[00:01] nym: ok
[00:01] nym: thanks SubStack 
[00:02] bingomanatee_: anyone who has experience with PayPal's APIs tends to sieze up. Then men come and take them to a special farm where their needs are cared for. 
[00:02] bingomanatee_: Or so I am told.
[00:03] SubStack: the farm with the drugs?
[00:03] SubStack: no wait that is another thing
[00:03] linac has joined the channel
[00:03] bingomanatee_: its the farm with all the dogs from your childhood. 
[00:09] alek_br has joined the channel
[00:18] ceej_ has joined the channel
[00:24] tibbetts has joined the channel
[00:24] piscisaureus has joined the channel
[00:25] MattJ has joined the channel
[00:27] pyrotechnick: net unsplit
[00:28] marienz has joined the channel
[00:28] maushu has joined the channel
[00:28] possibilities has joined the channel
[00:30] tg has joined the channel
[00:33] dthompson has joined the channel
[00:33] tim_smart has joined the channel
[00:33] tim_smart: micheil: Hey man, around?
[00:35] galaxywatcher has joined the channel
[00:37] wdperson has joined the channel
[00:43] possibilities has joined the channel
[00:48] janne has joined the channel
[00:52] jakehow has joined the channel
[00:52] mlins has joined the channel
[00:53] sudoer has joined the channel
[00:56] baoist has joined the channel
[00:57] bmizerany has joined the channel
[01:06] desaiu has joined the channel
[01:11] JojoBoss: whats the latest node that will run in cygwin?
[01:11] robotarmy has joined the channel
[01:13] davidwalsh has joined the channel
[01:13] isaacs has joined the channel
[01:13] ceej_ has left the channel
[01:28] leepsy has joined the channel
[01:29] hassox has joined the channel
[01:32] CrypticSwarm has joined the channel
[01:34] xSmurf has joined the channel
[01:34] hunterloftis has joined the channel
[01:34] jimt has joined the channel
[01:35] hunterloftis: Anybody got tips for ajax uploaders? I've tossed one together but I'd rather use one that's been vigorously tested
[01:37] ceej_ has joined the channel
[01:39] nym: does anyone know a good library for doing websockets with iphone
[01:41] polotek: nym: in mobile safari?  doesn't socket.io work there?
[01:42] nym: i mean within a native iphone a
[01:42] pyrotechnick: i think he means in objective-c
[01:42] nym: app
[01:42] nym: yes
[01:42] themiddleman has joined the channel
[01:42] pyrotechnick: nfi
[01:43] [[zz]] has joined the channel
[01:43] brianmario has joined the channel
[01:43] pyrotechnick: there was some dude in here the other day asking
[01:43] pyrotechnick: he stackoverflowed the answer
[01:43] nym: i set up sockt-chat and i'm trying to connect to it now
[01:44] slaskis has joined the channel
[01:45] nym: ok
[01:46] polotek: nym: https://github.com/esad/zimt
[01:46] polotek: looks promising
[01:46] nym: thanks
[01:47] polotek: also I've heard good things about http://pusherapp.com/
[01:47] nym: is this totally foreign btw?
[01:47] polotek: if you wanted to use a third party service
[01:47] Kudos has joined the channel
[01:47] jimt has joined the channel
[01:47] polotek: nym: what do you mean?
[01:48] nym: i mean, people coming in here wanting to connect to a node server from the iphone
[01:48] Kudos: so I open a connection to a database and am synchronously adding entries to the db
[01:48] polotek: never seen it before. but apparently it's not completely new
[01:48] nym: and does it make sense to want to use websockets to do so from obj-c?
[01:49] Kudos: problem is, i don't know when to close the connection to the db, so my app is just hanging
[01:49] Kudos: halp?
[01:49] polotek: Kudos: why are you doing it synchronously? what lib are you using?
[01:49] Kudos: mongo
[01:49] Kudos: node-mongodb-native to be specific
[01:49] polotek: what node client are you using?
[01:49] polotek: ah
[01:50] polotek: I didn't think it supported sync updates
[01:50] polotek: anyway, gist some code and I'll take a look
[01:50] Kudos: err
[01:50] Kudos: basiaclly, i am using the node-utils request class to scrape a site
[01:51] Kudos: that's what's driving the synchronous part of it
[01:52] polotek: there's a close method on the db connection.
[01:52] polotek:  But I'm pretty sure you should be doing the updates asynchronously
[01:53] Kudos: the request callback is called synchronously
[01:53] Kudos: so the mongo inserts happen synchronously
[01:54] polotek: that's not how it works. gist me some example code
[01:54] Kudos: and yeah, the close method is what i was using, but it closes it before e
[01:54] Kudos: any entry gets added
[01:54] polotek: yeah, I'm pretty sure I know what you're doing
[01:55] polotek: the mongo updates happen asynchronously
[01:55] polotek: but you're not passing a callback
[01:55] polotek: then you're calling close and it's killing the connection before your updates are done
[01:55] Kudos: correct
[01:55] polotek: you need to wait until all of the async updates to mongo are done
[01:56] Kudos: how do i wait
[01:56] polotek: there are lots of techniques
[01:56] polotek: start with a simple one so you can get a feel for how it works
[01:57] polotek: write a callback for the mongo updates. use the same one every time
[01:57] polotek: the callback should count down as each update finishes
[01:57] polotek: and when all of the updates are done, then you can call db.close()
[01:57] Kudos: hmm, i think i follow
[01:57] coffee has joined the channel
[01:58] polotek: try it with a simple example, take the request stuff out of the equation
[01:58] Kudos: count up for every update i expect to do, and then back down as it does them?
[01:58] polotek: just add some test data
[01:58] polotek: sure, if you like
[01:59] Kudos: balls, that won't work
[01:59] Kudos: that will introduce a race condition with the requests
[02:00] polotek: Kudos, not if you orchestrate things correctly
[02:01] Kudos: is there any way i can tell the request class to be asynch?
[02:01] Kudos: it's slower, but far less painful :P
[02:02] polotek: Kudos: everything in node wants to be async :)
[02:02] polotek: just pass it a callback
[02:03] polotek: it's a node convention that if the last argument is a function
[02:03] polotek: it's used as an async callback
[02:05] micheil: tim_smart: yup.
[02:05] Kudos: I am already passing it a callback which does nothing
[02:05] Kudos: i'll obfuscate a gist for you
[02:05] Kudos: just removing identifiable info ;P
[02:06] Ond has joined the channel
[02:06] polotek: cool
[02:07] Kudos: https://gist.github.com/b9a07c06a817da1bc354
[02:08] polotek: first of all I'm pretty sure you only have to jQuerify once
[02:09] polotek: and you only have to client.open once
[02:09] Kudos: pretty sure you need to jQuerify per document you open
[02:10] Kudos: it injects jquery into the dom of that document only
[02:10] polotek: could be right
[02:10] polotek: I don't use jsdom much
[02:10] polotek: but your request call should be async
[02:10] googol has joined the channel
[02:10] polotek: I mean the way it is here
[02:10] polotek: what makes you think it's firing synchronously?
[02:10] Kudos: it isn't though
[02:11] Kudos: if i stick process.exit() at the end of that code block, nothing inside the request fires
[02:12] polotek: at the end of what code block?
[02:12] polotek: also process.exit stops everything
[02:12] polotek: so it doesn't matter if it's async or not
[02:12] Kudos: yes, but if you console.log before a process.exit, you still get the output
[02:13] polotek: maybe that's a special case. I think it tries to flush the stdout buffer
[02:13] polotek: try a console.error after the code block
[02:13] polotek: the warn and error functions write to stderr and are synchronous
[02:14] omygawshkenas has joined the channel
[02:14] polotek: anyway, regardless of all of that
[02:14] polotek: this should be okay
[02:14] Kudos: no difference
[02:15] polotek: it could be more efficient, but there's nothing wrong with it at first glance
[02:15] Kudos: it's my first foray into node :P
[02:15] polotek: but it's going to be tough to follow the control flow here
[02:15] polotek: every time you call an async function with a callback
[02:15] aaronblohowiak has joined the channel
[02:15] polotek: you can't tell when that callback will be put on the event loop
[02:16] aaronblohowiak: grrr, every filter is failing in jade.  anyone else have the Array.prototype.toString is not generic issue?
[02:16] polotek: I would factor out the things that only need to happen once
[02:16] sivy has joined the channel
[02:16] polotek: I think you're right about jsdom.jQueryify
[02:16] polotek: but the client.open and client.connection I'm pretty sure you only need once
[02:19] forzan has joined the channel
[02:19] Kudos: client.collection i have to call each time
[02:20] Kudos: client.open i only have to call one time
[02:20] quest88 has joined the channel
[02:20] polotek: you can reuse the collection object too
[02:20] Kudos: tried it, it didn't work
[02:21] MattDiPasquale has joined the channel
[02:22] Kudos: client.collection() returns nothing
[02:23] polotek: factor out the open and collection calls
[02:23] polotek: wrap the whole thing in the callback for client.collection
[02:23] polotek: so client and connection are available for all subsequent callbacks
[02:23] pyrotechnick: v8: console.log('bo is cool');
[02:23] v8bot: pyrotechnick: CONSOLE: ["bo is cool"], OUTPUT: undefined
[02:25] pyrotechnick: pyrotechinck++
[02:25] v8bot:  pyrotechnick has given a beer to pyrotechinck. pyrotechinck now has 1 beers.
[02:25] pyrotechnick: ryah++
[02:25] v8bot:  pyrotechnick has given a beer to ryah. ryah now has 2 beers.
[02:25] pyrotechnick: ryah--
[02:25] v8bot:  ryah is getting too many beers.  Don't let ryah get drunk!
[02:26] ryah: yeah, careful
[02:26] coffee has joined the channel
[02:26] c4milo1 has left the channel
[02:28] Kudos: polotek, while that does make sense, i'm still left with my original problem
[02:28] devdrinker has joined the channel
[02:28] Kudos: my code still doesn't know when to close the db handle
[02:29] CIA-121: node: 03Ryan Dahl 07master * rfeb77ea 10/ (lib/repl.js test/simple/test-repl.js): 
[02:29] CIA-121: node: Fix REPL for named functions
[02:29] CIA-121: node: add some tests. - http://bit.ly/gbctmZ
[02:35] quest88: I'm reading about http.Client here: http://nodejs.org/docs/v0.3.2/api/http.html#http.Client -- I see http.createClient(80,'www.google.com') and the following line: request = google.request('GET', '/', {'host': www.google.com}). If I wanted to get search results would I swap out '/' witih '/?q=searchterms' ?
[02:38] tk has joined the channel
[02:39] ryah: please test http://nodejs.org/dist/node-v0.3.3.tar.gz
[02:41] ryah: % ./node -e process.versions
[02:41] ryah: { node: '0.3.3',
[02:41] ryah:   v8: '3.0.4.1',
[02:41] ryah:   ares: '1.7.3',
[02:41] ryah:   ev: '4.0',
[02:41] ryah:   openssl: '1.0.0a' }
[02:41] devdrink_ has joined the channel
[02:45] MattDiPasquale has joined the channel
[02:46] polotek: Kudos: I'm trying to run your example but there's a problem with jsdom
[02:46] pengwynn has joined the channel
[02:47] polotek: I don't use it so I'm not sure what the problem is
[02:48] Kudos: polotek
[02:48] Kudos: i'm throwing in the towel for the night
[02:48] Kudos: thanks for your help up to now :)
[02:49] polotek: no prob
[02:49] polotek: keep at it
[02:49] polotek: I think you're using too many libraries right off the bat
[02:49] polotek: too many places for things to go wrong
[02:49] polotek: try simpler examples to make sure you know what's happening and when
[02:54] xSmurf: any xmpp gurus in the house? I'm trying to implement 0065 (S5B file transfers) and running into some issues
[02:54] Aria: Fun! I'm no guru, but I can put eyes on it
[02:55] SubStack: oh no xmpp
[02:56] devdrin__ has joined the channel
[02:56] MattJ: xSmurf: Wireshark :)
[02:57] CIA-121: node: 03Ryan Dahl 07master * r57544ba 10/ (AUTHORS ChangeLog doc/index.html src/node_version.h): Bump version to v0.3.3 - http://bit.ly/e756Ev
[02:57] CIA-121: node: 03Ryan Dahl 07master * r2b5b128 10/ src/node_version.h : Now working on v0.3.4-pre - http://bit.ly/hz9o2Z
[02:58] xSmurf: MattJ: ohai ;)
[02:58] xSmurf: yes I'm wiresharking
[02:58] pyrotechnick: :O
[02:58] xSmurf: and that's the issue, I don't see anything wrong here
[02:59] xSmurf: the tcpdata doesn't include those illusive chars
[02:59] nym: hi, i have a client that i'm not sure is connecting properly to my socket.io server. do you have any idea how i could open up a socket from the command line to listen to see if anything tries to connect?
[03:00] polotek: nym telnet?
[03:01] nym: i'd need to set up a telnet server and listen to the output somehow, right?
[03:02] pyrotechnick: nym
[03:02] pyrotechnick: i got code
[03:02] pyrotechnick: its like 4 lines
[03:02] pyrotechnick: sec
[03:02] nym: sweet
[03:02] nym: thank you pyrotechnick 
[03:03] MattJ: xSmurf: and Psi is the recipient?
[03:03] polotek: nym: oh I read it wrong, you wanna test the client, not the server
[03:03] nym: right
[03:03] xSmurf: MattJ: Psi and/or Adium
[03:04] pyrotechnick: nym: https://gist.github.com/e715d99e0224e3009a13
[03:04] xSmurf: Adium saves the file with the extra chars at the beggining and truncated data and psi juste refused to receive the file
[03:04] pyrotechnick: thats a rootkit
[03:04] pyrotechnick: but i mean
[03:04] pyrotechnick: if u take out the exec
[03:04] MattJ: xSmurf: check your socks5 negotiation again then
[03:04] pyrotechnick: itll just dump the data
[03:04] MattJ: xSmurf: Wireshark probably has a decoder for it
[03:04] xSmurf: MattJ: it seems ok since Psi is sending me the iq
[03:05] MattJ: which iq?
[03:05] pyrotechnick: nym: is that what you need?
[03:06] nym: i think so, let me try
[03:06] pyrotechnick: i wudnt run that
[03:06] nym: why?
[03:06] xSmurf: MattJ: http://jabber.org/protocol/bytestreams stream-used
[03:07] pyrotechnick: why is there not an xmpp server in node
[03:07] quest88: I'm getting status 503 trying to use this: http://pastebin.com/JQ0nAwYz but when I'll use the same URL (http://services.digg.com/2.0/stream?types=digg&return_after=1) in browser and in a java program and I get data. Do I have something incorrect?
[03:07] jashkenas: 'congrats on 0.3.3
[03:07] nym: oh, it's coffeescript anyways
[03:07] pyrotechnick: nym do u want it in javascript?
[03:07] xSmurf: only then do I start writing data
[03:07] nym: sure
[03:08] pyrotechnick: https://gist.github.com/f389fee080feaf4fb55c
[03:09] MattJ: pyrotechnick: because I lost motivation when writing it
[03:09] polotek: pyrotechnick: I think there is
[03:09] MattJ: pyrotechnick: mainly because I already wrote one in a superior language :)
[03:09] pyrotechnick: which language
[03:09] MattJ: Lua
[03:09] pyrotechnick: polotek: i dont think there is
[03:09] pyrotechnick: prosody?
[03:09] polotek: https://github.com/astro/node-xmpp
[03:09] xSmurf: MattJ: let me pastie you some code... http://pastie.textmate.org/private/bchwurj312fvhdqazypdw
[03:09] evanmeagher has joined the channel
[03:09] pyrotechnick: thats not a server
[03:09] MattJ: pyrotechnick: Yes, Prosody
[03:10] pyrotechnick: WTF
[03:10] pyrotechnick: u wrote prosody?
[03:10] pyrotechnick: we use prosody...
[03:10] MattJ: Yes
[03:10] pyrotechnick: *bows*
[03:10] MattJ: Everyone cool uses Prosody :)
[03:10] xSmurf: ^^ +1
[03:10] pyrotechnick: yeah i got tired of erlang config
[03:10] pyrotechnick: id rather stab my eyes out
[03:10] MattJ: I actually ported some of its core libs to Javascript (Node.js) as a kind of fun side-project
[03:10] polotek: pyrotechnick: it says it has a server component
[03:10] polotek: I don't know anything about it
[03:10] polotek: just got it from the node modules page
[03:10] pyrotechnick: it is for xmpp components
[03:11] pyrotechnick: its not a server
[03:11] MattJ: if someone /really/ wanted a Javascript server, it wouldn't be hard at all
[03:11] pyrotechnick: yeah
[03:11] pyrotechnick: i started to write one
[03:11] pyrotechnick: i know
[03:11] pyrotechnick: its just
[03:11] pyrotechnick: it blows my mind
[03:11] pyrotechnick: why there's not one
[03:11] pyrotechnick: and lua = superior lol
[03:11] MattJ: ;)
[03:11] MattJ: Javascript comes in second, don't worry
[03:11] polotek: I always hear good things about lua
[03:12] MattJ: but Lua is kind of how Javascript /should (have) be(en)
[03:12] MattJ: /
[03:12] pyrotechnick: right
[03:12] pyrotechnick: i think coffee is a nice buffer from js
[03:12] brianmario has joined the channel
[03:12] xSmurf: but then again, other than people bitching about ejabberd I always heard good things of erlang :P
[03:12] pyrotechnick: is it the vm of js or the syntax?
