinkofdeath> Paprikachu: Same format as chat 13:25 < Paprikachu> thank you :) 13:38 < Paprikachu> what's the "username" field in the login success packet? is it a display name? 13:39 < Thinkofdeath> Should be the same as the one that was sent to the server in Login Start (might get case corrected) 18:54 < Not-003> [node-minecraft-protocol] roblabla pushed 1 commit to master [+0/-0/±1] http://git.io/GtiwSg 18:54 < Not-003> [node-minecraft-protocol] roblabla 2d40fec - Fix #73 - this string should be unlimited 20:13 < Yogu> Hi there! Has anyone of you implemented or seen an implementation of gravity / jumping? I think I used the values from vanilla, but my implementation is not quite right, and some servers don't like it (send position-and-look packets when I jump) --- Day changed lun. mars 17 2014 21:38 < winny> is there some ways one can reduce mc network lag? 21:38 < winny> especialy due to tcp related overhead 21:54 < Aronnn> Hello, can anyone help me with packing a varint in javascript? 21:55 <+Fador> Aronnn: https://github.com/Mineserver/mineserver.js/blob/master/lib/mcbuffer.js#L48 this might work 21:55 <+Fador> winny: I used TCP_NODELAY for my server to get things less laggy =) 21:56 <+Fador> winny: http://en.wikipedia.org/wiki/Nagle's_algorithm 21:56 < Aronnn> Thanks, I'll take a look at it 22:13 < Yogu> Aronnn: there's a neat little package: https://github.com/chrisdickinson/varint/ 22:15 <+sadimusi> Yogu: is this a minecraft specific implementation? iirc the handling of negative numbers does not follow the usual specs 22:15 < Yogu> oh, I never encountered negative varints. Don't think it's minecraft-specific. 22:17 <+sadimusi> it looks like all the fields which can be negative were reverted to ints 22:17 <+sadimusi> so that shouldn't be a problem after all --- Day changed mar. mars 18 2014 01:21 <+SirCmpwn> off topic, but I know lots of you have reddit accounts 01:21 <+SirCmpwn> http://www.reddit.com/r/sircmpwn/comments/20ocsd/please_upvote_this/ 01:22 < shoghicp> SirCmpwn: upvoted! 01:29 <+SirCmpwn> thanks, shoghicp 01:29 < shoghicp> what is it for? a bot account, and you need karma to bypass the captcha? 01:29 <+SirCmpwn> correct 02:24 < dx> SirCmpwn: she looks like an upvote 03:19 <+SirCmpwn> dx: I suppose she does 06:15 < cathode> hi there. whats a good way to find a range of item IDs that are 'free' if i need to remap some for a mod? 06:17 <+clonejo> I think there are id allocation lists 06:18 < cathode> oh NEI can dump a list 06:18 < cathode> had no idea... 06:53 < winny> Fador: I wish notch would have not been stupid and used udp 06:53 < winny> :) 06:53 < winny> maybe there is a proxy that encapsulates the payloads as udp 06:57 <+Amaranth> A lot of things in the protocol are dependent on packets being sent in order though 06:57 <+Amaranth> So you'd need a general protocol redesign as well 06:58 <+Amaranth> Some things would always require things sent in order but right now pretty much everything except chat is dependent on packets showing up in more or less exactly the order they are sent 06:58 < winny> oh snap, that limits what can be done 06:59 <+Amaranth> Telling the client entity metadata for an entity it doesn't know about yet is no good, for example 06:59 <+Amaranth> And you probably shouldn't tell the client about the entity until you've told it about the chunk that entity is on 07:00 < winny> Fador: do you enable TCP_NODELAY system wide aka echo 1 > /proc/sys/net/ipv4/tcp_low_latency 07:00 < winny> or do you use iptables or something to enable it just for mc 07:00 <+Amaranth> It's a socket option 07:01 <+Amaranth> You should still batch packets in your server though 07:01 <+Amaranth> Difference is you get to control the batching 07:01 <+Fador> winny: https://github.com/fador/mineserver/blob/master/src/sockets.cpp#L315 =b 07:02 <+Fador> winny: I thought you were talking about custom server..