12:33 <+Thinkofdeath> the first packet after a handshake for status is http://wiki.vg/Protocol#Request with an id of 0x00 and no fields
12:34 < Emile_> so the req is { 0 }
12:34 < Emile_> in raw array
12:34 < Emile_> and its not { 0 , 0 }
12:34 <+Thinkofdeath> er yes
12:38 < Emile_> hmm this is real odd
12:38 < Emile_> im getting the hand shake
12:38 < Emile_> then i just get a byte[] thats lenght is 0
12:38 < Emile_> and the it asks for old ping stuff
12:43 < Emile_> here is a raw dump of every packet i get
12:43 < Emile_> http://imgur.com/5Soc90S
12:46 < Emile_> The reson im developing this server is becuse, i wrote an comprestion algorithem 2 weeks ago and i realized i can make it work with minecraft chunck loading wiht a bit of pageing i can get the ram usage down to 50 mb for chunck loading
12:46 < Emile_> and if the server is in c#
12:46 < Emile_> i can use the windows iot core
12:46 < Emile_> and compile the project for rasber pi 2
12:47 < Emile_> this way the server can run as if it was an a fast vps on a simple small pi
13:21 < Not-f7bb> [SpockBot] nickelpro pushed 1 commit to master [+0/-0/±1] http://git.io/vZZqc
13:21 < Not-f7bb> [SpockBot] nickelpro dd60038 - Fix jumping and falling
16:07 < Not-f7bb> [SpockBot] gamingrobot pushed 1 commit to master [+0/-0/±1] http://git.io/vZnkT
16:07 < Not-f7bb> [SpockBot] gamingrobot 76b8127 - Update CONTRIBUTING.rst
16:22 < Not-f7bb> [SpockBot] gamingrobot pushed 2 commits to master [+0/-0/±2] http://git.io/vZnY2
16:22 < Not-f7bb> [SpockBot] gamingrobot 1e77fe1 - Update example to use vector methods
16:22 < Not-f7bb> [SpockBot] gamingrobot 9fab1eb - Merge pull request #121 from gamingrobot/update-example
18:29 < Emile_> Hello need some help, http://wiki.vg/SMP_Map_Format#Implementations
18:30 < Emile_> im trying to do 0x26
18:30 < Emile_> i need to wright the chang array
18:30 < Emile_> chunck
18:30 < Emile_> the "Chunk Data	Array of Chunk"
18:30 < Emile_> part
18:31 < Emile_> but i dont fully get the Chunk layout in the final bytes that get sent
19:06 < Not-f7bb> [SpockBot] nickelpro pushed 1 commit to master [+0/-0/±1] http://git.io/vZcCl
19:06 < Not-f7bb> [SpockBot] nickelpro e22b807 - Clarify naming to make implementation clearer
19:19 < morfin> hm
19:27 < Emile_> ?
19:30 < Emile_> o i think i get it
19:41 < Emile_> never mind still stuck
21:55 < Doxin> so I'm trying to convert an int to a VarInt, but I'm doing something wrong. can someone have a look at it: http://paste.pound-python.org/show/HEKKKuTXsa4YrKoZgens/ ?
22:25 < Doxin> so according to http://wiki.vg/Server_List_Ping once my server replies to the handshake, the client should send a ping packet. what I'm seeing instead is nothing, and then after a bit the client closes the connection.
22:25 < Doxin> am I misinterpreting something here?
22:37 < Fenhl> Doxin: client sends Handshake, then immediately follows up with Request
22:37 < Doxin> yes,
22:37 < Fenhl> server should first send Response
22:37 < Doxin> yes
22:37 < Fenhl> then client will send Ping
22:38 < Fenhl> at which point is something unexpected happening for you?
22:38 < Doxin> the client doesnt seem to send ping
22:38 < Doxin> but I'm sure I'm just parsing stuff wrong
22:42 < Doxin> yeah, comparing the packet captures it does /send/ the ping, I'm just not receiving it somehow
23:40 < Not-2be> [Charge] Wallbraker pushed 1 commit to master [+1/-0/±0] http://git.io/vZCde
23:40 < Not-2be> [Charge] Wallbraker 78fae80 - charge: Add Makefile for windows
--- Day changed jeu. sept. 10 2015
00:20 <+Thinkofdeath> o.O
00:27 < dx> dav1d: you're in that organization. i'm blaming you for that ^
00:28 < dx> actually, so is Prf_Jakob
00:28 < dx> yeah, Wallbraker == Prf_Jakob
00:28 < dx> nick mappings are hard
00:52 <+Prf_Jakob> Thats my bad.
00:56 < Doxin> so I'm getting a packet [0, 5, 68, 111, 120, 105, 110] , but I'm having trouble getting it parsed. seems like it'd be a 0x00 Handshake, but then the protocol version is 5, and there's a string of length 68 that can't be. what gives?
01:18 < dx> dav1d: are you a client or a server? what state are you in?
01:19 < dx> ...
01:19 < dx> i suck at irc
01:19 < dx> dav1d: nevermind
01:19 < dx> Doxin: are you a client or a server? what state are you in?
01:37 < Doxin> dx: I'm a server, I just switched to login
01:39 < dx> Doxin: if you're in login, 0 doesn't mean handshake, it means login start
01:39 < dx> 5 is a string length, the other 5 bytes are Doxin
01:39 < Doxin> oh
01:39 < Doxin> OH
01:39 < Doxin> oops
02:55 < Not-f7bb> [minecraft-data] rom1504 pushed 5 commits to 1.8 [+0/-0/±6] http://git.io/vZWVe
02:55 < Not-f7bb> [minecraft-data] roblabla 6118cd1 - Add bitfields, use it to define position and entityMetadata
02:55 < Not-f7bb> [minecraft-data] roblabla b260d7d - Protocol schema, implemented bitfield, anon struct, entityMetadata
02:55 < Not-f7bb> [minecraft-data] rom1504 05bbc0d - simplify the schema a bit by using oneOf instead of not and anyOf
02:55 < Not-f7bb> [minecraft-data] ... and 2 more commits.
03:21 < gamingrobot> Hi server I'm client
03:21 <+ammar2> :getout:
03:21 < gamingrobot> hehehe
03:31 < gurun> i don't feel comfortable opening an MC client .. and it starts a server, configure portfowarding in my router .. and let the world in.
03:35 < gamingrobot> gurun: disable upnp then?
03:36 < gamingrobot> wait does minecraft even deal with upnp?
03:36 < gurun> W10 seems to do it 
03:36 < gurun> i'm still trying to figure this out.
03:37 < gurun> i just started the client .. because i was testing the server (MiNET) .. and people started to poor in.
03:37 < gurun> so i have a new record .. full server on W10 :-)
03:37 < gurun> 5 players 40% CPU ... lol
03:41 < Not-f7bb> [SpockBot] gamingrobot pushed 1 commit to master [+0/-0/±1] http://git.io/vZWPf
03:41 < Not-f7bb> [SpockBot] gamingrobot 3dc9808 - move -> movement fixed
03:54 < Doxin> in 0x00 Encryption Request, what is the verify token? it's not mentioned anywhere else.
03:56 <+Amaranth> gurun: It's really not all that efficient compared to the desktop version
03:56 <+Amaranth> They just turned things down/off and have a much better renderer
03:57 < gurun> it is pretty ok really
03:57 < Doxin> the snapshots have a way more efficient renderer than latest stable too
03:57 < Doxin> it's crazy
03:58 <+Amaranth> gurun: PE is apparently 1.2 ported to C++ then with a new renderer and things turned off
03:58 <+Amaranth> So it probably still has the same issues with mob AI performance
03:58 < gurun> things turned off?
03:58 <+Amaranth> Yeah like the nether, caves, etc (not anymore but you get the idea)
03:59 <+Amaranth> And they don't tick as large of a radius
03:59 < gurun> is that fact?
03:59 < gurun> it's really not the impression i have
03:59 < jython234[away]> gurun: MCW10 upnp!
03:59 < jython234[away]> thats scary!
04:00 <+Amaranth> gurun: Didn't sound right to me either but it's what I was told
04:00 < gurun> yeah .. well .. it just looks like it. Just can't really really know for sure.
04:00 < jython234[away]> thats scary
04:01 < jython234[away]> anyway i'm still confused with the new "network rewrite" for PE
04:24 < Doxin> in 0x00 Encryption Request, what is the verify token? it's not mentioned anywhere else.
05:17 < gamingrobot> Doxin: uh 0x01 is Encryption Request
05:17 < Doxin> oh
05:17 < Doxin> duh
05:18 < Doxin> that's what I ment
05:18 < Doxin> gamingrobot: still, what's the token for?
05:19 < Doxin> the client isn't sending me the encryption response if it's wrong it seems
05:19 < gamingrobot> I think its accessToken but double checking now
05:26 < gamingrobot> nvm
05:27 < gamingrobot> Doxin: it just seems to be the verify_token sent by the server encrypted with the public key https://github.com/SpockBotMC/SpockBot/blob/master/spock/plugins/core/auth.py#L145
05:27 < Doxin> gamingrobot: I'm writing the server. what should I set verify token to?
05:31 < gamingrobot> it looks like it can be just random bytes as long as you store it and verify when you receive encryption response
05:32 < Doxin> right
05:32 < Doxin> then there's some other reason I'm not getting a response :D
05:35 < gamingrobot> Doxin: its a random 4 bytes
05:35 < Doxin> let's try "poop"
05:35 < gamingrobot> well it should be random for every person
05:35 < Doxin> well sure
05:35 < Doxin> for now this'll do
05:35 < Doxin> still no responser
05:35 < Doxin> *response
05:36 < gamingrobot> and your sending a valid public key?
