03:04 < tehme> now i'll try flat array version
03:05 < Drainedsoul> I don't know how the boost multiarray is implemented under the hood.  At the very least it has to have a pointer to some block of memory to hold everything
03:06 < Drainedsoul> it may be an array-of-arrays, which would be very slow indeed, lots of calls to the heap manager, lots of indirections
03:09 < tehme> you are a fujjen genius
03:09 < tehme> fukken*
03:09 < tehme> ~0,01 s per column
03:10 < tehme> i lied
03:10 < tehme> 0.008
03:10 < tehme> thank you very much man
03:10 < Drainedsoul> ah yes, all hail locality and vectorization
03:10 < Drainedsoul> netime
03:12 < tehme> good night
03:12 < tehme> what time is it at your location?
03:13 < Drainedsoul> 18:13
03:13 < Drainedsoul> Pacific Standard Time
03:13 < Drainedsoul> Victoria, British Columbia, Canada
22:27 < tehme> yo su
22:27 < tehme> p
22:27 < tehme> hey Drainedsoul, are you here?
22:29 < tehme> how should i use flat array 256*16*16 as 3d array?
22:29 < tehme> const BlockData& getBlockByRelCoords(int _relX, int _relY, int _relZ) const{return m_columnData[_relY * 256 + _relZ * 16 + _relX];}
22:33 < Not-002> [mc4p] sadimusi pushed 2 commits to master [+0/-0/±2] http://git.io/x4J8lQ
22:33 < Not-002> [mc4p] Simon Marti ede086f - Reverted license of authentication.py to WTFPL
22:33 < Not-002> [mc4p] Simon Marti 7d92a9d - Protocol version 75 (13w36a)
22:35 <+sadimusi> I don't seem to be able to figure out the new byte in 0x3E
22:36 <+sadimusi> it's usually 5 or 6 and when I set it to 1 the sound won't get played
22:36 <+sadimusi> all other values don't appear to have any effect
22:40 <+sadimusi> the source suggests sound category
22:43 < tehme> ask Grum?
22:44 <+sadimusi> where's the fun in that?
22:44 < tehme> lol
22:45 < tehme> i'm having fun with wrong coordinates
22:47 < Not-002> [mc4p] sadimusi pushed 1 commit to master [+0/-0/±1] http://git.io/1FtaMg
22:47 < Not-002> [mc4p] Simon Marti 66a8216 - New field in 0x3E is sound category
22:49 < tehme> omfg x is y
22:49 < tehme> shit
23:03 < Thinkofdeath> sadimusi: Added to the pre-release page http://wiki.vg/Pre-release_protocol
23:15 <+sadimusi> Thinkofdeath: looks good, but could you add 0xc8 as well?
23:18 <+ammar2> oh fyi guys
23:19 <+ammar2> what I did last time for changed packets is
23:19 <+ammar2> I bolded the change
23:19 <+ammar2> its better for readability and easier for when it needs to be moved back to the main Protocol page
23:21 <+sadimusi> ammar2: a few people asked about those fields and nobody here had any idea why they were bold ;)
23:21 <+sadimusi> ammar2: somebody figured it out eventually though
23:30 <+sadimusi> Thinkofdeath: where did you get the categories from btw?
23:31 < Thinkofdeath> sadimusi: adk.class from 13w36b
23:33 <+sadimusi> Thinkofdeath: somehow JD-GUI completely omitted the values() method
23:34 < Thinkofdeath> Yeah it doesn't like enums much
23:38 <+SpaceManiac> Doesn't values() come from enum anyways
--- Day changed sam. sept. 07 2013
01:59 < shoghicp> The Creeper uses the index 16 in its metadata for the fuse
01:59 < shoghicp> is the TNT entity using the same index too?
03:59 < BigBlackBoris> Hey guys, is there a regex to remove everything from the "text" value in the JSON array from the packet 0x03? To get like username: message and not §fusername§f: §fmessage for example?
03:59 <+md_5> hm?
03:59 <+md_5> not sure what you mean
04:00 < BigBlackBoris> This Bukkit server that I'm testing on likes to send §fusername§f: §fmessage in the text value from packet 0x03
04:00 < BigBlackBoris> I need to parse it to get just "username: message"
04:01 <+md_5> like without the colour codes?
04:02 < BigBlackBoris> Without anything alphanumeric. I think Minecraft would really benefit from sending separate values in the JSON array for sender and message 
04:02 <+md_5> so.. without colour codes?
04:03 < BigBlackBoris> Basically.
04:03 <+md_5> BigBlackBoris : (?i)§[0-9A-FK-OR]
04:04 <+md_5> if your regex engine supports (?i) for case insensitivity
04:04 < BigBlackBoris> Is that not the same as what I'm doing here? /\\xA7[0-9A-FK-OR]/i
04:04 < BigBlackBoris> When I print the result of that I get "cConsolef  Place all items in chests before logging off", though.
04:04 <+md_5> what language is this?
04:04 <+md_5> \\xA7 might not be escaped correctly
04:04 < BigBlackBoris> Java using replaceAll
04:05 <+md_5> msg.replaceAll("(?i)§[0-9A-FK-OR]", "");
04:05 <+md_5> works fine for me
04:05 <+md_5> in java
04:05 < BigBlackBoris> let me try that
04:05 <+md_5> you should be using \u00A7 
04:05 <+md_5> not \\xA7
04:06 <+md_5> msg.replaceAll("(?i)\u00A7[0-9A-FK-OR]", "");  <-- works for me
04:06 < BigBlackBoris> Oh, okay. Eclipse told me that wasn't valid. I had to escape the backslash to beable to even run my code
04:06 < BigBlackBoris> Yeah, that works, thanks.
04:07 < Kyle> md_5: another happy customer
04:07 < Kyle> he joined #mindcrack for some reason lol
04:07 <+md_5> Kyle can I get a raise now
04:07 < BigBlackBoris> I joined #mindcrack because I didn't know where else to go xD
04:08 < Kyle> BigBlackBoris: definitely not there, only a few devs in there,
04:08 < BigBlackBoris> I was going to ask on SO but I figured I would ask on IRC first, tried #minecraft and it said I couldn't send the message or something and the only other minecraft-related IRC I knew off the top of my head was Mindcrack
04:11 < BigBlackBoris> Eugh, I don't think this is going to work. I'm trying to make a bot know when somebody chats and there is ranks for each user. If I use .contains anybody could easily send "John: !help" for example
04:12 < BigBlackBoris> there are*
04:12 <+md_5> I dont understand your issue
04:13 <+md_5> also something tells me a plugin would be better for what you want
04:15 < BigBlackBoris> md_5: I'm making a bot that uses the Minecraft protocol to connect to a server and idle, if someone sends a message prefixed with '!' the bot will parse the message, Each user has a role and will only be allowed to use certain commands. The way I plan on doing this is using 'contains' to check if the message contains "name: !help" for example. Maybe it'll be best to make a regex to capture both parts of the ':' like (\w*): (\w*)
04:16 < BigBlackBoris> third-party servers, I can't install plugins on them
04:16 <+md_5> why not
04:16 <+md_5> just split it
04:16 < BigBlackBoris> by the delimiter ':'?
04:16 <+md_5> String[] split = message.split(":",1); String username = split[0]; String message = split[1];
04:18 < BigBlackBoris> ok, thanks. I'll try that. Just out of curiosity, if servers have some sort of plugin like ChatManager to change the format of the message such as [Rank] [Name]: Message does that modify the text value in the JSON?
04:18 < BigBlackBoris> if yes, i'm going to have to rethink all my logic
04:18 <+md_5> of course
04:18 <+md_5> think about what your client sees
04:19 <+md_5> thats what you bot sees
04:19 < Kyle> md_5: this is what we told him lol
04:19 <+md_5> have you actually coded a bot yet
04:19 <+md_5> because the minecraft protocol
04:19 <+md_5> is pretty damn complicated
04:19 <+md_5> and if you can figure it out
04:20 <+md_5> I really think youshould be able to figure these trifles out
04:20 < Kyle> complicated? not for just parsing chat
04:20 < BigBlackBoris> I have the protocol all down and sorted. It connects to the server, can send messages, walk around
04:21 <+md_5> then how can you not figure this out
04:21 <+md_5> o.O
04:21 < BigBlackBoris> What I don't understand is, why use a JSON array for one value -_-. Would it not be easy for Mojang just to implement some extra values like Bukkit does? Like a message value and a sender value?