=) 07:02 < winny> oh no i wan to speed up vanilla :D 07:03 <+Amaranth> Vanilla already uses TCP_NODELAY and custom batching 07:06 < winny> I see --- Day changed mer. mars 19 2014 04:45 <+SirCmpwn> does anyone have a machine-readable list of blocks and info like hardness, light, bounding boxes, texture coordinates, etc 04:45 <+SirCmpwn> burger stopped working at some point 17:29 < morfin> TCP_NODELAY? 17:30 < morfin> i readed about that 17:31 < morfin> is not that bad? you send tiny packets with headers(TCP and IP) that can be bigger than payload 17:36 < jast> it's a tradeoff. do you value immediacy more than efficiency? then TCP_NODELAY may be a reasonable choice. 17:39 < morfin> but that can be a problem for clients with not so good connedction 17:50 < jast> sure 17:52 < morfin> does vanilla disconnect after 1 notresponded keepalive? 17:52 < morfin> or there some limit for that? 17:55 < morfin> like did not respond 10 keep alives - dead 17:56 < shoghicp> well... it can even ignore the keepalives 17:57 < shoghicp> at least in my tests OP players could get to bypass that 17:57 < shoghicp> (long time ago) 17:57 < shoghicp> so, no packet received/sent at all, but it kept the TCP connection open 17:58 < morfin> :O 19:49 < kamcio96> how import other libraries in golang? 19:49 < kamcio96> eg: https://godoc.org/code.google.com/p/go-uuid/uuid 19:54 <+SirCmpwn> kamcio96: this is not the venue for your question 19:54 <+SirCmpwn> kamcio96: try #go-nuts 19:54 < kamcio96> i know ;) but im writing mc server in go 19:54 <+SirCmpwn> this is still not the venue for your question 19:54 <+SirCmpwn> general language questions go in #your-language-of-choice 19:54 <+SirCmpwn> minecraft questions go here 19:55 < kamcio96> so i dont understand it http://wiki.vg/Protocol#Map_Chunk_Bulk 19:56 <+SirCmpwn> here's some C# code that decodes that packet: https://github.com/SirCmpwn/Craft.Net/blob/master/source/Craft.Net.Client/Handlers/WorldHandlers.cs 19:56 <+SirCmpwn> the packet itself being read from the network here: https://github.com/SirCmpwn/Craft.Net/blob/master/source/Craft.Net.Networking/Packets.cs#L1797 --- Day changed jeu. mars 20 2014 20:47 < kamcio96> what json string i have to use to disconnect player? 20:47 < kamcio96> there: http://wiki.vg/Protocol#Disconnect_2 20:52 < Thinkofdeath> kamcio96: Its the same format as chat messages 20:52 < kamcio96> ty --- Day changed ven. mars 21 2014 06:31 < cathode> heyo. whats the reccomended development environment for java? eclipse? (on windows, i should clarify...) 06:32 < dx> i heard that's a matter of taste 06:32 < Eviltechie> I use eclipse and I'm fine with it. 06:32 < Eviltechie> Many people use intellij and love it. 06:32 < Eviltechie> For others, it's netbeans, which I beleive has some good gui building tools 06:33 < cathode> ok 06:33 < cathode> i'm not a java dev at all, but i want to hack on a mod so i just need something to compile it with 06:33 < cathode> i'm a 9-year C# vet, and java is similar enough it doesnt look difficult at all to pick up 06:34 < Eviltechie> Java and C# have a lot of similarities 06:35 < cathode> indeed 06:35 < cathode> my goal here (perhaps futile) is to try and improve the performance of RotaryCraft and ReactorCraft 06:36 < cathode> getting fed up with having hundreds of entities taking 50uS or more each, for no reason other than lazy coding 06:40 < cathode> this is not good. https://dl.dropboxusercontent.com/u/88948814/Minecraft/2014-03-19_23.59.45.png 06:41 < cathode> oops, 25uS+ 07:01 < cathode> ok so this source has no project file... 07:02 < Eviltechie> whatever ide you use, should be able to import source without a project file 07:03 < Eviltechie> generally speaking, you don't commit those anyway 07:03 < cathode> uh what? 07:03 < cathode> that makes no sense... how would you build without a project file if you're using CI or something? 