[03:12] polotek: I prefer js syntax to coffee
[03:12] pyrotechnick: or both
[03:12] xSmurf: + ... http://www.archive.org/details/ErlangTheMovie beat that!
[03:12] MattJ: xSmurf: Yeah, we have people using Prosody who are real erlang fans - they just can't stand ejabberd
[03:12] pyrotechnick: so u dont like python?
[03:12] polotek: I think we already has that discussion
[03:13] pyrotechnick: MattJ would you help with an xmpp in node?
[03:13] pyrotechnick: if it were written in js
[03:13] pyrotechnick: because we have the start of one in coffee
[03:13] pyrotechnick: but im willing to use js
[03:13] MattJ: pyrotechnick: I don't like the JS implementations much, the syntax is ok, though prototypes are a bit messed up
[03:13] pyrotechnick: if it means more ppl will help
[03:13] pyrotechnick: maybe ud like coffee
[03:13] MattJ: pyrotechnick: just look at node-xmpp and port more of Prosody's code over :)
[03:14] xSmurf: MattJ: If you have a min to overlook that code and maybe give some pointers that'd be really awesome I've been banging my head aaaall day with no result :/
[03:14] polotek: quest88: looked at your gist
[03:14] polotek: you probably need to set some more request headers
[03:14] nym: pyrotechnick: thanks, that did exactly what i needed
[03:14] coffee has joined the channel
[03:14] nym: my client is at least trying to connect
[03:15] polotek: the http client doesn't set some basic stuff, like HOST
[03:15] xSmurf: I must be missing something stupid
[03:15] pyrotechnick: polotek: use request
[03:15] xSmurf: it's completely oblivious to me :(
[03:15] MattJ: pyrotechnick: er, when I said node-xmpp I meant xmpp.js :)
[03:15] polotek: quest88: also ^
[03:16] nonnikcam has joined the channel
[03:16] quest88: polotek: I added host, no go too.
[03:16] pyrotechnick: nym: no problem.
[03:16] quest88: is request something someone else already made?
[03:16] pyrotechnick: yeah
[03:16] MattJ: node-xmpp is still the name of xmpp.js in my dev folder, but now the name of Astro's lib :)
[03:16] polotek: quest88:  yeah you should just use request
[03:16] pyrotechnick: it just makes requets a bit nicer
[03:16] polotek: npm install request
[03:16] xSmurf: (yes I'm aware it's shit code :p just trying to get it work before I refactor properly)
[03:16] pyrotechnick: yep
[03:16] MattJ: xSmurf: working on it :)
[03:16] quest88: npm?
[03:16] xSmurf: :o +1! thx
[03:16] quest88: oh, rpm
[03:17] pyrotechnick: node package manager
[03:17] polotek: quest88: no
[03:17] polotek: npm
[03:17] pyrotechnick: lol
[03:17] pyrotechnick: https://github.com/isaacs/npm
[03:17] polotek: npm should be the first thing people learn when they get to node
[03:17] polotek: how do we make that happen?
[03:17] pyrotechnick: ryah: you need to mention npm
[03:17] pyrotechnick: on the site
[03:17] xSmurf: yeah node-xmpp isn't exactly a complete library, even with node-xmpp-client on top... it lacks
[03:17] pyrotechnick: it's a problem.
[03:17] polotek: quest88: https://github.com/mikeal/node-utils/tree/master/request
[03:17] pyrotechnick: is there any reason you dont ryah?
[03:18] pyrotechnick: i think node needs an unofficial site
[03:18] ryah: pyrotechnick: hm?
[03:18] pyrotechnick: npm
[03:18] pyrotechnick: you need to mention it
[03:18] ryah: pyrotechnick: what about it?
[03:18] pyrotechnick: too many people dont know what it is
[03:18] quest88: thanks Prometheus, pyrotechnick 
[03:18] pyrotechnick: it needs to be somewhere on the front page
[03:18] polotek: jsdom doesn't work at all for me
[03:18] polotek: fortunately I don't need it to
[03:18] pyrotechnick: they keep breaking jsdom
[03:18] polotek: 
ACTION goes back to what he was doing before

[03:18] pyrotechnick: its unstable
[03:18] ryah: pyrotechnick: perhaps someday
[03:18] pyrotechnick: do you have a good reason why?
[03:19] pyrotechnick: or just havent got around to it?
[03:19] ryah: mostly that
[03:19] polotek: pyrotechnick: at one point there were other package managers
[03:19] pyrotechnick: would you prefer that kind of thing to be on a community page
[03:19] xSmurf: 
ACTION is full on xmpp'ing, writting some jabber bot in node

[03:19] pyrotechnick: right
[03:19] pyrotechnick: there were
[03:19] polotek: and ry didn't want to give one too much credit
[03:19] pyrotechnick: sometimes i wish kiwi one, i like the idea of 'seeds'
[03:19] polotek: but I think we're past that
[03:19] polotek: npm stands on the mountain of slain foes
[03:19] pyrotechnick: LOL
[03:19] pyrotechnick: i loled
[03:20] pyrotechnick: like ruby has gems
[03:20] pyrotechnick: and python has
[03:20] pyrotechnick: recipes or something
[03:20] pyrotechnick: cheese factory or watever
[03:20] MattJ: xSmurf: have you tried without the fast extension?
[03:20] polotek: eggs, sort of
[03:20] pyrotechnick: yeah
[03:20] noahcampbell has joined the channel
[03:20] jakehow has joined the channel
[03:20] polotek: but python is all messed up
[03:20] pyrotechnick: but we have
[03:20] xSmurf: MattJ: yep
[03:20] pyrotechnick: packages
[03:20] pyrotechnick: wow fun
[03:20] pyrotechnick: packages
[03:20] xSmurf: same result iirc, though I will try again
[03:20] MattJ: xSmurf: Psi supports that I think, so it might be simpler to drop it for now until you get the basics working
[03:20] ryah: i did hire npm's author
[03:21] ryah: i think that make it the official unofficial package manager :)
[03:21] pyrotechnick: lol
[03:21] polotek: indeed
[03:21] pyrotechnick: if i pull request it will you accept it
[03:21] xSmurf: MattJ: yeah I was aiming for proxy server at first but I down stepped to direct connection for now
[03:21] ryah: no, not right now
[03:21] pyrotechnick: is nodejs.org even in the repo
[03:21] ryah: there's work to be done still
[03:21] pyrotechnick: on npm?
[03:22] ryah: on both
[03:22] pyrotechnick: i cant say i understand but okay
[03:22] ryah: i.e. i don't think the module story is finished
[03:22] pyrotechnick: ill just add it to the wiki
[03:22] xSmurf: MattJ: without fast in adium, maybe it's a fluke, I don't get extra data, but it's still truncated
[03:22] ryah: pyrotechnick: ok
[03:22] pyrotechnick: right but i dont think the solution is to let people miss out on npm
[03:22] Me1000 has joined the channel
[03:22] polotek: pyrotechnick: they learn about it pretty quick
[03:22] ryah: pyrotechnick: http://nodejs.org/docs/v0.3.3/api/appendix_1.html
[03:22] pyrotechnick: well
[03:22] polotek: just could be quicker
[03:22] pyrotechnick: they sometimes do
[03:22] xSmurf: roh! now psi goes through... but same problem extra and truncated data
[03:23] pyrotechnick: i think npm is more worthy than being in an appendix but i dunno maybe its not to others
[03:23] marienz has joined the channel
[03:23] pyrotechnick: it certainly is for us end users of node
[03:23] Me1000 has joined the channel
[03:24] MattJ: xSmurf: does it ever log *** CONNECT *** ?
[03:24] MattJ: I don't see it in your output
[03:24] xSmurf: is it supposed to log another connection after the sock handshake?
[03:24] xSmurf: shoot don't mind the output I forgot it's there, it's old and stale
[03:24] MattJ: Oh, ok :)
[03:25] pyrotechnick: why do some people struggle so hard to keep wikis in alphabetical order
[03:25] xSmurf: MattJ: pm...
[03:25] xSmurf: that's current
[03:28] AAA_awright_ has joined the channel
[03:30] xSmurf: MattJ: no dice? :/
[03:32] bingomanatee_: how about node.us?
[03:32] MattJ: xSmurf: could you possibly print out the data you get?
[03:32] MattJ: in hex
[03:32] MattJ: or whatever
[03:33] ceej_ has joined the channel
[03:33] xSmurf: http://pastie.textmate.org/private/xgdknq0fssff1g8y7md62q
[03:34] aaronblohowiak: micheil: Evan is cool
[03:34] aaronblohowiak: micheil++
[03:34] v8bot:  aaronblohowiak has given a beer to micheil. micheil now has 4 beers.
[03:34] micheil: aaronblohowiak: it's pretty standard code though
[03:34] aaronblohowiak: that is a good thing.
[03:34] shinmei has joined the channel
[03:35] micheil: aaronblohowiak: also, evan may be cool, but node-eventstream is even more awesome
[03:35] micheil: https://github.com/miksago/node-eventstream
[03:35] micheil: (may not work with the latest nws)
[03:35] cafesofie has joined the channel
[03:36] xSmurf: MattJ: want a wireshark dump?
[03:36] bingomanatee_: I thought that github was the unofficial node site :D
[03:37] MattJ: xSmurf: yes please :)
[03:37] MrNibbles: gentlemen
[03:37] aaronblohowiak: micheil: i am building something similar, but based on uuids for objects and collections
[03:37] MrNibbles: happy new year you lot
[03:37] aaronblohowiak: MrNibbles: what about knaves?
[03:37] micheil: aaronblohowiak: my next project will probably be a way to more-easily scale event-emitters
[03:38] aaronblohowiak: micheil: scale in terms of ease of development or performance?
[03:38] MrNibbles: those scallywags can wish themselves a happy new year
[03:38] aaronblohowiak: ;)
[03:38] aaronblohowiak: happy new year, aaronblohowiak!
[03:38] pyrotechnick has joined the channel
[03:38] micheil: aaronblohowiak: in terms of being able to have two servers and have the events emit between them
[03:39] aaronblohowiak: micheil: ah. aka distributed mq
[03:39] aaronblohowiak: ;)
[03:39] MrNibbles: well, its 3:40am and i am still coding... must be the christmas holidays
[03:39] aaronblohowiak: :)
[03:40] micheil: or rather then having a central store of the events, you do internetworking
[03:40] aaronblohowiak: neat.
[03:40] unomi has joined the channel
[03:40] shaver: vector clocks! link-local addressing! redundant ARP! log-structured append-only data stores! RAID-Z!
[03:40] pyrotechnick1 has joined the channel
[03:41] micheil: shaver?
[03:41] shaver: sorry
[03:41] micheil: sarcasm?
[03:41] shaver: just got carried away
[03:41] aaronblohowiak: shaver: i like all those things
[03:41] shaver: yeah
[03:42] polotek: micheil: aren't people already doing that with redis pubsub?
[03:42] aaronblohowiak: polotek: SPOF
[03:42] polotek: there's always a SPOF
[03:42] aaronblohowiak: uncorrect
[03:43] aaronblohowiak: it is just much more complicated to avoid a spof
[03:43] MikhX has joined the channel
[03:43] polotek: always == for most practical purposes
[03:43] polotek: unless your company has much $$
[03:44] pyrotechnick1: oh god http://www.slideshare.net/creationix/what-does-node-need-from-us
[03:44] aaronblohowiak: pyrotechnick1: this is 3 months old
[03:44] pyrotechnick1: it's kind of timeless
[03:44] pyrotechnick1: it's hilarious
[03:45] pyrotechnick1: a) the people who care about that stuff aren't little kids and don't need to be told b) the people that don't, don't, so it's pointless
[03:45] MrNibbles: pff, that inflated london visiting stat is blatantly 80% me
[03:45] pyrotechnick1: lol
[03:46] aaronblohowiak: pyrotechnick1: no need to laugh at him, tim means well and is a huge asset to the community.
[03:46] polotek: I think saying that stuff out loud does help
[03:46] polotek: pyrotechnick1: I think you underestimate the immaturity of some
[03:47] pyrotechnick1: i think the people he's attempting to convert are babied too much and the solution isn't to baby them some more
[03:48] polotek: it's not for them
[03:48] polotek: it helps create an atmosphere where their shenanigans aren't tolerated
[03:48] pyrotechnick1: lol
[03:48] pyrotechnick1: what are you going to do?
[03:48] pyrotechnick1: gline them from the irc?
[03:48] pyrotechnick1: ban them from github?
[03:49] polotek: being shunned by peers has impact
[03:49] pyrotechnick1: blacklist their ips from nodejs.org?
[03:49] rsms has left the channel
[03:49] aaronblohowiak: verbalizing social disapproval of certain kinds of behaviour is a good way to influence behavior.  most people want to be accepted, and in communities that do not overtly shun bad behavior, the few people that reward it just serve to help it propagate
[03:50] JusticeFries has joined the channel
[03:50] pyrotechnick1: it's like saying dont feed the trolls, then going and saying exactly what trolls want to hear, that they're making an impact and getting a raise out of people
[03:50] aaronblohowiak: for example: the women-unfriendliness of many Rails conferences
[03:50] mikew3c_ has joined the channel
[03:50] pyrotechnick1: maybe it's different there in amercia
[03:50] aaronblohowiak: pyrotechnick1: there is one thing to not feed the trolls, it is another to completely ignore them
[03:51] pyrotechnick1: but here we don't need someone trying to be our mother and telling us how to conduct ourselves
[03:51] aaronblohowiak: pyrotechnick1: i envy your optimism
[03:52] pyrotechnick1: he's also discouraging competition in node.js
[03:52] MrNibbles: i think the node community are all downright upstanding citizens of refinement and intellect, in contrast, try spending 10 mins in #jquery
[03:52] pyrotechnick1: he's also suggesting we don't tag a node web framework as a X-killer when it's exactly what node needs
[03:53] micheil: actually, node doesn't need X-Killers
[03:53] pyrotechnick1: i dont know about you but id rather have my friends being paid to do rails in node than rails in ruby
[03:53] micheil: node isn't going to kill [insert your own library/framework/application here]
[03:53] pyrotechnick1: yeah it is
[03:53] tim_smart has joined the channel
[03:54] pyrotechnick1: http://en.wikipedia.org/wiki/Technological_singularity
[03:54] aaronblohowiak: pyrotechnick1: "rails in node" isn't going far enough.  i'd rather see lotus notes in node
[03:54] MrNibbles: liking the optimism
[03:54] pyrotechnick1: lotus notes
[03:55] micheil: I'd rather not see either of rails or lotus notes in node
[03:55] pyrotechnick1: what is lotus notes
[03:55] aaronblohowiak: lol
[03:55] pyrotechnick1: why do flowers need to remember things
[03:55] micheil: both were designed for specific contexts; rails for ruby, lotus notes for ...?
[03:55] aaronblohowiak: pyrotechnick1: the original groupware, goes back 30 years
[03:55] aaronblohowiak: micheil: mainframes
[03:55] MrNibbles: lotus notes was designed to demolish your will to live
[03:56] pyrotechnick1: i think it's extremely naive to say X wont be built for node.js
[03:56] pyrotechnick1: in 6 months you will eat those words
[03:56] pyrotechnick1: no matter what you say X is
[03:56] MrNibbles: if in 6 months someone has re-built lotus notes in node i will literally eat this coaster
[03:56] pyrotechnick1: it's not optimism it's trends
[03:56] pyrotechnick1: dude
[03:56] aaronblohowiak: pyrotechnick1: it didnt say that it wont be, i just dont think that porting the old stuff to the new platform is particularly interesting
[03:56] micheil: pyrotechnick1: it's not that X won't be built in node, but rather, when X is built in node will anyone apart from the ones who built it really care?