05:36 < Doxin> I sure hope so
05:36 < Doxin> might not be
05:36 < Doxin> the packet I send looks like this http://paste.pound-python.org/show/UKQopLkx2sz3Pyb3AXUD/
05:36 < Doxin> the RSA library I use is a bit undocumented
05:37 < Doxin> or at least: I couldn't find the relevant docs
05:38 < Doxin> keypair generation looks like this http://paste.pound-python.org/show/SkwAoJKZqGHo6LLi2eo1/
05:42 < gamingrobot> what language is this?
05:44 < gamingrobot> oh its c++ the writeln threw me off
05:44 < Doxin> it's very much not c++
05:44 < Doxin> it's in D
05:44 < Doxin> one of the pubkeys it generates is (hex encoded) MIGJAoGBAPCNpoi6UnKj16M4xMe9QsF8vAAPq3lVFCnYkAWKAtzIwtLMItrWyM4fMK746LIBrJ9khQ4kfuf0vsU9rTpu0k7Z4n7/hXGrNLBItwBSrIwnXTyRbYnQVf+O2rZ6m4j5X5QNGB0EB7P7thKTf+5xAsNtXmUnwPugpRSAwfH3E+nVAgMBAAE=
05:44 < Doxin> no idea if that's valid though
05:46 < gamingrobot> darn I knew it looked weird
05:46 < Doxin> :P
05:47 < Doxin> gamingrobot: so any clue where I cocked up?
05:53 < gamingrobot> so you send the request but never get a response?
05:55 < Doxin> yep
05:55 < Doxin> it just times out
05:59 < Doxin> gamingrobot: the log looks like this: http://paste.pound-python.org/show/deNVzfnph2MbfYMkTiXC/
06:02 < gamingrobot> Doxin: if I am reading your output correctly it seems you have a packet id as the first byte for the receiving packets but not outgoing
06:02 < gamingrobot> its been awhile since I did stuff with network code
06:02 < gamingrobot> minecraft network code*
06:02 < Doxin> gamingrobot: well the stuff breaks a bit oddly
06:03 < Doxin> gamingrobot: all packets come prefixed with length, but on incomming it first reads one byte, before reading the rest
06:03 < Doxin> >>> [15] >>> [0, 47, 9, 108, 111, 99, 97, 108, 104, 111, 115, 116, 99, 221, 2] is one packet
06:04 < gamingrobot> yea
06:04 < gamingrobot> I got that, I ment for outgoing
06:04 < Doxin> outgoing is the same, but it prints it in one chunk
06:04 < Doxin> so it goes [129, 21, 1, 0, 129...., where the first two bytes are the length
06:05 < gamingrobot> ah ok now I see how its layed out
06:06 < gamingrobot> Doxin: do you want me to connect with spock and the echo packet plugin enabled to see what breaks / what it tries to decode?
06:06 < Doxin> that'd be good, let me throw open a port
06:07 < gamingrobot> yea and you can pm me the address
06:07 < Doxin> my address isn't exactly secret :P
06:07 < Doxin> gimme a sec
06:08 < Doxin> gamingrobot: should be up on superdoxin.com:25565
06:09 < Doxin> gamingrobot: I got the login packet and it replied
06:10 < gamingrobot> I never got the response
06:10 < Doxin> really?
06:10 < Doxin> huh
06:10 < Doxin> lemme try explicitly flushing the socket
06:11 < gamingrobot> yea I sent login start and then spock didnt decode anything else
06:11 < Doxin> flushing doesn't change anything
06:11 < Doxin> I am most *definitely* writing stuff here :P
06:12 < Doxin> time to break out wireshark
06:14 < Doxin> gamingrobot: wireshark tells me it does send stuff. can you try it again?
06:14 < gamingrobot> yea lemee do some more debug info
06:18 < Doxin> gamingrobot: 200% sure I sent a response there :P
06:18 < gamingrobot> Doxin: https://gist.github.com/gamingrobot/a67c87e33b0f2693c332
06:18 < gamingrobot> yea I got it
06:19 < Doxin> okay so how am I formatting my packet wrong
06:21 < gamingrobot> Doxin: updated gist with what a spigot server spits at me
06:22 < Doxin> I can't really compare them is the thing, the pubkey is different, the token is different
06:22 < Doxin> I'd have done that otherwise
06:22 < gamingrobot> refresh again I added the public key
06:22 < gamingrobot> so you can eliminate that
06:28 < Doxin> what the heck
06:28 < Doxin> gamingrobot: I can't find the pubkey anywhere
06:28 < Doxin> as in
06:28 < Doxin> I convert the hex to a c literal
06:28 < Doxin> and it's just not in there
06:28 < Doxin> the base64 into a literal
06:30 < Doxin> oh the converter is being funny
06:30 < Doxin> ech
06:30 < Doxin> well then I dunno how to eliminate it
06:33 < Doxin> gamingrobot: I cannot get this figured out/
06:36 < Doxin> gamingrobot: I need help ;_;
06:36 < gamingrobot> ah I see
06:36 < gamingrobot> I get a buffer underflow exception. 
06:36 < Doxin> oh
06:36 < Doxin> is a count messed up somewhere?
06:36 < gamingrobot> that means you didnt send as many bytes as you have packet length
06:37 < Doxin> let me actually try and get my own decoder ran over it
06:38 < Doxin> ah!
06:38 < Doxin> gamingrobot: I get underflow too
06:40 < Doxin> which is odd, considering the parser and serializer are generated from the same data
06:41 < Doxin> oh god I bet I know what's going on here
06:42 < Doxin> actually no I dont
06:42 < Doxin> gamingrobot: so I suspect writeBytes is broken, but I don't see it. mind having a look: http://paste.pound-python.org/show/GDjwEpkmNQOGLxLLfpfM/ ?
06:46 < Doxin> I'm just stumped
06:49 < Doxin> > parsing string of length 2689
06:49 < Doxin> that can't be right
07:49 < Doxin> gamingrobot: in case you're still there: my varint implementation is broken :D
07:49 < gamingrobot> ah ok
07:51 < Doxin> now to figure out what I did wrong :P
07:52 < Doxin> gamingrobot: the thing is that it works for numbers  0=<n<128
07:52 < Doxin> so it took me a while to catch it
08:15 < Doxin> does someone have a varint implementation I can have a look at? this is driving me crazy
08:19 < yawkat> Bungee has one 
08:19 < yawkat> Also note that negative numbers are fubar in protocol varints
08:25 < Doxin> oh?
08:32 < gamingrobot> Doxin: https://github.com/SpockBotMC/SpockBot/blob/master/spock/mcp/datautils.py#L31
08:32 < Doxin> gamingrobot: I'm looking at googles implementation right now :P
08:33 < gamingrobot> Doxin: be aware I did not write that varint code
09:55 < Not-f7bb> [SpockBot] gamingrobot pushed 2 commits to master [+0/-0/±4] http://git.io/vZl96
09:55 < Not-f7bb> [SpockBot] gamingrobot 61977ff - Update readme to include features
09:55 < Not-f7bb> [SpockBot] gamingrobot 8fdab93 - Merge pull request #130 from gamingrobot/readme-update
09:58 < Not-f7bb> [SpockBot] gamingrobot pushed 1 commit to master [+0/-0/±1] http://git.io/vZlHZ
09:58 < Not-f7bb> [SpockBot] gamingrobot 2346001 - Fix slight editing oops
14:50 < Not-f7bb> [SpockBot] gamingrobot pushed 2 commits to master [+0/-0/±2] http://git.io/vZ8dA
14:50 < Not-f7bb> [SpockBot] luken 92bf8d6 - Don't send player position packets till connected to the server. On
14:50 < Not-f7bb> [SpockBot] gamingrobot 5cdece1 - Merge pull request #129 from luken/bug-connect_reliability
17:10 <+Thinkofdeath> So what broke this snapshot? :)
17:12 < gamingrobot> Thinkofdeath: cuddling
17:13 < gamingrobot> thats what broke
17:13 <+Thinkofdeath> Nooooo!
17:13 < XorBoole> do we still have huggling though
17:14 <+XorBoole> why, dear god, does freenode nickserv hate me
17:15 <+Thinkofdeath> XorBoole: because you apparently can't set up your bouncer correctly? :)
17:15 <+XorBoole> shhh
18:24 < dav1d> dx: I have an incomplete backlog
18:25 < dav1d> [00:27:19] <dx:#mcdevs> dav1d: you're in that organization. i'm blaming you for that ^
18:27 < TkTech> There may be temporary resolution errors (few minutes) for wiki.vg as the domain transfers to a new registrar.
18:27 < TkTech> Also, "I'm planning a large-scale wiki for everything video-game related, and was planning on naming it wiki.vg.
18:27 < TkTech> Are you willing to consider a different domain for your Minecraft wiki?"
18:27 < TkTech> ~_~
18:37 < TobiX> TkTech: You don't want to donate your domain to such a noble cause? ;)
18:38 < TkTech> Terrible, I know. Looked him up and it's nothing but turnkey sites and rehosting stolen material.
18:39 < TobiX> Isn't there .wiki as a new toplevel domain now?
18:43 < TkTech> Considering all the new random TLDs I'd be really surprised if there wasn't
18:43 < TkTech> I should get vg.wiki for shits'n'giggles
18:44 < TkTech> Not available :(
18:44 < TkTech> Domain Status: BANNED https://icann.org/epp#BANNED
18:59 <+ammar2> TkTech: so the topic says the channel is publicly logged, do these public logs actually exist?
19:00 < TkTech> Yes/no, more a disclaimer than anything else at the moment
19:00 < TkTech> They exist, just aren't public (yet?)