07:04 < cathode> or you'd need an ant buildfile separately or whatever 07:04 < dx> there's maven and ant 07:04 < cathode> he didnt include buildfiles for either of those 07:05 < dx> so maybe he doesn't use a CI and just imports the thing into the IDE 07:06 < cathode> seems like terrible software development practice. i'm used to being able to git clone and build immediately... all repos should be set up for this no exceptions, in my opinion 07:07 < dx> you're working with minecraft client mods 07:07 < dx> don't expect beauty 07:08 < cathode> i expect developers to have some semblance of pride and/or common sense. but yeah, whatever 07:09 < dx> also, "git clone, import project, build immediately" doesn't sound bad at all 07:10 < cathode> maybe i don't understand the import project part. how would you keep track of which files are designated for building or whatever, without some project metadata in source control? 07:10 < dx> the average maven pom file doesn't specify what files are built 07:11 < dx> it just has some metadata and dependency info 07:11 < dx> the part about building stuff is handled automatically for all the valid java files in the tree 07:11 < dx> so you're not missing a lot 07:11 < cathode> so if you want to temporarily disable a .java file from building...? 07:12 < cathode> blah, i dont think i have the patience to deal with this bullshit right now. should probably look at this tomorrow 07:12 < dx> uh, i don't know, read the docs 07:12 < cathode> no wonder nobody is seriously using java anymore except for legacy crap 07:12 < cathode> because of this shit. can't use lincoln logs to build enterprise apps 07:12 < dx> wat... 07:14 < dx> protip: learning a new language requires not just to know the syntax rules, but also to change the way you think 07:15 < cathode> neither of those involves "not complaining about the shitty way things are structured" 07:15 < dx> for most people, disabling compilation of a single .java file is a non-issue. surely there must a way to do that but i've never needed to do that, hence "read the docs" 07:15 < cathode> or perhaps it's just eclipse that is junk 07:15 < dx> yes, i said "changing the way you think" 07:16 < dx> the build system is not shitty, but the way you think about it make you get to that conclusion 07:16 < dx> *makes 07:20 < dx> and i just fried my usb-ttl converter. i'm blaming java for this 07:20 < cathode> seems legit. 07:21 < Eviltechie> oh my god 07:21 < Eviltechie> this kid does not know what a npe is 07:21 * Eviltechie stab 07:22 < dx> i know what a NPE is and NPEs don't smell like this 17:16 < shoghicp> heh, shogchips joined here 17:16 < shoghicp> that shouldn't happen 17:20 < MrARM> shoghicp what is shogchips? 17:20 < shoghicp> a common typo of my name 17:21 < MrARM> rela 17:21 < MrARM> really? :P 17:22 < shoghicp> yes 17:42 <+fragmer> TkTech: I just got the email from tumblr "The Minecraft Coalition turned 4 today!" 17:42 <+fragmer> Has it really been that long? Damn 17:43 < shoghicp> :o 18:47 < TkTech> fragmer: Yeah, saw that too. Crazy right? 18:47 < TkTech> It's 2014. 18:47 < TkTech> Can't believe it. 19:25 < mpa1212> Is there a mod/option to allow the normal functionality when minecraft loses focus? 20:37 <+Amaranth> mpa1212: F3+P disables pausing when the window loses focus 20:40 < mpa1212> Amaranth: Thank! Exactly what I was looking for 23:28 < Not-001> [bravo] MostAwesomeDude pushed 2 commits to master [+0/-0/±2] http://git.io/6X35lA 23:28 < Not-001> [bravo] Corbin Simpson b8fb345 - beta/protocol: Remove infamous "Ignoring unlocated position" message. Leave a comment, but seriously, this message is not exceptional and no longer worth logging. 