04:22 < BigBlackBoris> I can't figure this out because there's an infinite number of combinations that the format can be. I need to get the author and the message. What if the format was like [Admin] [John] Message, or John [Admin]: Message
04:23 <+md_5> dude
04:23 <+md_5> thats up to the server to decide
04:23 <+md_5> the server can send whatever it likes
04:23 <+md_5> eg I can make my server be
04:23 < Kyle> my question
04:23 <+md_5> %%%% OVERLORD %%% *** md_5 *** ::: This is My test message :::
04:23 < Kyle> why not just calibrate it per server?
04:23 < Kyle> have a regex configurable for each server it can connect to.
04:25 < BigBlackBoris> I'm trying to make it be as user friendly as possible without using any hacky solutions
04:25 <+md_5> dude
04:25 <+md_5> its a fact of life
04:25 < Kyle> it's minecraft
04:25 < BigBlackBoris> Like just running ./run 127.0.0.1 25565
04:25 <+md_5> that servers have diffferent chat styles
04:25 < Kyle> there *are no* hacky solutions
04:25 <+md_5> you cannot expect every server to be vanilla
04:25 < Kyle> ;)
04:25 <+md_5> plugins exist FOR customizing chat
04:27 < BigBlackBoris> I know, but it would be easy as shit for Mojang to just send the message and author as values in the JSON. It wont break any existing code and it will actually make use of the JSON. Dinnerbone's "new" chat system uses JSON for one value. Why not just send it plaintext? What's the point in using JSON for one value? It makes no sense to me
04:28 < BigBlackBoris> All the code is right there and ready for them to use. They just need to implement it
04:28 <+md_5> BigBlackBoris VANILlA DOES
04:28 <+md_5> BUKKIT DOES NOT
04:29 <+md_5>  / this server does not
04:29 < BigBlackBoris> sooo, why doesn't bukkit?
04:29 <+md_5> becaus
04:29 <+md_5> e
04:29 <+md_5> that 
04:29 <+md_5> would
04:29 <+md_5> break
04:29 <+md_5> every
04:29 <+md_5> single
04:29 <+md_5> plugin
04:29 <+md_5> in
04:29 <+md_5> existance
04:29 <+md_5> and
04:29 <+md_5> they would need to all be updated / patched/ rewritten
04:29 < BigBlackBoris> no
04:29 < BigBlackBoris> it
04:29 < BigBlackBoris> would
04:29 < BigBlackBoris> not
04:30 <+md_5> dude
04:30 < BigBlackBoris> explain how it would break every plugin?
04:30 <+md_5> because not EVERY message sent has a username
04:30 <+md_5> tell me how
04:30 <+md_5> you can make sendMessage(String)
04:30 < BigBlackBoris> send a blank author
04:30 <+md_5> suddenly include an arbitrary set of key value params
04:30 <+md_5> *head desk*
04:30 <+md_5> that defeats the entire purpose
04:31 <+md_5> you clearly do not think in the big scheme of things
04:31 < BigBlackBoris> I'm thinking as hard as I can
04:31 <+md_5> then every message becomes
04:31 <+md_5> <> test message
04:31 <+md_5> with blank author
04:31 <+md_5> do what we told you
04:31 <+md_5> and use a regex
04:31 < BigBlackBoris> Sending two extra values in a JSON array would break fuck all.
04:31 <+md_5> the vanilla translation files only contain <%s> for usernames anyway
04:31 <+md_5> so you would STILL need custom text
04:31 <+md_5> to get custom formatting
04:32 < BigBlackBoris> Exactly my point
04:32 < BigBlackBoris> Keep the text value
04:32 < BigBlackBoris> just send message and author along with it
04:32 <+md_5> that makes 0 sense
04:32 <+md_5> please leave now
04:32 < BigBlackBoris> No it fucking doesn't
04:32 < BigBlackBoris> 0x03 is a JSON array
04:33 < BigBlackBoris> with one value
04:33 < BigBlackBoris> "text"
04:33 < BigBlackBoris> send values "author" and "message"
04:33 < BigBlackBoris> explain how that would break anything
04:33 < BigBlackBoris> please, i'd love to know
04:38 < BigBlackBoris> md_5 http://wiki.vg/Protocol#Chat_Message_.280x03.29
04:42 < iBotPeaches> It would break everything
04:42 < iBotPeaches> and with that, im going to bed.
04:44 <+Amaranth> BigBlackBoris: See how the example has all of that other fancy stuff for translations and sending the text in parts?
04:44 <+Amaranth> That doesn't really work when the interface given to plugins is "this is the chat message going to the player"
04:45 <+Amaranth> And changing that interface means breaking the plugins
04:45 < BigBlackBoris> no, i mean, only send the author value when it's a player message
04:46 <+Amaranth> So send the whole raw text _and_ send the author?
04:46 < BigBlackBoris> yes but as different values in the array
04:46 <+Amaranth> Because the plugin interface it a literal "this is the raw text going to the client" not "here is the part the player wrote and we'll fill in the rest"
04:46 <+Amaranth> That is a breaking change
04:47 < BigBlackBoris> so what you're saying is it will only break chat modifying plugins
04:48 < BigBlackBoris> not every single plugin like md_5 said
04:48 <+md_5> dude
04:48 <+md_5> explain
04:48 <+md_5> how
04:48 <+md_5> you can alter the chat sending system
04:48 <+md_5> to use the new style
04:48 <+md_5> without breaking everything
04:48 <+md_5> you CANNOT
04:48 <+md_5> the new system doesnt even have colour codes
04:48 < BigBlackBoris> you managed to fucking do it when dinnerbone came along and changed everything to fucking JSON
04:48 <+md_5> rendering the ChatColor enum useless
04:49  * md_5 hits sadimusi and TkTech with a block
04:49 < iBotPeaches> The new JSON change took about 1.5 months for me, not something I want to do again
04:49 < iBotPeaches> not to mention it was borked and didn't have support for nested colors until a month later
04:49 <+Amaranth> BigBlackBoris: We managed it by pretending the new system is the old system and only using the text value in the JSON instead of the full capabilities :P
04:49 < iBotPeaches> Amaranth: I hope they never remove that :p
04:49 <+Amaranth> Well that's not quite true but close enough
04:49 < iBotPeaches> I just send all the stuff to old text field, formatted in old way :)
04:50 <+Amaranth> But the interface plugins are given is "here is the message the player is sending" and "here is the format string <%s> %s"
04:50 <+Amaranth> And the plugin is allowed to modify both in any way it wants and the server then combines them (if still applicable) and sends them off
04:51 <+Amaranth> Sending author and message separately and relying on the client to do the <%s> %s part doesn't let plugins put in prefixes and such
04:52 <+Amaranth> So yeah, changing to what you want would not only be a breaking API change for any plugin that is looking at chat it would also remove the ability to customize chat
04:55 < BigBlackBoris> well this is annoying
05:28 < ackpacket> So, if i'm understanding this correctly, a man in the middle attack will not work for a minecraft session stealer because a mitm involves intercepting keys and replacing with the attacker's keys to remain part of the conversation, but the mc client and server  authenticate with mohjang using a hash from the keys they receive, and a mitm will cause them to have different keys
05:29 < ackpacket> so they will contact mohjang with diferent hashes
05:42 <+md_5> ackpacket yup
05:43 < ackpacket> Well, that's good to know
05:43 < ackpacket> it would have been fun to make a session stealer, but i suppose the world is better off without one
05:56 <+md_5> ackpacket thats why encryption was implemented :p
06:09 < Drainedsoul> couldn't you alter a chat sending system without breaking anything by just implementing a parser?  Parse the old style to some intermediate representation, and then transform that into JSON
06:11 < Drainedsoul> having code just send string literals with the formatting codes in them though just smacks of a poorly-designed abstraction.  Code consuming an API shouldn't have to understand the way that API represents or transmits the data, that's the point of an API
11:30 < winny> http://jsfiddle.net/42EDX/10/
11:30 < winny> ^ to find a stronghold
13:36 < Morrolan> winny: Heh, nice.
13:36 < Morrolan> (Though I have to admit that I prefer just using the seed to calculate the positions :P)
13:36 < winny> where can I read about deriving stronghold positions from the seed?
13:40 < winny> are you referring to Skidoodle's work?
13:58 < Morrolan> I do not actually know how it's done. I've just stumbled upon a tool which calculated stronghol's position based on the seed, a while ago.
13:58 < Morrolan> AMIDST, yea, Skidoodle it is.
14:14 < shoghicp> Got explosions working on PocketMine-MP :D http://imgur.com/a/szAz5
14:22 <+sadimusi> shoghicp: that doesn't look right
14:25 < shoghicp> sadimusi: big explosions
14:25 < shoghicp> with power 80
14:25 <+sadimusi> cubic explosion?
14:25 < shoghicp> same as big PC Minecraft explosions
14:26 < shoghicp> let me find an image
14:26 < shoghicp> Sethbling had one in his videos
14:26 <+sadimusi> and what's up with those stripes?