19:00 < TkTech> The not bot at various times has had experimental logging
19:07 <+XorBoole> logs confirmed for nonexistance
19:07 < dx> dav1d: disregard that, Prf_Jakob accidentally configured notifico to send unrelated commits here and i didn't recognize his github name right away
19:08 <+ammar2> TkTech: I'll put in a botbot.me request, their logging is really pretty
19:08 <+ammar2> https://botbot.me/freenode/readthedocs/
19:08 <+ammar2> mmm
19:08 <+Prf_Jakob> dx: It is a port of this http://charged-miners.com/
19:08 <+Prf_Jakob> dx: So technically its not unrelated.
19:10 < dx> Prf_Jakob: oh, fancy
19:11 <+Prf_Jakob> But both projects are pretty, well not moving.
19:13 < dx> the name blends in well with the rest of the electricity puns
19:14 <+Prf_Jakob> Indeed :)
19:52 < shoghicp> botbot!
20:35 < Gjum> there isn't even an .api TLD :P
21:38 < Aikar> Dinnerbone / Grum - with all your work to use 8 on client, can you bump server requirement to 8 for 1.9 so everything can finally make some progress in utilizing 8 :P then you can start utilizing 8's benefits for the code itself while your doing all this rewriting
21:39 < Aikar> hell even netty 5 is considering making 8 required...
21:40 < Aikar> (see: https://github.com/netty/netty/issues/3534 )
21:46 < yawkat> ++
21:46 < rom15041> (*java8)
22:04 < gurun> So PC envy .. this is a new PE exlusive block (cobble-bush) http://prntscr.com/8evjv3
22:16 <+Dinnerbone> Aikar, not for 1.9 sorry
22:16 < yawkat> It looks like a leaf texture with broken biome
22:16 <+XorBoole> Dinnerbone ofc, you're going to use java 9
22:16  * XorBoole runs
22:17 < gurun> yawkat with an embedded slab...
22:17 <+XorBoole> gurun I bet it's FallingSand
22:17 < shoghicp> gurun: I have more, and this time intended!
22:17 < shoghicp> https://imgur.com/a/xcLZv#0
22:17 < shoghicp> ^ and no falling sand
22:18 <+XorBoole> shoghicp what have you done
22:18 < morfin> wtf is that
22:18 < shoghicp> well, 0.12 supports snow blocks inside grass
22:18 < shoghicp> it's extra block on a block
22:18 < morfin> dead bush in glass?
22:19 <+XorBoole> <shoghicp> it's extra block on a block
22:19 <+XorBoole> wat
22:19 < shoghicp> I just generalized it so it supports anything
22:19 < shoghicp> so you can send this kind of stuff
22:19 < morfin> but what for
22:19 <+XorBoole> does this mean that you can have multiple materials per grid location?
22:19 < shoghicp> or save it in the world
22:19 < gurun> ok, so the code for dual blocks is there, so maybe just a simple bug. I'll investigate what the world have in there.
22:19 < shoghicp> yep, but one of them is "fake"
22:19 <+XorBoole> define "fake"
22:19 < shoghicp> you won't collide
22:20 < shoghicp> and renders in an interesting way sometimes
22:20 < gurun> yeah, the bush is fake because the break action was on the slab. Breaking the slab, removed both.
22:20 <+XorBoole> hmm, interesting
22:20 <+XorBoole> how portable is this to PC?
22:20 < shoghicp> anyway, you can implement swapping so the fake one becomes the real one when you break it
22:20 < shoghicp> hmm, it's a hashmap per chunk
22:21 <+XorBoole> ew
22:21 < gurun> yeah well .. i have to investigate why this happens in this world. It is only that one block.
22:21 <+XorBoole> you can already do it in PC with fallingsand, sort-of
22:21 <+XorBoole> native support would be nice though
22:21 < shoghicp> I don't like it, but I just wanted to play with that code I found
22:21 < gurun> yeah, but it's cool with the snowlayer thing so
23:01 < Doxin> does http://paste.pound-python.org/show/QITeP8l8VB8Aevowl5FB/ look like a valid EncryptionRequest to you guys? sending that to the mc client makes it crash with a nullpointer exception.
23:21 <+Thinkofdeath> Doxin: empty server id, 140 byte public key and 4 byte verify token. Sound correct?
23:21 < Doxin> sounds right to me
23:21 < Doxin> minecraft seems to disagree
23:22 <+Thinkofdeath> openssl doesn't like your pub key
23:22 < Doxin> I see
23:22 < Doxin> well I haven't the foggiest how to get botan to produce me that damn key then
23:25 <+Thinkofdeath> how are you trying it now?
23:27 <+Thinkofdeath> Doxin: taking a quick look at the docs i'd guess 'RSA_PrivateKey (RandomNumberGenerator &rng, size_t bits, size_t exp=65537)' to generate it and 'x509_subject_public_key () const override' to encode it
23:27 <+Thinkofdeath> (complete guess)
23:29 < Doxin> Thinkofdeath: that's what I'm doing :P
23:29 < Doxin> Thinkofdeath: http://paste.pound-python.org/show/OFPPOwICYIkaeO9Cks8p/
23:35 <+Thinkofdeath> Doxin: X509::BER_encode try that
23:36 < Doxin> Thinkofdeath: I'm using the D version of the library which is slightly different. let me see if I can find that function
23:38 <+Thinkofdeath> https://github.com/etcimon/botan/blob/master/source/botan/pubkey/x509_key.d#L40
23:38 < Doxin> I can call BER_encode directly on the pubkey, no x509 involved.
23:39 < Doxin> aaaand still getting a nullpointer exception
23:40 < Doxin> oh how about I send the pubkey and not the privkey :P
23:40 < Doxin> oh, BER_Encode isn't defined for pubkeys ._.
23:40 < Doxin> oh
23:41 < Doxin> nvm
23:44 < Doxin> Thinkofdeath: yeah that definitely doesn't work.
23:44  * Thinkofdeath shrugs
23:45 <+Thinkofdeath> out of ideas
23:45 < Doxin> crypto sucks :I
23:46 < Doxin> time to try something else than botan
23:47 < Doxin> Thinkofdeath: happen to know how to do it with openssl?
23:49 <+Thinkofdeath> I work on a client so I only know the decode method d2i_RSA_PUBKEY
23:49 < Doxin> drat
23:49 <+Thinkofdeath> i2d_RSA_PUBKEY seems like a safe bet :P
--- Day changed ven. sept. 11 2015
00:08 < Doxin> how the toot do I even use that function
00:09 <+Amaranth> Grum: http://cube-drone.com/comics/c/the-singleton-fairy
00:10 < Doxin> Thinkofdeath: does 162 bytes sound like it'd be the right length for a pubkey?
00:11 < Doxin> and minecraft still gives me a nullpointer
00:11 <+Thinkofdeath> Doxin: yes
00:11 < Doxin> well it aint working
00:11 <+Thinkofdeath> got the dump of the packet again?
00:11 < Doxin> sure do
00:11 < Doxin> Thinkofdeath: http://paste.pound-python.org/show/9JIySoLhRWbS2XJgqvV2/
00:11 < Doxin> this thing has logging up the wazoo
00:12 <+Thinkofdeath> Amaranth: they managed to get rid of MinecraftServer.getServer if that counts? :)
00:12 <+Amaranth> hehe
00:13 <+Thinkofdeath> Doxin: openssl doesn't like it still, got the crash report from the client too?
00:14 < Doxin> where do I find that again
00:14 < Doxin> also the client didn't actually crash
00:14 <+Thinkofdeath> the launcher has a log
00:14 <+Thinkofdeath> if you set it not to hide itself
00:14 < Doxin> well lets get another crash then I guess
00:15 <+Thinkofdeath> although it may not actually dump the stack trace for errors like that
00:15 < Doxin> hmm
00:16 < Doxin> > [00:15:40] [Netty Epoll Client IO #0/ERROR]: Public key reconstitute failed!
00:16 < Doxin> well that narrows it down to the pubkey /still/ being wrong.
00:20 < Doxin> ech
00:52 < Doxin> how I feel about this crypto mess: https://i.imgur.com/QmTvkwn.jpg
03:57 < nickelpro> Doxin: crypto is easy with the right libs
07:20 <+Grum> Amaranth: hah indeed, thanks ;D
07:20 <+Grum> we're slowly removing some of those singletons, i think we completely got rid of the serverinstance one
16:23 < roblabla> Is 15w37a the same protocol version ?
16:23 < roblabla> As 15w36d that is
16:45 < rom15041> That's unlikely
16:48 <+Amaranth> Dinnerbone said he changed chunk handling so that's probably a protocol change
16:49 <+Amaranth> Although it could just be reordering the chunk packets to make sure the closest ones are sent first
16:49 <+Amaranth> That would be kind of disappointing since CraftBukkit had that forever and even did a spiral pattern to ensure turning around didn't look weird
16:49 <+Amaranth> And I'm hoping to see something amazing
16:52 <+Dinnerbone> Amaranth: Minecraft has had the spiral thing since like a month after cb did :p
16:52 < Aikar> Dinnerbone, any thoughts on improving it to be coned in front of player to get the visible chunks loaded faster? something I remember people looking into in past
16:53 <+Dinnerbone> We only generate x closest chunks to players per tick, and send the ones that are closest incrementally when they're ready & generated
16:53 < Aikar> well, a mixture but 'favor' forward
16:55 < Aikar> <Thinkofdeath> Amaranth: they managed to get rid of MinecraftServer.getServer if that counts? :)      
16:55 < Aikar> in favor of DI or static all the things?
16:55 <+Amaranth> Dinnerbone: You mean the spiral thing I wrote with the help of Wolvereness in 2012?
16:56 <+Amaranth> iirc the original chunk sorting in that file just sent the ones closest first, it didn't try to make a spiral
16:56 <+Thinkofdeath> Aikar: passing the instance around where required
16:58 < rom15041> That's DI (dependency injection)
17:00 < Aikar> and that comic is so stereotyping, not everyone codes on laptops 24/7!