23:28 < Not-001> [bravo] Corbin Simpson d724cf8 - inventory/windows: Don't use ListContainer. Since we don't need recursion, just use a regular list. --- Day changed sam. mars 22 2014 11:06 < morfin> hmm 15:30 < morfin> hmmm 15:30 < morfin> first varint in protocol is always packet length 16:15 < morfin> and i forgot about that.. 22:06 < Brottweiler> dav1d: waaaaaa 23:58 < cathode> lol 23:58 < cathode> http://www.minecraftforum.net/topic/1969694-16xforgesmp-reikas-mods-tech-worldgen-civilization-and-more/page__st__6380#entry29914975 23:58 < cathode> this post... --- Day changed dim. mars 23 2014 00:00 < cathode> people have cleverly figured out ways to delete radioactive waste from reactorcraft by putting it in void chests or stuff like that. so that apparently needs to be 'fixed' by getting all the authors of those mods to write code specific for reactorcraft ... 00:01 < cathode> when that doesnt work, i dont think you can prevent an item from going into a container where the container is not from your mod, can you? 00:22 < benbaptist> Does Minecraft now send chat packets when left/right clicking on blocks? Or is this some sort of weird parsing issue? 00:23 < shoghicp> I think it is a weird parsing issue 00:24 < shoghicp> never heard of that 00:24 < benbaptist> yeah, it doesn't seem to happen all the time. But everything else parses normally, but it's weird. 00:25 < shoghicp> well, that wouldn't happen on the old protocol. not parsing something correctly? boom! out of sync 00:25 < shoghicp> hmm 00:25 < shoghicp> what kind of chat packet, btw? 00:25 < benbaptist> yeah, I like the new packet parsing system now. not everything goes crazy now :P 00:25 < benbaptist> http://benbaptist.com:82/notepad/view.php?id=er1j1i1bj7zde8irsinjn20z 00:25 < benbaptist> the 0x02 packet, I guess. 00:26 < benbaptist> To be honest, it's been a while since I've used the protocol (I last worked with it in the 1.6 days) and so I don't know if there's different types of chat packets now :P 00:26 < shoghicp> "x03x01" 00:27 < shoghicp> are those hex characters? 00:27 < benbaptist> yeah, my website apparently stripped the \ out of that text. :P 00:27 < shoghicp> is it vanilla minecraft? 00:27 < benbaptist> Yup, oh, and it's in the snapshot too. 00:27 < shoghicp> or is bukkit (with plugins) 00:27 < benbaptist> 14w11b 00:27 < shoghicp> hmm 00:28 < shoghicp> and that is sent from the server, right? 00:28 < benbaptist> Also, I probably do have some sort of parsing issue, because occasionally I get kicked from the server. (I'm writing a man in the middle between the client and server, and usually I'm just letting most client packets right through to the server) 00:28 < benbaptist> No, from the client. 00:29 < benbaptist> and it *seems* like it sometimes happens when I left/right click, but not always. 00:29 < shoghicp> do you receive "{'player': 'ohai', 'message': 'x03x01'} "? 00:29 < shoghicp> as json 00:30 < benbaptist> No, I make up that JSON object myself. The player being named 'ohai' is temporary until I can interpret the packet that sends the username as well. I probably should check and see if maybe I'm doing something weird. 00:31 < benbaptist> but it shouldn't be printing that ever unless it receives a chat packet. 00:31 < shoghicp> Raw packets D: 00:31 < shoghicp> Can you give me the raw packets? 00:32 < benbaptist> Sure! 00:33 < shoghicp> I work with raw packets faster xD 00:33 < shoghicp> I worked with the Minecraft PC protocol since b1.6 or so 00:33 < shoghicp> and left it on Final 1.5 or so 00:33 < benbaptist> Heh awesome! 00:34 < benbaptist> http://benbaptist.com:82/notepad/view.php?id=pnwi05k25g0jg15gvu9neu4yq 00:34 < shoghicp> now I'm focused on the Minecraft: PE protocol 00:34 < shoghicp> let's see 00:34 < benbaptist> I basically just printed every packet as .encode('hex') (I'm in Python) and you can see all the other console output as well. 00:34 < benbaptist> I held right click down and it made an even weirder chat packet appear with lots of stuff. 