20:17 < dav1d> dx: the latter
20:18 < dx> dav1d: ..but why?
20:18 < dav1d> you'll get yelled at at first, but later on, meh I yelled at that guy isn't that trivial, let's help him
20:18 < dx> sounds good :D
20:18 < dav1d> others will start helping you because you dominate the channel with your problem
20:19 < dav1d> (my experience/explanation)
20:19 < dx> ha ha good luck dominating #archlinux
20:19 < dav1d> ^^
20:19 < dx> i always want to be "the guy who did as much homework as possible before asking", but evidently people go tl;dr
20:20 < dav1d> dx: yeah :(
20:20 < dav1d> #archlinux is often my last resort
20:20 < dav1d> and when I ask there I have quite some information gathered up
20:20 < dx> a stack exchange site might be a good option for this kind of question, too
20:21 < dx> i asked only one question ever in there, http://stackoverflow.com/questions/16874043/bash-command-substitution-forcing-process-to-foreground
20:21 < dx> after failing to get answers in most irc channels, someone who knew how to answer it appeared
20:23 < dav1d> ha SO!
20:23 < dav1d> I think I put a 500 reputation bounty on a question once
20:23 < dx> how did that go?
20:23 < dav1d> the one answering was me ... 48 hours later with no sleep
20:24 < dav1d> I forgot a matrix.transpose() :(
20:24 < dx> :(
20:25 < dav1d> http://stackoverflow.com/questions/12538361/frustum-culling-implementation
20:25 < dav1d> even 5 days later :(
20:25 < dav1d> and I did research that problem I bet at least 5 days before asking on SO :(
20:26 < dav1d> dx: haha thanks for the upvote^^
20:27 < dx> :3
20:29 < dx> damn, 3pm already
20:30 < dx> just saw a mention of google reader in dinnerbro's twitter feed... i was supposed to write a script to export the shared items of all my friends but i procrastinated it until today, when it dies
--- Day changed lun. juil. 01 2013
05:02 <+md_5> Download job 'Version & Libraries' started (8 threads, 1 files)
05:03 <+md_5> good jerb mojang
12:29 <+md_5>     } else if (((paramob.al().equals("Dinnerbone")) || (paramob.al().equals("Grumm"))) && (
12:29 <+md_5>       (!(paramob instanceof tz)) || (!((tz)paramob).bG()))) {
12:29 <+md_5>       GL11.glTranslatef(0.0F, paramob.P + 0.1F, 0.0F);
12:29 <+md_5>       GL11.glRotatef(180.0F, 0.0F, 0.0F, 1.0F);'
12:29 <+md_5> I dont getit
12:29 <+md_5> tz is the class for player
12:29 <+md_5> and bG() checks if a cape is hidden
12:29 <+md_5> but parammob is the thing that is name tagged
12:29 <+md_5> oooh
12:29 <+md_5> I guess you can nametag other stuff, just not in vanilla
12:30 < Dinnerbone> md_5: you can nametag any mob...
12:30 < Dinnerbone> In vanilla
12:30 <+md_5> yeah
12:30 <+md_5> but whats with the second condition
12:30 <+md_5> c(!(paramob instanceof tz)) || (!((tz)paramob).bG()
12:30 < dav1d> so Db and grum get special nametags?^^
12:30 < Dinnerbone> If it's a player, check if the cape is on
12:31 <+md_5> you cant name a player in vanilla :(
12:31 < dav1d> :(
12:31 < dav1d> would be funny :)
12:31 <+md_5> lets make a plugin!
12:31 <+md_5> ohwait bukkit wont give us an api for 5 years
12:31 < dav1d> well that won't work
12:31 < dav1d> the skin would change, wouldn't it?
12:51 < dx> jeb mentioned 1.6.2
12:51 < dx> it's our opportunity guys
12:52 < dx> we can have packet length headers this time
12:52 < dav1d>  I have the best idea ever
12:52 < dav1d> remove the packet id
12:52 < dav1d> replace it with a length header, find the correct header by guessing
12:52 < dav1d> s/header/packet/
12:52 < dx> loling
12:53 < dx> there's a few simple packets that look exactly the same though
12:54 < dx> dav1d: what happened with the UTF-8 change anyway?
12:55 < dav1d> no idea
12:55 < dav1d> someday, maybe
12:55 < dx> lol
12:55 < dx> that's something we can actually and realistically manage to get merged
12:55 < dx> maybe.
12:56 < dav1d> for big servers that would also quite reduce the bandwith, especially with json
12:56 < dx> yeah
12:56 < dav1d> 50% less in average I'd assume
12:56 < dav1d> at least for europe/us
12:56 < dx> i stopped reading the logs the last time we talked about it because some people had drunk weird stuff and were suggesting UTF-32 instead
12:58 < Calinou> <dx> we can have packet length headers this time
12:58 < Calinou> lol
12:58 < dx> :3
13:06 < dav1d> well, the only benefit of packetlength headers would be, you don't have to implement the complete protocol
13:06 < dav1d> and you can verify your networking code
13:06 < dav1d> but that's about it?
13:07 < dx> also much more lightweight proxies that can handle more traffic
13:07 < dx> and trivial support of multiple protocol versions for those
13:26 < Grum> < dx> we can have packet length headers this time <-- but who would listen to this channel anymore? ... oh wait....
13:27 < dx> hah what
13:28 < dx> nice to see you around
13:28  * Grum idles some more.
13:29 < dx> see you later :3
13:29 < Grum> TkTech: dont keep that ban on my account btw. do whatever you want
13:29 <+md_5> there is no ban o.O
13:29 < dx> ...did i miss something?
13:29 <+md_5> also
13:29 <+md_5> why is sir gone again
13:29 <+md_5> ._.
13:29  * md_5 finds logs
13:29 < Grum> also, we'll never add packet-length-prefixes in this major protocol version; just so you know; you can stop asking/demanding/begging
13:30 <+md_5> thats what you said about the ping stuff :3
13:30 < dx> Grum: "major protocol version"?
13:30 < Grum> yes, and you said it was a '9 line fix'
13:30 < Grum> look how wrong we both were.
13:30 < dx> you mean the protocol versions covered by minecraft 1.6?
13:30 < Grum> i know i'm right about this one though.
13:31 < Grum> dx: until we majorly overhaul the protocol (look at the end of my todolist) we wont change it.
13:32 < Grum> (and 'it' being the 'adding of length-prefix to packets)
13:32 <+md_5> Grum at the same time switch to NIO for networking :3
13:32 < dx> Grum: i assume this todolist will be complete before the release of 1.7
13:32 < dx> Grum: so we're going to start bothering you again by that time
13:33 < Grum> md_5: maybe
13:33 < Grum> dx: it wont be
13:33 < dx> Grum: shhhh
13:33 < dx> Grum: it will
13:33 < dx> Grum: have some faith in yourself
13:33 < Grum> nope. it wont
13:33 <+md_5> Grum dont worry, its not 10 lines
13:33 < Grum> this todolist is multiple years worth of work :)
13:33 <+md_5> its about 15000
13:33 <+md_5> *1500
13:34 < Grum> I'm actually not sure what a 'sane' way of doing the protocol is
13:34 < Grum> or rather the network implementation
13:34 < Grum> i've used netty, its nice
13:34 <+md_5> https://github.com/EcoCityCraft/Spigot/blob/master/CraftBukkit-Patches/0025-Netty.patch
13:34 <+md_5> yah
13:34 <+md_5> although been using netty 4 for all my stuff
13:34 <+md_5> means a lot of yelling to the devs
13:35 <+md_5> Even CR7 (20th or so release of netty4) has at least one bug I need fixed, 
13:36 < Grum> i'm not really looking forward implementing the encryption using netty
13:37 < Grum> also the whole different queues etc, we should just get rid of that BS
13:52 <+md_5> Grum Ive already implemented the encryption surprisingly simple and Ive perfected it over the months
13:52 <+md_5> still need to pull request it to them
13:54 < Grum> md_5: the way we're doing the round-cycling with packets to confirm state
13:54 < Grum> i do not like that :/
14:03 <+md_5> and all the mojang infrastructure is brning
14:03 <+md_5> burning
14:03 <+md_5> good job
14:07 < Grum> md_5: yeah we broke ourselves hehe
14:07 < Grum> seems the loadbalancers @ amazon are screaming
14:08 <+md_5> more $$$ of EC2 plz
16:55 <+sadimusi> does anybody have an idea what the first byte of the new plugin message sent after the server list ping is used for?