17:00 < Aikar> >_> I enjoy my 5 monitor beast :P http://aikar.co/gibson.jpg
17:01 <+Amaranth> Yeah, only hipsters run around with laptops all the time
17:01 < Aikar> pic is old :3 desk is so clean in it .... lol
17:01 < Aikar> such a mess atm ._.
17:01 <+Amaranth> It's not like a MBP is my main computer or anything
17:01 <+Thinkofdeath> rom15041: psh, I don't follow all the terms
17:02 < Aikar> i pretty much stack up statements into a pile, then shred them when it gets too big
17:02 < rom15041> Lol, when do you switch to coding with a virtual reality helmet Aikar :p ?
17:02 < Aikar> rom15041, id love that :P
17:02 < rom15041> Or maybe a video projector
17:03 < Aikar> i dont think projector would help, its more about the pixels
17:03 < rom15041> Yeah
17:03 < Aikar> and would have to move head more for a projector
17:03 < Aikar> and strain eyes looking farther
17:03 < Aikar> and compromise posture
17:04 < Aikar> i did a lot of work into workplace posture, it improved my effeciency so much
17:04 < rom15041> Yeah I guess coding in virtual reality is better :p
17:04 < Aikar> keyboard tray vs keyboard on desk, wrist supports for kb and mouse, proper upright chair
17:04 < rom15041> You could even code Minecraft stuff *in* Minecraft
17:04 < Aikar> i need the monitors at day job elevated more than I have them
17:06 < rom15041> Hmm, looking at the 2 monitors on top seems distracting to me
17:07 < Aikar> those i use for monitoring and terminals
17:07 < Aikar> but... im tall, so i line up well with them
17:07 <+Dinnerbone> Amaranth: the one erik wrote before then. I didn't know you changed it. It spirals around the player sorting the chunks to send
17:08 <+Amaranth> iirc the original wasn't actually a spiral and would jump around from side to side and corner to corner while the new one was actually a spiral
17:09 < Aikar> http://imgur.com/cCmqTC8 is my dayjob setup
17:12 < Aikar> >_> just realized calendar was in pic, but nothing important on it :P
17:16 < Aikar> hmm that comic you linked Amaranth just informed me of something important to know.... https://www.jetbrains.com/toolbox/
17:16 < Aikar> jetbrains having monthly subscriptions now
17:18 < Aikar> i guess i should get the all license since I already have IDEA and PHPStorm
17:18 <+Dinnerbone> They may be changing it again, as this significantly raises the prices for a lot of customers who aren't businesses
17:18 <+Dinnerbone> And there was a lot of uproar about that.
17:19 < Aikar> hm.. yeah that is a massive price increase 
17:20 < Aikar> pretty much removes the half off renewal
17:25 < Aikar> i think they are wanting to do that cause they know their customers are going to stick with thier products, and they would get more money off renewals than they do currently :/
17:28 < winny> Aikar: your desktop is sooo clean wow
17:28 < winny> do the gel rests help ?
17:29 < Doxin> is there anyone in here with a bit of knowledge of crypto? I'm having trouble getting my publickey formatted so the mc client accepts it.
17:30 < Aikar> dayjob work desk is cleanish yeah, cant trash it :P but yes the wrist supports def help prevent carpal
19:19 <+ammar2> Doxin: yeah, what format are you sending it in?
20:06 < Doxin> ammar2: oops, sorry for the slow reply
20:06 < Doxin> ammar2: well, supposedly x509
20:07 < Doxin> ammar2: but the openssl api is a bit odd so I'm probably doing it wrong
20:07 < Doxin> ammar2: all I know for sure is that netty is having troubles reconstuting the pubkey
20:07 < Doxin> *reconstituting
20:11 <+ammar2> Doxin: could you show your code
20:11 < Doxin> sure, gimme a sec
20:12 < Doxin> ammar2: at server start this runs: http://paste.pound-python.org/show/qE5hjbpoYZkLYkNqs3uB/ , encodedPublicKey then gets tossed into the encryptionRequest
20:31 < Doxin> ammar2: need anything else?
20:37 <+ammar2> Doxin: sorry, had to step away for a minute, looking now
20:38 < Doxin> np
20:41 <+ammar2> Doxin: you likely wanna take a look at i2d_X509
20:42 <+ammar2> instead of using i2d_RSA_PUBKEY
20:43 < Doxin> right
20:44 < Doxin> ammar2: I'll give that a go next time I work on it
21:15 < Aikar> Dinnerbone, mind a quick PM?
21:16 < shoghicp> ammar2: oh god don't remember me those D:
21:16 < shoghicp> just looking at i2d gave me headaches
21:16 <+ammar2> shoghicp: pls, openssl is the most beautiful api ever
21:17 < Aikar> ammar2, not as beautiful as you of course <3
21:17 < shoghicp> same API, and I need to do stuff in a different way every time
21:17 < shoghicp> some parts use a pre-allocated array
21:17 < shoghicp> others need a pointer that they allocate and you need to free later
21:18 < shoghicp> others do black magic and drop your const and modify it
21:18 <+ammar2> shoghicp: I have learned to accept it all
21:18 < shoghicp> some parts need you to hash the data before, others hash it themselves :P
21:18 <+ammar2> you need to accept openssl into your heart
21:19 < shoghicp> at least I made a nice interface and wrapper around it
21:19 < shoghicp> so I can swap it with another crypto library automagically :D
21:20 < shoghicp> this time we'll be using elliptic curves and ECDH :)
21:21 <+ammar2> sounds fancy
21:24 <+Thinkofdeath> ammar2: interestingly I use d2i_RSA_PUBKEY to decode the key on my client. You'd assume the reverse did... well the reverse
21:35 <+ammar2> Thinkofdeath: welcome to openssl
21:35 <+Thinkofdeath> .-.
21:39 <+ammar2> Thinkofdeath: actually yeah, that should work https://github.com/mkrautz/openssl-mirror/blob/beb1e1804912fac3f7818b922400031b35a81ff6/crypto/asn1/x_pubkey.c#L234
21:39 <+ammar2> beats me
21:39 <+ammar2> maybe Doxin made some other mistake
22:15 < Doxin> I probably did
23:52 < roblabla> While updating NMP and testing it on a server, it seems I tripped up an entityMetadata type 12
23:52 < roblabla> which isn't documented on wiki.vg
23:55 < roblabla> aha, it would seem Thinkofdeath already knows about it https://github.com/thinkofdeath/steven/blob/master/protocol/metadata.go#L115 :P
--- Day changed sam. sept. 12 2015
00:08  * Thinkofdeath *whistles*
00:09 < rom1504> I just added that in the wiki, if you know more than 'VarInt' add it though ;)
00:11 <+Thinkofdeath> rom1504: its a block id in combined format
00:12 <+Thinkofdeath> I got lazy with the implementation just for a quick fix
00:16 < roblabla> Thinkofdeath, you mean (id << 4) | data ?
00:16 <+Thinkofdeath> yep
00:16 < roblabla> kk
00:16 < roblabla> I guess it makes sense. It was attached to an enderman
--- Log closed sam. sept. 12 01:05:20 2015
--- Log opened sam. sept. 12 01:05:28 2015
01:05 -!- Irssi: #mcdevs: Total of 149 nicks [1 ops, 0 halfops, 15 voices, 133 normal]
01:08 -!- Irssi: Join to #mcdevs was synced in 204 secs
15:48 < Not-f7bb> [SpockBot] gamingrobot pushed 1 commit to master [+0/-0/±1] http://git.io/vZwXf
15:48 < Not-f7bb> [SpockBot] gamingrobot d768203 - Fix typo
18:53 < Not-f7bb> [SpockBot] gamingrobot pushed 2 commits to master [+0/-0/±4] http://git.io/vZrw5
18:53 < Not-f7bb> [SpockBot] gamingrobot 4883f12 - DefaultPlugins -> default_plugins
18:53 < Not-f7bb> [SpockBot] gamingrobot f89c031 - Merge pull request #136 from gamingrobot/default-plugins
21:21 < Aikar> http://www.hitbox.tv/aikar live streaming epic boss battle on difficulty 10 if anyone wants to watch :) 
21:32 <+XorBoole> Aikar > not instakill
21:32 <+XorBoole> wait, when'd you add chat highlighting? I should add this
21:38 < Aikar> XorBoole: been a while, users can add custom alert words :P
--- Day changed dim. sept. 13 2015
00:08 < Gjum> in 0x02 Chat Message, when is the Position set to 2 (above action bar)? I've never seen this
00:08 < Aikar> its a new mc feature that no api exists for yet
00:08 < Aikar> dunno why it doesnt have onem its so simple
00:09 < Aikar> >_> oh this isnt spigot channel lol
00:12 < Gjum> :)
00:13 < Gjum> so, does anyone know what Position = 2 is for?
00:18 <+Thinkofdeath> Gjum: locked chests
00:18 < Gjum> Thinkofdeath, thanks, that's a start :)
00:18 < angal> And custom servers :)
00:19 <+Thinkofdeath> Aikar: I'm still split on the api, most want to attach it to send message but it seems really out of place there
00:20 < Aikar> Thinkofdeath: I added a sep sendActionBar
00:20 < Gjum> Oooh, that's position 2! cool
00:20 <+Thinkofdeath> Aikar: that seems saner but i've yet to see anyone else agree :)
00:21 <+Thinkofdeath> Aikar: brightside, i'm trying to premptively add api for 1.9 so hopefully it wont be an issue with missing api like 1.8 
01:10  * XorBoole looks at Thinkofdeath 
01:10  * XorBoole looks at his villagers PR
21:54 < Doxin> I'm writing a mc server, and I'm having some trouble getting the client to accept the public key. I keep getting "[Netty Epoll Client IO #0/ERROR]: Public key reconstitute failed!". can someone have a look at the relevant code and tell me if I'm doing something obvious wrong? http://paste.pound-python.org/show/JSc67l0zTEzoRfvmsIvV/
22:40 < rom1504> Doxin: did you think about looking how an implementation of mc protocol is done in the language you are using ?