[03:57] micheil: it's also, node doesn't neccessarily need X
[03:57] pyrotechnick1: it's not about what node does and doesn't 'need' it's about what's going to happen
[03:57] pyrotechnick1: i wish i could be so surprised when it does happen
[03:57] pyrotechnick1: but unfortunately i would have seen it coming a mile away
[03:58] polotek: pyrotechnick1: the point is you're quibbling over opinions expressed in that presentation
[03:58] polotek: you can disagree but it doesn't help to disparage
[03:58] micheil: pyrotechnick1: think of it as "if you build a solution, would you clone an existing solution when it isn't entirely applicable to node?"
[03:58] pyrotechnick1: why does everyone assume people want to make straight ports to node
[03:58] pyrotechnick1: it's not even possible to make straight ports to node
[03:59] pyrotechnick1: you _have_ to do things async
[03:59] micheil: because, that's what a rails in node would be.
[03:59] micheil: otherwise it's no longer what the original idea was, is it?
[03:59] pyrotechnick1: you're just making the blanket statement that noone on this planet is good enough to port rails to node properly
[03:59] pyrotechnick1: it's incredibly foolish so have fun with that
[04:00] micheil: rather, if you port things to other platforms, you will always get differences; but before porting you should be asking does this solution actually apply to node?
[04:00] pyrotechnick1: rails isn't a solution
[04:00] pyrotechnick1: it's a web framework, and a good one
[04:00] pyrotechnick1: you make solutions with it
[04:00] pyrotechnick1: you hack it
[04:01] pyrotechnick1: you use it out of the box
[04:01] pyrotechnick1: however you want
[04:01] pyrotechnick1: but the point is you have rails to start with
[04:01] micheil: rails is a solution: you want to easily build dynamic websites with ruby
[04:01] aaronblohowiak: pyrotechnick1: have you seen express? it might be up your alley
[04:01] pyrotechnick1: yes we use zappa
[04:01] micheil: rails isn't a solution in terms of what you present to your client, but it does solve a problem and allow you to do your job more easily.
[04:02] pyrotechnick1: right
[04:02] polotek: I'm not sure I understand the debate anymore
[04:02] pyrotechnick has joined the channel
[04:03] MrNibbles: i think i got lost at lotus notes
[04:04] pyrotechnick has joined the channel
[04:04] aaronblohowiak: MrNibbles: haha, sorry. i should have left lotus notes out of it and simply said "platforms for creating realtime groupware"
[04:04] micheil: the whole branding of things as "X Killer" is quite wrong, because people will always still use X, they won't just magically convert to javascript because node has a rails killer. They use rails because they like ruby (often)
[04:05] micheil: or because they like ruby over all the other languages that have similar things.
[04:05] jashkenas: micheil is an X-Killer killer.
[04:06] MrNibbles: aaronblohowiak: hah
[04:06] MattJ: 
ACTION considers whether killing micheil is worth it to be an X-Killer killer killer

[04:06] pyrotechnick: merb was a rails killer
[04:06] pyrotechnick: but it got killed
[04:06] pyrotechnick: now it's just an empty shell of what it should have been around rails 3
[04:07] pyrotechnick: if i was making a rails killer id blatently brand it as such
[04:07] pyrotechnick: but hey, thats just me, im a direct person
[04:07] pyrotechnick: not out to be everyones friend
[04:07] pyrotechnick: in fact the very labelling of it as a rails killer would help keep rails people away
[04:08] vborja has joined the channel
[04:08] aaronblohowiak: pyrotechnick: there is no scarcity of frameworks, why pick fights?
[04:08] polotek: I'm out folks
[04:08] polotek: take it light
[04:08] pyrotechnick: saying you're a X-killer isn't starting a fight, you have a very loose definition of a fight
[04:08] polotek: especially pyrotechnick 
[04:08] aaronblohowiak: =)
[04:08] pyrotechnick: a fight involves blood and ambulances and pain etc
[04:09] MrNibbles: me too, gn gents
[04:09] aaronblohowiak:  Thomas Hagan is an X-Killer (sorry for the bad joke)
[04:09] polotek has left the channel
[04:09] pyrotechnick: yeah im done anyway
[04:09] aaronblohowiak: night
[04:09] pyrotechnick: its out of my system
[04:09] fangel has joined the channel
[04:09] bingomanatee_: only if you fight fair...
[04:11] bingomanatee_: The only way I'd be concerned about being a rails killer would be if any site of significant traffic used rails. 
[04:11] pyrotechnick: errr
[04:11] pyrotechnick: twitter?
[04:11] pyrotechnick: their view layer is still rails
[04:11] bingomanatee_: really.
[04:12] pyrotechnick: yellow pages
[04:12] pyrotechnick: github
[04:12] pyrotechnick: shopify
[04:12] pyrotechnick: lighthouse
[04:12] pyrotechnick: groupon
[04:12] bingomanatee_: wow - I had no idea it went that deep. 
[04:12] pyrotechnick: a list apart
[04:12] bingomanatee_: You should really jump on that freight train. 
[04:12] pyrotechnick: blinksale
[04:12] jchris has joined the channel
[04:13] pyrotechnick: penny arcade
[04:13] pyrotechnick: soundcloud
[04:13] mikeal has joined the channel
[04:13] pyrotechnick: urban dictionary
[04:13] pyrotechnick: http://www.kongregate.com/
[04:13] pyrotechnick: ruby website
[04:14] pyrotechnick: hulu
[04:14] pyrotechnick: etc
[04:14] pyrotechnick: list goes on
[04:15] pyrotechnick: id rather jump in front of a freight train
[04:15] ceej_ has joined the channel
[04:16] devnieL has joined the channel
[04:18] aaronblohowiak: pyrotechnick: what is your pre-node language of choice?
[04:18] pyrotechnick: we used jaxar for a while
[04:18] pyrotechnick: but before that mono
[04:18] pyrotechnick: c# is a great language
[04:18] pyrotechnick: despite its father
[04:19] pyrotechnick: before that rails and merb
[04:19] pyrotechnick: before that zend with php5
[04:19] aaronblohowiak: agreed, c# is excellent.  delegates ftw!!!!
[04:19] pyrotechnick: before that spagetti php
[04:19] pyrotechnick: yeah we love c#
[04:19] pyrotechnick: we did xna dev for a bit
[04:19] pyrotechnick: while we did mono
[04:19] aaronblohowiak: any unity3d dev?
[04:19] pyrotechnick: nah but there was ALOT of that at my uni
[04:20] pyrotechnick: we're doing WebGL now
[04:20] pyrotechnick: it's similar
[04:20] pyrotechnick: well i mean
[04:20] pyrotechnick: it feels similar
[04:20] pyrotechnick: same quality of games
[04:20] aaronblohowiak: yea
[04:20] pyrotechnick: dynamic, managed languages
[04:20] pyrotechnick: web plugin for unity vs webgl
[04:20] pyrotechnick: it will be quite the battle
[04:20] pyrotechnick: but i see webgl trumping it
[04:20] pyrotechnick: im biased as hell
[04:20] pyrotechnick: but i like to think of myself as a reasonably objective thinker
[04:21] pyrotechnick: webgl has alot going for it
[04:21] pyrotechnick: there's a reason we use node.js and it's not because js is a great language
[04:21] pyrotechnick: it does okay but its not incredibly incredible to use
[04:21] aaronblohowiak: so what is that reason?
[04:21] pyrotechnick: evented async
[04:21] aaronblohowiak: 
ACTION likes js

[04:22] pyrotechnick: doing realtime in evented async is actually pretty dreamy
[04:22] pyrotechnick: it's weird as hell
[04:22] aaronblohowiak: pyrotechnick: what do you do for client-side events?
[04:22] pyrotechnick: and if you suggest it to any of kind real 3d programmer they almost laugh in your face
[04:22] pyrotechnick: javascript clientside has events...
[04:22] pyrotechnick: v8 is in chrome
[04:22] v8bot: pyrotechnick: v8 <code> is no longer supported (except in PM).  Try v8: <code> or v8> <code>
[04:22] aaronblohowiak: pyrotechnick: i mean, do you have an event system for your js objs ?
[04:22] pyrotechnick: where it makes sense
[04:23] micheil: pyrotechnick: you know that node.js isn't the first evented framework?
[04:23] pyrotechnick: not everything is message passing
[04:23] pyrotechnick: yes
[04:23] micheil: or the first thing with event driven nature?
[04:23] pyrotechnick: i am mildly aware of that
[04:23] pyrotechnick: yes
[04:23] pyrotechnick: nginx for instance
[04:23] pyrotechnick: runs circles around node.js
[04:23] pyrotechnick: because its node but in c
[04:23] pyrotechnick: but
[04:23] micheil: actually, it's not node at all really.
[04:23] pyrotechnick: you know what ims aying
[04:25] aaronblohowiak: EventMachine, Twisted, Tornado, Aleph, Jetty
[04:28] pyrotechnick: the reason i got into node
[04:28] pyrotechnick: is because we were doing some eventmachined proxies in ruby
[04:28] pyrotechnick: for reasons i wont go into
[04:28] pyrotechnick: it wasnt exactly legal
[04:28] pyrotechnick: bypassing iframe restrictions etc
[04:28] pyrotechnick: but it was just such a PITA to deal with
[04:29] pyrotechnick: none of those tack on evented solutions are as elegant as node
[04:29] pyrotechnick: even if node is slightly slower
[04:29] pyrotechnick: and maybe js isnt your thing
[04:29] pyrotechnick: but youll have a much more pleasent time in node
[04:29] possibilities has joined the channel
[04:29] pyrotechnick: even given its immaturity
[04:29] pyrotechnick: anyway
[04:29] pyrotechnick: found node
[04:29] pyrotechnick: wrote an evented proxy in 20 lines
[04:30] pyrotechnick: about 80% performant as the twisted one
[04:30] pyrotechnick: much faster than the event machine one
[04:30] pyrotechnick: we were sold.
[04:30] shaver: but your princess was in another castle :-/
[04:30] herbySk has joined the channel
[04:30] pyrotechnick: meaning?
[04:30] pyrotechnick: i get the reference
[04:30] possibilities has joined the channel
[04:30] pyrotechnick: are you just trolling or do you mean something by that
[04:31] shaver: nothing, it just sounds like an epic battle
[04:31] shaver: sorry, I'm in a weird mood
[04:31] shaver: definitely not trolling
[04:32] pyrotechnick: well i will not get all creationix on your arse and stop you
[04:32] pyrotechnick: troll / be weird away
[04:32] pyrotechnick: but yes
[04:32] aaronblohowiak: =/
[04:32] pyrotechnick: our princess was in another castle
[04:32] aaronblohowiak: okay, i am going to follow my nose
[04:32] aaronblohowiak: night
[04:32] pyrotechnick: wudnt that be into your monitor?
[04:32] lin_ has joined the channel
[04:33] pyrotechnick: is that like TRON?
[04:33] mlins has joined the channel
[04:33] pyrotechnick: you entering the grid?
[04:33] hunterloftis: Anybody have experience using XMLHttpRequest uploads with node?
[04:34] pyrotechnick: sorry hunterloftis
[04:34] pyrotechnick: not i
[04:34] pyrotechnick: bingomanatee_: just had the greatest idea ever
[04:35] pyrotechnick: google.com/ just loads "public/index.js
[04:35] pyrotechnick: and the browser can execute the js
[04:35] pyrotechnick: includes done thru xhr
[04:35] bingomanatee_: Reanimate the bodies of hot checks who die early from heart attacks and program them like bioorganic sexbots. 
[04:35] bingomanatee_: ... uh o yeah that one. 
[04:35] aaronblohowiak: was tron:legacy any good?
[04:35] pyrotechnick: yeah it was aight
[04:35] pyrotechnick: im only 21 so i dunno the old ones
[04:35] bingomanatee_: it was ... a sequel.
[04:35] pyrotechnick: yeah it was a sequel apparently
[04:35] pyrotechnick: it was shiney
[04:36] aaronblohowiak: the soundtrack is pretty amazing, i've been listening to it over and over, but i <3 daft punk
[04:36] pyrotechnick: although it was half in 3d and half in 2s and there was no "its time to you put on your 3d glasses…now" bit
[04:36] pyrotechnick: so i was kind of dissapointed
[04:36] aaronblohowiak: so you wore your glasses the whole time? or do you recommend seeing it in 2d?
[04:36] pyrotechnick: yes
[04:37] pyrotechnick: it was hard to tell if the start was 3d or 2d or not
[04:37] pyrotechnick: at one point i thort it was but everyone else had their glasses off
[04:37] pyrotechnick: u can just leave them on
[04:37] pyrotechnick: but the ones here are really uncomfortable
[04:37] pyrotechnick: especially the new ones
[04:37] hunterloftis: Sounds like it's not much of an effect
[04:37] bingomanatee_: Somehow for me, the 3D was not especially effective. I think the large amounts of black kind of made the 3D effect a little unimpressive.
[04:37] pyrotechnick: yeah
[04:37] bingomanatee_: Oh! see True Grit. 
[04:37] pyrotechnick: i agree
[04:37] bingomanatee_: THAT was an amazing movie. 
[04:38] aaronblohowiak: bingomanatee_: yes, best movie i have seen in a very long time
[04:38] pyrotechnick: really guys?
[04:38] bingomanatee_: Its a Cohen bros. movie. 
[04:38] pyrotechnick: oh badass
[04:38] pyrotechnick: its a western
[04:38] bingomanatee_: Very effective performances. 
[04:38] aaronblohowiak: the dialog is also amaaaaazing
[04:38] bingomanatee_: Interestingly enough - it has the same lead actor  as Tron. 
[04:38] aaronblohowiak: i loled, sometimes i was the only one in the theatre loling
[04:38] pyrotechnick: haha
[04:39] pyrotechnick: is that the original tron?
[04:39] aaronblohowiak: bingomanatee_: the lead actor of True Git is a 14 year old girl
[04:39] pyrotechnick: the other dude is tron
[04:39] aaronblohowiak: lol
[04:39] bingomanatee_: given. 
[04:39] aaronblohowiak: True Grit is appropriately epic, but never boring. 
[04:39] bingomanatee_: But Jeff Bridges was in both movies. 
[04:40] pyrotechnick: :(
[04:40] pyrotechnick: http://www.jspp.io/
[04:40] pyrotechnick: php :( :(
[04:40] pyrotechnick: it's happening
[04:40] pyrotechnick: people are going to think that's real node
[04:40] aaronblohowiak: buying .io domains is the new status symbol 
[04:40] pyrotechnick: yeah
[04:40] pyrotechnick: we are about to get one lol
[04:41] bingomanatee_: I don't suppose Oh.io is up for grabs...
[04:41] pyrotechnick: doubt it
[04:42] jamescarr: bleh
[04:43] vborja has joined the channel
[04:44] masahiroh has joined the channel
[04:47] arlolra has joined the channel
[04:47] bingomanatee_: wow - you can't even get an io domain on godddy.
[04:47] pyrotechnick: yeah 
[04:47] pyrotechnick: im trying to get a .co
[04:47] gabeh has joined the channel
[04:47] mjr_: But you can get Danica Patrick
[04:48] pyrotechnick: theyve gone up in price
[04:48] pyrotechnick: godaddy front-run you know
[04:48] pyrotechnick: i wouldnt use them
[04:48] pyrotechnick: like if you come up with a sweet domain
[04:48] pyrotechnick: never type it into godaddy
[04:48] pyrotechnick: ever
[04:48] pyrotechnick: under any circumstances
[04:49] pyrotechnick: i had one i went to hand register
[04:49] pyrotechnick: that was available
[04:49] pyrotechnick: straight after i type it in it's mysteriously owned by godaddy
[04:49] Aria: Heh, yeah. Speculative registration, then returned before the grace period's up.
[04:49] aaronblohowiak: whaaaat?
[04:49] pyrotechnick: i dont think theres grave anymore
[04:49] Aria: Though ICANN has cracked down on that lately.
[04:49] pyrotechnick: because of the fee
[04:49] pyrotechnick: yeah
[04:50] Aria: Yeah, exactly.
[04:50] pyrotechnick: but they still front run
[04:50] pyrotechnick: if they know its worth more than the fee
[04:50] Aria: (I must say, being a Tucows/OpenSRS reseller has been a boon. They're super smooth to work with, even if they don't /do/ polish.)
[04:50] pyrotechnick: the fee is only like 99 cents to them
[04:50] Aria: (and I hear good about Joker.com for individual registrations)
[04:51] bingomanatee_: i don't see an io option on joker.com
[04:52] pyrotechnick: anyone got a .co?
[04:52] cardona507 has joined the channel
[04:53] bingomanatee_: okay - so - who would you go to to get a domain.io
[04:53] eee_c has joined the channel
[04:53] aaronblohowiak: nic.io
[04:53] micheil: personally I like namecheap..
[04:54] Aria: .co I can do via opensrs, I know.
[04:54] dthompson: name.com 
[04:54] devnieL has left the channel
[04:54] riven` has joined the channel
[04:59] pyrotechnick has joined the channel
[05:02] ossareh has joined the channel
[05:03] aaronblohowiak: when developing client libraries, how great a sin is it to require jquery?