16:55 <+sadimusi> so far I only encountered the value 49
16:59 <+sadimusi> o.O it looks like the client only sends the plugin message and the magic (or shrödinger) byte the first time you click on "Multiplayer"
17:00 <+sadimusi> refreshing the list only sends the 0xfe and the server happily replies
17:00 <+sadimusi> Dinnerbone: ^
17:01 <+sadimusi> and the server complains with "[SEVERE] Reached end of stream for /127.0.0.1"
17:03 <+sadimusi> (never mind about the byte, it was of course the protocol version and I printed it out in hex)
17:22 <+sadimusi> maybe there's something wrong with my proxy
17:22 <+sadimusi> but I'm pretty sure the server gets a NullPointerException when an older client pings it
17:22 <+sadimusi> i.e. lower protocol version in the plugin message
17:23 <+sadimusi> higher values on the other hand are ok
17:25 < dx> odd
17:26 < dx> you must be doing something wrong, since getting errors in older servers was the reason grum scrapped the original extended 0xfe and used plugin messages instead
17:26 < dx> oh wait - older client to new server?
17:26 <+sadimusi> well, older servers won't send the plugin message at all
17:26 <+sadimusi> *lient
17:26 <+sadimusi> +c
17:26 < dx> oh, so it's an issue that doesn't exist yet
17:26 <+sadimusi> yes
17:26 < Not-001> [Craft.Net] SirCmpwn pushed 8 commits to master [+11/-0/±49] http://git.io/f_Eo1w
17:26 < Not-001> [Craft.Net] SirCmpwn 28052d4 - Update protocol version, add new blocks/items
17:27 < Not-001> [Craft.Net] SirCmpwn e187742 - In-progress protocol updates
17:27 < dx> got it
17:27 < Not-001> [Craft.Net] SirCmpwn b43d7dc - Added another changed packet
17:27 < Not-001> [Craft.Net] SirCmpwn 78cff86 - Add new packets
17:27 < Not-001> [Craft.Net] SirCmpwn ff1eac9 - Merge master into snapshot
17:27 < Not-001> [Craft.Net] SirCmpwn 5580ab9 - Basic 1.6 support
17:27 < Not-001> [Craft.Net] SirCmpwn 6eab3ef - Fixed walking/flying speed for 1.6, fixed another changed packet
17:27 < Not-001> [Craft.Net] SirCmpwn 3a3ac91 - Update README
17:27 <+sadimusi> my proxy probably just didn't pick up on further plugin messages, I assume they are sent every time you click on refresh
17:28 <+sadimusi> because if you don't send it you get an old-style reply
17:28 <+sadimusi> i.e motd§0§20
17:30 <+sadimusi> and the protocol byte is of course signed, so the highest thing you can send is 127
17:31 < dx> yep. that's going to be fun
17:34 < dx> some day i should read up on signedness in java, what i've seen so far doesn't make much sense.
17:35 <+sadimusi> why? everything is signed, period.
17:35 < Calinou> sadimusi: suddenly: negative versions
17:36 < Calinou> or just a reset :P
17:36 < dx> sadimusi: i heard that & 0xFF turns it into 'unsigned'
17:36 < dx> and that's where it stops making sense
17:37 <+sadimusi> you can get the unsigned value of a byte this way, yes
17:37 <+sadimusi> but you have to store it in an int
17:37 < dx> oh
17:39 < dx> is it something like the bitwise operators always working with unsigned? because doing & 0xFF over an eight bit number sounds like a NOOP to me
17:40 <+sadimusi> I'm not quite sure how it works since the VM handles bytes as ints anyway
17:43 < Yoshi2> shouldn't the only difference between signed and unsigned lie in how they are interpreted?
17:45 < dx> that would make sense in a lower level language
17:45 < dx> but it seems you can't just tell java to 'interpret this as unsigned', you have to have some int in the middle
18:17 < Grum> dx: there is some issue with the 1.6 server reading the requests, it will send the 1.3 one back
18:17 < Grum> already fixed; deployed when we update
18:17 < dx> Grum: woo!
18:17 < dx> also sadimusi ^
18:17 < Grum> again, shows even more how annoying it is to fix the 'super small issue' in a good way ;)
19:31 < ellisvlad> Hey everyone ;)
19:33 < Yoshi2> good day, ellisvlad 
19:35 < ellisvlad> Hai Yoshi ;) Anyone tried to play around with the new 1.6 update yet? :P
19:38 < ellisvlad> I'm trying, but so much has changed! :P
19:38 < ellisvlad> Like Chat messages 0.o
19:38 < Yoshi2> the 1.6 update is out already? Whew, time flies by fast when I don't pay attention to it
19:38 < ellisvlad> Yeah, it came out today ;)
19:38 < dexter0> Craft.Net has 1.6 protocol support iirc.
19:39 < ellisvlad> you will need the new launcher
19:39 < SinZ> people have calculated that the PluginAPI will be released in 1.9
19:39 < SinZ> as 1.3 changed how singleplayer completelly worked
19:40 < SinZ> 1.6 is the new launcher
19:42 < ellisvlad> aha, go the chat working ;)
19:42 < ellisvlad> *got
19:43 < ellisvlad> You need to send this string now, {{"text":"<ellisvlad> Hello World"}}
19:43 < ellisvlad> :/
19:44 < Grum> ohnoes, bytes
19:44 < iBotPeaches> I wonder how long we can still use the old chat w/ new json
19:45 < ellisvlad> lol
19:46 < ellisvlad> grr.. nvm... didn't work -.-
19:47 < Grum> iBotPeaches: until we break it ;)
19:47 < ellisvlad> oh nvm got it
19:47 < ellisvlad> {"text":"<ellisvlad> Hello World"}
19:47 < ellisvlad> sinlge set of curly braces
19:49 < iBotPeaches> It was easy to strap the old chat into the new json packet, which worked to get a working 1.6 update out, now can rest and work on full json integration
19:49 < iBotPeaches> using all the pretty bold, color, translate, using, etc
19:51 < ellisvlad> :P
19:55 < Grum> for now ;)
19:57 < Yoshi2> I do hope that the "now" will last for a while :)
19:58 < iBotPeaches> 1.6.x comes out, its removed :o
19:59 < iBotPeaches> would be my luck :)
20:01 < ellisvlad> http://puu.sh/3sHsz.jpg ;)
20:09 < Grum> like making the chat work is that hard lol :P
20:11 < Drainedsoul> I'd think making things work would be pretty easy when you're being paid and don't have to reverse engineer.
20:12 < ellisvlad> :P
20:15 < ellisvlad> ugh, now the player movement speed x.x
20:15 < ellisvlad> The player goes insainly fast xD
20:15 < ellisvlad> Hmm, using floats now..
20:24 < ellisvlad> ugh... my 0xCA Player Abilities packet seems to do nothing :/
20:24 < ellisvlad> The flags are set and got correctly, but the speeds are just wrong -.-
20:33 < iBotPeaches> ellisvlad: Loop the entity_count, reading string then float
20:33 < iBotPeaches> s/entity_count/properties_count in 0x2c
20:42 < ellisvlad> Hmm, The annoying thing is that when I set the fly and walk speeds, the player goes really fast (0.6 blocks per step, instead of 0.25) but if the client starts flying, the server gets sent the walk and fly speeds, which come as what I set them -.-
20:44 < ellisvlad> what is even stranger, is that the FOV changes as I send lower walk and fly speeds, but the player stll moves just as fast
20:51 < Grum> Drainedsoul: except that we documented it and told people about it some weeks ago
20:55 < ellisvlad2> wait, what?!
20:55 < ellisvlad2> can I have a link please? ;)
20:56 < Grum> yeah sure, go look in dinnerbones twitter history
20:56 < Drainedsoul> https://gist.github.com/Dinnerbone/5631634
20:57 < iBotPeaches> I wonder if this new system supports newlines, without manually sending new packets for each newline
20:57 < Drainedsoul> the old system did
20:58 < Drainedsoul> ...
20:58 < iBotPeaches> hm, then I was doing something wrong on the old system
20:58 < Drainedsoul> http://i.imgur.com/T75BUHj.png that's all one packet
20:58 < iBotPeaches> are you sending \n ?
20:59 < Drainedsoul> I'm just using my special string Newline, I think it's CRLF
20:59 < Drainedsoul> when it implicitly converts itself to a string
20:59 < Drainedsoul> I'll check sec
20:59 < Drainedsoul> oh no apparently it's just LF
21:00 < iBotPeaches> I'll give it a shot tonight, thanks for the info
21:00 < Drainedsoul> netime
21:01 < Yoshi2> does the server send one or two packets for when a player writes a very long message?
21:02 < Drainedsoul> what do you mean "very long"
21:02 < ellisvlad2> It can all be one packet
21:02 < Yoshi2> so long, that it is broken into two lines
21:02 < ellisvlad2> Up to 32767 characters xD
21:03 < ellisvlad2> but it's easier just to use \n
21:03 < Drainedsoul> there's no race conditions with newline either
21:03 < Yoshi2> I'm happy if it does not send two packets for long player messages
21:04 < ellisvlad2> player messages are a slightly different story
21:04 < Yoshi2> that was always a pain on my old bot which accepted chat commands
21:04 < ellisvlad2> Client -> server is still limited to 200 chars or somthing
21:04 < Drainedsoul> is that limit client or server imposed?