22:45 < Doxin> rom1504: I don't think there are any. I did look.
22:45 < Doxin> rom1504: note it's D, not C(++)
22:50 < rom1504> okay
22:50 < rom1504> (I didn't know which language it was)
22:52 < Doxin> now you do :D
22:52 < Doxin> I compared my packet to the official minecraft client
22:52 < Doxin> and it seems the same except for the token and the pubkey
22:52 < Doxin> so my guess is I'm using openssl incorrectly.
23:41 < gamingrobot> how do I write in a book?
23:41 < gamingrobot> from a protocol standpoint
23:42 <+Amaranth> It's a plugin channel message because reasons
23:42 <+Amaranth> http://wiki.vg/Plugin_channel#MC.7CBEdit
23:51 <+AndrewPH> yolo , that's why
23:52 <+Thinkofdeath> I assume it was because they only had a max of 256 packets shared between all states in 1.6<. No reason not to change it now though
--- Day changed lun. sept. 14 2015
00:04 < Doxin> Thinkofdeath: the reason of "don't touch it, it works."
00:08 < jast> protocol design 101: always change everything to make the structure look neater, especially if there's lots of software already using the protocol
00:09 <+ammar2> don't touch it, because everything is shit
00:09 <+ammar2> I think they've been refactoring for like a solid 2.5 years now
00:09 < Gjum> > don't touch shit
00:09 < Gjum> reasonable
01:28 < MonkeyJamboree> hey all. i have an idea for a simple application to intercept pings or status requests to my minecraft server, and send a startup request to the azure vm  when it gets one if the vm isn't started, and otherwise to just pass requests through to the server. any thoughts on the plausibility of doing this from the minecraft network/protocol side of things?
01:32 <+ammar2> MonkeyJamboree: its possible, there are applications that do this already
01:32 <+ammar2> the easy way to do it is to disconnect the person when they log in 
01:33 <+ammar2> with like a "please reconnect in a minute, server starting up"
01:33 <+ammar2> and then run a tcp proxy between the server and your "server"
01:33 <+ammar2> once its actually started up
01:34 < MonkeyJamboree> oh, i hadn't realized there were apps to do that
01:34 <+ammar2> off the top of my head I think mc my admin does it
01:34 <+ammar2> but the example was just to say that it is possible
01:34 < MonkeyJamboree> how would they log in if the vm isn't started yet though and thus won't show on their available status list
01:34 < MonkeyJamboree> ?
01:34 < MonkeyJamboree> oh alright
01:35 <+ammar2> well the proxy intercepts the status requests and the initial log in request
01:35 < MonkeyJamboree> ahah, i gotcha
01:35 <+ammar2> so like user <==(status)==> proxy <---> server
01:35 <+ammar2> and once the server is actually up
01:35 <+ammar2> it can just forward all the tcp packets through both ways
01:35 <+ammar2> and everything should work fine
01:35 < MonkeyJamboree> makes sense
01:37 < MonkeyJamboree> i've never done anything with proxies or much with network traffic outside of making http requests to APIs and such, so i'll have to look into that
21:48 < morfin> hello again
21:49 < nickelpro> Hello
21:49 < morfin> so returning to question with synchronization between worlds
21:49 < morfin> how critical if they would have small difference?
21:50 < nickelpro> Hugely critical if an entity collides with a block and the world's disagree on the physics response
21:51 < nickelpro> You're talking about sharing or something along those lines? Civcraft had been trying to make it work for awhile
21:51 < morfin> no not sharing
21:52 < nickelpro> sharding** autocorrect messed me up
21:52 < morfin> as example i have overworld and enderworld and netherworld(same server) but they have tiny difference because i can't 100% synchronize them
21:53 < nickelpro> What do you mean synchronize them? They are completely different dimensions, there is nothing to synchronize
21:54 <+ammar2> probably passing stuff between portals
21:57 < morfin> i assume physics, AI etc works when tick happen
21:57 < morfin> about sharding i am not sure it is even possible
21:58 < morfin> ammar2, what will happen if i run tick handlers of different worlds(on one server) with difference
21:59 < morfin> and drop something into portal(as example overworld=>enderworld)
21:59 <+ammar2> no idea lol, all I know is some redstone devices assume that the world ticking is synchronized
22:00 < morfin> hmm
22:00 < morfin> maybe i should have one global timer
22:24 < Not-2be> [mineflayer] rom1504 pushed 1 commit to master [+0/-1/±3] http://git.io/vZSlm
22:24 < Not-2be> [mineflayer] rom1504 deleted branch use_minecraft_wrap
22:27 < Not-2be> [mineflayer] rom1504 pushed 1 commit to master [+0/-0/±1] http://git.io/vZS8T
22:27 < Not-2be> [mineflayer] rom1504 366f494 - remove dependencies only used in minecraft-wrap and move minecraft-wrap to dev dependencies
--- Day changed mar. sept. 15 2015
00:30 <+md_5> 7777777777777777777777777777777777777777777777777777777777777/clr
00:30 <+md_5> 7777777777777777777777777777777777777777777777777777777777777/clr
00:30 < Gjum> uh, hi?
00:31 < nickelpro> Everyone quiet, I think it's trying to communicate
00:31 <+md_5> oops
00:31 <+md_5> key must've got stuck
00:40 <+XorBoole> 10/10 would seven again
00:58 < dx> 7/7
01:04 < Not-f7bb> [minecraft-data] rom1504 pushed 1 commit to 1.8 [+2/-0/±2] http://git.io/vZ9fJ
01:04 < Not-f7bb> [minecraft-data] rom1504 5d86a19 - add version.json
01:49 < Aikar> lol
10:29 < johni0702> roblabla, what's the purpose of the 'anon' property in the protocol.json of minecraft-data?
10:30 < johni0702> it's only used once to create some sort of anonymous inner container. but then the question is what's the purpose of having that inner container?
14:19 < roblabla> johni0702, entityMetadata uses a special type of inner container, a bitfield
14:19 < roblabla> The point of anon is that it will take all the members of that inner container into itself
14:20 < roblabla> The point is that the new entityMetadata using bitfield is compatible with the old one field-wise (I don't have to suddenly to metadata.keyType.key, I can just use metadata.key
14:20 < roblabla> wow, that sentence wasn't english.
14:28 < roblabla> johni0702, another use-case off the top of my head, though I haven't used it yet, is to avoid having an inner container and/or repeating myself when using a switch. For instance, https://github.com/PrismarineJS/minecraft-data/blob/1.8/enums/protocol.json#L1566 I keep repeating the condition.
14:37 < johni0702> ah, thx that makes sense
16:42  * yawkat prods Grum
16:42 < yawkat> what's the server infinisleeper for?
16:43 < yawkat> it's a daemon thread that... sleeps
16:43 < yawkat> eternally
16:50 < Aikar> this is mojang, do you need more of an answer?
16:50 < Aikar> :P
16:50 < Aikar> (jk mojang!)
16:50 < Aikar> had to give the bandwagon answer >_>
16:51 < Aikar> but id assume there was some old quirk in an older jvm that that was a dirty hack to make up for
17:09 <+XorBoole> yawkat it's destined to some day rise up and destroy the world
17:10 <+XorBoole> beware the Infinisleeper, creature of the deep!
17:11 <+ammar2> yawkat: http://stackoverflow.com/a/824472
17:11 < yawkat> nice.
17:24 < Aikar> gotta love windows....
17:25 < Aikar> interesting note on that about parkNanos
17:25 < Aikar> might play with using that on my fork
17:26 < Aikar> cause Thread.sleep inaccuracy causes so many ticks to under sleep, causing them to immediately re-enter another super short sleep, then that one of course oversleeps. I believe thats the cause of the server actually floating around 19.92~ tps
17:35 < roblabla> pfft
17:35 < roblabla> SO there's this bug, you have this hacky alternative, oh and btw, it triggers another bug
17:50 < Aikar> whats the other bug? i was reading on it seems it may randomly wake up. as i said "play with it" :P
17:51 < Aikar> ultimately aiming to make my catch up system fine tuned to ignore super small sleeps, as I already have the buffer rolling both ways, so if it under sleeps, itll just add onto the next sleep
18:35 <+Grum> yawkat: its an ancient bugfix for something, i think we might not need it anymore
18:36 < yawkat> ammar linked something
18:36 <+Grum> ah yeah that
18:57 < Aikar> Grum, in the realm of performance improvements, thoughts on moving Tile Entity to not be a world tick list, and schedule them to do work on the interval that they actually do work, like beacons etc? using the next tick list thing?
18:57 < Aikar> and please remove the need for chests to tick ._. 
18:58 < Aikar> 1.8 removing sign ticking was a major improvement, just need chests removed now
18:59 <+Amaranth> Aikar: iirc chests need to tick on the client still
18:59 < Aikar> and my goal i want to do as a patch but would be awesome if you did it by default - move entity (and tile entity if it must stay this way) to be chunk based, so things like 'is chunk loaded' checks can be done once per patch instead of once per tile entity, and also to benefit from better cache hits working in the same chunk at a time
18:59 <+Amaranth> So even if they dropped their server tick logic (which they could easily do) they'd have to put in a "is this the server?" hack to stop them from ticking entirely
19:00 < Aikar> could have a separate client tick list?