00:34 < shoghicp> 020301 00:34 < benbaptist> {'player': 'ohai', 'message': 'x05xc2x05x82xf6Bx0bt_x01'} 00:34 < shoghicp> that is not a proper chat packet 00:35 < shoghicp> in fact 00:35 < shoghicp> shouldn't a string be prefixed by the length? 00:35 < benbaptist> I believe that includes the length and ID too. 00:35 < shoghicp> hmm 00:35 < shoghicp> oh, right 00:35 < benbaptist> I 00:35 < shoghicp> the VarInt thing 00:36 < benbaptist> yeah, which I've hated a lot :P since it's not as easy as throwing it at struct.unpack. :P 00:37 < shoghicp> well, I think your parsing is not right 00:37 < benbaptist> There's a chance that maybe my packing of the varInts isn't good at times or something. I don't know. 00:37 < shoghicp> any other packet that you get? 00:37 < benbaptist> Nothing else, I'm just reading for chat packets. 00:37 < benbaptist> and since the server randomly kicks you after a while (sometimes instantly after connecting, or sometimes after a few minutes) I might be doing something weirdly. 00:37 < shoghicp> Bad packet id 6465 00:37 < shoghicp> yep 00:38 < benbaptist> yeah 00:38 < benbaptist> I just realized, when I'm packing bytes, are they supposed to be unsigned or signed? 00:39 < shoghicp> unsigned 00:39 < shoghicp> but depends on how you handle it 00:41 < benbaptist> this is my packing and unpacking code, I got it from someone else here. http://benbaptist.com:82/notepad/view.php?id=37o0ezuf2e0un5s1b45h3cfml&raw 00:42 < benbaptist> https://github.com/barneygale/authserver/blob/master/authserver.py#L59 00:44 < shoghicp> can you get other mitm proxy and dump the packets? 00:44 < shoghicp> or use the Minecraft client logger 00:45 < shoghicp> I think it includes one now 00:45 < shoghicp> so you see what it does send, and what you see 00:48 < benbaptist> oh yeah, I forgot about that 00:48 < benbaptist> yeah, let me try to enable that 00:59 < benbaptist> I have to go for now shogchips, I'll be back later. Sorry I couldn't get the better packet dump :P 01:01 < shoghicp> oh, shogchips is still in here 01:02 < shoghicp> done :) 23:01 <+SirCmpwn> https://github.com/SirCmpwn/Craft.Net/pull/197/files 23:02 <+SirCmpwn> I am at a loss for words 23:06 < Thinkofdeath> SirCmpwn: I think a few packets do that now, KeepAlive is int one way and varint the other in the snapshots 23:06 <+SirCmpwn> o_o 23:12 < shoghicp> SirCmpwn: the Minecraft protocol is worsening again... 23:12 <+SirCmpwn> they should just hire me to design the thing 23:12 < shoghicp> https://bugs.mojang.com/browse/MCPE-6865 23:13 < shoghicp> at least they are fixing this one :) 23:35 <+Amaranth> shoghicp: Good thing the mods seem to keep themselves to Minecraft and not MCPE otherwise that'd be closed as "we don't support mods or third party software" 23:35 < shoghicp> well 23:36 < shoghicp> MCPE has mods too 23:36 < shoghicp> they use that message to close lots of issues 23:36 < shoghicp> but PocketMine-MP is accepted as a half-vanilla server 23:37 < shoghicp> so, issues that have been tested using PocketMine-MP are valid, but not issues with PocketMine-MP 23:37 < shoghicp> https://bugs.mojang.com/browse/MCPE-6790 23:39 < shoghicp> also, https://bugs.mojang.com/browse/MCPE-6334 23:39 < shoghicp> other bugs are reported directly to Johan 23:45 <+Amaranth> I can understand issues caused by one half of the client/server being modded or third party being given lower priority but if you're crashing on invalid network input that's kind of scary 23:45 <+Amaranth> Crashing when loading a map should be a valid bad no matter how the map was made too 23:46 <+Amaranth> IO should be paranoid and resilient 23:47 < shoghicp> well, the network is fine 23:47 < shoghicp> just that it can cause problems 23:47 <+Amaranth> Oh, I was talking