21:05 < ellisvlad2> Server -> Client is unlimited (for most purposes xD)
21:05 < ellisvlad2> Client
21:05 < Drainedsoul> yeah so it's not really limited
21:05 < ellisvlad2> And vanilla Server
21:05 < Drainedsoul> : P
21:05 < Drainedsoul> ah okay
21:05 < iBotPeaches> server -> client has no limit ive reached
21:05 < ellisvlad2> 32767 ;)
21:05 < Drainedsoul> in a sane world it'd be 65535 :(
21:05 < iBotPeaches> so size of short :)
21:05 < ellisvlad2> xD
21:06 < ellisvlad2> signing where pointless ftw
21:06 < Drainedsoul> life wouldn't be fun without if (len<0)
21:06 < Yoshi2> got to love java for that
21:06 < ellisvlad2> although, even with a signed short, you will never reach that huge number :P
21:06 < whatupdave> hey does anyone have the new 0xFE packet?
21:06 < Drainedsoul> there's a now 0xFE packet?
21:07 < ellisvlad2> yeah I got it, kinda
21:07 < whatupdave> yeh, it includes the server address in it. AWESOME
21:07 < Drainedsoul> what do you mean "address"?
21:07 < iBotPeaches> ah was that in the 1.6.1 update?
21:07 < ellisvlad2> http://b.wiki.vg/1.6...1.6.1
21:08 < whatupdave> oh awesome, thanks ellisvlad 
21:08 < ellisvlad> np
21:09 < Drainedsoul> writeString(MC|PingHost); <== what is this
21:10 < ellisvlad> I'm not entirly sure
21:10 < Drainedsoul> like are "MC" and "PingHost" constants and they're being OR'd together, or did someone forget to put quotes around a string?
21:10 < whatupdave> it's a straight up string
21:10 < ellisvlad> I keep just getting 157.0..0.0..0 in my packet dump
21:10 < whatupdave> it appears just like that in there
21:11 < whatupdave> i'm guessing it acts like a user-agent
21:14 < ellisvlad> Oh wow, I'm reading from the wrong buffer...
21:14 < Drainedsoul> wut
21:14 < ellisvlad> p->writebuf->readbyte() was meant to be p->readbuf->readbyte()
21:14 < Drainedsoul> : |
21:14 < ellisvlad> no wonder I was getting random-ish results xD
21:14 < Drainedsoul> you don't
21:15 < Drainedsoul> abstract the protocol parsing...
21:15 < Drainedsoul> ?
21:15 < ellisvlad> "abstract the protocol parsing", What do you mean? :P
21:16 < Drainedsoul> I mean use some facility to abstract the protocol away from you
21:16 < Drainedsoul> I use a "Packet" class
21:16 < Drainedsoul> with some metaprograms
21:16 < Drainedsoul> and type erasure
21:18 < Grum> <Drainedsoul> like are "MC" and "PingHost" constants and they're being OR'd together, or did someone forget to put quotes around a string? <-- quotes
21:18 < ellisvlad> Oh no, I do :P I'm just doing this so I can see what the ping packet says :P
21:19 < Drainedsoul> thanks, that clears that up at least
21:20 < ellisvlad> http://puu.sh/3sKPu.png :P
21:22 < Drainedsoul> C?
21:22 < Grum> Drainedsoul: for backwards compat we 'fake' a 'custom payload packet' around the extra data
21:22 < Grum> else ancient servers will terminate the connection before you get a reply after reading a 'bad packet id'
21:23 < ellisvlad> Drainedsoul: c++
21:23 < Drainedsoul> ah okay, that would make sense
21:24 < Drainedsoul> but you're using printf not std::cout!
21:28 < ellisvlad> :P
21:28 < ellisvlad> To whom it may concern: http://puu.sh/3sLaY.png (Sorry it is a little ugly on the eyes)
21:29 < Drainedsoul> this is what the client sends correct?
21:29 < ellisvlad> ugh, my internet likes to be dodgy -.-
21:29 < Drainedsoul> this is what the client sends correct?
21:30 < ellisvlad> Yeah Client->Server
21:30 < ellisvlad> Server->Client is the same as it was
21:31 < Drainedsoul> the last ~half of the packet, what's it a representation of, and how does it represent that thing
21:32 < ellisvlad> Opps I filtered a little too much http://puu.sh/3sLmk.png
21:33 < ellisvlad> So, using http://b.wiki.vg/1.6...1.6.1#packets
21:33 < ellisvlad> "MC|PingHost" is the user agent of minecraft
21:33 < ellisvlad> b is the ip of the server, according to the client
21:34 < ellisvlad> That's quite cool
21:34 < Drainedsoul> wait after host, they write a short, and then they write a string of bytes
21:35 < Drainedsoul> oh nvm
21:35 < ellisvlad> So using this, we could now, in theory, code many virtual servers into one very easily
21:36 < ellisvlad> Just by checking which IP/hostname the client thinks they are connecting to
21:36 < Drainedsoul> what does the 0x0E mean?
21:36 < Drainedsoul> before the IP starts
21:36 <+sadimusi> protocol version
21:36 <+sadimusi> wait, no
21:36 < ellisvlad> :P
21:36 <+sadimusi> that's probably the length of the string
21:36 < ellisvlad> yeah
21:36 <+sadimusi> the protocol version should be 0x34
21:37 < ellisvlad> it sends the length of b twice?
21:37 < Drainedsoul> what is the 0x49 for then?
21:37 < ellisvlad> 1
21:37 < ellisvlad> lol
21:37 <+sadimusi> it's a regular plugin message packet
21:37 < ellisvlad> the start of the IP
21:38 < Drainedsoul> ah okay
21:38 < Grum> it sends: FE, 01 (backward compat part); 0xFA String:"MC|PingHost" ShortPayloadSize, Byte:protocolVersion, String:"Host", int:port
21:38 <+sadimusi> I currently build it like this in my ping script: http://pastie.org/8100945
21:39 < Grum> so a server who has no idea abou the extra part, will read the 0xFA packet without problems
21:39 <+sadimusi> but I guess I'll remove it again when the 1.3 response is fixed
21:39 < Grum> oh that is fixed
21:39 < Grum> just not pushed live
21:40 < Drainedsoul> okay that seems pretty straightforward, thanks
21:41 <+sadimusi> there are already 591 servers on 1.6.1 on my list, so I have to wait until that drops to 0 before changing anything
21:41 < whatupdave> thanks Grum! Stoked this got added to the protocol
21:41 < Drainedsoul> does it always send the IP, or is it however the client entered the server on the list?
21:42 < whatupdave> Drainedsoul it's the address the client added to the server list
21:42 < ellisvlad> can be a hostname
21:42 < Drainedsoul> yeah that's super useful
21:42 < Drainedsoul> it's like HTTP all over again ;)
21:42 < ellisvlad> lol
21:43 < whatupdave> yep, this is going to make the experience at Minefold a lot better
21:44 < ellisvlad> :P
21:45 < ellisvlad> :( still can't get the player speeds right..
21:46 <+sadimusi> I still don't get why anybody would use this for virtual hosting over a simple SRV record
21:46 < ellisvlad> lol
21:46 < ellisvlad> SRV can take a few mins to update xD
21:47 <+sadimusi> not necessarily
21:48 < ellisvlad> My DNS always takes forever, even when I flush it :P
21:49 <+sadimusi> well, if this is really an issue for somebody I guess virtual hosting is an option
21:49 <+sadimusi> but it's just so much more work
21:49 < ellisvlad> :P
21:50 <+sadimusi> plus it uses some of your precious resources
21:50 < ellisvlad> Bungee could make a thing of it
21:51 < ellisvlad> ..idk :P just throwing random ideas out there
21:51 <+sadimusi> using bungee only for virtual hosting wouldn't make much sense, but maybe you could directly connect to certain servers this way
21:51 < whatupdave> what you really want is a way of redirecting the client, so bungee could send a packet to the client to say reconnect to this host
21:52 < whatupdave> then bungee wouldn't have to proxy the connection
21:52 < ellisvlad> yeah
21:52 <+sadimusi> if you don't want to proxy, just use DNS
21:52 < ellisvlad> but to merge servers with portals or chat commands is a nice feature
21:53 < ellisvlad> lol, anyone else have a minecraft launcher with this many errors yet? http://puu.sh/3sMbH.png :P
21:54 <+sadimusi> Grum: did you also fix the nullpointerexception you get when sending a protocol version < 73?
21:54 < Grum> ?
21:55 <+sadimusi> it happens when the client sends a protocol version < 73 in the 0xfa packet
22:00 < Grum> sadimusi: no idea
22:00 < Grum> also dont care
22:00 < Grum> no client does that
22:01 <+sadimusi> it might become a problem when 1.6.2 comes out
22:25 < Grum> sadimusi: not sure how it would send a lower version? :D
22:26 <+sadimusi> Grum: well, I assume the error would then occur if someone sends a version < 74
22:26 <+sadimusi> which a 1.6.1 client would do
22:26 <+sadimusi> I could be wrong of course, maybe just try it out before shipping 1.6.2
22:54 < ffmdr> does anyone know the origin of map_client.txt?