14:26 < shoghicp> sadimusi: it is a single explosion, not multiple TNT blocks
14:27 < shoghicp> sadimusi: rays
14:27 <+sadimusi> I assumed as much
14:27 < shoghicp> http://img404.imageshack.us/img404/2990/20121103230905.png
14:28 < shoghicp> I'm using the same algorithm bukkit / mc uses
14:28 < shoghicp> http://www.youtube.com/watch?v=8ow5GzfM_dI
14:28 < shoghicp> just at the start
14:28 <+sadimusi> I guess that works for smaller explosion radii
14:29 < shoghicp> yep
14:29 < shoghicp> they look like vanilla
14:29 < shoghicp> TNT power 4, this was a power 80 explosion
14:35 < shoghicp> sadimusi: normal TNT http://i.imgur.com/wnRxBqa.png
14:35 <+sadimusi> that look more like it
14:39 < shoghicp> force 20: http://i.imgur.com/amSJqWd.png
14:39 < shoghicp> power*
20:33 < BigBlackBoris> Has anyone got any experience with https://github.com/Steveice10/mc-protocol-lib ?
20:33 < BigBlackBoris> I've decided to use that instead of using the one I made
20:35 < BigBlackBoris> I can't seem to send movement packets
20:36 <+pdelvo> if you show us some code we might be able to help
20:37 < BigBlackBoris> I have no code to show. I can't get it to work. I'm checking if the packet is 0x0D in my Listener in onPacketRecieve, if it is run onPositionLook((PacketPlayerPositionLook) packet);
20:37 < BigBlackBoris> in there I'm trying to change x and z and then using client.send to send the packet but it doesn't seem to send to the server
20:41 <+pdelvo> have you tried a proxy to check if it sends it?
20:42 < BigBlackBoris> how would I go about proxying requests?
20:45 < iBotPeaches> https://github.com/sadimusi/mc4p
20:45 <+pdelvo> you can use mc4p: https://github.com/sadimusi/mc4p
20:45 <+pdelvo> lol :D
20:45 < iBotPeaches> :p
21:04 < BigBlackBoris> error: Setup script exited with error: Unable to find vcvarsall.bat
23:22 < TkTech> BigBlackBoris: You'll need to follow a python + windows guide.
23:22 < TkTech> BigBlackBoris: You're either missing VC++ or haven't finished setting everything up.
23:26 < BigBlackBoris> I've done it, I had to install it on Ubuntu though. I wouldn't get any packets sending apart from chat so I gave up
23:31 < Drainedsoul> Ugh, VC++, pls more MinGW :(
23:42 < dav1d> :(
23:42 < dav1d> I wonder why I needed a BIOS reset :(
23:43 < dav1d> luckily I guessed, that it's a BIOS issue. still damn windows 8, I blame YOU
23:47 < Drainedsoul> why are you using Windows 8, I think that's the more notable thing here ._.
--- Day changed dim. sept. 08 2013
00:21 < dav1d> Drainedsoul: got it cheap and yay CS:GO works like a charm
00:22 < dav1d> since the last steam update steam was unusable with my XP
00:22 < dav1d> also, everyone said it's bad, but I never really used it
00:22 < dav1d> now I can officially confirm, it sucks
00:22 < dav1d> I really like how you shutdown windows 8, you go into deksotp mode, press windowskey+r, open cmd
00:22 < dav1d> shutdown -s -t 00
00:28 < Drainedsoul> why didn't you just go to win7 ._.
00:55 < Drainedsoul> anyone have an idea of how long it takes the vanilla server to save a chunk or column to disk?  Or how I could go about benchmarking this myself?
07:25 < ahabs_ghost> sorta n00b question... what's [???] Invalid argument
07:26 < ahabs_ghost> i know python, i've read some of the mark2 code...
07:26 < winny> happens in socket programming, I don't recall exactly why
07:27 < ahabs_ghost> it fires about every 5 secs. not on my laptop, but on a "server"
07:27 < ahabs_ghost> is it something i can suppress via python logging config?
07:28 < winny> usually invalid argument is something worth investigating further
07:28 < winny> can represent a bug
07:30 < ahabs_ghost> it could easily be a permission thing with the setup... pip install of mark2 as root, but running as regular user?
07:33 < ahabs_ghost> oh, it looks like the previous one i had going under tmux hadn't acually stopped
07:39 < ahabs_ghost> that wasn't it... there isn't a lot of memory on this critter. could be lots o problems
07:39 < Drainedsoul> does the client actually do anything with the first field in 0x04?
07:39 < TkTech> ahabs_ghost: Is there a stack trace with that message?
09:21 < Drainedsoul> how does the client derive the phase of the moon
16:51 < ahabs_ghost> TKTech: no, no stack trace.
17:10 < ahabs_ghost> interesting, the service ping interval is 5 seconds, .. "grep -r 5 *" is a crude approach, but...
17:29 < ahabs_ghost> looking like maybe configuration/permissions error... i set up a new user to run the server, but a virtualenv as my own user
17:33 < ahabs_ghost> nope, i had been using MARK2_CONFIG set to a local dir
17:34 < ahabs_ghost> but changing the ping interval to 6 seconds shows that is the source of the Invalid Argument
17:53 < ahabs_ghost> hmm. server.log appears to have this:  [SEVERE] Reached end of stream for /127.0.0.1
17:56 < ahabs_ghost> and on shutdown i do have a java trace...  [???] java.net.SocketException: Socket closed
17:59 < ahabs_ghost> feels like a jdk related problem, but hitting a bug in the logging...
18:00 < ahabs_ghost> well, thats all for now, it's a great looking system, ill just have to get back to it another day
20:27 < Not-002> [mineflayer] zuazo pushed 4 commits to master [+0/-0/±4] http://git.io/qUJsUA
20:27 < Not-002> [mineflayer] pratham2003 03c57bf - Update entities.js Added swing boolean parameter. function attack(target, swing)
20:27 < Not-002> [mineflayer] pratham2003 1875f8d - Update entities.js Added swing parameter to attack function.
20:27 < Not-002> [mineflayer] pratham2003 cb72781 - Fixed indentation.
20:27 < Not-002> [mineflayer] zuazo 43dd5a3 - Merge pull request #160 from pratham2003/master Added boolean swing arm parameter to attack
21:15 < Not-002> [fCraft] fragmer * r2203 4 files : Every log message now contains a 2-letter code identifying the log type (e.g. RC for RankChat, EE for error, IC for IrcChat, etc). Changed capitalization of IrcChat and IrcStatus logtypes (should not affect anything - config has case-insensitive parsing for this stuff).
21:31 < Not-002> [fCraft] fragmer * r2204 6 files : Separated BuildingCommands from DrawCommands, since that class was getting way too huge.
23:06 < tehme> yoyoyo i finally made world representation that works
23:06 < tehme> http://imgur.com/6xls0Yy
23:06 < tehme> column height map
23:08 < tehme> http://imgur.com/Okx77EV
23:08 < tehme> og this place
23:08 < tehme> of*
23:19 < shoghicp> tehme: my client, early development stage: http://imgur.com/a/G32k7
23:38 < tehme> shoghicp: wow
23:38 < tehme> how early?
23:42 < Kassick> hey guys
23:43 < Kassick> has anyone ever tried netcatting into an MC server? I can't seem to figure this stuff out myself
23:45 < Kassick> My only goal here is to send a player list query to the server and get back the number of players online
23:46 < shoghicp> tehme: one year ago
23:46 < Thinkofdeath> Kassick: http://wiki.vg/Server_List_Ping might help
23:46 < tehme> i have a question
23:47 < Kassick> Thinkofdeath: I've been looking at that page but I feel like I may be doing something else wrong
23:47 < tehme> 0x38 is sent when player is teleported to admin
23:47 < tehme> wiki says:  you send this packet to update area surrounding player in the middle of the game, entities within the updated area may become unattackable
23:47 < tehme> oh
23:48 < tehme> i got it
23:48 < Kassick> Thinkofdeath: It's hex, so I'm assuming I need to escape it or format it in someway for NC to accept it
23:48 < Kassick> I get no response back and the server kills the connection whenever I try it
23:49 < tehme> shoghicp: do you plan 3d?
23:50 < shoghicp> tehme: It was a PHP client, but it is outdated since 1.5
23:50 < shoghicp> You can add anything to it
23:50 < tehme> why don't you update it?
23:51 < shoghicp> We used it on a server here in SPain to manage users, catch X-Rayers, FAQ bot, Player vs Bot, etc
23:51 < shoghicp> but I lacked the time, and somewhere it is failing on 1.5
23:51 < shoghicp> even after updating it
23:52 < tehme> shoghicp: how did you catch x-rayers?