[05:05] bingomanatee_: Mission accomplished via name.com
[05:05] bingomanatee_: thanks all
[05:06] aaronblohowiak: wow name.come is far cheaper than nic.io
[05:07] pyrotechnick1 has joined the channel
[05:10] coffee has joined the channel
[05:13] _1marc has joined the channel
[05:23] noahcampbell has joined the channel
[05:24] saikat has joined the channel
[05:25] rasactive has joined the channel
[05:28] zentoooo has joined the channel
[05:32] bingomanatee_: name.come? for porn domains?
[05:35] bingomanatee_: yeah -- I went there. 
[05:39] hunterloftis: Any recommendations on the best simple node graphics lib for basic stuff like resizing uploaded images into thumbs?
[05:39] _1marc has left the channel
[05:40] mfernest: bingomanatee: you're going to frighten these kids
[05:40] pyrotechnick1: hunterloftis: node-canvas
[05:40] hunterloftis: pyrotechnick1: Yeah I'm looking at that now, good with speed and a familiar interface, but for something as simple as resizing to 128x128 it might be overkill
[05:43] Aria: I'm still not sure why people are afraid of overkill when it brings in a dependency that people use frequently. It's like saying "unix is overkill, use DOS"
[05:43] hunterloftis: Aria: because I would prefer not to write a bunch of png loading, canvas drawing, etc operations designed for a browser when what I need is whatever.resizeTo(128, 128, callback)
[05:44] Aria: That sounds like underkill then ;-)
[05:44] hunterloftis: Aria: not functional overkill, but complexity overkill
[05:44] hunterloftis: ;)
[05:46] hunterloftis: gm seems to be the tool for this task: gm(filename).resize(128,128).write(file, callback)
[05:46] amerine has joined the channel
[05:50] coffee has joined the channel
[05:50] bingomanatee_: Unix is a bit heavy. 
[05:50] aaronblohowiak: lol
[05:50] pyrotechnick1: i prefer bios
[05:50] bingomanatee_: Perhaps a js version would be better?
[05:51] possibil_ has joined the channel
[05:51] possibi__ has joined the channel
[05:53] benburkert has joined the channel
[05:53] Ond has joined the channel
[05:53] possib___ has joined the channel
[06:06] jpld has joined the channel
[06:15] robotarmy has joined the channel
[06:16] pyrotechnick has joined the channel
[06:16] sudoer has joined the channel
[06:18] TheEmpath has joined the channel
[06:18] TheEmpath: hail nodites
[06:18] TheEmpath: any known way to emulate connections with socket.io?
[06:19] bingomanatee_: nick you tere?
[06:19] pyrotechnick has joined the channel
[06:20] pkrumins: TheEmpath: what kind of connections?
[06:20] pyrotechnick: how do i get node.js on port 80 without running it as root
[06:20] TheEmpath: any really.  anything that simluates connections my server can communicate with and vice versa
[06:20] pyrotechnick: tried authbind but it doesnt work on latest ubuntu
[06:21] pkrumins: TheEmpath: well socket.io is "connections"
[06:21] TheEmpath: hrmm
[06:21] TheEmpath: how can i simulate clients without a browser?
[06:22] markstory: curl
[06:22] TheEmpath: curl your face!
[06:22] TheEmpath: i need persistant connections
[06:23] markstory: well if you don't have a browser, you would need some other application to make connections.
[06:23] markstory: You could make another node app, talk to your first one.
[06:23] TheEmpath: ya
[06:23] TheEmpath: im thinking along those lines as well
[06:25] pyrotechnick1 has joined the channel
[06:26] mjr_: pyrotechnick1: what's wrong with running as root?
[06:26] richcollins has joined the channel
[06:26] pyrotechnick1: hahaha
[06:27] pyrotechnick1: is that a joke?
[06:28] sideshow has joined the channel
[06:28] mjr_: Only 25% joking
[06:28] Aria: pyrotechnick1: You start as root, then drop privilege. Setuid!
[06:28] pkrumins: mjr_: did you get my messages? i sent you some. i am so sorry about my behavior.
[06:29] mjr_: In this day of cheap and easy VPS where hardly any other users are on an instance, root is about the same as any other user.
[06:29] mjr_: pkrumins: yes, I did.  It's cool.  Glad you are OK
[06:29] Aria: Except that root can erase the whole filesystem.
[06:29] pkrumins: thanks to you for that!
[06:29] markstory: pyrotechnick1: you could put something else running as root like a loadbalancer?
[06:30] pkrumins: great that it's cool. :)
[06:30] mjr_: pyrotechnick1: also, yes, you can change you userid after you get the listening event.
[06:30] Aria: (Hell, I on my own machines use a half dozen uids just to keep things from breaking each other)
[06:30] mikeal has joined the channel
[06:30] mjr_: But the whole root part of the filesystem is just some stock OS install.  The valuable data is that owned by the users.
[06:31] mikew3c has joined the channel
[06:31] mjr_: Anyway, if it's bad, then it's bad.  Either authbind, or drop privs after you've bound.
[06:31] Aria: Yes, which I keep apart from each other. (And often have a daemon between the parts of)
[06:31] mjr_: I increasingly don't see the point though.
[06:31] Aria: It's saved my ass a bunch.
[06:35] mjr_ has joined the channel
[06:36] hunterloftis: Does node-canvas only support PNG input?
[06:38] mikew3c has joined the channel
[06:39] pquerna: dance.
[06:40] benburkert has joined the channel
[06:41] pquerna: its unfortunate that the number of local root exploits has largely made running as non-root largely ineffective.
[06:41] pquerna: however the growth of cheap cloud/vps/whatever means its easy to find machiens of the perfect size for a specific task
[06:41] pquerna: limiting the need to have multiple apps on one machine
[06:41] cognominal has joined the channel
[06:42] pquerna: however, its still not a bad best practice. security is all about layers, onions, ogres, picket fences, etc.
[06:42] trotter has joined the channel
[06:43] shinmei has joined the channel
[06:46] cromape has joined the channel
[06:47] Aria: I haven't had a local root exploit on my own servers in .. man, six or seven yers.
[06:47] Aria: With a thousand active users.
[06:51] pyrotechnick1: dah
[06:51] pyrotechnick1: never use sudo
[06:51] Aria: Nah?
[06:52] TheEmpath: doing a net.createConnection to a socket.io server is insufficient :(
[06:53] googol: in node, how can I get the home dir, as ~ seems to not be working
[06:53] Aria: process.env['HOME'] or look it up in /etc/passwd
[06:54] Aria: ~ is a shellism. It's not a part of the path.
[06:57] pquerna: Aria: huh, I meant, like, the linux kernel local root exploits, of which there have been several in 2010
[06:57] Aria: Yes, if not properly mitigated.
[06:57] pquerna: Aria: yes, sometimes you have 'nice' users, but unless you are using ksplice or rebooting more than once a month with a new kernel
[06:57] MikhX has joined the channel
[06:57] Aria: (also, if you run a configuration that's tighter than usual.)
[06:58] pyrotechnick1: i have a dev user i was using to host node apps on not port 80 right
[06:58] Aria: And ksplice rocks my world for things actually running $wellknowndistro.
[06:58] pyrotechnick1: now i want to just make it work quickly on 80 with sudo
[06:58] pyrotechnick1: but if i sudo it it screws with the paths and cant find node
[06:58] _1marc has joined the channel
[06:58] pquerna: sudo -E
[06:58] pyrotechnick1: yeah doesnt help
[06:58] Aria: IF your sudo is new, you can add to the path on the commandline
[06:59] Aria: sudo PATH=$PATH commadn
[06:59] Aria: If not, sudo env PATH=$PATH command
[07:01] pquerna: Aria: even with things like ksplice -- they are good, to me it just feels like russian roulete -- trusting that there aren't 0 day local kernel exploits -- virtuatlization is so cheap and.. well. effective enough.  there have been relatively few (any recently?) break outs of xen doms compared to the number of kernel exploits that a local user could do;  why even take the risk now days?
[07:01] Aria: Depends on your uses! Also, cost of administration.
[07:01] Aria: Also, RAM.
[07:02] pquerna: i buy cost of administration far more than ram
[07:02] pquerna: ram keeps getting cheaper, people don't
[07:02] Aria: Motherboards that hold gobs of RAM don't.
[07:04] Aria: (And actually, on my own systems, I haven't actually been vulnerable to any of the local root exploits in the past few years. grsecurity is awesome.)
[07:04] Aria: (I keep reading notes, checking and then realizing that something broke the exploit.)
[07:06] bartmax has joined the channel
[07:13] RichardJ: good morning
[07:13] pyrotechnick1: hi
[07:13] benburkert has joined the channel
[07:14] SvenDowideit_ has joined the channel
[07:14] muhqu has joined the channel
[07:16] sveimac has joined the channel
[07:18] TheEmpath: if i start my node server, and go to http://184.106.93.156:8000/ it works.  if i put in the request to var client = net.createConnection(8000, 'localhost'); after my node server is up... i get a Error: ECONNREFUSED, Connection refused
[07:18] TheEmpath: the same for 127.0.0.1, the internal IP and the external IP
[07:21] ErikCorry has joined the channel
[07:28] ezmobius has joined the channel
[07:29] evanmeagher has joined the channel
[07:31] possibilities has joined the channel
[07:34] simme has joined the channel
[07:35] JojoBoss: can you change the port its running on and try again? I don't know if that would affect it or not, but i'd make sure its not just that port or something
[07:35] JojoBoss: also, i think if you leave out localhost it assumes it
[07:36] admc has joined the channel
[07:37] bingomanatee_: http://www.youtube.com/watch?v=kWiB38jayIs
[07:41] fangel has joined the channel
[07:41] herbySk has joined the channel
[07:43] masahiroh has joined the channel
[07:43] MikhX has joined the channel
[07:51] beawesomeinstead has joined the channel
[07:53] opengeard_ has joined the channel
[07:56] pquerna: http://twitter.com/#!/wadey/status/21822413851336704
[07:58] derren13 has joined the channel
[08:05] devdrin__: having trouble extending the Object.prototype in node from a require
[08:05] devdrin__: is the convential way okay
[08:05] devdrin__: ?
[08:06] mAritz has joined the channel
[08:06] femtoo has joined the channel
[08:07] pyrotechnick has joined the channel
[08:08] ttpva has joined the channel
[08:14] desaiu has joined the channel
[08:15] mraleph has joined the channel
[08:18] SamuraiJack has joined the channel
[08:20] herbySk has joined the channel
[08:23] ezmobius_ has joined the channel
[08:27] devdrin__: alt+3
[08:27] xSmurf: is stream.write completely blocking?
[08:28] thinkingpotato has joined the channel
[08:28] xSmurf: if I try to write my data one byte at a time in nextTicket the connection drops after the first byt
[08:28] xSmurf: e
[08:29] Aria: Not at all.
[08:30] CrypticSwarm has joined the channel
[08:31] xSmurf: weird my xmpp connection drops when I transfer file on an socks proxy
[08:31] vineyard has joined the channel
[08:32] pyrotechnick1 has joined the channel
[08:34] Druid_ has joined the channel
[08:40] sveimac has joined the channel
[08:41] xSmurf: oh something unrelated it seems
[08:42] fangel has joined the channel
[08:42] pyrotechnick1: wats with all the dcs today
[08:43] TheEmpath has joined the channel
[08:43] TheEmpath: 80 users... 400+ requests per second... 67% CPU usage... 96MB of RAM... oh yeah, baby. Oh hell yeah.
[08:43] pyrotechnick1: wats that
[08:44] TheEmpath: imhotep!
[08:44] TheEmpath: https://github.com/cainiac/imhotep
[08:44] TheEmpath: an extended socket server that im trying to design for massive load usage for things like games or apps
[08:45] pkrumins: not bad
[08:45] yozlet has joined the channel
[08:46] TheEmpath: that was just hammering the server with rapid server connects
[08:46] TheEmpath: 1000 per second
[08:46] TheEmpath: it was socket.io that went under first lol
[08:46] TheEmpath: once i can simulate connections reliably, i'll be able to brute it with more elaborat ethings
[08:46] TheEmpath: things*
[08:46] pyrotechnick1: Tell them what else they've won, Al!
[08:49] pietern has joined the channel
[08:50] romeo_ordos has joined the channel
[08:50] romeo_ordos has left the channel
[08:52] markwubben has joined the channel
[08:55] beawesomeinstead has joined the channel
[08:55] beawesomeinstead has joined the channel
[08:57] d0k has joined the channel
[08:57] Gruni has joined the channel
[08:58] Gruni has joined the channel
[08:58] riven has joined the channel
[09:01] saikat has joined the channel
[09:02] Ond has joined the channel
[09:04] desaiu: also relevant on this channel...
[09:04] desaiu: what open source javascript projects can i get involved with?
[09:07] MikhX has joined the channel
[09:08] ezmobius has joined the channel
[09:08] pkrumins: there are a million, desaiu.
[09:09] pkrumins: i'd suggest just going through latest js projects on github
[09:09] fangel has joined the channel
[09:09] pkrumins: and seeing which one you like
[09:10] desaiu: okay, i will do this i suppose when i don't suck 
[09:10] pkrumins: dont worry about it, just start hacking on something
[09:11] sveimac has joined the channel
[09:11] ezmobius has joined the channel
[09:13] romainhuet has joined the channel
[09:13] frodenius has left the channel
[09:14] micheil has joined the channel
[09:16] matjas has joined the channel
[09:20] herbySk has joined the channel
[09:21] herbySk: v8: [3,1,4].concat()
[09:21] v8bot: herbySk: [3, 1, 4]
[09:22] herbySk: v8: var a = [3,1,4]; a.concat() === a
[09:22] v8bot: herbySk: false
[09:22] adambeynon has joined the channel
[09:22] oberhamsi has joined the channel
[09:25] micheil: mikeal: jspp is yours?!
[09:26] mikeal: yup
[09:26] mikeal: i spent a few days during my forced vacation on it :)
[09:26] micheil: *shudders slightly*
[09:26] mikeal: been talking about it for a while with issacs and a few other people
[09:26] mikeal: it's only superficially like php :)
[09:27] micheil: it looks soo damn wrong though
[09:27] mikeal: haha
[09:27] WaterCooled has joined the channel
[09:27] mikeal: ignore the <?jspp and use the <script type="application/jspp">
[09:27] micheil: even so
[09:28] mikeal: hahaha
[09:28] micheil: it takes all the logic that was separate and merges it back in
[09:28] mikeal: it was never really seperate
[09:28] mikeal: that's an artificial abstraction 
[09:28] micheil: true, but it's a good abstraction imho.
[09:28] andrewfff has joined the channel
[09:28] mikeal: and you have the node.js module system
[09:29] pyrotechnick1: github have broken their new repo button
[09:29] pyrotechnick1: fags
[09:29] hunterloftis has joined the channel
[09:29] mikeal: so if you want to stick all your code in models
[09:29] mikeal: you can do that and just require the module
[09:29] mikeal: then only stick the presentation logic for server side processing the jspp blocks
[09:29] herbySk: v8: (function () { return [].concat.apply(arguments) })('foo', 3, 'bar')
[09:29] v8bot: herbySk: [{"0": "foo", "1": 3, "2": "bar"}]
[09:29] micheil: mikeal: it also disables the idea of routing to files that don't exist
[09:29] mikeal: but the filesystem as the API can be wonderful
[09:29] romainhuet_ has joined the channel
[09:29] mikeal: yeah, it's also a huge advantage
[09:29] pyrotechnick1: he's trolling
[09:30] micheil: mikeal: we should talk later; :P
[09:30] mikeal: easily cachable, hot reloading
[09:30] micheil: pyrotechnick1: who's trolling?
[09:30] mikeal: i'm also working on something to allow "routes" like stuff
[09:30] pyrotechnick1: are routes just symlinks lol
[09:31] mikeal:  /dir/user.jspp
[09:31] micheil: pyrotechnick1: fwiw, I wouldn't try trolling mikeal :P
[09:31] micheil: bbl
[09:31] mikeal: is an html page with jspp htat defines routes in the init block
[09:31] pquerna: mikeal: have you looked at xhp much?
[09:31] mikeal: so you could do /dir/user/blah
[09:31] mikeal: and that'll route to the user.jspp
[09:31] mikeal: still get all the caching and hot reloading benefits
[09:31] mikeal: what's xhp?