22:57 < ffmdr> vanilla mentions it in crash reports and it is referred to in mcp/forge as proguard map file(not retroguard one used by mcp)
22:57 < Krenair> google proguard
22:57 < ffmdr> ...
22:58 < ffmdr> I am not asking about proguard
22:58 < ffmdr> it looks like mcp/forge devs get original mojang obf mappings after minecraft releases
22:59 < ffmdr> and are allowed to use them to some extent
22:59 < brainiac94> indeed
23:00 < ffmdr> are you confirming that it looks like that or confirming that they receive the mappings?
23:01 < brainiac94> i'm pretty sure they do get the mappings
23:02 < ffmdr> I am pretty sure that Searge on livestream is working with mojang mappings and then removes package names
23:02 < Krenair> I've heard that MCP gets the obfuscator mappings, yeah..
23:03 < Krenair> According to @SeargeDP: @D!nnerbone knows that changes in the obfuscation are usually no big problem for us.
23:03 < ffmdr> I readed that tweet as: are usually big problem for us
23:03 < ffmdr> fail
23:04 < ffmdr> was removing package names requirement by Mojang to not make source code too pretty?
23:04 < ffmdr> Forge does have its own package mappings but what's the point then o.O
23:07 < ffmdr> https://github.com/MinecraftForge/FML/blob/master/update_packages.py#L27
23:07 < ffmdr> and https://github.com/MinecraftForge/FML/blob/master/update_packages.py#L89
23:09 < brainiac94> has anyone looked at the new packets in the protocol?
23:10 < ffmdr> spout have updated to 1.6, some of new packets have proper names
23:10 < ffmdr> while other have fields named unknown
23:10 < brainiac94> i'm just asking because the spec of 0x2c in the burger vitrine is confusing
23:11 < brainiac94> is it just transmitting a Map<String, Double> ?
23:11 < ffmdr> lemme look
23:11 < brainiac94> http://b.wiki.vg/1.6.1...1.5.2#packets:0x2c
23:12 < ffmdr> looks like that one isn't in spout yet
23:13 < ffmdr> let's look at mcp
23:13 < ffmdr> it is named PacketUpdateAttributes and it transmitting a collection
23:15 < ffmdr> collection of 'AttributeInstance's...let me look deeper
23:15 < brainiac94> how does it have getKey() and getValue()?
23:16 < ffmdr> first int in packet could be entity id
23:16 < ffmdr> then goes string => double hashmap
23:17 < brainiac94> that's what i thought. I'll just try and see whether it gives me any useful information
23:17 < brainiac94> thanks very much!
23:17 < ffmdr> and here is 0x1b
23:17 < ffmdr> https://github.com/Vanilla/Vanilla/blob/631322cfa15a033482dd575964f1d95ccd2b3836/src/main/java/org/spout/vanilla/protocol/codec/entity/SteerVehicleCodec.java
23:19 < brainiac94> thank you
23:21 < ffmdr> health is not float, other changes are pretty clear
23:21 < ffmdr> now float*
23:22 < ffmdr> confirmed 0x2c int is entity id
23:24 <+sadimusi> uhm... 0x2c has been documented in the wiki for quite some time now http://wiki.vg/Pre-release_protocol#Entity_Properties_.280x2C.29
23:24 < brainiac94> .. I completely forgot to check that.
23:24 < brainiac94> thank you
23:25 < ffmdr> sadimusi: oh
23:26 < ffmdr> sadimusi: it should now go to main protocol page
23:32 < ffmdr> just found READ_ME_I_AM_VERY_IMPORTANT file from Dinnerb0ne :D
23:33 < Krenair> Are you guys updating the Protocol page?
23:33 < Krenair> if not I will
23:36 < ffmdr> is minecraft using scala?
23:37 < ffmdr> or is it just scala support in mcp
23:37 < ffmdr> for mods
23:38 < Krenair> Apparently LexManos added Scala support in MCP so people can make Forge mods with it or something
23:38 < ffmdr> ah
23:38 < Krenair> So yeah now you get '"scalac" is not found on the PATH.  Scala files will not be recompiled' every time you run a script
23:39 < ffmdr> FML .gitignore: map_client.txt
23:39 < ffmdr> heh
23:49 < Krenair> ffmdr, where was that?
23:51 < ffmdr> Krenair: which gitignore was that?
23:51 < Krenair> yes. from what repo?
23:51 < ffmdr> forge mod loader
--- Day changed mar. juil. 02 2013
00:06 < Krenair> :/
00:06 < Krenair> Protocol History page doesn't include 1.5 releases
00:12 < ffmdr> You are not allowed to: Release the decompiled source code of Minecraft in any way.
00:12 < ffmdr> there are 383 github repos with minecraft source code
00:12 < ffmdr> lololo
00:12 < ffmdr> *no one cares*
00:33 < brainiac94> .. I just found out about the new chat formatting
00:33 < brainiac94> this is going to be interesting for my bot
00:35 < ffmdr> {text: blabla}
00:36 < brainiac94> is it just that for general messages?
00:37 < ffmdr> I think yes, dunno about colors
00:37 < ffmdr> https://github.com/ElasticPortalSuite/BungeeCord/blob/1296783d9bd9ce13dfcd665935e7c28f8bf43ba1/proxy/src/main/java/net/md_5/bungee/BungeeCord.java#L491
00:37 < ffmdr> (example)
00:38 < brainiac94> so what if i send formatted text - i've read that they ditched § altogether, so is it going to be an array of formatted text objects?
00:40 < ffmdr> I have no idea about colors, I haven't seen them impl anywhere
00:41 < Drainedsoul> is there any kind of complete documentation on the chat formatting/JSON/etc.?  Or just the gist that Dinnerbone threw up?
00:41 < brainiac94> i havent seen one
00:42 < Drainedsoul> okay, would be nice to know which strings are supported for "color", and whether "bold" is supported, etc.
00:43 < ffmdr> oh found dinnerb0ne gist about colors
00:43 < ffmdr> https://gist.github.com/Dinnerbone/5631634
00:44 < ffmdr> oh you can now throw translations into chat
00:44 < ffmdr> that's nice
00:44 < Drainedsoul> this is the gist I was just looking at, it doesn't give a complete enumeration of supported colors
00:46 < ffmdr> it is in EnumChatFormatting class
00:46 < ffmdr> hmm weird there is code in it for reading  old chat style
00:50 < Drainedsoul> where can I see this EnumChatFormatting class
00:50 < ffmdr> http://pastebin.com/Nma1t4HB
00:50 < ffmdr> I guess it maps names to enum by replacing _ with space
00:50 < ffmdr> or just with _
00:52 < Drainedsoul> so we can probably stick like "bold", "italic", "strikethrough", etc. as JSON properties?
00:52 < Drainedsoul> will have to test.
00:52 < ffmdr> yeah, also test "dark blue" or "dark_blue"
00:53 < Drainedsoul> yeah I'll have to convert my stack-based formatting over to generating JSON instead of generating section sign codes
01:07 < ffmdr> yay, just finished my script which adds package names to MCP
01:07 < ffmdr> it was a pain
01:09 < brainiac94> ffmdr, the pastebin URL is invalid...
01:10 < ffmdr> brainiac94, I set it to 10 minutes
01:10 < brainiac94> nevermind then
01:11 < nopresnik> :O MC protocol changed. Pinging the version returns motd
01:14 < ffmdr> nopresnik: :0
01:15 < nopresnik> :O
01:24 < ffmdr> does anyone know what's bukkit status?
01:28 < nopresnik> ffmdr, yep. They are not moving
01:29 < ffmdr> you mean not doing anything?
01:31 < nopresnik> ffmdr, They will, but they aren't just yet
02:09 < brainiac94> is there any documentation on this "Yggdrasil" protocol yet?
02:13 < ffmdr> brainiac94: nope
02:13 < ffmdr> just look at launcher code
02:13 < brainiac94> meh. Thanks
02:16 < brainiac94> wait - does it actually affect the way a client connects to the game server after acquiring a session ID? Currently, my bot manages to connect and receive the 0x01 packet, but then the server says client end-of-streamed
02:17 < brainiac94> Which is odd, because I know for a fact it's the server closing the socket
02:22 < ffmdr> >  wait - does it actually affect the way a client connects to the game server after acquiring a session ID?