19:01 < Aikar> i namely really want the chunk based ticking. with the getChunkAt cache I put back in, really reduced that hotspot, and that cache would be more effeciently used with a chunk based order for ticking
19:01 < Aikar> just need to maintain when the last time an entity ticked to prevent double ticking when moving between chunks
19:06 <+Grum> Aikar: nextticklist is absolutely utterly shit and i do not want to ever use it
19:06 < Aikar> i agree :P
19:07 < Aikar> but, some form of only tick when we know something is scheduled to tick
19:07 < Aikar> one thing I was working on was bucketing tile entities into their tick interval
19:07 < Aikar> so you only had to tick the bucket for TE's you know need to tick this tick, and to also spread them over multiple ticks
19:07 <+Amaranth> I mean, nextticklist itself might be awful but some sort of scheduler is needed
19:08 <+Amaranth> Aikar: btw I had an interesting idea I thought you might enjoy, an addition or alternative to your entity activation range stuff
19:08 < Aikar> i had it mostly working in 1.7 but didnt have time to figure out a few bugs
19:08 < Aikar> oh?
19:09 <+Amaranth> You always tick players and their mounts (recursively!) then sort the remaining entities by their distance from a player and tick them until you run out of time
19:09 <+Amaranth> You'd have to guess about running over because you can't measure time that often or you'll spend too much time doing that instead of ticking things
19:09 < Aikar> oh you mean measuring it like spigot currently is that I had to remove? lol
19:10 <+Amaranth> Eh?
19:10 <+Amaranth> I'm talking about doing a System.nanoTime every 1000 entities or something so you can see when you've gone longer than a tick can go
19:11 <+Amaranth> That way you're always doing as much work as possible, it's like a dynamic activation range
19:14 < Aikar> spigot added a tick limiter to cap time spent per tick on entities, but its measuring it near every entity (current bug, it meant 1 in 10 entities but reversed the check)
19:14 < Aikar> but problem with your idea is that entities would need to be last
19:14 <+Amaranth> Yeah 1 in 10 is still way too much
19:14 <+Amaranth> But hey maybe someone was listening after all :P
19:14 < Aikar> cause if you push it up to 50ms, theres more code after entities (and also, ticking multiple worlds), so itll go over
19:15 <+Amaranth> Right you have to define timeslices for everything
19:15 < Aikar> also that method runs risk of not ticking some entities for long periods of time, specially in high load scenarios
19:16 < Aikar> Grum, are you aware of my activation range stuff?
19:16 < Aikar> if mojang did it native, people might bitch but theyd get over it :P
19:17 < Aikar> just like everything else you guys do
19:17  * Amaranth still thinks it should be chunk based
19:18 < Aikar> as a vanilla level maybe, but I have my granularity at less than a chunk :P
19:18 <+Amaranth> You could combine it with some existing chunk checks that way
19:18 < Aikar> 8 to 16 for items can be a major reduction
19:18 <+Amaranth> If item drops are enough of a problem for you that you need to drop them from 16 to 8 you need to tell your players to pick up their trash
19:19 < Aikar> but yeah it would simplify some of the isActive checks
19:19 <+Amaranth> Or just start deleting drops to teach them
19:19 < Aikar> though the main slowness of isActive is the isChunkLoaded check
19:19 <+Amaranth> Which you don't need
19:19 <+Amaranth> How would you be ticking the entity if the chunk wasn't loaded?
19:19 < Aikar> i need to look at the code, i tried to remove it before but i remember it bing needed
19:20 < Aikar> maybe its neighbor check
19:20 <+Amaranth> Hell vanilla won't even tick them if chunks near them aren't loaded (which is expensive to check, btw)
19:20 <+Amaranth> Although I made that cheaper in CraftBukkit, was a pretty huge win
19:20 < Aikar>         Chunk chunk = entity.world.getChunkIfLoaded( x >> 4, z >> 4 );
19:20 < Aikar>         if ( isActive && !( chunk != null && chunk.areNeighborsLoaded( 1 ) ) )
19:21 <+Amaranth> Right the neighbors loaded thing, I made that fast
19:21 < Aikar> that getChunkIfLoaded, wouldnt be needed if entity had a ref to the chunk its currently added to....
19:21 < Aikar> yep, its the getChunkIfLoaded portion thats hot
19:21 < Aikar> i need to extend my cache to that method too
19:21 <+Amaranth> Vanilla is already doing that check
19:21 < Aikar> where i placed it ended up not using the cache
19:21 <+Amaranth> Why are you doing it too?
19:21 < Aikar> right, this replaces vanillas check
19:21 < Aikar> i took vanillas out
19:22 < Aikar> so it doesnt even do that check if it fails the other checks
19:22 < Aikar> reduces the neighbor checks to only as a final 'ok, last check to determine active'
19:24 <+Amaranth> btw Grum wants to hide Chunk entirely so you only work with World
19:24 <+Amaranth> Opposite of what you're asking for :P
19:24 < Aikar> how do you hide the chunk class?
19:25 < Aikar> store chunksections in the world instead?
19:25 <+Amaranth> Well obviously he chunk loader and such would have to know about them
19:25 <+Amaranth> And so would World
19:25 <+Amaranth> But entities and blocks don't
19:25 <+Amaranth> And World wouldn't have any methods you can call to get a Chunk
19:26 < Aikar> well good thing were in a position to add needed things :P
19:26 < Aikar> if mc can access it, so can my patches!
19:26 <+Amaranth> Chunk, ChunkSection, these would be a hidden implementation detail
19:26 < Aikar> i didnt mean api wise
19:26 < Aikar> i meant instead of doing getChunkAt to find entities current chunk as it does in many places, just use the ref
19:27 <+Amaranth> Right but Entity wouldn't know Chunk exists
19:27 <+Amaranth> I'm not talking about an API either, except within the game itself
19:28 <+Amaranth> Anywhere that currently works with Chunk or ChunkSection imagine it working with World instead and thus doing repeated getChunkAt calls
19:34 < Aikar> Grum, thats a horrible idea! :P
19:34 < Aikar> need to be reducing those dang calls lol
19:35 < Aikar> its no concern for smaller servers, but larger ones that stuff becomes a big hot spot :/
19:35 < Aikar> specially with 5k+ chunks
19:43 <+Grum> what is a hotspot?
19:43 <+Grum> getting a random chunk?
19:43 <+Grum> there are other ways of solving that
19:44 <+Grum> by having the thing that 'requests a chunk' become an 'owner of chunks' with a certain range
19:44 <+Grum> when they are owner, they would have an array of 'the chunks they own' (own as in usable)
19:44 <+Grum> and then you can for each of these questions just redirect it straight into those arrays
19:46 <+Grum> basically constant access times at the cost of rewritign everything and it wouldn't work for commands doing /setblock etc
19:53 < Aikar> Grum, the getChunkAt accessing the map to get a chunk. I added a cache that requesting same chunk multiple times in a row to skip the map lookup and greatly improved that (with busting the cache on unload / every tick)
19:53 <+Grum> i quite dislike that in all honesty
19:53 < Aikar> why?
19:53 < Aikar> all the blocks calling getType() results in so many getChunkAt calls
19:53 <+Grum> because when you solve a problem with a cache you now got another problem
19:54 < Aikar> well we know the cases a chunk can go invalid, and wipe out the cache the start of every tick to be safe anyways
19:54 <+Grum> and that is gross code to have
19:55 <+Grum> and its not enough
19:55 <+Grum> and if this ever bugs ..... impossible to find
19:55 < Aikar> well, could override the remove method in the chunk map to force remove there :P then never worry about missing lol
19:56 <+Grum> also 100% sure absolutely not threadsave in any remote way
19:56 <+Grum> s/v/f
19:56 < Aikar> its not safe to call getChunkAt async in first place....
19:56 < Aikar> so thats moot
19:57 <+Grum> agreed
--- Day changed mer. sept. 16 2015
19:47 < Aikar> Grum, Dinnerbone can you please fix signs for 1.9 to let users write past the visible render (at least 20~+ chars). People with different resource packs have different max limits per line of a sign :/ and people with 16 char names that use fatter characters cant even write full name even on vanilla RP
19:48 <+XorBoole> Aikar does the resource pack *actually* affect the number of chars on a sing
19:48 < Aikar> (fixing for 1.8 would be even more awesome)
19:48 <+XorBoole> that seems strange
19:48 < Aikar> XorBoole, yep
19:48 <+XorBoole> wat
19:48 <+XorBoole> dammit searge
19:48 < Aikar> had one user complain to me, then switching to vanilla let them write full name
19:48 < Aikar> cause the RP had bigger font
19:49 <+XorBoole> how is this behavior even sane
19:49 < Aikar> its not, is why im asking them to fix :P
19:49 <+XorBoole> did you open a bug report?
19:49 <+XorBoole> inb4 no because it's a feature
19:49 < Aikar> you know how bug reports work, if you cant report how its impacting gameplay behavior it gets closed >_>
19:50 < Aikar> "i need it for my custom server > closed
19:50 <+XorBoole> " people with 16 char names that use fatter characters cant even write full name even on vanilla RP"
19:50 <+XorBoole> is that not a sufficient case to make a report?
19:50 < Aikar> maybe
19:50 <+XorBoole> I agree that the mods are overly heavy-handed on mojira but I don't think that'd get thrown out
19:50 <+XorBoole> then again they made my client-crashing villager bug public and no one has commented on it...
19:51 <+XorBoole> for the record, send villager trades with a null result to crash the client
19:51 <+XorBoole> (it's public now, so I don't care)
19:52 < Aikar> https://bugs.mojang.com/browse/MC-76380?jql=text%20~%20%22resource%20pack%20sign%22
19:52 < Aikar> this doesnt give me hope
19:53 <+XorBoole> well, you could go find searge and make him fix it
19:53 <+XorBoole> it's his feature
19:56 < Aikar> https://bugs.mojang.com/browse/MC-88858
21:47 < morfin> i remember you said something about sharding
21:48 < morfin> which type you meant - sharding between worlds?