23:54 < shoghicp> the bot ran commands to get block break stats, and created a list of the most ovbious ones. Then, when it was in the "mining" world, teleported to it being invisible and followed him analyzing the path followed (straight line, or putting fake blocks and logging what he did)
23:54 < shoghicp> then, it made a table with all the data, and humans reviewed it
23:55 < shoghicp> we catched ~5 per day, and the bot warned us when someone was in the action
23:55 < shoghicp> so we teleported too there
23:56 < shoghicp> it was really nice :D
23:56 < tehme> wow
23:56 < tehme> yep, some of them are easy to catch
--- Day changed lun. sept. 09 2013
00:19 < tehme> it is time-consuming, but if you want it, you can break anything having < 15 distance from wall to private end
00:20 < Kassick> Thinkofdeath: nevermind, I got it working, thanks
00:24 < woder> quick question on the way to auth for a client, what is "profile id" in token:<accessToken>:<profile ID>
00:30 < tehme> yo i have a question. server sends Block Change when vine grows. it seems to replace block type with type of vine
00:31 < tehme> is block being replaced, or just attached somehow?
00:31 < tehme> how do server and client know what block vine is on?
00:32 < tehme> and i saw replacement of air with air
00:32 < tehme> what the hell is this?
02:07 <+sadimusi> Kassick: you have to wait 5 seconds before connecting again
02:12 < Kassick> sadimusi: netcat wasn't staying open long enough to receive a response back
02:12 <+sadimusi> oh, ok then
08:09 < Shagrat> Hello
08:11 < Shagrat> Please http://wiki.vg/Pre-release_protocol 0xC8 explain format. Bad description
08:20 < Shagrat> Sorry understood
08:20 < Shagrat> 0xC8 = 0 - is reset statistic?
08:23 < Drainedsoul> what's your question?
10:45 <+Amaranth> clonejo: I thought I recognized that nick :D
10:46 <+clonejo> Amaranth: same here, I just wasn't quick enough to be the first to answer (mobile connection)
10:52 <+md_5> I feel I'm missing some context
10:52 <+clonejo> md_5: Amaranth discovered me in #rust@irc.mozilla.com
10:53 <+md_5> bloody rust
10:53 <+md_5> stupid language
10:53 < eddyb> lol
10:54 < eddyb> my biggest complaint is the Ruby-like block lambdas
10:54 <+md_5> lets just not talk about go
10:55 <+md_5> all these new fangled languages make me sad
10:55 < eddyb> Rust > Go. it just needs more love
10:57 < eddyb> it's typed, yet you don't see any explicit types in the example here http://www.rust-lang.org/
10:57 < eddyb> apart from the damn lambdas, it's pleasant to me, like ES6
10:57 < eddyb> and it doesn't go to the crazy end of the rainbow, like Haskell
10:58 <+md_5> I should really really learn another language proficiently
10:58 < t92> python
10:58 <+md_5> nothnks
10:58 < t92> lolwhy
10:58 <+md_5> realistically probably C++
10:59 < eddyb> lol python
10:59 <+md_5> since I can tie it in nicely with java
11:00 < eddyb> md_5: I hope you don't end up hating C++, like all the people who don't understand it
11:01 <+md_5> I wish there was a language which had ides as good as java :p
11:01 < t92> c#?
11:01 <+md_5> also dynamically typed languages are poopy
11:01 <+md_5> count them all out
11:02 < eddyb> kdevelop with C++, if you want to give it a try
11:02 <+md_5> netbeans with c++ m8
11:02 <+md_5> its actually really good
11:02 < eddyb> it has C++11 support now and it's almost as good as a Java IDE (which are good because Java is crap)
11:03 < t92> learn c/c++ md_5 
11:04 < eddyb> C/C++ is a bad thing to say
11:04 < SinZ> learn brainfuck md_5
11:04 < jast> can't wait to see Rust go stable
11:05 < t92> learn both though
11:05 <+md_5> it sucks that I dont know enough C that I have to get someone else to write this native code for me: https://gist.github.com/md-5/16d0bc3cdf2a8c1cbf95
11:05 <+md_5> thats mainly cause JNI is confusing though
11:05 < eddyb> C is pretty much useless these days
11:05 < t92> not true
11:05 <+md_5> C(++)[11]
11:05 < jast> nah
11:05 < eddyb> the only reason to use C is because bindings, and that's some fucked up thing
11:06 < t92> c++ abi's are not consistent so c is used
11:06 < jast> I have to admit that C++ is slowly becoming usable
11:06 < jast> as long as I don't have to use the STL
11:06 < SinZ> md_5: purebasic
11:07 < eddyb> jast: I've worked in a hobby OS project with a C++ kernel :D
11:07 < jast> good for you
11:08 < eddyb> apart from the lack of a simple C++ FFI, it's strictly superior
13:06 < tehme> oh, i see discussion about c++ *slowpoke.jpg*
13:48 < piracetam> **:00:55         eddyb | C is pretty much useless these days
13:48 < piracetam> air is pretty much useless these days
13:50 < piracetam> md_5: java is emulation of language, not language
13:50 < piracetam> you don't actually code program, you just describe jit how to code the program
13:50 < piracetam> unnecessary abstraction layer
13:51 <+sadimusi> piracetam: please don't start another senseless language debate
13:52 < dx> air is useless since adobe dropped linux support
13:52 < SinZ> ^
13:52 < eddyb> dx <3
13:53 < t92> i breathe air
13:53 < t92> .-.
13:53 < piracetam> sadimusi: ok, but he started it - "**:57:08         +md_5 | I wish there was a language which had ides as good as java :p"
13:53 < piracetam> t92 ++
13:55  * Fador is coding HEVC encoder in C89 and Assembly at work
13:55 < dx> Fador: fun!
13:55 <+Fador> dx: sure ;)
13:56 < dx> Fador: so you're in charge of making it suck less than the reference implementation?
13:56 < dx> i know very little about it, but mostly heard that it's slooooow
13:56 <+Fador> dx: well there's also that x265 project ;D
13:57 <+Fador> we are mostly doing research on HEVC..
13:57 < dx> also fun!
13:58 <+Fador> https://bitbucket.org/multicoreware/x265/wiki/Home <- they also use the reference implementation as the base of their encoder
16:10 < Drainedsoul> Why would you use C++ and not use the STL
16:32 < tehme> great c++ adept is here!
17:31 < TkTech> Drainedsoul: Because internally most STL implementations are a mess, some have obscure bugs (looking at you VC++), often don't perform optimally, platform inconsistencies, and debugging them is a total bitch.
17:38 < eddyb> "VC++" it's not like you're supposed to use it :P
17:41 < tehme> oh lol resease is turbomode
22:22 < Drainedsoul> why are you debugging the STL unless you're an STL developer?
23:08 <+clonejo> That moment when you read man pages and you know the people in the author's section
23:18 <+clonejo> s/author's/authors/
23:19 < brainiac94> greetings
23:42 < tehme> yoyoyo
23:45 < brainiac94> have you heard of the authentication glitch in CB?
--- Day changed mar. sept. 10 2013
00:04 < tehme> tell us
00:44 < brainiac94> I cannot tell you much about how it works, but it abuses 0x02 packets and allows the attacker to login with any user name.
00:44 < brainiac94> It works on a CB without plugins and the only evidence it leaves is this line in the server log:
00:44 < brainiac94> class net.minecraft.server.v1_6_R2.PlayerConnection wasn't prepared to deal with a class net.minecraft.server.v1_6_R2.Packet2Handshake
05:36 < Drainedsoul> can anyone give me some details on time?
05:36 < Drainedsoul> i.e. at what tick is midnight?
08:34 < benbaptist> Drainedsoul, I believe 12000 is the precise midnight time, if that's what you mean. 
10:14 <+md_5> hm
10:15 <+md_5> has anyone here ever used any 'other' protocols, for networking, ie not udp/tcp
10:15 <+md_5> or some framework ontop of udp with support for ack'ing + connections
11:50 <+md_5> jast yeah, or UDT
11:50 <+md_5> SCTP too
14:31 < amadornes> Hello. Is there any way to deactivate "java.cli.D.jline.terminal=jline.UnsupportedTerminal" and activate "--nojline" from Bukkit/Spigot? It makes my server use 40% CPU vs the 2% with the Bukkit "--nojline"
14:52 < amadornes> Any ideas?
18:58 < tehme> http://www.youtube.com/watch?v=Fqx7zQJvLVU
19:12 < Kaspars_Raoulfr> Hello everyone, Kaspars_Raoulfrhere and i'm here and ready to DDoS.
19:19 < tehme> sup
19:31 -!- Kaspars_Raoulfr was kicked from #mcdevs by TkTech [Your behavior is not conducive to the desired environment.]