02:22 < ffmdr> I guess not
02:22 < ffmdr> so if you are a server you are fine
02:24 < Drainedsoul> whoo
02:25 < ffmdr> btw no more lastlogin stelears
02:25 < ffmdr> *yay*
02:26 < Drainedsoul> wut
02:26 < ffmdr> stealers
02:26 < Drainedsoul> why/how is that
02:26 < ffmdr> http://www.mpgh.net/forum/422-minecraft-tutorials/647873-how-make-last-login-stealer-noobs-visual-basic.html
02:26 < ffmdr> (random example)
02:27 < Drainedsoul> how have they gotten around that?
02:50 <+sadimusi> the only way you could possibly prevent this is by not offering a "keep me logged in" option...
02:55 < ffmdr> nope
02:56 < ffmdr> after logging in with password it receives a token from server and saves it
02:56 < ffmdr> after logging in with token it is invalidated and you receive a new one
03:07 < Drainedsoul> so what if they steal your token and log in before you do?
03:23 < ffmdr> [Ljava.lang.String;@398f573b left the game
03:23 < ffmdr> lol
03:23 < ffmdr> MCP bug?
03:26 < brainiac94> ! isn't that an object identifier?
03:27 < brainiac94> i've had that happen a lot when i was unexperienced and tried to convert a byte[] into a String using .toString()
03:28 < ffmdr> it looks like I screwed up varargs somewhere in translation classes
03:29 < ffmdr> yeah it was that nevermind
03:29 < ffmdr> I had a warning related to varargs when recompiling so I removed it
03:30 < ffmdr> reverted and warning is magically gone too
06:38 < SinZ> Drainedsoul: on the event someone got a copy of your access token, you could simply re-generate it by logging in normally
06:49 < Drainedsoul> I was just getting at the fact that someone running software on your computer to steal your password is -- in my opinion -- being on the other side of the airtight hatchway
06:50 < Drainedsoul> http://blogs.msdn.com/b/oldnewthing/archive/2012/12/07/10375415.aspx <== if you haven't heard that figure of speech before
07:02 <+md_5> Grum 
07:02 <+md_5>    } catch (Throwable throwable) {
07:02 <+md_5>             this.a = 0;
07:02 <+md_5>             this.b = "";
07:02 <+md_5>         }
07:02 <+md_5> why u so catch throwable
07:02 < Grum> because that is what the old code did, i didn't feel like ensuring that all exceptions where of type ioexception
07:10 < gonewest> hello all.  question about the new 1.6.1 server release and the 0xFE server list ping protocol
07:10 < AlphaBlend> Grum: Did you guys name leads in the mojang code LEASHes?
07:10 < AlphaBlend> i'm curious :P
07:10 < Drainedsoul> what about it?
07:11 < AlphaBlend> the unobfuscated bukkit code says LEASH in the NMS Item class
07:11 < AlphaBlend> so i'm curious
07:13 < gonewest> for me anyway, a new server running 1.6.1 is replying to 0xfe 0x01  with the old v1.3 response: [motd, #players, maxplayers]  rather than the newer response including server version and protocol version...  why?
07:14 < AlphaBlend> gonewest: 1.6.2 --> MC-19592 - 1.5.2 clients show server 1.6.1 as 1.3 in Multiplayer screen
07:16 < AlphaBlend> so yes, that got fixed in the next patch
07:19 < gonewest> ah,  thanks.  I didn't see that.
07:31 < Grum> AlphaBlend: yes we renamed the item after it was added (leash != good name)
07:31 < AlphaBlend> oh
07:31 < AlphaBlend> rope? :P
07:48 < Not-001> [Craft.Net] SirCmpwn pushed 1 commit to master [+1/-0/±3] http://git.io/mt0DXQ
07:48 < Not-001> [Craft.Net] SirCmpwn c54daeb - Added packet events to client
10:55 < ffmdr> yay craftbukkit 1.6 \o/
11:01 < SpaceManiac> aww, the game doesn't respond kindly to being maximized while loading anymore
11:01 < ffmdr> before it was just lagging 
11:01 < ffmdr> what changed?
11:02 < SpaceManiac> game appears entirely in bottom-left corner
11:04 <+md_5> ffmdr spigot 1.6 \o/
11:05 < ffmdr> spigot <3
11:05 < ffmdr> https://github.com/EcoCityCraft/Spigot/commit/6e333e5df4345ab43d98e550e2ee1aed5793560f
11:05 < ffmdr> hahahahahaah
11:05 < ffmdr> nice
11:06 <+md_5> ffmdr I always end up leaving that trace in
11:06 <+md_5> and then the issue is
11:06 <+md_5> when someone disconnects by clicking x on the client
11:06 <+md_5> it gets spammy
11:06 <+md_5> and people come crying
11:09 <+ammar2> md_5: if cause isn't end of stream or other unimportant network errors, print trace?
11:10 < ffmdr> just found lastloginstelear in some 1.6 compatible launcher
11:10 < ffmdr> it even uses UDP
11:11 < ffmdr> no more custom launchers...
11:12 <+md_5> can a mojangsta please tell me about sponges
11:12 <+md_5> werent they meant to be implemented like 8 versions ago
11:12 <+md_5> o.O you can leash *anything*
11:21 < ffmdr> https://gist.github.com/ffmdr/f632db50354c15c151fd - SEEMS LEGIT
11:23 <+md_5> ffmdr I dont even
11:23 < ffmdr> md_5: I nearly used that launcher
11:23 <+md_5> where
11:23 <+md_5> dafaq
11:23 <+md_5> did
11:23 <+md_5> you get it from
11:24 < ffmdr> http://minecraft.zyczu.pl/
11:24 < ffmdr> lol
11:24 <+md_5> ffmdr was the mojang one so sexy it hurt?
11:25 < ffmdr> yeah maybe a little
11:25 < ffmdr> just found author's email/name/home address, gonna make his life a pain now
11:26 < ffmdr> does anyone checked if MagicLauncher is clear? it was made by optifine author iirc but is obf'd and such
11:28 <+md_5> ffmdr does it support 1.6?
11:28 < ffmdr> yeah
11:28 <+md_5> I use SKMCLauncher, no 1.6 yet though
11:28 < ffmdr> I use magic launcher generally
11:28 < ffmdr> for modding
11:28 <+md_5> guise I'm doing 1.6 update for wiki
11:28 < ffmdr> it shows conflicts nicely and stuff
11:28 <+md_5> plz dont touch page till I iz done
11:28  * edk makes 100 whitespace edits
11:37 <+md_5> done
11:46 < ffmdr> will minecraft run from chroot or something like that?
11:46 < nopresnik> Anyone know how to fix "name has protected access in EntityHuman" when trying to get "name" from final EntityPlayer entityplayer = ((CraftPlayer) player).getHandle();??
11:47 < ffmdr> nopresnik: it is protected for a reason
11:47 < nopresnik> ffmdr, only since 1.6
11:47 < ffmdr> just use player.getName() ?
11:47 < nopresnik> Cant
11:47 < ffmdr> try to get player by its entity?
11:48 < nopresnik> have a look, im trying to update this
11:48 < nopresnik> https://github.com/NerdNu/ModMode/blob/master/src/nu/nerd/modmode/ModMode.java#L127-L176
11:50 < ffmdr> is it adding colors tags to nick?
11:50 < nopresnik> pretty much, yeah
11:50 <+md_5> probably gonna have to use reflection
11:50  * md_5 checks
11:51 < ffmdr> yeah field.setAccessible(true)
11:51 <+md_5> yah
11:51 <+md_5> reflection required
11:55 < nopresnik> Well what's the best way of doing that?
12:07 < ffmdr> health is now float right?
12:08 < ffmdr> is not 10 hearts 10.0 health?
12:09 < ffmdr> is now*
12:09 < nopresnik> I did hear that in #bukkitdev
15:13 <+sadimusi> if I generate a new access token, does my old one get invalidated? I certainly hope not
15:14 < SinZ> I think they only get invalidated if you log out with them
15:15 <+sadimusi> alright, so if I steal one I just have to use it once so the actual user has to get a new one :)
15:25 < ffmdr> sadimusi: it is
15:25 < ffmdr> you receive a new one
15:26 <+sadimusi> of course I receive a new one, the question is wether I can still use the old one to log in
15:26 < ffmdr> ohh
15:26 < ffmdr> thats a good question
15:26 < ffmdr> give me your token, I'll check it
15:26 <+sadimusi> :p
15:27 <+sadimusi> where is this thing stored?