[09:32] pquerna: mikeal: http://www.facebook.com/notes/facebook-engineering/xhp-a-new-way-to-write-php/294003943919 https://github.com/facebook/xhp
[09:32] mikeal: i really can't read PHP anymore
[09:32] mikeal: because whenever I see $ i think "jQuery"
[09:33] mikeal: this is interesting
[09:33] mikeal: the stuff that makes jspp a template system I was kinda forced to add by isaacs
[09:33] mikeal: I didn't want you to actually output from inside the syntax, and i settled on that write function, but you don't have it HTML
[09:33] mikeal: you just have dom and jQuery
[09:34] mikeal: and eventually you can swap jQuery for whatever crazy thing you want
[09:34] mikeal: YUI people only use YUI ;)
[09:35] pquerna: yes, i'm still not quite convinced having a full server-side dom for html generation is really what i want.
[09:35] mikeal: it's REALLY expensive
[09:35] pquerna: i do like xhp mostly for its auto-escaping stuff
[09:35] mikeal: that's why i'm so rigorous about caching
[09:35] pquerna: making it hard to mess up
[09:35] pquerna: anti-xss stuff just needs to be on by default etc
[09:35] pquerna: or people won't do it right
[09:36] mikeal: jQuery does all the auto-escaping for you as well
[09:38] tanepiper: mikeal: ditto on the PHP
[09:38] mikeal: someone on reddit actually said "I can't tell what the example does, what is $?"
[09:39] tanepiper: looks like chinese to me now
[09:39] mikeal: i couldn't figure out how to comment on reddit to explain that it's jquery
[09:39] pyrotechnick1 has left the channel
[09:41] sveimac has joined the channel
[09:41] fangel has joined the channel
[09:43] pyrotechnick1 has joined the channel
[09:43] ezmobius_ has joined the channel
[09:43] pyrotechnick1: anyone know about something like https://github.com/mnutt/hummingbird
[09:44] saschagehlich has joined the channel
[09:44] fangel: depends what you mean by "something like" :)
[09:44] altamic has joined the channel
[09:47] mikeal: that new thing mnot did
[09:47] mikeal: there it is
[09:47] mikeal: https://github.com/mnot/htracr
[09:47] mikeal: that's "like hummingbird"
[09:48] pyrotechnick1: you have an extremely loose definition of 'like'
[09:50] mikew3c_ has joined the channel
[09:50] davidc_ has joined the channel
[09:50] fangel: I'm working on a not-yet-released real('ish)-time event-tracking system, that can be used to track page-views.. :)
[09:50] fangel: so I know if it, but nobody else knows of it ;)
[09:51] pyrotechnick1: why dont u release it
[09:51] fangel: because it's not done yet.. I'm about 70-80% there.. 
[09:52] pyrotechnick1: cool
[09:52] pyrotechnick1: make an empty repo
[09:52] pyrotechnick1: so i can watch it plz
[09:52] matjas_ has joined the channel
[09:53] fangel: https://github.com/fangel/Mellow - as empty as can be :)
[09:55] tim_smart has joined the channel
[09:57] pgte has joined the channel
[09:58] saschagehlich: did anyone get socket.io running on opera 10?
[09:59] Ond: Yep
[09:59] saschagehlich: Ond: what did you have to do to get it working?
[09:59] Ond: It worked without any modifications
[09:59] Ond: Why what is yours doing
[10:00] saschagehlich: I guess the problem is that the socket is running on another host than the web page
[10:00] saschagehlich: which causes a security issue
[10:00] zemanel has joined the channel
[10:01] claint` has joined the channel
[10:02] claint` has left the channel
[10:02] femtooo has joined the channel
[10:11] sveimac_ has joined the channel
[10:14] b_erb has joined the channel
[10:14] opengeard has joined the channel
[10:15] crohr has joined the channel
[10:16] opengeard has joined the channel
[10:16] claint` has joined the channel
[10:17] claint` has left the channel
[10:22] herbySk has joined the channel
[10:23] Phyllio has joined the channel
[10:30] tisba has joined the channel
[10:32] shinmei has joined the channel
[10:34] ezmobius_ has joined the channel
[10:34] saschagehlich: ircretary: remind rauchg_ to have a look at socket.io security issues with opera 10. (crossdomain stuff)
[10:34] ircretary: saschagehlich: I'm not sure what to do with that command.  Ask for help in PM.
[10:35] saschagehlich: ircretary: tell rauchg_ to have a look at socket.io security issues with opera 10 (crossdomain stuff)
[10:35] ircretary: saschagehlich: I'll be sure to tell rauchg_
[10:35] ezmobius has joined the channel
[10:36] b_erb: is it good practice to prefer 'once()' over 'on()' when applicable or just syntactic sugar?
[10:36] pyrotechnick has joined the channel
[10:37] saschagehlich: on() is just syntactic sugar I guess. usually you use addEventListener
[10:37] pyrotechnick has left the channel
[10:38] pyrotechnick has joined the channel
[10:38] jetienne has joined the channel
[10:39] pyrotechnick: does anyone with node debugging experience want to help with https://github.com/feisty/node-bullet
[10:41] teemow has joined the channel
[10:41] devcorkvince has joined the channel
[10:41] cafesofie has joined the channel
[10:42] sveimac_ has joined the channel
[10:50] hellp has joined the channel
[10:51] pgte has left the channel
[10:52] pgte has joined the channel
[10:52] tfcoding1 has joined the channel
[10:53] zorzar has joined the channel
[10:57] pgte: does anyone knows a good and fast crc / md5/sha1 hash generation module that can be installed via npm?
[10:57] pgte: s/knows/know
[10:59] SvenDowideit has joined the channel
[11:00] micheil has joined the channel
[11:05] jetienne: pgte: md5 is alraedy in nodejs
[11:05] jetienne: http://nodejs.org/docs/v0.2.6/api.html#crypto-createhash-240
[11:05] pgte: yup, any alternative you know so I can benchmark against?
[11:06] jetienne: pgte: no, but i dont think you need it. it is using openssl in the background
[11:07] Ond: hash?
[11:07] Ond: npm view hash
[11:07] Ond: and rbytes
[11:07] Ond: Maybe not rbytes
[11:07] pgte: > require('hash');
[11:07] pgte: ReferenceError: Class is not defined
[11:08] pgte: jetienne: maybe a quick CRC will be faster, don't know
[11:11] Vertice has joined the channel
[11:12] sveimac has joined the channel
[11:20] mr_daniel has joined the channel
[11:22] tfcoding1 has left the channel
[11:22] jetienne: pgte: crc is MUCH faster than md5
[11:23] fermion has joined the channel
[11:23] jetienne: pgte: to get a comparison of hash function speed you can use openssl cmdline
[11:23] pgte: jetienne: thanks
[11:24] case__: pgte, is security a concern, or just speed ?
[11:24] pgte: case__: speed
[11:24] jetienne: "openssl speed" and just wait a long time :)
[11:25] pgte: case__: it's for local file corruption detection
[11:25] pgte: case__: thanks!
[11:25] mpoz2 has joined the channel
[11:26] case__: pgte, i personnaly will go for sha1 even if you only want speed
[11:27] case__: pgte, crc32 is just a joke, and the speed difference between md5 and sha1 is not worth it
[11:27] pgte: case__: thanks for the advise! :)
[11:28] maushu has joined the channel
[11:28] case__: (and, it's already in node.js)
[11:29] maushu: Anyone knows if anyone did an attempt to add metadata to javascript code?
[11:30] maushu: I'm guessing that there isn't an standard, so... better follow someone else instead of reinventing the wheel.
[11:30] ajpiano has joined the channel
[11:34] matjas_: using the latest Node.js, I get process.platform === 'PLATFORM'
[11:34] matjas_: anyone else having this problem? is this a known bug?
[11:35] mape: Anyone looked at using node.js with the google analytics api?
[11:36] maushu: mape, no. They steal your soul!
[11:36] mape: maushu: :& Well.. I still wanna
[11:37] mape: Guess I'll take a look at it later
[11:37] maushu: ...your soul must be very cheap.
[11:37] mape: Don't worry, I have others to sell
[11:38] herbySk has joined the channel
[11:39] fangel has joined the channel
[11:40] mraleph has joined the channel
[11:42] sveimac has joined the channel
[11:43] ewdafa has joined the channel
[11:43] xla has joined the channel
[11:57] derren13 has joined the channel
[11:57] oberhamsi has joined the channel
[12:04] SamuraiJack_ has joined the channel
[12:05] saschagehlich: does anyone know to convert html to plain text, so like replacing all html special chars with utf8 characters?
[12:12] fly-away has joined the channel
[12:13] sveimac has joined the channel
[12:13] Rv has joined the channel
[12:16] oal has joined the channel
[12:29] devdrinker has joined the channel
[12:32] sivy has joined the channel
[12:33] markwubben has joined the channel
[12:35] saschagehlich: node-faq: commands
[12:35] saschagehlich: node-faq: help
[12:35] saschagehlich: hm.
[12:36] maushu: wat
[12:37] kolor has joined the channel
[12:43] sveimac has joined the channel
[12:52] stepheneb has joined the channel
[12:53] Rv has joined the channel
[12:53] fermion has joined the channel
[12:53] mr_daniel has joined the channel
[12:53] teemow has joined the channel
[12:53] rasactive has joined the channel
[12:53] forzan has joined the channel
[12:53] tg has joined the channel
[12:53] shachaf has joined the channel
[12:53] lstoll has joined the channel
[12:53] PyroPeter has joined the channel
[12:53] freeformz has joined the channel
[12:53] Tidwell has joined the channel
[12:53] Twelve-60 has joined the channel
[12:53] cnu has joined the channel
[12:53] aconran__ has joined the channel
[12:53] mfernest has joined the channel
[12:53] Kai` has joined the channel
[12:53] tanepiper has joined the channel
[12:53] zedas has joined the channel
[12:53] rjack has joined the channel
[12:53] killfill has joined the channel
[12:53] dispalt has joined the channel
[12:53] EGreg_ has joined the channel
[12:53] kkaefer has joined the channel
[12:53] aakour has joined the channel
[12:53] thedjinn has joined the channel
[12:53] beppu_ has joined the channel
[12:53] Gregor has joined the channel
[12:53] SubStack has joined the channel
[12:53] BHSPitMo1kay has joined the channel
[12:53] wao has joined the channel
[12:53] vileda has joined the channel
[12:53] Prometheus has joined the channel
[12:54] fjakobs has joined the channel
[13:02] crohr has joined the channel
[13:07] shripadk has joined the channel
[13:11] mikedeboer has joined the channel
[13:11] lwille has joined the channel
[13:13] sveimac has joined the channel
[13:22] ianward has joined the channel
[13:27] derren13 has joined the channel
[13:28] modular has joined the channel
[13:30] cwo has joined the channel
[13:32] adambeynon has joined the channel
[13:35] hellp has joined the channel
[13:44] sveimac has joined the channel
[13:44] hellp: Anyone using/hacking zombie.js? Does "npm link <zombiejsdir>" work for you? It doesn't for me. Using npm and zombie latest; node 0.2.5. Not sure if it's my install or a general problem.
[13:51] malkomalko has joined the channel
[13:51] heavysixer has joined the channel
[13:51] fermion has joined the channel
[13:52] kriszyp has joined the channel
[13:52] m0rph3v5 has joined the channel
[13:52] m0rph3v5 has left the channel
[13:52] m0rph3v5 has joined the channel
[13:52] m0rph3v5: hi
[13:53] ianward has joined the channel
[13:53] thinkingpotato has joined the channel
[13:55] fumanchu182 has joined the channel
[13:55] m0rph3v5: anyone here?
[13:59] pgte: yo
[14:01] RichardJ: little post about JS: http://goo.gl/NHQLw
[14:01] RichardJ: replies are welcome, the discussion may be interesting :)
[14:02] mraleph: RichardJ: function () {} () is not a valid JS afaik
[14:03] mraleph: v8: function () { } ()
[14:03] v8bot: mraleph: SyntaxError: Unexpected token (
[14:03] mraleph: RichardJ: ^^^^ see
[14:03] dothebart has joined the channel
[14:03] dothebart: hy everyone
[14:03] wink_ has joined the channel
[14:03] shripadk: hello
[14:04] RichardJ: mraleph: pay attention to "var a=" before it.
[14:04] oberhamsi has joined the channel
[14:04] RichardJ: this is valid and will alert 5.
[14:05] dothebart: i'm trying to adopt c-ares lookup on another project, and have been gdb'ing node_cares.cc to find out where it actualy sends / receives the query...
[14:05] dothebart: and i'm a little lost here. node handles the i/o via libev here?
[14:06] mraleph: v8: var a = function (b) { console.log(b); } (1)
[14:06] v8bot: mraleph: CONSOLE: [1], OUTPUT: undefined
[14:06] jherdman has joined the channel
[14:06] mraleph: v8: var a = function (b) { console.log(b); } (function () { return 1; } ())
[14:06] v8bot: mraleph: CONSOLE: [1], OUTPUT: undefined
[14:06] mraleph: jessus christ
[14:06] shripadk: v8bot: (function(){return true;}())
[14:06] v8bot: shripadk: Use v8: <code> to evaluate code or "`v commands" for a list of v8bot commands.
[14:06] beta_ has joined the channel
[14:07] Imperion has joined the channel
[14:07] RichardJ: v8: alert(void 5);
[14:07] v8bot: RichardJ: ReferenceError: alert is not defined
[14:07] mraleph: abomination 
[14:07] RichardJ: of course
[14:07] RichardJ: v8: void 5;
[14:07] v8bot: RichardJ: undefined
[14:08] mraleph: javascript is one of the worst languages i have ever seen. 
[14:08] shripadk: v8bot: `v commands
[14:08] v8bot: shripadk: Use v8: <code> to evaluate code or "`v commands" for a list of v8bot commands.
[14:08] beta_ has joined the channel
[14:08] shripadk: `v commands
[14:08] v8bot: shripadk: Commands: Type `v <command>.  Optionally, type `v <command> @ <nick> to send to a specific user. 				`v Commands are: about, beers, commands, git, google, help, macro. Other commands: v8, `re, `pcre, `ref.  Type `v help <command> for more information.  Join #v8bot for more support.
[14:09] SamuraiJack__ has joined the channel
[14:09] RichardJ: mraleph: you just ignore the bad parts, then it's a pretty good language
[14:09] mraleph: RichardJ: good language should have no parts to ignore
[14:09] RichardJ: mraleph: name one language with no bad parts.
[14:10] mraleph: RichardJ: C
[14:10] RichardJ: and yes, difficult syntax is a bad part.
[14:10] mraleph: Lua, Haskell
[14:10] mraleph: these three languages have no bad parts.
[14:10] RichardJ: C definitely has a bad part.
[14:10] RichardJ: it's hard to learn and hard to use
[14:10] RichardJ: that is a very bad part
[14:11] RichardJ: lua and haskell are indeed very good languages
[14:11] RichardJ: but I'm sure they have bad parts as well
[14:11] RichardJ: other way around: you don't seem to know them if there are any. that's good.
[14:11] RichardJ: I wish that was true of most JS-users
[14:12] RichardJ: just because a "feature" exists, doesn't mean it's good.
[14:12] mraleph: I like Lua's consistency (though I despise certain changes in upcoming Lua 5.2, like abomination called _ENV)
[14:13] RichardJ: I mean, C has fallthrough in the switch statement, which I consider a bad part
[14:13] RichardJ: it's *very* dangerous and easy to look over
[14:13] chrischris has joined the channel
[14:14] sveimac has joined the channel
[14:14] b_erb has joined the channel
[14:18] romanb has joined the channel
[14:28] shripadk has left the channel
[14:31] augustl has joined the channel
[14:34] MattJ has joined the channel
[14:37] stepheneb has joined the channel
[14:38] stepheneb has joined the channel
[14:39] davidsklar has joined the channel
[14:43] fermion has joined the channel
[14:48] fumanchu182 has joined the channel
[14:50] c4milo has joined the channel
[14:54] dnolen has joined the channel
[14:55] davglass has joined the channel
[14:56] themiddleman has joined the channel
[14:57] JusticeFries has joined the channel
[15:00] figital has joined the channel
[15:01] paulrobinson has joined the channel
[15:03] SamuraiJack has joined the channel
[15:03] saschagehlich has joined the channel
[15:08] akahn has joined the channel
[15:09] dmcquay has joined the channel
[15:11] davidwalsh has joined the channel
[15:13] galaxywatcher has joined the channel
[15:15] Ezku\ has joined the channel
[15:15] cgutierrez has joined the channel
[15:17] masahiroh has joined the channel
[15:17] booths has joined the channel
[15:17] ceej has joined the channel
[15:19] tilgovi has joined the channel
[15:20] booths: How would one forcefully disconnect a client w/ socket.io... I've tried deleting them from socket.clients, but that seems to have no effect on client side.