21:49 < morfin> or even more insane idea - split world to shards
21:54 < roblabla> Splitting a single world has  already been done,  not that insane ^^
21:56 < roblabla> I remember lazertester showing off something similar to sharding a world
21:57 < morfin> it is because of redstone
21:58 < morfin> but it allows maintain world partially hmm
22:07 < morfin> wait tick is 1/20 of second right
22:07 < morfin> for redstone it's 1/10 - that's very big delay 
22:10 < morfin> to compare i have latency 100ms to server in UK from Ural(it's almost 4000 kilometers), so if servers are close enought that means no problems?
22:12 < Aikar> XorBoole, and a mod closed it as duplicate to a ticket its not a duplicate of lol...
22:40 <+XorBoole> Aikar 10/10
22:40 <+XorBoole> would report again
22:40 <+XorBoole> who knows, one day I might be jadded enough to moderate mojira
22:46 <+ammar2> Aikar: I saw a duplicate loop the other day
22:46 <+ammar2> 3 tickets
22:46 <+ammar2> each marked as duplicates of each other
22:46 <+ammar2> with no root ticket
22:50 < angal> No ticket - no bug. Problem resolved.
--- Day changed jeu. sept. 17 2015
00:04 < rom1504> duplicate loop, nice concept
06:02 < Not-f7bb> [SpockBot] gamingrobot pushed 3 commits to master [+2/-0/±3] http://git.io/vnes4
06:02 < Not-f7bb> [SpockBot] Gjum e4feb8d - Add chat plugin
06:03 < Not-f7bb> [SpockBot] Gjum cf8ab42 - Fix file not found error for Python 2.7
06:03 < Not-f7bb> [SpockBot] gamingrobot 471c8e9 - Merge pull request #137 from Gjum/chat
19:28 < Techdaan> Hey! I am wondering something
19:28 < Techdaan> Does anyone know something about the logging-in protocol, specifically when the client sends his username to the server after the client sent the handshake with state 2
19:29 < Techdaan> I cannot seem to receive this packet; I am training with networking, so I figured out I could try to do something like this, but this's really bugging me ":)
19:44 < gamingrobot> Techdaan: server or client?
19:53 < Techdaan> Client, it is
19:53 < Techdaan> Gamingrobot: The username should be sent after the state 2 handshake packet, but I cannot receive the username somehow. It simply doesn't send
19:57 < gamingrobot> what do you mean by receive the username somehow?
19:58 < Techdaan> Like
19:58 < rom1504> did you send http://wiki.vg/Protocol#Login_Start N
19:58 < rom1504> ?
19:58 < Techdaan> The client first sends the handshake packet with state 2
19:58 < Techdaan> That is the packet I have to receive :p
19:58 < Techdaan> But I do not receive it
19:58 < rom1504> no
19:58 < rom1504> that is the packet you have to send
19:59 < Techdaan> Then how am I going to receive the username of the user?
19:59 < rom1504> http://wiki.vg/Protocol#Login_Success
19:59 < Techdaan> I guess I have been completely confused
19:59 < gamingrobot> so you are writing a server? or a client? you said client but your talking about not receiving packets sent by the client, implying your writing a server.
20:00 < Techdaan> Oh crap
20:00 < Techdaan> I mis-typed that, sorry
20:00 < Techdaan> Writing a server
20:01 < gamingrobot> so you receive the handshake packet but not the login start, correct?
20:02 < Techdaan> Correct
20:03 < rom1504> are you parsing the handshake packet properly ?
20:03 < Techdaan> I am quite sure I am
20:03 < Techdaan> In fact, when the client sends handshake with state 2 there is no response
20:06 < rom1504> exactly, so it's likely the client send you the handshake packet and the login start packet "together"
20:07 < Techdaan> Hmm, I thought about that too
20:08 < Techdaan> And that may be true, if I read a string after everything I do read the username
20:08 < Techdaan> But there's a [] box and a tab before the name
20:08 < Techdaan> The [] box is like unknown UTF symbol
20:09 < Techdaan> Urgh, I have to go :(
20:10 < rom1504> people reinventing 1/100 of a mc protocol parser everytime ;)
20:12 < Aikar> Grum, can you reopen https://bugs.mojang.com/browse/MC-88858
20:12 < Aikar> i couldnt find any previous issue for it
20:14 < gamingrobot> rom1504: half this channel is asking how to implement the protocol :P
21:38 < morfin> why not
21:38 < morfin> of course i know NIH is bad but if another implementations is not fitting in your program design
21:52 < rom1504> yeah but most don't even look at other implementations (or know how about them)
22:07 < morfin> btw i do not know what's so special in protocol implementation
22:07 < morfin> NBT exist outside of protocol(with gzip), other things seems to be trivial 
22:08 < rom1504> well if the protocol implementation is trivial, then the whole minecraft must be like trivial+trivial
22:09 < morfin> i think hmm there is lots of things harder than just protocol
22:10 < rom1504> a bit harder yes
22:10 < angal> Like explain client, what server wants from him by using this trivial protocol.
22:11 < morfin> client is waaay harder for me 
22:15 < angal> Cause it doing lots of server staff...
22:18 < morfin> what do you mean
22:21 < morfin> client just display
22:22 < angal> Vanilla?
22:24 < angal> It should calculat physics for some entities. Calculate result of (mostly) all block interactions.
22:25 < angal> The whole creative is client-side.
22:26 < angal> Vanilla client is a slave server, which only send some data to real one to validate results and receive back accepts and some entities movements/world changes.
22:31 < morfin> what?
22:32 < morfin> what about server-side physics?
22:33 < angal> It used for most of entitity types (all living).
22:35 < angal> For players + some other (items, etc) used client side physics calculation.
22:35 < Gjum> items are on server I believe, but the client calculates them too, so they glitch
22:37 < angal> Vanilla server send updates for items once upon a while, so without client predication items will move... glitchy.
22:38 < Gjum> but the server still simulates the "official" item, and handles picking them up
22:39 < angal> Ou... And pickuping to handled by clients. Server only validates it.
23:07 < morfin> yes i know
23:08 < morfin> i meant it's just prediction which can be totaly wrong if i changd something on server side
23:10 < angal> It's not prediction. Protocol enforces you to implement this on client, or this willn't happent at all. (Except items movement)
23:12 < morfin> i mean things like falling send, flying tnt, arrows etc
23:13 < morfin> as example i changed something so tnt fly not like in vanilla(on server-side), but client will still predict wrong movement until synchronize to server
23:14 < morfin> that would be very glitchy i guess?
23:14 < angal> Yup.
23:14 < angal> And this willn't work on players.
23:15 < morfin> why 
23:15 < angal> Cause you can't smothly move players, where server want without teleportation packet.
23:15 < morfin> sent by client?
23:17 < angal> Smoth movement for players only relative teleportation and velocity. But that can't gurantee, that clients receive it at point, when server want, nd couse of this might appear desync.
23:18 < morfin> shit
23:18 < angal> So the only packet to overide client player physics is absolute telepotation, which not lokk cool from player, and players willn't play on your server if you send it to often.
23:19 < morfin> so for players it's more than just prediction
23:20 < morfin> and without user sending moving/teleportation packet i can't do anythin
23:21 < angal> In vanilla client calculate physics and server only check if it not to far from origin and there no walls in path...
23:21 < morfin> lol
23:21 < morfin> so in theory i can fall trought wall if server say no walls 
23:21 < morfin> :D
23:22 < morfin> well, i see now - client modifications required
23:22 < morfin> to be compatible with custom physics
23:22 < angal> Realy, if you create both server and client, you can bypass this in current protocol... But this willn't be compatible with vanilla client/server.
23:24 < angal> Yeah, 'someone' here was asking about fails in protocol. There ^ one which can be fixed without (big) changes to client/server.
23:29 < morfin> btw i was thinking how hard it will be implement walking on different surfaces
23:29 < morfin> :)
23:30 < angal> :)
23:32 < angal> If you do not want implement other living entitities, than player - you are not required to do this.
23:32 < morfin> do what?
23:33 < angal> Think about walking on surfaces.
23:33 < morfin> why?
23:34 < morfin> for other living entities prediction will do shit too right?
23:34 < angal> You are server?
23:34 < morfin> yes
23:35 < angal> You going to implement living entities?
23:35 < morfin> so i can implement on server side for living things but i will have to modify client anyway
23:36 < angal> Wor living you not need to modify client.
23:36 < angal> It can be handled fully serverside.
23:36 < morfin> hmmm
23:36 < morfin> ok then
23:37 < morfin> basically i still try to design something scalable
23:37 < angal> Thats fun :)
23:37 < morfin> was thinking to just implement server but in this case i would not be able modify without modifying original code
23:37 < morfin> which is bad
23:39 < angal> Maybe when our servers will be ready steven will be done and there will be some protocol tweaks :)
23:39 < morfin> what server?
23:40 < morfin> i was talking not just about protocol, but about something more like hmm
23:40 < morfin> custom behaviours, creating custom entities using scripting only(no recompiling required) etc
23:40 < morfin> of course for custom entities modification of client will be required(sadly)
23:41 < angal> I understand what you want... You want customise gamplay features serverside.
23:41 < angal> Yeah.
23:42 < morfin> basically i like idea how Gmod implemented that
23:43 < morfin> well, not just Gmod but Source engine too(you can precache resources in mod code too), Gmod just provides API to do that
23:44 < angal> Extending resourcepacks can solve this problem at some point.