15:29 < ffmdr> o.O
15:29 < ffmdr> can't find it
15:29 < ffmdr> got it
15:29 < ffmdr> launcher_profiles.json
15:29 < ffmdr> there are accessToken and clientToken
15:32 <+sadimusi> yep, it's still valid
15:32 < ffmdr> sadimusi: send me yours, I'll try to login as you and check
15:32 < ffmdr> :P
15:32 <+sadimusi> wait, an old access token is also still valid? o.O
15:33 <+sadimusi> ah no, the client just doesn't complain
15:33 < ffmdr> http://www.reddit.com/r/Minecraft/comments/fq8zi/why_does_logging_onto_a_smp_server_require/
15:34 <+sadimusi> the issue would be invalidating those signatures when you change the passowrd
15:35 < ffmdr> hmm
15:35 < ffmdr> so checking if key was revoked
15:35 < ffmdr> but this way it would be same traffic load on auth servers
15:35 < ffmdr> I guess
15:36 <+sadimusi> a centralized system isn't that bad, the auth servers aren't down nearly as much recently as they used to
15:37 < ffmdr> afk
15:38 < TkTech> They're up more often then they're down at least, which is a change :)
15:38  * TkTech still can't believe he home-rolled his database
15:39  * sadimusi still can't believe minecraft.app doesn't want to use the pretty jdk 7 he installed :/
15:40 < TkTech> Did they set a fixed version in the plist?
15:40 <+sadimusi> not sure, it might be OS related
15:40 <+sadimusi> I'm using the mavericks preview, so it's totally my fault
15:41 <+sadimusi> running it from the console works fine though :)
15:42 <+sadimusi> oracle removed all the mac links for jdk 6 o.O
15:43 < barneygale> I wonder /why/ mc.net is up more now. afaik part of the problem was it being a java app running on a single machine keeping all its data in memory
15:43 < barneygale> Have they improved the code, or the architecture?
15:44 <+sadimusi> iirc they switched from jsp to play
15:44 < barneygale> play? /me googles
15:44 < barneygale> ah nice, that's good
15:44 <+sadimusi> mollstam once mentioned it and I think there is still a play fork in their github org
15:45 < barneygale> for all the shit the mc devs get, it's a lot better than it used to be
15:45 < barneygale> Did SirCmpwn get banned for his tirade against grum last week?
15:45 < SinZ> yup
15:46 < barneygale> good
15:46 < barneygale> I was hoping that would happen - it was pretty unacceptable
16:09 < ffmdr> barneygale: again?
16:10 < barneygale> yeah you'd think he'd learn
16:10 < ffmdr> > * sadimusi [14:39:44] still can't believe minecraft.app doesn't want to use the pretty jdk 7 he installed :/
16:10 < ffmdr> sadimusi: apple tools don't play well with oracle jdk
16:50 < Drainedsoul> well assuming he values actually speaking his mind about the various developmentally-challenged parts of Minecraft over being in this channel, there's nothing to learn
16:50 < dav1d> I guess it's time to update!
16:50 < dav1d> burger, yay
16:50 < Drainedsoul> "you'd think he'd learn" is condescending and implies there's some kind of "lesson", that the moderators/administrators of this channel are some arbiters of moral rectitude
16:55 < Grum> * sadimusi [14:39:44] still can't believe minecraft.app doesn't want to use the pretty jdk 7 he installed :/ <-- just make it use it?
16:56 < dav1d> y u osx :(
16:56 < Grum> also saying 'minecraft.app' means nothing :P
16:56 < Grum> which of the two? ;D
16:57 < dav1d> chromium cache is pissing me off right now, updating documentation and I have to wget it to check if it updated
17:01 < ffmdr> Grum: last time I checked Apple java wrapper for .app doesn't seen oracle java
17:01 < dexter0> Apple recommends that developers bundle Java with the .app
17:02 < dexter0> and they must do so if they submit to the Mac App Store (which I know you guys don't)
17:14 < AlexDaDu> Anybody here who knows the new available fields of the ServerPingPaket?
17:17 < TkTech> Was the wiki not already updated?
17:17 < mbaxter> Didn't seem to be when I looked
17:17 < dx> probably not, it was a really late addition
17:18 < TkTech> Grum: dexter0 has a point actually, do you happen to know why it isn't bundled? Just not a priority at the moment or is there a technical blocker?
17:20 < barneygale> Drainedsoul: what are you talking about? http://mcdevs.org/irc/rules "Do not harass Mojang employees." is why he was banned, I assume.
17:20 < barneygale> IRC channels need rules yo
17:20 < barneygale> well, when they're big enough anyway
17:21 < Drainedsoul> which has nothing to do with what I said.  What I said was that following the rules is only worthwhile so long as the benefits do not outweigh the costs.  This is true of everything in life.
17:21 < barneygale> Your argument is "rules are made to be broken"?
17:21 < barneygale> Fascinating
17:22 < Drainedsoul> No, my argument is that everything has a cost/benefit calculus
17:22 < Drainedsoul> and that valuations are subjective
17:22 < barneygale> Sure. And the people who actually /run/ the channel thought it was out of line
17:22 < Drainedsoul> so where the benefits -- if any, to you -- of harassing Mojang employees may not outweigh the costs -- being banned -- the same may not be true for SirCmpwn
17:22 < barneygale> As did basically everyone else who was there
17:22 < dx> AlexDaDu: parse everything after the constant 1 byte that was there before as a "plugin message" packet, the contents of that message are a byte that is the protocol version, a string that is the host the client is connecting to, and an int that is the port AFAIK
17:22 <+ammar2> AlexDaDu: the client sends the normal \xFE ping packet, followed by a plugin message packet containing a byte, string and int (unsure about the order jsut yet)
17:23 < dx> ammar2: ha ha i win
17:23 <+ammar2> good on you
17:23 < dx> ha ha.
17:23 < Drainedsoul> http://puu.sh/3sLmk.png here's the packet, for reference
17:23 < AlexDaDu> ok, thanks :) I'll try it.
17:24 < barneygale> So speaking of which I'm updating the Server List Ping page, and was wondering what the channel thinks of this example that was added: http://wiki.vg/Server_List_Ping#Example
17:24 < barneygale> I think it's a bit verbose and duplicates the example above it.
17:24 < Drainedsoul> since I added it I think it's good
17:25 < barneygale> oh you did?
17:25 < barneygale> well, any other thoughts? tis a wiki after all. I don't mind if it stays.
17:25 < dx> "PROTOCOL ANALYSIS" lol
17:25 < mbaxter> I dunno why it sends it as FA custom message instead of just, you know, sending the String
17:25 <+ammar2> seems ok I guess, a bit too verbose for my liking though
17:25 <+ammar2> mbaxter: because the server continues to reads packets as normal
17:26 <+ammar2> so you can't just send more stuff without it being interpreted as a new packet
17:26 < dx> mbaxter: old servers stop parsing when they get 0xFE 0x01
17:26 <+ammar2> dx: they don't, which is what the problem is.
17:26 < dx> er.
17:26  * barneygale wonders what extra data format will be tacked on the next time they want to change server list ping ;D
17:26 < mbaxter> ammar2: pfffffffffft don't need your logic
17:26 <+ammar2> after you send that 0x01 its waiting on a keep alive id
17:26 < dx> yeah i said it backwards.
17:26 < dx> ammar2: keepalive is 0x00
17:26 < mbaxter> correct.
17:26 < dx> none of that is sent
17:26 <+ammar2> well whatever the the thing is
17:27 < mbaxter> 0x01 is login request, lol
17:27 < mbaxter> silly amar
17:27 < mbaxter> +m
17:27 <+ammar2> yeah so its waiting on the rest of that packet
17:27 <+ammar2> which is why the hacky plugin message approach
17:27 < dx> that's *really* old servers
17:28 < dx> the ones mojang doesn't care about
17:28 <+ammar2> yup, gr.um was fine with them being broken but not the ones which try to read that one byte out
17:28 < Drainedsoul> well I think the issue is that at some point they did care about them, and now they're stuck with that legacy
17:29 < dx> yes, they cared about them when they weren't too old
17:29 < dx> lol.
17:29 < Drainedsoul> it's kind of like why WinAPI still has relics of old 16-bit Windows in it
17:31 < AlexDaDu> So, is the new information, such as host and port, included in the ServerPingPacket or in an extra packet?
17:31 < barneygale> Drainedsoul: windows has older stuff than that
17:31 <+ammar2> extra packet
17:31 < mbaxter> I believe it'd be read as a 250 immediately after the ping packet?
17:32 < barneygale> try creating a file called AUX
17:32 < mbaxter> 250/FA
17:32 < dx> the official implementation sends everything as part of 0xFE
17:32 <+ammar2> mbaxter: riht
17:32 < mbaxter> ammar2: yay
17:32 < barneygale> You won't be able to do it, and that restriction came from DOS 1.0 (iirc)
17:32 <+ammar2> dx: wot m8
17:32 < dx> you're free to interpret 0xFE 0x01 as one thing, and 0xFA and the rest as another packet
17:32 < dx> ammar2: check burger m8
17:32 < dx> http://b.wiki.vg/1.6...1.6.1#packets:0xfe
17:32 <+ammar2> dx: regardless of how its sent, at the other end it looks like a \xFA after a \xFE
17:33 < dx> yeah
17:33 < dx> well
17:33 < barneygale> Actually I was intending to write up on Server_List_Ping as if it were one packet
17:33 < dx> i think the read part of this is symmetrical too
17:33 < Drainedsoul> I think that's the best approach
17:33 < barneygale> As it's good to have that page readable without understanding the Protocol page
17:33 <+ammar2> certainly makes sense
17:34 < AlexDaDu> In the original minecraft code it seems to be send in one packet
17:34 < dx> barneygale: hmm sounds good, but also mention that it can also be interpreted as 0xFA for backwards compat, since people will wonder what those fields are for
17:34 < barneygale> Yep, sure
17:36 < dx> should burger vitrine show obfuscated class names? the json output includes it, and in this case i'd like to look at that code
17:37 < dx> i mean, are there any reasons to hide that kind of info?