[15:21] galaxywatcher has joined the channel
[15:21] BillyBreen has joined the channel
[15:22] matt_c has joined the channel
[15:24] muhqu_ has joined the channel
[15:26] elijah-mbp has joined the channel
[15:26] hendrikcdn has joined the channel
[15:27] tibbetts has joined the channel
[15:27] ajpiano has joined the channel
[15:29] rsms has joined the channel
[15:32] davidsklar has left the channel
[15:32] sveimac has joined the channel
[15:32] amacleod has joined the channel
[15:33] davidsklar has joined the channel
[15:34] dmcquay has joined the channel
[15:37] bingomanatee has joined the channel
[15:37] fangel has joined the channel
[15:37] bingomanatee: Hey Nick you up?
[15:40] ben_alman has joined the channel
[15:44] quest88 has joined the channel
[15:45] unomi has joined the channel
[15:45] markwubben has joined the channel
[15:46] derren13 has joined the channel
[15:46] boaz_ has joined the channel
[15:47] mape: mikeal: around?
[15:49] alek_br has joined the channel
[15:49] mape: mikeal: Any know reason this doesn't work? require('request')({'uri': 'https://www.google.com/accounts/AuthSubSessionToken'}, function(){console.log(arguments)});
[15:50] mikew3c_ has joined the channel
[15:50] mape: *known
[15:51] sugardave has joined the channel
[15:51] darthdeus has joined the channel
[15:51] jakehow has joined the channel
[15:51] m0rph3v5 has joined the channel
[15:52] jamund has joined the channel
[15:52] tibbetts has joined the channel
[15:53] ben_alman has joined the channel
[15:54] siong1987 has joined the channel
[15:58] mape: Hmm seems only options.client.on('close',function(){}) is triggered, not end
[15:58] xSmurf: mroing
[15:58] figital has joined the channel
[16:00] stepheneb_ has joined the channel
[16:00] JojoBoss: anyone know if 0.3.3 will work in cygwin?
[16:01] JojoBoss: getting the following error when trying to compile it:
[16:01] JojoBoss: Build failed:  -> task failed (err #2):
[16:01] JojoBoss:         {task: libv8.a SConstruct -> libv8.a}
[16:01] JojoBoss: make: *** [program] Error 1
[16:01] bartmax has joined the channel
[16:01] sonnym has joined the channel
[16:01] JojoBoss: I get the same error for 0.3.2 however 0.3.1 works.
[16:01] mraleph: JojoBoss: I really recommend getting some nice *nix box not poor man's imitation called CygWin.
[16:02] JojoBoss: ehhh, i run a dual boot with ubuntu
[16:02] arpegius has joined the channel
[16:02] JojoBoss: anything serious I do off their, but I like haveing it running in cygwin regardless
[16:03] jamund has joined the channel
[16:03] jchris has joined the channel
[16:03] mraleph: cygwin is not one of the platforms v8 is being developed on; so build is destined to break every now and then.
[16:08] cafesofie has joined the channel
[16:10] sveimac has joined the channel
[16:10] fly__ has joined the channel
[16:11] jamund_ has joined the channel
[16:13] pietern has joined the channel
[16:14] Phyllio has joined the channel
[16:14] romainhuet has joined the channel
[16:14] briznad has joined the channel
[16:16] Ari-Ugwu has joined the channel
[16:18] wilken has joined the channel
[16:19] omni5cience has left the channel
[16:22] devdrinker has joined the channel
[16:26] JusticeFries has joined the channel
[16:27] pengwynn has joined the channel
[16:29] shazmoh_ has joined the channel
[16:30] wilken has joined the channel
[16:33] strmpnk has joined the channel
[16:33] psi0nik has joined the channel
[16:34] micheil has joined the channel
[16:36] jpld has joined the channel
[16:36] jpld has joined the channel
[16:37] 36DAAX62I has joined the channel
[16:37] 17SAAOBW0 has joined the channel
[16:37] pHcF has joined the channel
[16:42] temp02 has joined the channel
[16:43] sveimac_ has joined the channel
[16:43] malkomalko has joined the channel
[16:44] tjholowaychuk has joined the channel
[16:44] Max-Might has joined the channel
[16:45] Guest1586 has joined the channel
[16:45] nonnikcam has joined the channel
[16:48] estrathmeyer has joined the channel
[16:49] masahiroh has joined the channel
[16:49] peritus_ has joined the channel
[16:49] peritus has joined the channel
[16:51] modular has joined the channel
[16:52] eml-mobile has joined the channel
[16:52] eee_c has joined the channel
[16:53] daniellindsley has joined the channel
[16:54] ryah: i'll try to bring the cygwin build up to speed today 
[16:54] ryah: *sigh*
[16:57] adambeynon has joined the channel
[16:57] amerine has joined the channel
[16:57] vineyard has joined the channel
[16:59] jakehow has joined the channel
[17:09] hunterloftis has joined the channel
[17:10] dguttman has joined the channel
[17:11] shinmei has joined the channel
[17:11] d0k has joined the channel
[17:14] marcab has joined the channel
[17:15] tahu has joined the channel
[17:15] booths has joined the channel
[17:16] sivy has joined the channel
[17:18] JusticeFries_ has joined the channel
[17:20] temp02 has joined the channel
[17:20] zomgbie has joined the channel
[17:21] noahcampbell has joined the channel
[17:24] jakehow has joined the channel
[17:26] rjack: what's the last version of node.js with https actually working? 0.2.6?
[17:26] arpegius has joined the channel
[17:27] Fullmoon has joined the channel
[17:27] hobodave has joined the channel
[17:28] nook has joined the channel
[17:29] tahu1 has joined the channel
[17:29] ryah: rjack: there's never been a version where https worked correctly
[17:29] ryah: but 0.3.1 and 0.2.6 have broken implementations
[17:32] rjack: ryah: ok, thanks
[17:34] robotarmy has joined the channel
[17:35] qFox has joined the channel
[17:36] cafesofie has joined the channel
[17:40] jamund_ has joined the channel
[17:43] benburkert has joined the channel
[17:44] siong1987 has joined the channel
[17:46] zomgbie has joined the channel
[17:48] CrypticSwarm has joined the channel
[17:48] jchris has joined the channel
[17:50] EyePulp has joined the channel
[17:51] unomi has joined the channel
[17:51] sveisvei has joined the channel
[17:52] rpflo has joined the channel
[17:52] davidascher has joined the channel
[17:52] markstory has joined the channel
[17:53] spetrea has joined the channel
[17:53] aklt has joined the channel
[17:53] jamund_ has joined the channel
[17:53] EyePulp: yo ho ho.  happy new year, etc.
[17:55] sveimac: Happy new node :O
[17:56] EyePulp: heh
[17:57] ossareh has joined the channel
[17:58] gkatsev: so, what do people thinkg about http://www.jspp.io/ ?
[17:58] fmeyer has joined the channel
[17:58] romainhuet_ has joined the channel
[18:02] EyePulp: gkatsev: well... it looks like a horrible regression under the banner of *now with node goodness!*
[18:02] springmeyer has joined the channel
[18:02] gkatsev: EyePulp: yep. Anything that claims that php may be good is by definition terrible.
[18:02] tjholowaychuk: gkatsev: I thought it was a joke
[18:04] stepheneb has joined the channel
[18:05] EyePulp: 
ACTION spent too long dealing with classic ASP to see any need to go back for any scale of dev work.

[18:06] cardona507 has joined the channel
[18:06] EyePulp: So can anyone point me to the underlying collection reference within the object created by mongoose on top of the native mongo driver?
[18:07] EyePulp: I need to run something directly against the native collection versus as part of the mongoose wrapper.
[18:07] TheEmpath has joined the channel
[18:08] TheEmpath: jewhaha
[18:08] TheEmpath: 1.1Ghz CPU, 256MB of RAM.... my server used 67% of a single core of the CPU and 92MB of RAM to sustain 100 connections broadcasting 500 requests per second
[18:09] quest88 has joined the channel
[18:11] EyePulp: jspp is done by mikeal - who I see on here a lot.   I suppose you could ask him directly what his motivations were. =)
[18:12] Me1000 has joined the channel
[18:12] richcollins has joined the channel
[18:12] tisba_ has joined the channel
[18:13] sveimac: Yeah, there should be a why section
[18:13] tjholowaychuk: the bad knockoff of the vows site made me think it was a joke
[18:13] tjholowaychuk: along with the bad idea
[18:14] mikeal has joined the channel
[18:16] jherdman has joined the channel
[18:16] sveimac: tjholowaychuk: btw, I recall you had some cms/group managment going on earlier... played more with it :)?
[18:16] sveimac: might be that it was in ruby, dont remember :P
[18:16] tjholowaychuk: sveimac: nah it is really low priority for me
[18:16] tjholowaychuk: yeah I had started one called "evolution" in ruby
[18:16] EyePulp: In truth, it's not as though thousands of sites or more haven't been built completely successfully on the php/asp model.  There are just some better approaches these days.
[18:16] sveimac: I figure :)
[18:16] tjholowaychuk: a while back
[18:17] mape: Anyone here worked with the google AuthSub api's and didn't use php?
[18:17] dthompson has joined the channel
[18:17] markwubben has joined the channel
[18:22] admc has joined the channel
[18:22] jherdman has joined the channel
[18:23] steffkes has joined the channel
[18:26] jherdman has joined the channel
[18:27] dguttman has joined the channel
[18:29] Yuffster_work has joined the channel
[18:29] isaacs has joined the channel
[18:33] quest88 has joined the channel
[18:36] bingomanatee_ has joined the channel
[18:38] Max-Might has joined the channel
[18:39] sudoer has joined the channel
[18:41] qwebirc5493 has joined the channel
[18:42] davidsklar has joined the channel
[18:44] romainhuet has joined the channel
[18:47] beta__ has joined the channel
[18:47] dnolen has joined the channel
[18:48] mjr_ has joined the channel
[18:51] isaacs_ has joined the channel
[18:55] dmcquay has joined the channel
[18:57] mikeal has joined the channel
[19:01] teddy has joined the channel
[19:03] aslakhellesoy has joined the channel
[19:04] quest88 has joined the channel
[19:04] aslakhellesoy: does anyone know if expresso works on node 0.3.3? I keep getting ECONNREFUSED errors
[19:05] skohorn has joined the channel
[19:05] sveimac has joined the channel
[19:06] gkatsev: tjholowaychuk: ^
[19:06] tjholowaychuk: aslakhellesoy: I have not tried recent versions but it worked with recent commits
[19:06] tjholowaychuk: I dont develop for 0.3.x though
[19:07] aslakhellesoy: tjholowaychuk: I had the same problems on 0.3.1 FWIW. Running expresso's own tests.
[19:09] modular has joined the channel
[19:09] zemanel has joined the channel
[19:09] tjholowaychuk: aslakhellesoy: probably
[19:09] tjholowaychuk: we dont use 0.3.x in production so there is little reason for me to develop for it until 0.4.x
[19:10] aslakhellesoy: tjholowaychuk: i hear you. only reason i'm on 0.3 is broken base64 in 0.2
[19:10] tjholowaychuk: ah gotcha
[19:10] gkatsev: aslakhellesoy: which is why 0.3.x is considered unstable, even though lots of people say it's stable.
[19:10] aslakhellesoy: i'll try to downgrade - maybe the newest 0.2.x has base64 fixed...
[19:12] altamic has joined the channel
[19:12] jherdman has joined the channel
[19:14] Yuffster_work has joined the channel
[19:14] gkatsev: aslakhellesoy: here is the changelog: http://nodejs.org/docs/v0.2.6/changelog.html fwiw
[19:16] pekim: sass with express - How can I provide additional functions to the context that is used to evaluate Javascript? In other words I want to be able to do "... {myfunc(...)} ..." in a sass file.
[19:16] stagas has joined the channel
[19:16] aslakhellesoy: gkatsev: thanks. looks like the base64 fixes that went into 0.3.0 didn't go into 0.2.x :-/
[19:17] tjholowaychuk: pekim: it is pretty limited
[19:17] tjholowaychuk: you cant do much right now
[19:17] tjholowaychuk: but its super old / basic
[19:17] gkatsev: aslakhellesoy: submit a fix patch to expresso?
[19:17] ryah: aslakhellesoy: what's the error?
[19:17] aslakhellesoy: 0.2.x' base64 decode is broken 
[19:18] aslakhellesoy: ryah: I'll come up with a simple test case shortly
[19:18] aslakhellesoy: there is no error, but the decoding is wrong
[19:18] aslakhellesoy: (and correct on 0.3.x)
[19:18] ryah: it seems all the recent base64 changes are in v0.2
[19:19] jchris has joined the channel
[19:19] ryah: but i must be missing something
[19:19] ryah: a test would help.
[19:19] ryah: i can bisect to find the fix
[19:20] aslakhellesoy: ryah: oh? http://nodejs.org/docs/v0.3.0/changelog.html lists a base64 fix that I don't see in any of the 0.2.x changelogs
[19:20] aslakhellesoy: but I haven't looked for the commits in git yet.
[19:20] ryah: changelogs are done by hand
[19:20] ryah: so - easy to lose stuff :)
[19:21] ryah: c735b4663acad20c4d8ccd7a469de77927af22bb <-- recent base64 changes in v0.2
[19:21] jakob has joined the channel
[19:22] isaacs has joined the channel
[19:22] tfe1 has joined the channel
[19:23] bingomanatee_ has left the channel
[19:23] skohorn has joined the channel
[19:23] teemow has joined the channel
[19:23] kjeldahl has joined the channel
[19:24] tahu has joined the channel
[19:24] pekim: tjholowaychuk: Ok, fair enough.
[19:24] pekim: What I actually wanted was to be able to do was to somehow achieve something like
[19:24] pekim:     a: 6em
[19:24] pekim:     b: 0.5em
[19:24] malkomal_ has joined the channel
[19:24] pekim:     c: {a + b}   // 6.5em
[19:24] pekim: whereas I currently have to do something
[19:24] pekim:     a: 6
[19:24] pekim:     b: 0.5
[19:24] pekim:     c: {parseFloat(a) + parseFloat(b)}em
[19:24] pekim: which isn't as clean, especially if the expression gets more complicated.
[19:24] pekim: I was looking for a way to inject a function in to the context that would allow me to evaluate arithmetic expressions on values that all had the same units.
[19:24] cafesofie has joined the channel
[19:25] dylang has joined the channel
[19:25] tjholowaychuk: pekim: yeah no worries, sass.js will be replaced shortly
[19:26] pekim: tjholowaychuk: With what? A new sass implementation? A richer one?
[19:26] tjholowaychuk: pekim: secret
[19:26] tjholowaychuk: :)
[19:26] pekim: tjholowaychuk: Tease!
[19:26] tjholowaychuk: lets just say, it is awesome
[19:26] tjholowaychuk: very awesome
[19:27] pekim: I can't wait. But obviously I'll have to.
[19:27] tjholowaychuk: yeah if you need arithmetic n whatnot yup
[19:27] V1 has joined the channel
[19:28] jpick has joined the channel
[19:30] hunterloftis has joined the channel
[19:30] CIA-121: node: 03Ryan Dahl 07master * r6593a96 10/ (10 files in 5 dirs): Merge branch 'debugger' - http://bit.ly/dNzQC1
[19:31] brianmario has joined the channel
[19:31] ph^ has joined the channel
[19:32] ryah: does anyone have some code to make an absolute path relative?
[19:32] steffkes: ryah, based on a given directory or what?
[19:32] ryah: yeah
[19:32] jamesarosen has joined the channel
[19:36] V1: is it node debug or node --debug? or are they both they same?
[19:37] ph^ has joined the channel
[19:38] chrischris has joined the channel
[19:38] booths has joined the channel
[19:39] hunterloftis: tjholowaychuk: Is there a roadmap for node-canvas being stable (error reporting from Cairo, no segfaults on jpg load, etc), and how could I help with that?
[19:39] tfe1 has joined the channel
[19:40] pietern has joined the channel
[19:41] tfe1 has joined the channel
[19:41] tjholowaychuk: hunterloftis: just the issue queue
[19:41] tjholowaychuk: patches :)
[19:42] hunterloftis: tjholowaychuk: k, will-do. Tried to use it last night for some thumbnail generation but it dies pretty roughly with certain jpeg inputs on image.src = 
[19:42] tjholowaychuk: shit :(
[19:42] tjholowaychuk: I admittedly did not try to many jpegs
[19:42] tfe1 has joined the channel
[19:43] maushu has joined the channel
[19:44] zomgbie has joined the channel
[19:45] aslakhellesoy: ryah: https://gist.github.com/763844
[19:46] aslakhellesoy: this outputs gibberish on 0.2.6, but works fine on 0.3.1 and 0.3.3 (the only 0.3.x I've tested with)
[19:47] opengeard has joined the channel
[19:47] MikhX has joined the channel
[19:48] opengeard has joined the channel
[19:49] aslakhellesoy: ryah: want me to file a bug report?