about how tickets for the desktop version are handled 23:47 <+Amaranth> https://bugs.mojang.com/browse/MC-46082 for example 23:48 <+Amaranth> Given an otherwise valid map, it loaded successfully, after all, you can crash the client or server with a map generation bug 23:48 <+Amaranth> But until someone could reproduce with the custom flat world options in vanilla it was closed 23:49 < shoghicp> :S 23:51 <+Amaranth> There was another one that was a bug in chunk packet serialization when you had an extended block id array that was all zeros which you can never generate in vanilla so I just worked around it 23:54 < shoghicp> well, in the MCPE case, bugs found using ModPE (equivalent of modding) are not accepted in the tracker 23:54 < shoghicp> And now I've the samme issue on my tracker 23:54 < shoghicp> lots of reports of plugins crashing things --- Day changed lun. mars 24 2014 00:11 <+Amaranth> shoghicp: Absolutely, if you're modifying the software you get to handle all the problems 00:12 <+Amaranth> That's not the same thing as talking to a non-standard server or loading a map made in mcedit 00:12 <+Amaranth> Again, lower priority is understandable but not ignoring 00:13 < shoghicp> well, the people that post the crash report with +15 plugins, with the title of "Fix it" and nothing else have the issue closed directly 00:13 < shoghicp> but you get people that have done their research ;D 00:14 < shoghicp> well, besides basic things, misconfiguration and such, we forward them to the support page that each plugin has in the forums 00:14 < shoghicp> at least the new tool that parses reports will do that automatically 00:14 < shoghicp> and give them a nice ID to paste instead of a long text 01:10 < erai> for strings sent over the wire, is the length field the number of utf8 chars or bytes? 01:10 < erai> wow I am stupid. sorry and thank you. --- Day changed mar. mars 25 2014 00:05 < benbaptist> Not sure if this is an appropriate question here, but any idea how I might be able to hide console/in-game spam from commands run from the server console? (I wrote a wrapper that runs commands and it spams ops and the server logs) 00:05 < benbaptist> I might just filter out the [Server: blah] chat packets manually if I have to, but I dunno. 00:06 <+sadimusi> is your wrapper also a proxy? and is it wrapping vanilla minecraft? 00:10 <+sadimusi> benbaptist: ^ 00:10 < benbaptist> Not yet - but I will eventually make it a proxy soon. And yeah, vanilla Minecraft. 00:10 < benbaptist> I'm doing pretty bad on the proxy side, though. I'm not really experienced with the new 1.7 protocol just yet, so my code organization is a bit weird. 00:10 <+sadimusi> I don't think there's another option than manual filtering then 00:11 < benbaptist> yeah, I figured so. that's not too bad, because I want to do it anyway to make advanced bukkit-like plugins for vanilla servers 00:14 <+sadimusi> I was maintaining a wrapper/proxy for a few years and found the limitations for such advanced plugins very tiring 00:15 < benbaptist> yeah 17:19 < MrARM> Hello 17:19 < ellisvlad> Hi 17:19 < MrARM> Any channel where I can get help with tcp packet loss? 17:19 < MrARM> on my own project? 17:20 < MrARM> I am not sure is it packet loss or issue at my side though 17:20 < ellisvlad> Is the loss through a network or your code? 17:21 < MrARM> At localhost 17:21 < MrARM> In my code, yes 17:21 < shoghicp> ^ 17:22 < ellisvlad> Try using RawCap to capture localhost traffic, and then check in WireShark to see if anything is missing. 17:22 < MrARM> I tried using relative position update in my game - it sends 2 bytes every 0.02 second 17:23 < MrARM> I am writing the game for Android, so i cant use wireshark 17:23 < shoghicp> tcpdump, then 17:23 < MrARM> hmmm 17:23 < ellisvlad> A common problem is people miss the fact that packets may come in bulk. Aka, more than one packet comes at once. Make sure that you are checking if there is any additional data on the end of the first packet and parsing it like a second packet. 