17:38 < TkTech> sadimusi: ^
17:39 < dx> (mostly for convenience. getting burger working locally was trivial, other than the fact that the dependency on solum wasn't mentioned)
17:39 < dx> "pip install git+https://etc" was enough to solve that
17:51 < AlexDaDu> I've solved my problem.
17:54 < AlexDaDu> 3 + 2 * this.b.length() + 4, what is 3 for a length?
17:54 < ffmdr> where
17:55 < AlexDaDu> In the 0xFA PingHost paket
17:55 < ffmdr> it is short and byte
17:55 < AlexDaDu> https://github.com/Bukkit/mc-dev/blob/master/net/minecraft/server/Packet254GetInfo.java#L36
17:55 <+ammar2> 1 (protocol version) + 2 (string short prefix)
17:57 < AlexDaDu> ah ok, thanks 
17:57 < AlexDaDu> Makes sense
17:57 < TkTech> dx: https://github.com/mcdevs/Burger/issues/9
17:58 < TkTech> Only 12 months ago.
17:58 < dx> TkTech: lol
17:59 < TkTech> dx: There's an edit button on github's UI now…no excuses!
17:59 < TkTech> Think it should be there, add it :P
17:59 < dx> lol.
17:59 < dx> you have no excuses either :D
17:59 < TkTech> Burger + Solum are both deprecated to me, it's sadimusi's project now.
18:00 < dx> excuses!
18:00 < dx> dammit.
18:01 < Yoshi2> there is an excuse for everything ;)
18:01 < TkTech> Solum is looooong dead to me, I haven't made a commit in…oh my god it's been two years?!
18:04 <+ammar2> barneygale: btw I updated your gist for the new stuff https://gist.github.com/ammaraskar/5910348
18:04 <+ammar2> if it seems hacky, thats because it is.
18:06 < barneygale> Oh, I got distracted doing something else bah
18:18 < barneygale> alright, here you go
18:18 < barneygale> when did minecraft get
18:19 < barneygale> err, clipboard fail
18:19 < barneygale> http://wiki.vg/Server_List_Ping
18:19 < barneygale> ^ thar
18:19 < barneygale> Formatting is a bit pants at the moment - not sure if I should move it into a table or perhaps make it all monospace
18:19 < Not-001> [Craft.Net] SirCmpwn pushed 1 commit to master [+1/-0/±4] http://git.io/qaMQ4w
18:19 < Not-001> [Craft.Net] SirCmpwn f07ba3f - Added 1.6.1 support
18:20 <+ammar2> barneygale: XX XX - port the client is connecting to, as a big-endian short.
18:21 <+ammar2> burger says int
18:21 <+ammar2> my code works with int
18:21 < barneygale> oh, my mistake
18:21 <+ammar2> and the length adds a 4 on, which is the length of an int
18:21 < barneygale> Hm, really?
18:22 <+ammar2> yup
18:22 < barneygale> Its 4 too big? :/
18:22 < barneygale> That seems like a bug
18:22 <+ammar2> wat
18:22 < barneygale> also why is mojang encoding ports as ints...
18:22 <+ammar2> because they can?
18:23 < barneygale> aren't port numbers restricted to 16 bit? I thought everything uses shorts for em
18:24 <+ammar2> seems so
18:24 < Grum> the other port was also transmitted as int; just keeping it identical
18:24 <+ammar2> barneygale: well whatever the case, they've used int for ports and you should fixitfixitfixit
18:24 < barneygale> yeah fixing it now
18:25 < Not-001> [PartyCraft] SirCmpwn pushed 1 commit to master [+0/-0/±2] http://git.io/J8w4kQ
18:25 < Not-001> [PartyCraft] SirCmpwn 90c8c2a - Update to 1.6.1
18:26 < Yoshi2> maybe because port numbers go up to 65 535, but a signed short can only store half of that
18:26 < Not-001> [SMProxy] SirCmpwn pushed 1 commit to master [+0/-0/±2] http://git.io/YYbYpQ
18:26 < Not-001> [SMProxy] SirCmpwn 8b7f436 - Update to 1.6.1
18:26 < barneygale> ah Yoshi2 that'll be it
18:27 < barneygale> updated the page btw
18:28 <+ammar2> barneygale: minor stuff: no need to specify big endian, thats the network order. May wanna elaborate a bit on 'length of the rest of the packet, as a big-endian short'
18:28 <+ammar2> since rest of the packet is gonna confuse people if they haven't looked at the plugin message packet page
18:30 < barneygale> "rest of the data" would be better I suppose
18:33 < Drainedsoul> why do people insist on saying "short", "int", "long" instead of "16-bit signed integer"
18:34 < dx> because saying short is shorter
18:34 < Drainedsoul> it's ambiguous
18:34 < SinZ> because short,int,long are better than int16,32,8,64
18:35 < dx> it's defined clearly in the data types wiki page
18:35 < Drainedsoul> so if people happen to read the data types page, then they'll know.  The issue is that "short", "int", "long", etc. are all actual data types.
18:36 < Drainedsoul> and they don't have a specific, fixed width in all instances
18:36 < barneygale> their length is obvious from the page
18:36 < dx> they do. in java.
18:36 <+ammar2> in the context of the minecraft protocol they have a particular meaning
18:36 < barneygale> and yeah, in java they do have those lengths
18:36 < barneygale> have you taken over as knit-picker general now
18:37 < Drainedsoul> yeah but the world doens't revolve around Java, as evidenced by the fact that the wiki gives code examples in a variety of languages
18:37 < barneygale> ammar2: made the changes you suggested
18:37 <+ammar2> barneygale: thanks <3
18:37 < dx> the minecraft protocol revolves around java
18:37 < barneygale> Drainedsoul: the wiki also gives example values every time it mentions "short" or whatever
18:37 < barneygale> there is no ambiguity
18:37 < barneygale> stop complaining
18:38 < Drainedsoul> I don't think pointing out that something is potentially ambiguous is "complaining"
18:38 < barneygale> It's not potentially ambiguous
18:38 < barneygale> only to complete cretins
18:39 < Drainedsoul> it's actually deliberately ambiguous in certain contexts
18:39 < barneygale> Where
18:39 < dx> it's only ambiguous if you're talking about C code that will be ported to several architectures
18:39 < Drainedsoul> the C and C++ standards deliberately avoid fixing the widths of short/int/long, giving them only minimum widths
18:39 < barneygale> dx: even then, there are example values and I'm intending to give a full packet dump.
18:40 < barneygale> if confusion enters the mind of a reader for a split second it can be easily resolved by checking either of those
18:40 <+ammar2> or you know, reading the Data types page
18:40 < dx> this is not C. we assume people coding C will be aware that their language makes no guarantee on the lengths of those data types. it's not our problem if they don't
18:40 <+ammar2> which is a pre-requisite to understanding the protocol anyway
18:40 < dx> ^
18:40 < barneygale> ammar2: I try to keep the Server List Ping page readable to people who don't want to understand the rest of the protocol
18:40 < Drainedsoul> or you could just make communication unambiguous
18:40 < barneygale> hence why it doesn't say "send an 0xFE, then a 0xFA"
18:41 < barneygale> Drainedsoul: then start your own wiki, with your own formal notation system
18:41 <+ammar2> barneygale: yeah but the data types page isn't exactly a large block of text like the protocol page. May as well link to that too
18:41 < barneygale> because using english is inherently ambiguous
18:41 <+ammar2> just to keep grounds covered
18:41 < barneygale> ammar2: sure
18:42 < barneygale> done.
18:42 < TkTech> barneygale: Because they have to.
18:43 < barneygale> TkTech: sorry, what was that in response to?
18:43 < TkTech> barneygale: The max port is 65535, which is an unsigned short.
18:43 < dx> oh, that.
18:43 < TkTech> barneygale: No unsigned types in Java, so you either have to & 0xFF it, or just store it as a larger type.
18:43 < barneygale> Right, yeah
18:43 < dx> context: "why is the port field int not short"
18:44 < barneygale> Yoshi2 explained that to me
18:44  * barneygale isn't very familiar with java
18:44 < Drainedsoul> lol no unsigned data types
18:44 < Drainedsoul> good language
18:44 < barneygale> Drainedsoul seems oddly familiar.