[19:49] aslakhellesoy: or is it serious enough that you'll start hacking on it right away ;-)
[19:51] tfe1 has joined the channel
[19:52] isaacs has joined the channel
[19:53] p0ns has joined the channel
[19:53] p0ns: hello guys
[19:53] herbySk has joined the channel
[19:53] mjr_: V1: "node debug" is ryah's new debugger thing.
[19:53] mjr_: node --debug is the regular V8 debugger option.
[19:53] gnrfan has joined the channel
[19:54] V1: mjr_: node --debug and node debug seem to trigger the same thing
[19:54] mjr_: Dunno, I haven't tried it myself.
[19:54] p0ns: i'm having some trouble with node.js on mac, i compiled it with OpenSSL, and it also has the -DHAVE_OPENSSL=1 flag, but I can't require the "tls" module, it says "Error: Cannot find module 'tls'". Any advices? Other dependences i should check out?
[19:54] gnrfan has left the channel
[19:54] AAA_awright_ has joined the channel
[19:55] Me1000 has joined the channel
[19:55] bingomanatee_ has joined the channel
[19:56] akahn has left the channel
[20:00] pyrotechnick has joined the channel
[20:02] V1: p0ns you using node 0.3.2~? 
[20:02] p0ns: 0.2.6
[20:02] V1: tls is only in the 0.3.2
[20:02] p0ns: oh
[20:02] p0ns: shit
[20:02] p0ns: i just realized
[20:02] p0ns: lol
[20:03] V1: :D
[20:03] JimBastard has joined the channel
[20:04] pandark_ has joined the channel
[20:04] p0ns: thanks
[20:04] p0ns: and how's 0.3.3? how unstable could it be
[20:05] ROBOd has joined the channel
[20:05] pyrotechnick: im using it quite happily
[20:05] pyrotechnick: our projects aren't small
[20:05] pyrotechnick: they touch on most of node.js, socket.io and express
[20:06] p0ns: i'm thinking of a secure communications between a main server and other media servers, for control and file sync
[20:06] pyrotechnick: very easy to do in node
[20:06] p0ns: that's why i'm diggin into it
[20:06] p0ns: a main app in Rails
[20:06] pyrotechnick: you are going to love it
[20:06] p0ns: and node.js
[20:06] femtoo has joined the channel
[20:06] p0ns: so 0.3.3 is the way
[20:06] pyrotechnick: yeah we got into node trying to solve problems of rails
[20:06] pyrotechnick: just try it
[20:07] SubStack: nodin'
[20:07] pyrotechnick: if you find yourself running into incompatibilities when loading in modules from npm then drop back to 0.2.6
[20:07] pyrotechnick: but you probably wont if you still to newish ones
[20:07] p0ns: i'm just gonna start this module
[20:07] pyrotechnick: you know about npm right?
[20:07] p0ns: yea
[20:07] pyrotechnick: cool
[20:07] p0ns: i've been using coffee
[20:07] pyrotechnick: because some newbies seem to miss that whole thing
[20:07] pyrotechnick: OH NICE
[20:07] p0ns: so i had node installed
[20:07] pyrotechnick: github.com/feisty
[20:07] pyrotechnick: check it
[20:07] SubStack: 
ACTION gets back to payment gateways

[20:07] pyrotechnick: everything we do is in coffee
[20:08] pyrotechnick: u might like some tools we made
[20:08] p0ns: but never tried node, i'll compile the last version
[20:08] p0ns: nice, i'm gonna check it out
[20:08] p0ns: :)
[20:08] pyrotechnick: try nave if you screw around with a few versions
[20:08] pyrotechnick: watch what you like even if you dont use it
[20:08] pyrotechnick: because we're working on the most watched / viewed things
[20:08] pyrotechnick: videos on my youtube youtube.com/user/pyrotechnick
[20:08] pyrotechnick: putting our website up today ^w^
[20:10] p0ns: i'll check that node JS server, seems cook
[20:10] p0ns: cool*
[20:12] skohorn has joined the channel
[20:13] pyrotechnick: if anyones interested in getting bullet physics into node please read this http://groups.google.com/group/nodejs/browse_thread/thread/85b60f18a54546f2
[20:14] broofa has joined the channel
[20:16] wink_: PyroPeter: do you have a stack trace for the crash?
[20:16] wink_: er
[20:16] wink_: pyrotec:
[20:16] wink_: fuck nick completion :<
[20:17] pyrotechnick: yes
[20:17] pyrotechnick: umm
[20:17] pyrotechnick: no
[20:17] wink_: pastebin?
[20:17] pyrotechnick: if you let me know how to make one
[20:17] pyrotechnick: ill do that now
[20:17] wink_: ok, run your application like so:
[20:17] pyrotechnick: that's all the errors the crash actually outputs
[20:17] pyrotechnick: sure
[20:17] wink_: gdb node_g
[20:17] wink_: once you get a prompt
[20:17] wink_: run <script name>
[20:17] pyrotechnick: run or something
[20:17] wink_: let it run until crash
[20:18] pyrotechnick: yeah
[20:18] pyrotechnick: ok
[20:18] pyrotechnick: sec
[20:18] wink_: then type bt
[20:18] wink_: take that and pastebin it
[20:18] pyrotechnick: its all in coffee so let me compile it
[20:18] wink_: remove as many layers as you can :>
[20:18] pyrotechnick: yeah thats the idea
[20:19] rburhum has joined the channel
[20:19] pyrotechnick: ok its running
[20:19] pyrotechnick: that's weird its way slower than normal
[20:19] pyrotechnick: like 100 times slowr
[20:19] pyrotechnick: this may take a while
[20:19] wink_: yeah, that'll happen :>
[20:20] pyrotechnick: this may take hours
[20:20] pyrotechnick: ill give you a bit of background
[20:20] pyrotechnick: the problem is related to a unit i call object seconds
[20:20] pyrotechnick: the integral of the time and number of objects that are in the simulation over its runtime
[20:21] pyrotechnick: the more objects, the longer the simultion, the sooner it crashes
[20:21] pyrotechnick: okay it died
[20:21] pyrotechnick: bt
[20:21] desaiu has joined the channel
[20:21] pyrotechnick: OH WOW
[20:21] pyrotechnick: why didnt they show me bt the other day...
[20:21] wink_: <3
[20:22] pyrotechnick: wink_ my friend
[20:22] pyrotechnick: what is your name
[20:22] pyrotechnick: its definately a problem with my bindings
[20:22] pyrotechnick: https://gist.github.com/a30153ea9e367f004ddb
[20:22] pyrotechnick: do you know much about them
[20:22] pyrotechnick: it's a scope thing i imagine
[20:22] isaacs has joined the channel
[20:22] pyrotechnick: what i have done is destroyed the bindings so that they are very very simple
[20:23] pyrotechnick: if u look in test/master.js
[20:23] pyrotechnick: im only using DiscreteDynamicsWorld
[20:23] pyrotechnick: and RigidBody
[20:23] pyrotechnick: so theyre the only bindings you should be concerened with if you have 5 to look over them for me
[20:23] pyrotechnick: is there a way to kick the gc in
[20:24] pyrotechnick: from js?
[20:24] pyrotechnick: or even from c++?
[20:24] wink_: im looking at rigidbody's destructor ;>
[20:24] wink_: i dont believe so unfortunately
[20:24] pyrotechnick: damn
[20:24] pyrotechnick: because that would let me decide whether its a scope thing or something else
[20:24] pyrotechnick: like i said
[20:24] pyrotechnick: it seems random
[20:24] pyrotechnick: but im not convinced it is random
[20:24] pyrotechnick: if the gc is deterministic
[20:24] pyrotechnick: then it's not random its just the gc kicking in
[20:24] pyrotechnick: does that stack look like that to you?
[20:25] pyrotechnick: theres no way in hell bullet would be cleaning up a rigidbody
[20:25] wink_: it definitely seems like the gc is aggravating it
[20:25] pyrotechnick: ok
[20:25] wink_: the gc isnt going to be totally random, although calling it deterministic would be a little strong
[20:25] pyrotechnick: yeah i see it
[20:25] pyrotechnick: #10 0x000000010010fe2f in v8::internal::Heap::CollectAllGarbage (force_compaction=false) at /Users/pyrotechnick/node/deps/v8/src/heap.cc:447
[20:26] pyrotechnick: so it thinks that RB is loose
[20:26] pyrotechnick: but it shouldnt be
[20:26] pyrotechnick: so its a scope thing
[20:26] pyrotechnick: im just not great enough with bindings to know where the problem lies
[20:26] pyrotechnick: lays
[20:26] pyrotechnick: if i just nuked the destructor
[20:26] wink_: just for cleanliness, go ahead and null out _btRigidBody after the free
[20:26] pyrotechnick: i am
[20:27] pyrotechnick: i thort i did that the other day
[20:27] pyrotechnick: but maybe i didnt
[20:27] wink_: https://github.com/feisty/node-bullet/blob/master/src/RigidBody.cc#L43
[20:27] pyrotechnick: yep
[20:27] pyrotechnick: done
[20:27] wink_: yeah i dont see it in there
[20:27] pyrotechnick: trying now
[20:27] devinus: what do you guys think makes more sense
[20:27] pyrotechnick: wait null it out
[20:27] devinus: Number("123")
[20:27] pyrotechnick: or comment it out
[20:27] devinus: or parseInt("123")
[20:27] wink_: null it out
[20:27] wink_: _btRigidbody = NULL; on L46
[20:27] pyrotechnick: like set it to (RigidBody*)null
[20:28] pyrotechnick: ok
[20:28] pyrotechnick: RigidBody::~RigidBody() {
[20:28] pyrotechnick:   if (_btRigidBody) {
[20:28] pyrotechnick:     free(_btRigidBody);
[20:28] pyrotechnick:     _btRigidbody = NULL;
[20:28] pyrotechnick:   }
[20:28] pyrotechnick: }
[20:28] pyrotechnick: like so
[20:28] wink_: yes
[20:28] pyrotechnick: cool
[20:28] pyrotechnick: run it again?
[20:28] Phyllio has joined the channel
[20:29] wink_: sure
[20:29] sveimac has joined the channel
[20:30] nym_ has joined the channel
[20:30] nym: hey
[20:31] stride: hi
[20:31] nym: i am working with a websocket client for obj-c called ZTWebSocket.c (Zimt)
[20:31] pyrotechnick: nym there's been a few people looking for that
[20:31] pyrotechnick: check the logs, they may want to help
[20:32] nym: ok thanks
[20:32] pyrotechnick: err wink_
[20:32] pyrotechnick: i dont want to get excited here
[20:32] nym: my question is that the client is connecting to socket.io, and i get the following message "~m~4~m~~h~1"
[20:32] pyrotechnick: but it's running for alot longer
[20:32] wink_: dont get excited yet :p
[20:33] nym: i believe this is a data frame, and that my client isn't decoding it correctly
[20:33] V1: nym:  you need to decode the socket.io messages
[20:33] wink_: because realistically if that fix does anything, its hiding a problem ;>
[20:33] pyrotechnick: wink_ this is the longest i've seen it run
[20:33] pyrotechnick: right so it might be leaking?
[20:33] pyrotechnick: well
[20:33] pyrotechnick: its not really
[20:33] wink_: which is that the gc seems to be cleaning up already cleaned up things
[20:33] V1: nym: see https://github.com/LearnBoost/Socket.IO-node/blob/master/lib/socket.io/utils.js#L22-52
[20:33] pyrotechnick: TBH i dont care if it leaves old RB
[20:33] wink_: its not going to leave them
[20:33] pyrotechnick: right
[20:33] wink_: but for some reason its trying to clean them twice
[20:33] pyrotechnick: ok
[20:34] pyrotechnick: it died anyway
[20:34] wink_: same stack trace?
[20:34] pyrotechnick: ill paste it just in case
[20:34] pyrotechnick: but it looks similar
[20:34] nym: V1: sweet
[20:34] nym: V1 is this a socket.io thing or a websocket thing?
[20:35] pyrotechnick: wink_: https://gist.github.com/a30153ea9e367f004ddb/c6c87813df67c5fe2e900edf75266483c994efc9
[20:35] nym: V1: that is, should this be part of my websocket lib, ZTWebSocket.c ?
[20:35] nym: V1: or part of my client
[20:35] V1: nym: a Socket.io thing
[20:35] nym: got it
[20:35] nym: thank you
[20:35] wink_: pyrotechnick: you commented out the free completely, and it still crashes?
[20:35] V1: yw
[20:35] pyrotechnick: not yet i will try that
[20:35] admc has joined the channel
[20:36] pyrotechnick: i did that in the dynamics world
[20:36] wink_: that'll likely cause it to leak like hell, but do that to verify that this is in fact the crash
[20:36] pyrotechnick: well like i said
[20:36] pyrotechnick: we actually dont need anything to be disposed
[20:36] pyrotechnick: once something is in the game it shudnt go until the server is restarted
[20:36] pyrotechnick: but this is no good for other people
[20:36] pyrotechnick: so it needs to be fixed
[20:36] pyrotechnick: but if we can get this prototype up
[20:37] pyrotechnick: we have funding and i can put someone on to fix them
[20:38] m64253 has joined the channel
[20:38] pyrotechnick: bullet physics in node would be amazing
[20:38] pyrotechnick: itll let us take our mmo to alpha
[20:38] wink_: is the idea to handle server side physics?
[20:38] pyrotechnick: and god knows what other people are going to do with it
[20:38] pyrotechnick: can you imagine
[20:38] pyrotechnick: yes
[20:38] wink_: it is in fact :p
[20:39] pyrotechnick: we were doing it on the client
[20:39] pyrotechnick: for the tech demo
[20:39] pyrotechnick: that was ok
[20:39] aslakhellesoy: is there a way to block and wait for a child process to exit?
[20:39] pyrotechnick: but when you try to do 20 000 objects like we need
[20:39] pyrotechnick: it kinda dies in js
[20:39] pyrotechnick: aslakhellesoy: i dont think so, but theres a callback
[20:40] pyrotechnick: so if u just chain from that
[20:40] pyrotechnick: then you have the same functionality
[20:40] pyrotechnick: without "blocking" even though you wont be blocking anything
[20:40] aslakhellesoy: I need to shell out to the base64 program to decode strings since I can't find a working base64 implementation for 0.2.x
[20:40] jamund has joined the channel
[20:40] pyrotechnick: yeah was that your post on the ML?
[20:40] pyrotechnick: i cant say for sure if we were doing base64 in 0.2 but we're just using buffers
[20:40] pyrotechnick: want to try the code?
[20:41] pyrotechnick: its in my xmpp server on GH
[20:41] pyrotechnick: but i can find it if you like
[20:41] cafesofie has joined the channel
[20:41] pyrotechnick: yeah wink_
[20:41] pyrotechnick: this is using 80% cpu and only 30megs ram
[20:41] pyrotechnick: 3 minutes of cpu time so far
[20:41] pyrotechnick: its looking good for us
[20:42] pyrotechnick: i seriously thought i tried killing that destructor
[20:42] pyrotechnick: so you think it's a scope problem?
[20:42] aslakhellesoy: pyrotechnick: I'm using this mailparser library that has a function for decoding base64: https://github.com/andris9/mailparser/blob/master/mime.js#L233
[20:42] mr47 has joined the channel
[20:42] pyrotechnick: id use a buffer
[20:42] pyrotechnick: TBH
[20:42] mr47: hi)
[20:42] pyrotechnick: wait
[20:42] pyrotechnick: thats how we are doing it
[20:42] wink_: PyroPeter: well, i doubt its a scoping issue per say, although that may be triggering this
[20:42] mr47: people who can help me 
[20:42] aslakhellesoy: since the native base64 code is broken I want to modify that method to shell out to base64 (http://www.fourmilab.ch/webtools/base64/)
[20:42] dothebart: ryah: maybe you want to add something like "dns.resolveMx('nodejs.org', function(error, data)      { console.log('Done'); console.dir(data);}); " to tests/simple/test-c-ares.js?
[20:43] pyrotechnick: i mean like a scope.Close thing
[20:43] wink_: if i clone your github, will it build and let me run the test?
[20:43] pyrotechnick: u need bullet
[20:43] pyrotechnick: are u on mac?
[20:43] pyrotechnick: but yes it will
[20:43] wink_: pyrotechnick: yes, i think its the objectwrap stuff
[20:43] wink_: im not on a mac 
[20:43] pyrotechnick: u on mac?
[20:43] pyrotechnick: linux?
[20:43] wink_: yes
[20:43] pyrotechnick: ok
[20:43] pyrotechnick: what distro
[20:43] wink_: ubuntu
[20:43] pyrotechnick: sec
[20:43] mikeal has joined the channel
[20:43] pyrotechnick: i need to know how to do it on ubu anyway
[20:43] pyrotechnick: so let me try
[20:44] wink_: ok, any specific version of bullet i need?