--- Day changed mer. sept. 11 2013
04:05 < Serin_> Hi
04:06 < Serin_> What's the point of transactions for the UI?
04:06 < Serin_> They seem to be completely meaningless
04:06 < Serin_> Nothing responds to a success/failure on transactions, the server just follows up with 0x68 SetSlot to sync the inventory state
04:07 <+sadimusi> doesn't 0x6A get sent anymore?
04:08 < Serin_> It's sent, but nothing actually uses it
04:08 < Serin_> Nothing happens regardless of what data is in that packet
04:08 < Serin_> Afaik
04:08 <+sadimusi> iirc the server doesn't send any 0x68 for most transactions
04:08 <+sadimusi> it was originally meant to handle conflicts when multiple players access the same container
04:08 <+sadimusi> but it might not be used anymore
04:09 < Serin_> >X might not be used anymore
04:09 < Serin_> >minecraft protocol in a nutshell
04:09 <+sadimusi> pretty much
04:12 <+sadimusi> yeah, it really looks like it just sends back a packet in some cases
04:12 <+sadimusi> and does nothing else
04:13 < Serin_> There's this on the server side, but I'm not really sure what the point is
04:13 < Serin_> this.playerEntity.openContainer.setPlayerIsPresent(this.playerEntity, true);
04:13 < Serin_> You would think it would be inferred which players are there by which ones you let open the inventory in the first place...
04:15 <+sadimusi> playerEntity.openContainer is always a container the player has opened, so I'm not sure either what it does
04:17 < Serin_> Oh well, my implementation was already working, I just wanted to double check that doesn't do anything
04:17 < Serin_> Thanks for your help
04:17 < Serin_> I have one other question
04:17 < Serin_> What actually produces the beacon beam effect?
04:17 < Serin_> Is that a BlockAction?
04:17 <+sadimusi> isn't it just the block itself?
04:18 < Serin_> No
04:18 < Serin_> You have to make a stack of iron/gold/diamond blocks and set it up first
04:18 < Serin_> Otherwise there's no beam
04:18 < Serin_> Block action would be my shot in the dark, but if it is it's not listed on wiki.vg
04:18 < Serin_> Just thought I'd ask while I happen to be here
04:18 <+sadimusi> well it's certainly easy to find out if you're right
04:20 < Serin_> Doesn't look like a block action, I'll look into it real quick
04:22 <+sadimusi> definitely not block action
04:25 <+sadimusi> might be 0x84
04:26 <+sadimusi> yes, with action set to 3
04:26 <+sadimusi> I assume the effects are listed as NBT in the data field
04:27 < Serin_> Yep, that's it
04:28 < Serin_> Do you know of a decent program to view Minecraft traffic in human readable form?
04:28 < Serin_> I've been dumping it with an MCP mod and reading the dump with a Python parser on shell
04:30 <+sadimusi> there are a few proxies
04:30 <+sadimusi> I use and maintain mc4p, but its log plugin is still not that great
04:30 <+sadimusi> there's also SMProxy
04:30 <+sadimusi> not sure if that's up to date though
04:31 < Serin_> I think it isn't
04:31 < Serin_> Seems to be 1.6.1
04:31 < Serin_> Shouldn't be hard to update though
04:31 <+sadimusi> what os are you using?
04:31 < Serin_> Windows on my dev machine
04:32 < Serin_> With Cygwin
04:32 <+sadimusi> then smproxy might be a bit easier to install
04:32 < Serin_> What's mc4p written in?
04:32 <+sadimusi> python
04:32 < Serin_> Python is my main language, so that's preferable
04:32 <+sadimusi> https://github.com/sadimusi/mc4p
04:32 < Serin_> I was under the impression that was designed for servers
04:33 < Serin_> Neat
04:33 < Serin_> Thanks
04:33 <+sadimusi> I'm not sure what the original author designed it for
04:33 <+sadimusi> but I think it would be too slow for a decent server
04:34 < SinZ> SMProxy can work for servers iirc
04:34 < SinZ> I doubt mc4p can though
04:35 <+sadimusi> SMProxy's scripting capabilities are probably not rich enough for a server
04:36 < SinZ> yoshis work on making mc4p work for connecting to classic servers via smp client is nice though
04:36 < SinZ> shame the conversion of maps take time
05:02 < Serin_> for that matter
05:02 < Serin_> is there any reason to have window IDs besides the separation of player inventory from everything else?
05:02 < Serin_> from open/close window packets server knows what window object is open, that seems unnecessary
06:29 < benbaptist> My string packing implementation seems to be really flakey, it seems it crashes the game sometimes at random when sending 0x03 packets
06:30 < benbaptist> Are the actual sizes of strings always double the "length" value (the short you receive before a string), or is it not consistent?
06:32 < Serin_> it uses two bytes for each character
06:32 < Serin_> so length will always be the number of characters * 2
06:32 < benbaptist> yeah
06:32 < benbaptist> I wasn't sure if maybe some characters were only one byte at times or not
06:33 < benbaptist> I noticed that occasionally the client crashes with a JSON error while sending the 0x03 packet despite my JSON being completely fine, and while I receive the chat messages just fine, sometimes various strings from the client such as the 'plugin message' packet are cut off. 
06:34 < benbaptist> e.g. when the client sends the packet for polling the MOTD and stuff, the channel name is "MC|PingHos" with the t cut off
06:34 < benbaptist> I'm in python using the struct module, not sure if .unpack('>h') is enough or not to receive the short or if something else is broken
06:51 < libminecraft> [libminecraft] yuri-sevatz pushed 1 new commit to master: https://github.com/yuri-sevatz/libminecraft/commit/6a2b02fb42e666a5b5b785adcf6151adba870098
06:51 < libminecraft> libminecraft/master 6a2b02f Yuri Sevatz: Add gitignores for KDevelop projects
07:05 < SinZ> wow, a github bot, rather than notifico
07:10 < dx> what's so "wow" about it?
07:11 < SinZ> suprising, for this channel atleast
07:11 < dx> oh
07:11 < dx> it's just someone who didn't read the channel rules, that's not surprising to me :D
14:06 < Yoshi2> SinZ: I think the conversion of classic maps into SMP format is slow because of my pure python code
14:35 < barneygale> Looking for advice on a problem I've never tackled before. I want to add incremental backups of minecraft worlds to a new server wrapper I'm writing. Any advice on a scheme? Simplest would be per-region, but per-chunk might also be possible if I used the first n bytes of the compressed data as a checksum, maybe?
14:36 < barneygale> wondering about file format, options/capabilities/config
14:37 < dx> http://rdiff-backup.nongnu.org/ ?
14:39 < barneygale> uses hardlinks iirc, which would limit it to per-region
14:39 < barneygale> that might not be so bad though
14:40 < barneygale> ah maybe I'm getting mixed up, nvm that ^
14:40 < dx> yeah, not sure what you were thinking :D although it is sort of limited per-region
14:40 < dx> per-file, rather
14:45 < dx> (i haven't actually used this by itself, instead i regularly overwrite my remote backups with rsync, which is the same thing for network transfer but without saving incremental backups)
15:38 < Blagun_Cibran> Hello everyone, Blagun_Cibranhere and i'm here and ready to DDoS.
15:39 -!- Blagun_Cibran was kicked from #mcdevs by sadimusi [Your behavior is not conducive to the desired environment.]
16:08 < shoghicp> Who was that?
16:09 < shoghicp> let me guess, LOIC?
16:09 <+sadimusi> another one of those bots
16:09 <+sadimusi> possibly
16:10 < shoghicp> :s
16:14 <+sadimusi> shoghicp: weren't you around when this happend? http://d.pr/n/jbfS
16:14 < shoghicp> lol, no
16:15 < shoghicp> Litespeed was the manager?
16:15 <+sadimusi> yes
16:15 <+sadimusi> and Sonjay
16:17 < shoghicp> :s
16:18 < shoghicp> That reminds me of people evading a ban on #mcpedevs to request the lif of their ban
16:18 < shoghicp> lift*
23:45 < ZenovaCreations> you should check out my new animation
23:45 < ZenovaCreations> http://www.youtube.com/watch?v=Zx5Mngfsezo
--- Day changed jeu. sept. 12 2013
00:15 <+Prf_Jakob> TomyLobo: http://i.imgur.com/px6Z1za.gif
00:15 <+Prf_Jakob> err
00:15 <+Prf_Jakob> ops
00:15 <+Prf_Jakob> TkTech: ^
00:26 < SinZ> wow
11:59 < tehme> yo
11:59 < tehme> help me please
11:59 < tehme> what is the general word for widht and height? bounds?