18:44 < dx> better than having every value as a double
18:44 < Drainedsoul> that's undoubtedly true
18:44  * dx read a 15 minute lua tutorial last week
18:45 < Drainedsoul> it's also better than having only 1 fixed width integer type, which is signed
18:45 < Drainedsoul> doesn't JavaScript do all numbers are doubles too?
18:45 <+ammar2> dx: oh god the one on site with comments in code explaining everything?
18:46 < dx> ammar2: ye, wat about it?
18:46 <+ammar2> dx: nil
18:46 < barneygale> Drainedsoul: http://blogs.msdn.com/b/oldnewthing/archive/2012/10/17/10360184.aspx
18:46 < dx> ammar2: hmm, i think that wasn't mentioned..
18:47 < dx> ammar2: WELP
18:47 < Yoshi2> catching exceptions in lua is fun because it often involves checking if the result of a function is nil
18:47 < TkTech> dx: Python does numerics pretty well (at least with 3.3+'s Decimal module updates)
18:47 < TkTech> dx: Dynamic typing + evolving types
18:48 < Drainedsoul> barneygale: http://wiki.vg/NBT#Specification
18:48 < TkTech> Where a number will change types to suit its value and eventually if necessary turn into an arbitrary-precision value.
18:48 <+ammar2> oh yeah, its really handy to have something just increase its capacity instead of overflowing
18:48 < barneygale> So change it if you want, I won't revert it back
18:48 < barneygale> someone who might feel more strongly than me might do so
18:49 < barneygale> you can have your bottle.
18:49 < Drainedsoul> my point is that it's been changed, and has been for a while
18:49 <+ammar2> TkTech: one of the many reasons why its great for quick prototyping
18:49 < Drainedsoul> well automatic promotion is great except where it isn't, I don't know how I feel about that
18:50 < Drainedsoul> I'm personally a fan of checked arithmetic, where you throw on overflow
18:54 < dx> TkTech: python handles it nicely (although not caring about efficiency), but i think what lua does is justified since their VM is tiny
18:55 < TkTech> dx: Wah? Python's AP is very efficient. 3.*+ has made incredible (over 300x speedups) leaps
18:55 < TkTech> dx: http://docs.python.org/3/whatsnew/3.3.html#new-decimal
18:57 < dx> TkTech: oh, i haven't followed on 3.3, but i meant that some operations will get inconsistent performance due to reallocations... or they did, before this update.
18:57 < dx> TkTech: it's not like most people can realistically use 3.x yet, though :(
18:57 < TkTech> dx: Sure they can, hell it's the default on Ubuntu.
18:58 < TkTech> dx: And the allocation issue is solved in part or in whole by pypy or one of the many inline tools for CPython
18:59 < dx> TkTech: and it's been the default in arch for a long while and all i do is typing python2 or changing hashbang lines of script, because most stuff isn't ported yet
18:59 < TkTech> Pfft, lazy bones. You're a dev and the project is probably open source!
19:00 < TkTech> 99% of my old tools were a quick 2to3 away from porting.
19:00 < dx> TkTech: the port of flask/werkzeug to py3 was finished last month actually, i could port some of my stuff to it now, but i believe i have other deps that are stuck in py2 still
19:01 < dx> TkTech: it's all about deps. it's rarely as simple as 2to3 for them
19:46 < ellisvlad> hey guys, quick question..
19:46 < ellisvlad> Have I got it right that Servers need to handle sprinting now?
19:47 < ellisvlad> e.g. when the player's speeds change
19:50 < TkTech> Yar, but it's apparently getting un-done.
20:36 < ellisvlad> oh kk
20:36 < ellisvlad> thx
20:37 < ellisvlad> I just spent the last half hour coding a sprint system to give 20 blocks of sprint with 0.3blocks/sec regen
20:37 < ellisvlad> lol
21:52 < Drainedsoul> yesterday someone pasted me the vanilla Minecraft's chat formatting enum.  Super useful but unfortunately the paste expired. :/  Anyone happen to have that?
21:56 < iBotPeaches> ah darn, i want that too. Instead of manually figuring out what all these possible "translate:" values are
22:12 < ffmdr> Drainedsoul: you mean my paste?
22:15 < ffmdr> Drainedsoul: https://gist.github.com/265374b55a14c08e39e3
22:17 < Drainedsoul> yes, sweet, thanks
22:19 < ffmdr> how does colors in entity names work now??
22:19 < ffmdr> is it also json?
22:20 < iBotPeaches> yer, "color": "yellow"
22:21 < iBotPeaches> I don't know all the colors yet, but all the basic ones seem to work
22:21 < iBotPeaches> and don't know how to color portions of strings anymore, because it seems the color effects entire string
22:23 < ffmdr> iBotPeaches: could you test dark_blue and "dark blue" for me
22:24 < iBotPeaches> ffmdr: at work now, can't be spinning up minecraft windows :)
22:24 < Drainedsoul> I'm pretty sure you can embed objects, so you can do
22:24 < ffmdr> iBotPeaches: :P
22:25 < Drainedsoul> {"text":[{"color":"blue","text":"hello"}," ",{"color":"yellow","text":"world"}]}
22:26 < ffmdr> what about entity names
22:27 < Drainedsoul> I have no idea about entity names, I'm trying to make my chat implementation compliant
22:27 < iBotPeaches> Drainedsoul: I'll give it a shot, too bad its a lot of bloat if you wanna throw a lot of colors together
22:29 < ffmdr> I can do a quick converter from & color codes
22:30 < iBotPeaches> if dark blue == navy, I have navy confirmed working
22:33 < Drainedsoul> is it actually called "navy"?
22:38 < ffmdr> what the heck is "navy"?
22:38 < Drainedsoul> a kind of dark blue
22:38 < Drainedsoul> "navy blue"?
22:40 < Drainedsoul> so are bold/italic/underline/strikethrough/random combinable now?
22:40 < Drainedsoul> like can you set "bold":true,"italic":true ?
22:41 < Drainedsoul> and is it "underline" or "underlined"?
22:42 < iBotPeaches> I'm still working my json thing through, havent made it to those yet
22:47 < Yoshi2> Drainedsoul: if I remember correctly, making italic/bold/etc. combinable is one of the reasons why the new chat system was created
22:49 < Drainedsoul> my biggest problem with this system is that it's a lot harder to make it gracefully fail
22:49 < Drainedsoul> previously if you had a string longer than 32267 characters, you could just slice it down to 32267, check for a trailing section sign, and send it
22:49 < Drainedsoul> what am I supposed to do with this JSON object, just not send it at all?
22:53 < iBotPeaches> Drainedsoul: thats what I'm dealing with :/ (I'm reconstructing the json, except the split text part)
22:53 < iBotPeaches> which is failing still miserably due to invalid json, since I keep accidentally splitting mid json format
22:54 < Drainedsoul> the only thing I can think of, which would be ridiculously expensive, is take your JSON object, reparse it, remove later nodes from it, reserialize it, and see what it serializes to
22:54 < Drainedsoul> either that or start again from scratch and cut out nodes or segments of text
22:54 < Drainedsoul> or just not send overlong serializations at all
22:55 < Drainedsoul> just send {"text":""} or something
22:55 < iBotPeaches> yeah gonna stick with not sending anything, till I have a lot of time
23:24 < Drainedsoul> so with JSON chat, is U+00A7 representable as a literal now?
23:49 < ffmdr> does anyone know good pastebin alternative with delete option which isn't gist?
23:50 < Drainedsoul> you can delete from pastebin?
23:51 < ffmdr> yeah
23:51 < ffmdr> with an account
23:51 < Drainedsoul> yeah
23:51 < Drainedsoul> ? what's the problem with that ?
23:52 < ffmdr> problem is that it is pastebin
23:52 < Drainedsoul> how is that a problem
23:52 < ffmdr> I don't like it
23:52 < Drainedsoul> oh, okay.  I misunderstood your question.
23:52 < ffmdr> pastebin stores password as unstalted md5
23:52 < ffmdr> passwords*
23:53 < Drainedsoul> I just log into pastebin with my facebook : P
23:53 < ffmdr> >_<
23:53 < ffmdr> yeah that perfectly solves the problem
23:53 < Drainedsoul> I'm not sure if that's sarcasm or not.
23:53 < ffmdr> recently pastebin was hacked and all emails/passwords/ips was leaked
23:54 < dav1d> also pastebin is ugly
23:54 < dav1d> use you rown
23:54 < dav1d> ffmdr: https://github.com/Dav1dde/vp
23:55 < ellisvlad> lol
23:55 < ellisvlad> nice
23:55 < ffmdr> meh I know some cool open source ones but I don't want to host my own
23:55 < ellisvlad> or use puush Much better ;)
23:56 < ellisvlad> you can send screenshots, files and text super quickly :P
23:56 < ellisvlad> look
23:56 < ellisvlad> http://puu.sh/3tEoJ.png
23:56 < Drainedsoul> in fairness I could've hosted that screenshot on imgur about as quickly