17:24 < MrARM> Hmm 17:24 < shoghicp> MrARM: or you can enable an option in the TCP socket to send the packets directly without waiting 17:24 < shoghicp> I don't remember how it was called 17:24 < ellisvlad> ^ non-blocking 17:24 < MrARM> shoghicp I am doing it in queue 17:25 < shoghicp> ellisvlad: no, a option 17:25 < MrARM> And I am doing the position update by doing lastSentX-newX 17:25 < MrARM> but let's check it 17:25 < MrARM> idk, this can be my issue too 17:25 < shoghicp> TCP_NODELAY to 1 17:25 < MrARM> shoghicp 17:26 < MrARM> It is easier in Java 17:26 < shoghicp> well, I know the C flag 17:26 < ellisvlad> Nagle 17:26 < shoghicp> yep 17:26 < shoghicp> non-blocking only returns directly, on read/write 17:26 < shoghicp> it won't send the packet itself 17:28 < MrARM> I am using Java 17:28 < MrARM> I can pastebin the code for sending and receiving it of 17:28 < MrARM> if needed 17:28 < MrARM> It's called setTcpNoDelay 17:28 < MrARM> btw I am not using nio 17:28 < MrARM> so i do direct calls to readXXX 17:30 < MrARM> SetTcpNoDelay = still the problem exists 17:30 < MrARM> The character is having an offset 17:31 < MrARM> It is important, I'm running it on Android 17:32 < MrARM> hmmm 17:32 < MrARM> I just logged at my logcat 17:32 < MrARM> Looks like issue with my code: 17:33 < MrARM> Move: 0,0 17:33 < MrARM> Move: 0,1 17:33 < MrARM> Received: 0,0 17:33 < MrARM> *0,1 17:33 < shoghicp> well, fix it :P 17:33 < MrARM> trying to do so :P 17:34 < MrARM> However, I had the same issue, earlier, before I rewrote some IO stuff 17:35 < MrARM> (its why I rewrote it) 17:37 < MrARM> wait 17:37 < MrARM> I know why it is 0 0 17:38 < MrARM> hmmm. 17:38 < MrARM> but it is non blocking 17:41 < MrARM> nah 17:41 < MrARM> I think I saw the logcat from previous attempt 17:42 < MrARM> I have to do some testing on it 17:46 < MrARM> http://pastebin.com/edY1YbNK 17:46 < MrARM> I don't see anything bad here 17:46 < ellisvlad> http://i.ellisvlad.com/35T7s.png lol 17:47 < MrARM> Ellis host not defined 17:47 < MrARM> When I try to open it 17:48 < ellisvlad> Well, it says This paste has been removed! 17:48 < MrARM> hmmm 17:48 < MrARM> http://pastebin.com/edV1YbNK 17:48 < MrARM> Ok 17:48 < MrARM> I wrote a letter wrong 17:48 < ellisvlad> yup 17:49 < ellisvlad> I'd be very careful when using threads like that, but otherwise no problems there that I can see? 17:51 < ellisvlad> With non blocking sockets, do not attempt to write to them from multiple threads at the same time.. Otherwise you could corrupt packets. Use Mutexs or do your SocketIO in one thread 17:52 < MrARM> ellis these get into a one synchronized call at the end 17:53 < MrARM> so no problem here (: 17:53 < ellisvlad> Double check that, wherever it is :P 17:53 < MrARM> ellis no, my tcp connection doesn't get corrupt 17:54 < MrARM> It is very sensible at this point 17:54 < MrARM> If we get at least one byte wrong we are out of sync 17:54 < shoghicp> :P 17:54 < shoghicp> make a better protocol 17:55 < MrARM> shoghicp how better? 17:55 < shoghicp> one that accepts corruption without getting out of sync 17:55 < MrARM> If we get one byte wrong in any protocol we are out of sync most times 17:55 < shoghicp> no 17:56 < MrARM> Ok, so what does one byte wrong mean - if it is, let's say player x 17:56 < MrARM> the connection won't close, but the server will literally bug out, but still will be hearing for next packets 17:56 < MrARM> I wonder where the player will be in such case xD 17:57 < shoghicp> have a way to get them in sync again ;) 17:57 < MrARM> shoghicp I'm gonna send the full pos every second 17:59 < MrARM> Ok, I see the packet loss here: 17:59 < MrARM> 0 -1