11:59 < tehme> width*
11:59 < jast> it's the word I'd use, at least
12:02 < zutto> dimensions?
12:03 < zutto> bounds, boundaries, dimensions.. someone can probably toss in few others
12:05 < tehme> well, let it be "bounds" for now
12:06 < tehme> programming is not building, change is easier:)
12:06 < zutto> i'm more used to "dimensions" word, but bounds is fine if you feel comfortable with it
12:08 < eddyb> "bounds" make sense in a context such as AABB (axis-aligned bounding box)
12:54 < winny> 10 DIM ARY(10)
16:08 < tehme> http://imgur.com/kN8MpUj
16:09 < tehme> first attempts to make interface
16:09 < tehme> with lua api
16:10 < eddyb> lol, lua
16:17 < tehme> yep
16:18 < tehme> do you like it?:)
21:53 < breezeyboy> Hi
21:54 < breezeyboy> How do I change the minecraft version?
21:59 <+pdelvo> of what?
21:59 < breezeyboy> the server
22:01 < dx> use a hex editor and replace all entries of 1.6.2 with 1.7
22:01 < dx> (that was a dumb joke don't take it seriously)
22:02 < dx> also, hi #mcdevs, i was away because my internet connection sucks today. did i miss any discussion on the new json chat extensions?
22:03 < dx> i sure wish the 'public logs' were actually public
22:07 < breezeyboy> No help?
22:08 <+AndrewPH> dx: I don't think you missed anything at all
22:08 < dx> cool, thanks.
--- Day changed ven. sept. 13 2013
02:16 < mbaxter> sadimusi: ammar2 tells me you're the one to poke about b.wiki.vg not having latest snapshot. So uh, poke. :)
02:17 <+sadimusi> mbaxter: gr.um refactored quite a lot of the codebase, it'll take some time before burger works again
02:18 <+sadimusi> but if you're just interested in the protocol, that part still works fine
02:19 <+sadimusi> the protocol didn't change in the snapshot today
02:20 <+sadimusi> however, the version is now 76
02:22 < tehme> omg changes changes
02:22 < tehme> my client will never be finished
02:23 < Kyle> tehme: the reality of minecraft eh
02:23 < tehme> i hope they will add message size field
06:37 < SinZ> I quite like the new stuff we can do in the chat json now ^.^
06:48 <+fragmer> Happy programmers' day
06:48 <+fragmer> https://en.wikipedia.org/wiki/Programmers%27_Day :D
06:49 < SinZ> yay, an excuse for why I didn't go to school today
11:31 < dx> so #mcdevs
11:31 < dx> my minecraft server is getting even more dead than before
11:33 < dx> a few people still play, but it's idling most of the time, and since it relies entirely on donations, i won't be able to keep it up for much longer
11:34 < dx> so i had this GENIUS AND COMPLETELY ORIGINAL IDEA. pay for the server by hour, and use a proxy server (in my other cheaper VPS) to get players to turn the server on when they want to play.
11:36 < winny> maybe you could hook it into inetd haha
11:37 < dx> it's more about nuking the instance and re-starting it from a snapshot, then having 'something' so the players can wait for it to start
11:37 < dx> but i can't help but think that this has been done before. has it?
11:48 < SomeoneWeird> what was the website that showed differences between protocol versions?
11:49 < dav1d> b.wiki.vg
11:49 < dx> but it's not updated for the snapshot due to refactoring of the minecraft code
11:51 < dx> but i think you could run burger yourself to get only protocol changes
11:52 < Yoshi2> dx: I do think that your idea is not entirely original, I've heard of at least one instance where somebody made something like that
11:52 < Yoshi2> but it was very long ago, sooo
11:52 < Yoshi2> go for it! :D
11:53 < dx> :D
11:55 < SomeoneWeird> dav1d, cheers!
11:59 <+md_5> dx 
11:59 < dx> md_5
11:59 <+md_5> a) servers charge when they are off
11:59 <+md_5> b) generally in 1 hour slices
12:00 < dx> 06:37 < dx> it's more about nuking the instance and re-starting it from a snapshot, then having 'something' so the players can wait for it to start
12:00 <+md_5> c) even the fastest servers take 60 seconds to provision (DO)
12:01 < dx> DO = digitalocean?
12:01 <+md_5> yah
12:02 < dx> well that doesn't sound too bad
12:02 < winny> it should be easy to write a simply inetd style service that shuts down servers that are inactive for say 5 minutes
12:02 < winny> simple*
12:02 < dx> why inetd
12:02 <+md_5> he means the entire vps
12:02 < winny> style
12:02 < dx> and ^md_5
12:03 < zutto> md_5: clouds dont charge per hour ;l
12:03 < dx> not java, but removing the entire instance and the allocated disk
12:03 < winny> clouds charge for actual usage :P
12:03 < zutto> ^
12:04 < zutto> so if your server is active for lets say 2 hours per day
12:04 < zutto> you could keep it up for quite cheap
12:04 < dx> >Amazon Elastic Compute Cloud (Amazon EC2) and Amazon Relational Database Service (Amazon RDS) instances accrue hourly charges as soon as you launch the instance, and continue accruing until you explicitly terminate the instance. If you launch an instance and then sign out of your account without terminating the instance, it will continue running and accruing charges
12:05 < zutto> wait, they moved away from the execution cycles?
12:05 < dx> digitalocean: >Your diskspace, CPU, RAM, and IP address are all reserved while your Droplet is powered off. If you want to save a Droplet for future use take a snapshot of it and then destroy it
12:06 < dx> rackspace: >If I am not using my server or do not have traffic to it, do I still have to pay for it?  Yes. Currently the server would either be running or you would have to delete it altogether
12:07 < dx> aaand that's all the 'cloud' hosting services i know
12:08 < zutto> i could guess that lw also offers some sort of cloud, and ovh does too
12:13 < dx> md_5: so yeah, it's probably going to be ugly, but i can't afford anything else (and sounds like a fun project, too)
12:20 <+md_5> dx if you have people on your server at least once an hour
12:20 <+md_5> you wont save any money
12:21 <+md_5> if you are short of csah
12:21 < dx> md_5: i have a total of 2 active players, only one of them logs in regularly
12:21 <+md_5> I highyl suggest digital ocean
12:21 <+md_5> I run Minecraft instances for 6 hours or so on them
12:21 <+md_5> works out great
12:21 < dx> neat.
12:21 <+md_5> $10/month if you want to go full time
12:21 <+md_5> 1gb ram is more than enough for 2 players
12:22 < dx> thing is, i had peaks of 6 players... during summer.
12:23 <+md_5> can you afford 3 cups of coffee a month?
12:23 < dx> everything pretty much died after classes started
12:23 < dx> lol
12:23 < dx> actually, personally i can't!
12:24 <+md_5> :(
12:24 < dx> i have no reliable way to pay for it myself
12:24 < zutto> md_5: how much are you paying for that 6 hours or so for the servers on DO?
12:24 <+md_5> zutto let me grab you my bill
12:25 < dx> but i got $40 left in the server donations box, so i'm trying to spread the butter reaaally thin
12:25 <+md_5> (s)
12:25 <+md_5> I dont use it much
12:25 <+md_5> zutto I have spent 16c in total
12:25 < zutto> 16c?
12:25 <+md_5> sorry + 29
12:25 < zutto> 0.29$?
12:25 <+md_5> 45 cents
12:25 <+md_5> yeah
12:25 <+md_5> 0.45$
12:25 < dx> what period of time?
12:26 < dx> just one day, 6 hours?
12:26 <+md_5> http://files.md-5.net/s/yd8Gb.png
12:26 <+md_5> http://files.md-5.net/s/ZBsGV.png
12:26 <+md_5> http://files.md-5.net/s/JZ38U.png
12:26 <+md_5> http://files.md-5.net/s/Mjrie.png
12:27 < dx> oh, so, basically, 45 cents on all your servers ever
12:27 < dx> lol.
12:27 < zutto> thats pretty cheap, i gotta admit
12:27 <+md_5> I use it when I want to test stuff
12:27 < dx> noticed
12:27 < zutto> gotta remember that site, i have times when i need to rent computing power from people ;l
12:28 <+md_5> or on the odd occasion portscan a host or two
12:28 <+md_5> :s
12:28 <+md_5> dont tell them that
12:28 < zutto> why not? :P
12:28 < zutto> they're servers ment to compute stuff
12:28 < dx> speaking of which, RIP http://instantserver.io/
12:28 <+md_5> > portscan
12:28 <+md_5> > legal
12:28 < dx> portscans aren't legal?