23:44 < morfin> so you download stuff required for client when connect(models, sounds, textures, some client-side code) and then play
23:44 < angal> Maybe not in vanilla but only in modded... But possible.
23:45 < morfin> and code separated(mostly) to server-side and client-side
23:45 < morfin> not sure it's possible to do that properly
23:46 < morfin> all "models" are just code - only thing exist on client side are sounds and textures as i know
23:47 < morfin> but that design where models are just code allows creating custom animations in perspective
23:53 < angal> Sometimes we will have this in mc.
--- Day changed ven. sept. 18 2015
00:00 < angal> But if the models will be code described, i don't think there will appear any new custom mc clients... Maybe only as plugin...
00:02 < davidjotta> Heya:) Could anyone recommend me a starting point to learning packets? I'm trying to understand how I could implement wiki.vg/Protocol in my own Java-based minecraft server, but I just can't find the right docs for it. (TL;DR, want to learn packets pls)
00:02 < morfin> wait what a heck
00:02 < morfin> chin.addBox(-1F, -2F, -3F, 2, 2, 2, 0.0F);
00:03 < morfin> body.addBox(-3F, -4F, -3F, 6, 8, 6, 0.0F);
00:03 < angal> Why not to use any existing library/server?
00:03 < morfin> boxes?
00:03 < morfin> ModelRenderer i assume is do all shit
00:03 < morfin> *doing
00:04 < davidjotta> Well, I found MCProtocolLib, but again no docs, so don't know where to start from
00:05 < angal> Are you sure you want to create your own server, not to join any existing project?
00:06 < davidjotta> I'm not really sure tbh. I would join an existing one, but I'm unsure because I don't know anyone, and I don't know if I have the enough skills to be helpful
00:06 < angal> :)
00:07 < angal> Have you expirience in programming/java?
00:07 < davidjotta> Yup
00:08 < angal> Hm...
00:08 < morfin> there is loooots of shit i should learn
00:08 < morfin> because i am trying to do C++ 
00:09 < angal> davidjotta, may i pm you?
00:09 < davidjotta> Sure
00:10 < morfin> angal, what language you use?
00:10 < morfin> jsut interested
00:11 < angal> java...
00:12 < morfin> oh
00:12 < angal> :)
00:12 < angal> Why not?
00:17 < morfin> just asked
00:22 < angal> :)
00:23 < angal> At least it fast to write. Tons of easy to use libraries :)
00:27 < morfin> funny
00:27 < morfin> i know at least 3(!) libraries for event-based programming for C(not C++)
00:28 < morfin> libev, libuv, libevent
00:49 < angal> ping?
01:17 < nickelpro> davidjotta: what don't you understand about the protocol docs?
01:19 < davidjotta> I was talking about the implementation of those, I mean, how can I use all of that data to actually write a functional minecraft server. The part I'm lacking is packets, and I wanted you to recommend a great starting point to learn them
01:19 < nickelpro> The wiki documents the whole protocol, every packet
01:20 < nickelpro> Start with parsing data types and the packet header and work your way from there, that's how most do it
01:20 < nickelpro> Not that it's necessary, there are over a dozen protocol implementations spanning every common and many less-than-common languages
01:22 < davidjotta> Oki, thanks for that. I'll get into it and probably come back asking when I just don't know anymore what I'm doing :)
01:22 < nickelpro> The protocol structure is a little unusual for a game but its still very simple to grok
04:00 < Not-f7bb> [SpockBot] gamingrobot pushed 1 commit to master [+0/-0/±1] http://git.io/vnkZd
04:00 < Not-f7bb> [SpockBot] gamingrobot eaa39d1 - Fix docs badge
06:03 < Aikar> Amaranth: you alive
06:03 <+Amaranth> Yo
06:03 < Aikar> do you know why craftworld.loadChunk is loading neighbor chunks it looks like?
06:03 < Aikar> oh nvm thats vanilla
06:03 < Aikar> why o.o
06:04 < Aikar> this Chunk.loadNearby
06:05 <+Amaranth> *shrug*
06:06 <+Amaranth> Sounds like a 1.8 thing to me, I don't remember a method named that
07:32 < morfin> without proper design code can become mess
17:47 < ekaron> is crash landing still actively developed?
18:31 < Aikar> Dinnerbone, why you remove the bukkit plugin generator :(
18:34 < angal> Couse it's no longer required. Bukkit is get out. :( Users should switch to some other server.
18:52  * Thinkofdeath stabs vanilla
18:54 <+Thinkofdeath> Grum: if you want an easy bug fix: you reuse the chunk packet in the class you use to send them to clients (PlayerChunk for us) however the buffer is released after the first send causing all other clients to disconnect and netty to complain
18:58 <+SinZ> that sounds like a major bug
20:33 < Aikar> angal, much as 'other server devs' want that, its not going to happen. bukkit is here to stay
20:34 < Aikar> the glory of open source, you cant kill it
20:49 < angal> What protocol does minecraft use to share singleplayer server game on lan?
20:49 < angal> Which used to determine game port.
21:29 < Not-f7bb> [SpockBot] Gjum pushed 3 commits to master [+2/-0/±8] http://git.io/vnmN5
21:29 < Not-f7bb> [SpockBot] gamingrobot c670e68 - Fix #140
21:29 < Not-f7bb> [SpockBot] gamingrobot a0f4d19 - Update example
21:29 < Not-f7bb> [SpockBot] Gjum 722dcda - Merge pull request #141 from gamingrobot/chat-move
21:31 < Not-f7bb> [SpockBot] gamingrobot pushed 2 commits to master [+0/-0/±2] http://git.io/vnmAX
21:31 < Not-f7bb> [SpockBot] gamingrobot e7c462a - Add varint and varlong tests
21:31 < Not-f7bb> [SpockBot] gamingrobot 5955398 - Merge pull request #142 from gamingrobot/datautils-tests
--- Day changed sam. sept. 19 2015
05:53 < Not-f7bb> [SpockBot] nickelpro pushed 2 commits to master [+0/-0/±2] http://git.io/vnO7e
05:53 < Not-f7bb> [SpockBot] Gjum 395e6ae - Add inv_ready event to signalize when the inventory can be used
05:53 < Not-f7bb> [SpockBot] nickelpro 5525f0d - Merge pull request #135 from Gjum/window-open-feedback
09:20 < DemiPixel> Hey, I was looking for some help on generation if anybody here knows anything...
16:44 < Jckf> Any neat scripts out there tha'll dump the content of a player data file for me? I have one for a player that suddenly was unable to connect, and removing the file solved the issue. Want to see what happened so that I can prevent it in the future
16:52 < Gjum> Jckf, http://wiki.vg/NBT#Libraries and utilities
17:04 < Jckf> Weird. NBT Explorer had no problems with the file. Renamed it to my own UUID, and I had no issues with my client either. Moved it back to his UUID, and the problem was gone.
17:04 < Jckf> He claims all connection attempts before that resulted in a kick with a message containing "DecodingException"
--- Day changed dim. sept. 20 2015
02:49 < Not-f7bb> [SpockBot] nickelpro pushed 6 commits to master [+0/-0/±20] http://git.io/vnZ1E
02:49 < Not-f7bb> [SpockBot] gamingrobot e15c35d - Make event names more useful
02:49 < Not-f7bb> [SpockBot] gamingrobot 0328a84 - Fix find and replace blunder
02:49 < Not-f7bb> [SpockBot] gamingrobot ea0252c - Fix another find and replace blunder
02:49 < Not-f7bb> [SpockBot] ... and 3 more commits.
02:50 < Not-f7bb> [SpockBot] nickelpro pushed 3 commits to master [+0/-0/±5] http://git.io/vnZ1z
02:50 < Not-f7bb> [SpockBot] Gjum 65ee8ed - Rename all inv_* events to inventory_*
02:50 < Not-f7bb> [SpockBot] Gjum 02b6881 - Rename inventory_ready to inventory_synced
02:50 < Not-f7bb> [SpockBot] nickelpro fc66fdd - Merge pull request #146 from Gjum/event-names-inv-to-inventory
02:52 < Not-f7bb> [SpockBot] nickelpro pushed 2 commits to master [+0/-0/±2] http://git.io/vnZ1D
02:52 < Not-f7bb> [SpockBot] Gjum 7c18180 - Let interact.look use the improvements from smart_look_at
02:52 < Not-f7bb> [SpockBot] nickelpro ef6ffc6 - Merge pull request #144 from Gjum/smart-look_at
03:05 < Not-f7bb> [SpockBot] nickelpro pushed 3 commits to master [+0/-0/±4] http://git.io/vnZDv
03:05 < Not-f7bb> [SpockBot] Gjum 2e951a3 - Make entity metadata a dict
03:05 < Not-f7bb> [SpockBot] Gjum abf929d - Allow all entities to have metadata
03:05 < Not-f7bb> [SpockBot] nickelpro 9af0455 - Merge pull request #143 from Gjum/entitymeta-dict
03:09 < nickelpro> MERGE ALL THE PRs
08:11 < AlphaBlend> switching users on the game launcher should reset the session key
08:11 < AlphaBlend> ought to*
08:12 < AlphaBlend> even if you switch then go back to your account
08:12 < AlphaBlend> so many times sessions expire for me at certain times
08:12 < AlphaBlend> at the wrong times, rather
08:12 < AlphaBlend> like it should reset its TTL on connection to a minecraft server
08:35 < Jckf> Making the launcher fetch a new token would only make the problem worse. You should instead identify which application you use that causes your tokens to expire, and fix that.
08:35 < Jckf> IIRC you can have 3 active tokens. If they keep expiring, it's because something keeps fetching new ones
09:01 < AlphaBlend> Jckf: I wouldn't think other applications cause the minecraft launcher token to expire