12:28 < zutto> oh right
12:28 < dx> that's news to me
12:29 <+md_5> well
12:29 <+md_5> > portscan
12:29 < zutto> i dont think thats illegal
12:29 <+md_5> > AUP / TOS
12:29 < dx> heh, okay then
12:30 <+md_5> http://imc.io/
12:30 <+md_5> :o
12:30 <+md_5> they use spigot
12:31 < dx> the price to extend for 6 hours is "f"
12:32 < dx> is that 15 or a "fuck you"?
12:32 <+md_5> dx looking at the src
12:32 <+md_5> its meant to be a fcebook like
12:33 < dx> oh
12:33 < dx> so the latter
12:33 < dx> md_5: also, wait, can you really do 24 players on 512mb ram?
12:33 <+md_5> not quite
12:33 < dx> okay, cool. there was no way spigot would be that good
12:33 <+md_5> you could get very close
12:33 < zutto> dx: you can if you have no plugins ;l
12:33 <+md_5> maybe 18
12:33 < dx> md_5: ...lol
12:34 <+md_5> but 24 is a stretch
12:34 <+md_5> sure if they were all at spawn
12:34 <+md_5> or view distance was 3
12:34 <+md_5> but with a view distance of 10, and all spread out, no
12:35 < dx> sadly my 'most active' server is FTB, and it's pretty bad at resource usage
12:39 < zutto> damn you md_5 
12:40 < zutto> i got all excited for these hourly servers now
12:40 < zutto> just thinking about all the stuff i could do with just few cents
12:42 < dx> :D
12:44 < zutto> i'm in process of building load balanced c# framework, these servers would be perfect for testing ;l
12:44 < dividuum> hi. I wrote a minecraft client (based on mc4p) and use Yggdrasil authentication. Now each time this client logs in, my normal minecraft clients ask for credentials when starting up. why?
12:45 < dividuum> according to the Authentication page on the wiki, logging in should only invalidate other accesstokens, when i'm not providing a client token myself (which i do)
12:45 < dividuum> any hints?
12:47 <+md_5> are you using the same client token each time?
12:47 < dividuum> not yet (i'm not keeping state at the moment)
12:48 <+md_5> you might have to
12:48 <+md_5> not sure
12:48 < dividuum> but even if i log in only once with my client, the normal minecraft client cannot log back in.
12:49 < dividuum> for me that looks like access tokens are invalidated every time i use the yggdrasil /authenicate url
12:53 < dividuum> just tested: using the same client token didn't help. the normal minecraft client still says "You're already logged in as xxxx in another profile"
12:56 < dividuum> hm. in authentication.py, there is an api endpoint /signout (no documentation on this on the wiki). maybe i have to signout one client before using the other. i'll test that
13:05 < tehme> how can i determine which blocks do i need to render?
13:05 < tehme> is there any algorithm better that brute force?
13:08 <+md_5> tehme http://en.wikipedia.org/wiki/Viewing_frustum
13:08 <+md_5> you should have a good understanding of opengl and stuff like culling before attempting to re create a game such as minecraft
13:09 <+md_5> http://zeuxcg.org/2009/01/31/view-frustum-culling-optimization-introduction/
13:09 <+md_5> http://fgiesen.wordpress.com/2010/10/17/view-frustum-culling/
13:09 <+md_5> alternatively for now, you can just let it do it for you: http://r3dux.org/2011/02/how-to-perform-manual-frustum-culling/
13:09 <+md_5> but it will be slow
13:18 <+Amaranth> Eh, a chunk based system makes such things simple
13:20 <+Amaranth> Sending a little extra data isn't a big deal so just check if any part of the chunk is in bounds instead of trying to do something more fine grained
13:24 <+sadimusi> dividuum: do you refresh your access token?
13:25 < dividuum> no
13:25 < dividuum> i didn't do that yet, since the normal minecraft client already fails to login after my client once
13:25 <+sadimusi> there must be something wrong with your code
13:25 < dividuum> *after my client logged in once
13:26 < dividuum> probably :-}
13:26 <+sadimusi> I use mc4p with the -a flag all the time and mc logs in fine afterwards
13:27 < dividuum> i guess that uses the credentials stored by the normal client
13:27 <+sadimusi> yes
13:27 < dividuum> without having looked what happens exactly, i guess that reuses the clienttoken and accesstoken generated by the normal minecraft client
13:28 <+sadimusi> yes
13:28 < dividuum> unfortunatelly this is not possible for my project
13:28 <+sadimusi> right
13:28 < dividuum> i'm trying to login to minecraft on behave of a user
13:28 <+sadimusi> so you generate a client token yourself?
13:28 < dividuum> yes
13:29 <+sadimusi> is your code available anywhere?
13:29 < dividuum> no. but i can put a minimal version of it
13:29 < dividuum> you're the author of mc4p, right?
13:29 < dx> *cough* gpl *cough*
13:30 <+sadimusi> the original author is mmcgill, but I've been maintaining it for some time now
13:30 < dividuum> thanks for that :-)
13:32 < dx> oh hey, pre-release of 1.6.3, i wasn't expecting that
13:32 < SinZ> the pre-release is just as a world conversion basically for 1.7, right?
13:33 < dx> seems to be, but i wonder if it has other interesting bugfixes
13:34 < dx> well nevermind
13:34 < dx> the tracker says it's just one ticket https://mojang.atlassian.net/browse/MC-15547
13:35 <+sadimusi> burger doesn't detect any changes either
13:35 <+sadimusi> expect for the protocol version which is now 77
13:36 < dx> supicious!
13:36 < dx> *suspicious
13:37 < SinZ> huh, why would the protocol jump
13:37 < SinZ> you would think they would try to keep it compatible
13:38 < dx> sadimusi: what is burger checking? protocol packets only?
13:38 < dx> or are there other functional toppings?
13:39 < zutto> md_5: how does this DO actually charge you?
13:39 < zutto> i made account and transfered 10$ to it, started new droplet for 3 minutes and shut it down -> made snapshot and destroyed it
13:39 < zutto> i cant see that it would have charged me anything
13:39 < zutto> yet
13:40 <+sadimusi> dx: all toppings are working
13:40 <+sadimusi> dx: just a few strings in the language files changed
13:40 < dx> sadimusi: i thought stuff was broken due to a refactoring, or did you fix everything already?
13:40 <+sadimusi> there's a new "/setidletimeout <Minutes until kick>" commad
13:40 <+sadimusi> *command
13:40 <+sadimusi> dx: it's broken in the snapshots, but 1.6.3 still uses the old codebase
13:40 < dx> ..oh, right
13:41 < dx> cool, then
13:43 < dividuum> sadimusi, https://gist.github.com/anonymous/6078924aae090d0ba4c8
13:50 <+sadimusi> dividuum: can't see anything wrong with it
13:51 < dividuum> try login in with it. then open your minecraft launcher. it will ask you for your password
13:55 <+sadimusi> I can reproduce it
13:58 < dividuum> your minecraft launcher doesn't ask you for credentials after logging in with the python program?
14:09 <+md_5> zutto it takes an hour or two before it goes through
14:10 <+sadimusi> dividuum: I think it's just a stupid mistake on my side
14:12 < zutto> ah i see
14:13 < dividuum> so the laucher asks for credentials?
14:13 <+sadimusi> [13:55:21] +sadimusi: I can reproduce it
14:13 <+sadimusi> yes
14:14 < dividuum> oops. ok. misread that. sorry
14:14 <+sadimusi> my fix doesn't seem to help much :/
14:14 <+sadimusi> did you already fix the _random_token method yourself? it didn't return anything
14:14 < dividuum> yeah. sent you a pull request 4 minutes ago :-)
14:14 <+sadimusi> but for some reason my token is still getting invalidated
14:15 < dividuum> maybe only one client token can every be used? so every time /authenticate is called, all other tokens/access tokens get invalidated?
14:16 <+sadimusi> no, at least that's not how it used to work
14:16 < dividuum> ok. strange
14:16 <+sadimusi> Dinnerbone: any idea why access tokens are getting invalidated when logging in with a new client token?
14:17 <+sadimusi> dividuum: I'll try loggin in with multiple instances of the vanilla client
14:18 < dividuum> ok. this is wierd. my launcher on my laptop works. my launcher on the desktop got invalidated. now i'm confused :)
14:19 < SinZ> iirc, can only run on one client token at a time
14:19 < SinZ> a new client token invalidates all existing stuff
14:19 < dividuum> sorry: had my girlfriends account on the laptop. that explains that :-}
14:19 < dividuum> SinZ, looks like it
14:20 <+sadimusi> indeed
14:20 <+sadimusi> but that was definitely not the case before and according to Dinnerbone's earlier statements isn't intended behaviour
14:20 <+sadimusi> time for another bug report
14:31 <+sadimusi> https://mojang.atlassian.net/browse/MCL-1513
14:33 < dividuum> sadimusi, thanks
14:36 < dividuum> btw, if you havn't seen it already: http://www.youtube.com/watch?v=a_RbDDVhYJA (my ugly ugly hack for long view distances)
14:37 <+sadimusi> interesting
14:39 < dividuum> it's a voxel renderer that uses a generated low-res 3d texture
14:40 < dividuum> and using mcp, a tiny c lib (called by JNI) that includes a python interpreter i run shader code in. ugly as hell :-)
14:42 < Not-002> [mc4p] sadimusi pushed 4 commits to master [+0/-0/±4] http://git.io/bkjApA
14:42 < Not-002> [mc4p] Florian Wesch 120d283 - actually create a client token
14:42 < Not-002> [mc4p] Florian Wesch 52f4374 - added metadata field to 0x14
14:42 < Not-002> [mc4p] Florian Wesch 226e7e5 - added particle effect paket (0x3f)
14:42 < Not-002> [mc4p] Florian Wesch 56a579e - make old authentication service available (without using lastlogin)
14:43 < dividuum> hey, i didn't even make a pull request for the last one :-)
14:43 <+sadimusi> I know, but I liked it
14:44 < dividuum> looks like i'll use the old authentication server, until this yggdrasil thing is somehow changed
14:44 <+sadimusi> you might want to hard reset your repo, I had to edit your commits a bit
14:45 < dividuum> yeah. noticed that. will do
14:45 <+sadimusi> had to get rid of that merge commit somehow :D
14:46 <+sadimusi> and you violated pep8 a bit with your first commit
14:47 < dividuum> d'oh :}
14:47 <+sadimusi> most of the original code written by mmcgill violates pep8 pretty heavily, but I'm trying to at least keep the new and completely rewritten files clean :) 
14:48 < dividuum> i can confirm: i looks better now
14:55 <+sadimusi> dividuum: I've just looked over your website a bit, you might like https://github.com/sadimusi/jinja2compact/
14:56 < dx> sadimusi: nice!
14:56 < dividuum> interesting. bookmarked for the future
14:57 < dx> really simple too
16:01 < tehme> yo people
16:01 < tehme> can you do a good thing for me?
16:02 < tehme> i can't register on a server because of ip limit, can you do it?
17:41 <+clonejo> tehme: what server?
23:18 < Brandon15811> dx, http://aws.amazon.com/big-data/powerof60/
23:19 < Brandon15811> $100 in aws credit for ec2 and emr
23:22 < tehme> lol
23:23 < tehme> dat evil kvirc set some aphorism on my channel exit
23:23 < tehme> well, better than ads
--- Day changed sam. sept. 14 2013
02:45 < tehme> yo
02:45 < tehme> what packet is used to move item from inventory to hotbar?
02:57 <+clonejo> tehme: it's just a regular inventory move
02:58 <+clonejo> the hotbar is part of the inventory
02:59 < tehme> clonejo: i can't find packet
02:59 < tehme> 0x66?
03:01 <+clonejo> are you doing a client or a server?
13:43 < tehme> yo sup
13:43 < tehme> do bow enchants work on sword?
--- Day changed dim. sept. 15 2013
01:52 < Not-002> [Miners] Astelyn pushed 1 commit to master [+0/-0/±7] http://git.io/17bjog
01:52 < Not-002> [Miners] astelyn 442b558 - mc: Added ClassiCube support
09:33 < winny> so is oracle java really faster?
09:35 <+md_5> not in my experience
16:49 < tehme> yo
16:49 < tehme> i've just hacked some noob's account and stole sword with enchants of 1000 lvl
16:49 < tehme> how is it done?
16:50 < winny> lol
16:50 < winny> poor password?
16:50 < tehme> "tututu"
16:50 < tehme> same as region name:D
16:51 < winny> roflmao.
17:05 < tehme> so how is this shit done? is it admin magic or cheats?
17:13 < dx> "hacked some noob's account"
17:13 < dx> so classy
17:18 <+Prf_Jakob> tehme: this channel is not the forum for bragging about exploits and hacking, this is your final warning.
17:19 < tehme> Prf_Jakob: ok
19:11 < eddyb> he could've said "I found lvl 1000 enchants, how is this possible?"
19:13 < eddyb> tehme: not sure if /enchant worked with unobtainable enchants/levels before, /give sure works now, any kind of external editors, and there was a way to put any enchants on a book, when signing it (the server didn't check what NBT fields the client sent)
19:14 < tehme> eddyb: was a way?:)
19:14 < tehme> or is?
19:15 < eddyb> if nobody removed the books created back then, they would still be usable (not sure if that exploit was fixed before you could use anvils to apply enchanted books, but I think so)
21:19 < tehme> HELL YEAH i reworked my buffer for protocol
21:19 < tehme> and it does not crash
21:19 < jast> yay! :)
--- Day changed lun. sept. 16 2013
00:04 < tehme> time to try opengl
00:09 < tehme> omg
00:10 < tehme> sdl2 opengl header is 11k lines long and full of macros
00:10 < dexter0> have you never used OpenGL?
00:10 < tehme> nope
00:10 < dexter0> b/c yes, there are a lot of macros
00:10 < tehme> i didn't
00:12 < tehme> where should i get dev libs for windows?
00:16 < tehme> is glut ok?
00:16 < tehme> and are there any alternatives?
00:17 < dexter0> glut is a helper library for GL, all you need is something that will give you a GL context you can issue commands to.
00:17 < dexter0> No idea how it works under Windows.
00:21 < tehme> found it
00:22 < tehme> headers are built into vs
00:22 < tehme> and libs too
00:22 < tehme> mb i should update them...
00:26 < tehme> my exampledoesn't work:(
00:47 <+md_5> anyone here got JNI experience :c
00:54 <+sadimusi> burger's block topping is almost working again :)
00:54 <+sadimusi> only one block still isn't quite right
00:55 <+sadimusi> "2 files changed, 267 insertions(+), 88 deletions(-)" :/
01:00 < tehme> almost
01:01 <+sadimusi> well, then I still have to fix items, recipes and biomes
01:03 < tehme> will they add message size?:)
01:03 <+sadimusi> of course not
01:03 < tehme> this is so sad
01:04 < tehme> sadimusi: do you know about lvl 1000 enchantments?
01:04 <+sadimusi> iirc you can't go over 10 in vanilla minecraft
01:04 <+sadimusi> even with editors and stuff
01:05 < tehme> i don't know if that server is vanilla
01:05 < tehme> but i sto... got a sword with full lvl 1000 enchantments
01:05 < tehme> one shots diamond-armored player
01:06 <+sadimusi> can the client even display that?
01:06 < tehme> i want to be able to make swords like this one
01:06 < tehme> yep
01:06 < tehme> wait a minute
01:07 < tehme> http://i.imgur.com/TBIjz5r.png
01:07 < tehme> like this
01:08 <+sadimusi> so it can't really display it
01:08 < tehme> heh
01:09 <+sadimusi> anyway, I'm sure there are dozens of plugins out there to build those
01:09 < tehme> so this is admin's work
01:09 < Drainedsoul> writing token pasting macros, how bad should I feel about myself?
01:10 < tehme> Drainedsoul: wut?
01:10 < Drainedsoul> #define GET(x,y) x( get_ ## y ( x ## _key , x ## _default ))
01:10 <+sadimusi> tehme: obviously
01:10 < Drainedsoul> #define GET_INT(x) GET(x,int)
01:10 < tehme> wow
01:11 < tehme> now you are a sinner too:D
01:11 < Drainedsoul> there was really no other way to accomplish this that didn't involve all sorts of typing. :\
01:11 < tehme> do you want to see my code? you will feel better
01:12 < Drainedsoul> lol
01:12 < Drainedsoul> do you use macros instead of templates/inline functions or something
01:13 < tehme> nope
01:13 < tehme> i use generated code
01:13 < tehme> and until today i used flat vector as message buffer
01:13 < tehme> offsets everywhere
01:14 < Drainedsoul> I use a flat vector as a message buffer, but I use clever templating tricks so you don't see it
01:14 < tehme> but when i look at those tricks my eyes bleed
01:14 < tehme> right?:D
01:15 < Drainedsoul> they're not especially difficult to understand
01:15 < tehme> link?
01:16 < tehme> and whu not deque?
01:16 < tehme> why*
01:17 < tehme> https://github.com/meandmark/SDLOpenGLIntro/tree/sdl2
01:17 < tehme> what do you rhink about this code style>
01:17 < tehme> ?
01:17 < tehme> think*
01:17 < tehme> fuck
01:18 < Drainedsoul> https://github.com/RobertLeahy/MCPP/blob/master/include/packet.hpp#L84
01:19 < Drainedsoul> allman gtfo
01:20 < tehme> wut?