[20:44] pyrotechnick: not really im using 2.77 so ud b safe with that
[20:44] pyrotechnick: u got instructions?
[20:44] mr47: how to fix the bug ([main] python 7120 C:\cygwin\bin\python.exe: *** fatal error - unable to remap \\?\C:\cygwin\lib\python2.6\lib-dynload\_functools.dll to same address as parent: 0x2D0000 != 0x610000 ), rebaseall dont help
[20:45] pyrotechnick: just cmake and build them
[20:45] wink_: im sure i can figure it out, i've built bullet for android, its gotta be easier than that :p
[20:45] mr47: i run configure
[20:45] pyrotechnick: yeah im sure it is
[20:45] yozlet has joined the channel
[20:45] pyrotechnick: umm and then install them
[20:45] pyrotechnick: so that waf can pick them uop
[20:45] pyrotechnick: then just
[20:45] pyrotechnick: node-waf configure build
[20:45] pyrotechnick: then if the js in is the repo its just node test/master.js
[20:45] pyrotechnick: but ill check
[20:45] mr47: i am on windows building
[20:45] pyrotechnick: cuz it might only have coffee
[20:46] mr47: cygwin 
[20:46] pyrotechnick: wink_: that's definitely suppressed the problem which is good enough for us
[20:46] pyrotechnick: wink_: but id like to fix them
[20:46] pyrotechnick: for others
[20:47] wink_: any special options for the build?
[20:47] pyrotechnick: hmm turn off demos
[20:47] pyrotechnick: or it takes all day
[20:47] pyrotechnick: https://github.com/yosilove/homebrew/blob/master/Library/Formula/bullet.rb
[20:48] pyrotechnick: and i think u can turn off extras
[20:48] jimt_ has joined the channel
[20:49] mraleph has joined the channel
[20:49] wink_: yeah
[20:49] pyrotechnick: ill run it overnight without the destructors just to test but lets try fix them ;)
[20:49] pyrotechnick: its baking my mac lol
[20:49] mr47: somebody can help me
[20:49] wink_: crunch crunch crunch crunch
[20:49] markwubben has joined the channel
[20:50] mr47: Сука 
[20:50] pyrotechnick: thats 500 cubes of sides (1, 1, 1) starting at (0, 0, 0)
[20:50] mr47: пидары 
[20:50] pyrotechnick: just flying apart and down into the world
[20:50] pyrotechnick: its a good stress test
[20:51] pyrotechnick: i think if i clocked it in the c it would be even faster
[20:51] pyrotechnick: but im clocking it in js
[20:51] pyrotechnick: 1000 times a second
[20:51] davidascher has joined the channel
[20:51] pyrotechnick: the fastest js can go
[20:51] pyrotechnick: without blocking
[20:51] quest88 has joined the channel
[20:52] pyrotechnick: wink i pushed a nicer version for you to use
[20:52] wink_: pyrotechnick: do you have an uncoffee'd version of the test.js?
[20:52] pyrotechnick: yep
[20:52] pyrotechnick: pull
[20:52] pyrotechnick: lol
[20:53] wink_: haha ok
[20:53] pyrotechnick: we both were thinking the same thing
[20:53] isaacs has joined the channel
[20:53] wink_: it built <3
[20:53] pyrotechnick: yay
[20:53] pyrotechnick: just
[20:53] pyrotechnick: node
[20:53] pyrotechnick: test/master.js
[20:53] pyrotechnick: shud run like
[20:53] pyrotechnick: forever
[20:53] pyrotechnick: if u want
[20:53] pyrotechnick: you can clock it down
[20:53] pyrotechnick: by changing the timeout > 0
[20:54] pyrotechnick: just so it doesnt completely nuke ur cpu
[20:54] wink_: haha
[20:54] wink_: it doesnt seem to be pissing it off too bad
[20:54] pyrotechnick: are you doolin?
[20:55] pyrotechnick: on gh?
[20:55] tahu has joined the channel
[20:55] pyrotechnick: so yeah
[20:55] pyrotechnick: problems in rigidbody or discretedynamicsworld
[20:55] pyrotechnick: or like maybe in init.cc but i doubt that
[20:55] pyrotechnick: theyre the only files being used
[20:55] wink_: trying to get the crash back
[20:55] pyrotechnick: just put that 
[20:55] pyrotechnick: destructor back in
[20:55] pyrotechnick: and maybe change the number of objects to 10000
[20:56] wink_: yeah, now im just waiting for it
[20:56] pyrotechnick: i mean
[20:56] pyrotechnick: 1000
[20:56] pyrotechnick: yeah
[20:56] pyrotechnick: if u add more objects
[20:56] pyrotechnick: it happens sooner
[20:56] pyrotechnick: thats why i say its related to 'object seconds'
[20:56] wink_: 1000 it is
[20:56] pyrotechnick: that may or may not crash bullet lol
[20:56] pyrotechnick: let me see
[20:56] pyrotechnick: nah it works
[20:56] pyrotechnick: only 777 frames on my i7 though
[20:56] wink_: its spinning
[20:57] pyrotechnick: hey ill hook it up to the game again today
[20:57] wink_: doing better than my core2 :p
[20:57] pyrotechnick: if you're around tonight ill give you a look
[20:57] pyrotechnick: lol
[20:57] pyrotechnick: what are you getting?
[20:57] pyrotechnick: im interested in the benchmarks
[20:57] pyrotechnick: our servers arent great
[20:57] wink_: ~280 +/- 5
[20:57] pyrotechnick: wow
[20:57] pyrotechnick: still
[20:57] wink_: that's at 1k
[20:57] pyrotechnick: you only need 50 or 60
[20:57] pyrotechnick: yeah my i7 is about 800
[20:57] pyrotechnick: at 1k
[20:57] pyrotechnick: but its SO HOT
[20:57] pyrotechnick: it burns my legs
[20:58] wink_: haha
[20:58] jimt has joined the channel
[20:58] pyrotechnick: ill try it on our vps
[20:58] pyrotechnick: theyre not going to be happy with physics on their vps lol
[20:58] wink_: however, its not crashing on me though
[20:58] pyrotechnick: even with the destructor back?
[20:58] wink_: yeah
[20:58] pyrotechnick: wtf
[20:58] pyrotechnick: you know what
[20:58] JusticeFries has joined the channel
[20:58] pyrotechnick: let me try
[20:59] pyrotechnick: ok
[20:59] pyrotechnick: see line 31 of rigidbody.cc
[21:00] pyrotechnick: set the origin to 0 0 0 then try it
[21:00] pyrotechnick: i have a feeling the problem was related to everything sitting on itself at the start, then flying off at the speed of light
[21:00] pyrotechnick: or faster
[21:01] pyrotechnick: tansform.setOrigin(btVector3(0.0, 0.0, 0.0));
[21:01] wink_: yeah, got it
[21:01] wink_: holy shit, it's crawling now
[21:01] wink_: lol
[21:01] Fullmoon has joined the channel
[21:01] wink_: the first few iterations took forever
[21:01] wink_: now its back up to speed
[21:01] pyrotechnick: yeah
[21:02] pyrotechnick: but mines still not crashing
[21:02] pyrotechnick: okay now im really confused
[21:02] pyrotechnick: let me check the commits
[21:02] pyrotechnick: maybe ive changed something else
[21:02] wink_: kk
[21:03] pyrotechnick: it can sometimes take hours to crash
[21:03] pyrotechnick: but 99% of the time it doesnt
[21:03] pyrotechnick: especially with so many objects
[21:03] jimt has joined the channel
[21:03] pyrotechnick: and yeah it would have been crawling because everything would have failed the broadphase and have to had been actually solved properly
[21:04] wink_: yeah
[21:04] TheEmpath has joined the channel
[21:04] pyrotechnick: but after a few iterations everythings BLOWN UP
[21:04] pyrotechnick: and flying out in space
[21:04] pyrotechnick: some things would go upwards
[21:04] pyrotechnick: anyway
[21:04] pyrotechnick: we're going to mirror the physics in two instances
[21:04] augustl has joined the channel
[21:04] pyrotechnick: and just have one not solving
[21:04] pyrotechnick: then if the other crashes
[21:04] desaiu has joined the channel
[21:04] pyrotechnick: it just starts solving the other
[21:04] wink_: nice
[21:04] pyrotechnick: and switches the socket over
[21:04] pyrotechnick: its going to run in a seperate node
[21:05] pyrotechnick: then also we can do like distributed physics
[21:05] pyrotechnick: quite easily
[21:05] wink_: you should have them communicating via 0mq
[21:05] pyrotechnick: and some other cool stuff
[21:05] wink_: and really distribute it
[21:05] pyrotechnick: do you think thats fast enough
[21:05] pyrotechnick: wats 0mq
[21:05] pyrotechnick: is it native stuff?
[21:06] wink_: http://www.google.com/url?sa=t&source=web&cd=1&ved=0CBYQFjAA&url=http%3A%2F%2Fwww.zeromq.org%2F&ei=UzoiTfLJC8P48Aay0Ij2DQ&usg=AFQjCNFp86g20I0EsBmcfeg-RwXo4KOsJA&sig2=haQ2QGPbvqYiruAiYNpEcw
[21:06] wink_: er
[21:06] pyrotechnick: i found their site
[21:06] pyrotechnick: reading now
[21:06] wink_: stupid google
[21:06] pyrotechnick: u think its fast enough
[21:06] pyrotechnick: i mean we are using websocket to the clients
[21:06] pyrotechnick: so its certainly not going to be the bottleneck
[21:07] pyrotechnick: websocket are slow
[21:07] jamund has joined the channel
[21:07] wink_: it'd be fast enough to communicate between node processes at 1khz im sure
[21:07] femtooo has joined the channel
[21:07] pyrotechnick: well like i said we only need 60
[21:07] pyrotechnick: or less
[21:07] pyrotechnick: so that sounds like a great idea wink
[21:07] pyrotechnick: do you know if theres bindings for node
[21:08] pyrotechnick: it looks like something we might use eslewhere too
[21:08] wink_: i believe there are, although i did hear about them being a little flakey
[21:08] wink_: although i dont know firsthand
[21:08] wink_: that said, the bindings are pretty straightforward
[21:08] pyrotechnick: https://github.com/JustinTulloss/zeromq.node
[21:08] herbySk: v8: Date.now()
[21:08] pyrotechnick: those?
[21:08] v8bot: herbySk: 1294088922046
[21:08] pyrotechnick: v8: Date.now() - Date.now()
[21:08] v8bot: pyrotechnick: 0
[21:09] pyrotechnick: wow
[21:09] pyrotechnick: v8: setTimeout()
[21:09] v8bot: pyrotechnick: ReferenceError: setTimeout is not defined
[21:09] pyrotechnick: lame
[21:09] pgte has joined the channel
[21:09] markstory: pyrotechnick: I've worked with a rabbitmq server, it is crazy fast
[21:09] wink_: yeah rabbit is similar in function
[21:09] pyrotechnick: fast enough for realtime markstory
[21:09] markstory: Using a python listener I drained 32000 events in ~5seconds
[21:09] pyrotechnick: wow
[21:09] markstory: and I was writing them to a mysql database
[21:09] pyrotechnick: hmm
[21:10] pyrotechnick: we'll def have to look into that
[21:10] pyrotechnick: we were going to roll our own
[21:10] markstory: as long as you keep your message sizes small its super fast
[21:10] wink_: the low latency messaging libs are going to run circles around the code you have in js-land
[21:10] pyrotechnick: how small is small
[21:10] wink_: they'll definitely not be your bottleneck
[21:10] markstory: pyrotechnick: ours are in <5kb
[21:10] pyrotechnick: yeah hence native physics
[21:10] pyrotechnick: yeah ours are
[21:10] pyrotechnick: bytes
[21:10] pyrotechnick: like
[21:10] pyrotechnick: 24 bytes would be one of the biggest
[21:10] markstory: rabbitmq will be more than fast enough.
[21:11] pyrotechnick: cool
[21:11] pyrotechnick: wats it written in
[21:11] markstory: erlang
[21:11] pyrotechnick: FUCK
[21:11] pyrotechnick: i hate compiling erlang
[21:11] wink_: lol
[21:11] pyrotechnick: i dont mind it
[21:11] pyrotechnick: it just takes so long
[21:11] markstory: its unkillable though.
[21:11] wink_: pyrotechnick: rabbit/0mq are pretty similar beasts
[21:11] pyrotechnick: wink_ which is more lightweight 
[21:11] pyrotechnick: we dont need queues
[21:12] pyrotechnick: this stuff is frame-based drop and forget
[21:12] markstory: its the same protocol for both I think.
[21:12] pyrotechnick: id prefer something in c that can do frames but we're quite open to anything that seems fitting
[21:12] pyrotechnick: we were just going to go udp between the nodes
[21:13] wink_: 0mq is cpp
[21:13] markstory: I like using things that I didn
[21:13] markstory: t have to write and ar good
[21:13] pyrotechnick: yeah
[21:13] markstory: wow I suck at typing.
[21:13] pyrotechnick: lol
[21:14] wink_: pyrotechnick: yeah it looks like justintulloss' github is a forked version of the original
[21:14] pyrotechnick: ok
[21:14] mikeal has joined the channel
[21:14] chrischris has joined the channel
[21:14] pyrotechnick: this looks cool
[21:14] pyrotechnick: https://github.com/networkimprov/websocket.MQ
[21:15] ryah: mraleph: ping
[21:15] mraleph: ryah: pong
[21:15] ryah: mraleph: how do i access the mirror objects?
[21:15] ryah: mraleph: src/mirror-debugger.js
[21:15] mraleph: ryah: whoa. dunno.
[21:15] mraleph: ryah: and nobody around to ask
[21:15] pyrotechnick: ryah: is your amqp client any good?
[21:16] maushu: >_>
[21:16] maushu: 
ACTION hides.

[21:16] ryah: mraleph: can't you guys work at night instead of day?
[21:16] ryah: ;)
[21:16] maushu: I wish.
[21:16] ryah: pyrotechnick: it's okay
[21:16] mraleph: :-)
[21:16] ryah: mraleph: they look pretty advanced - but i can't figure out how to get to them in d8
[21:17] ryah: 
ACTION reads the source

[21:17] maushu: Until the end of the week, I'm going to resign.
[21:17] maushu: So that I can follow my dreams!
[21:17] aslakhellesoy: ryah: did you see my little base64 test?
[21:17] ryah: maybe they'r eonly internally used in v8
[21:17] ryah: aslakhellesoy: no, sorry where it is?
[21:18] aslakhellesoy: https://gist.github.com/763844
[21:18] aslakhellesoy: so that stuff will work fine on 0.3.x and not at all on 0.2.6
[21:19] romainhuet_ has joined the channel
[21:19] ryah: aslakhellesoy: would it be possible to have a test with a less personal message.txt? :)
[21:19] ryah: aslakhellesoy: and would you mind if i included this in the testS?
[21:19] aslakhellesoy: ryah: I guess, but it's not consistently misbehaving
[21:20] ryah: 
ACTION tries

[21:20] mraleph: ryah: http://code.google.com/p/v8/source/browse/branches/bleeding_edge/test/mjsunit/debug-referenced-by.js?r=5893
[21:20] tim_smart has joined the channel
[21:21] mraleph: ryah: MakeMirror is visible through debug object it seems
[21:21] aslakhellesoy: node: You can use this one. They are both colleagues of mine and there is nothing secret there :-)
[21:21] aslakhellesoy: s/node/ryah LOL
[21:22] Insanity5902 has joined the channel
[21:23] AAA_awright_ has joined the channel
[21:23] isaacs has joined the channel
[21:24] ryah: mraleph: rad
[21:24] ryah: http://code.google.com/p/v8/source/browse/branches/bleeding_edge/test/mjsunit/debug-referenced-by.js?r=5893#57 <-- pretty useful
[21:25] jae has joined the channel
[21:25] mraleph: ryah: there is also C++ API GetMirror v8-debug.h
[21:26] mraleph: ryah: but frankly speaking i do not know how mirrors are used in real world. probably you need to look into dev tools sources to figure that out.
[21:27] ryah: getting a ref count on an object seems pretty great.
[21:27] jae: Hello, I'm having trouble with node.js https client requests. Can somebody run this code on node and paste what you get? http://pastebin.com/raw.php?i=fqALHCP9
[21:28] ryah: jae: https client is broken
[21:28] jae: oh. what should i do?
[21:28] ryah: jae: wait until it's fixed :(
[21:28] cafesofie has joined the channel
[21:28] mjr_: We need to coerce pquerna into fixing https.
[21:29] jae: do we know where the bug is? maybe i can monkeypatch it
[21:29] jae: i need to use it for oauth
[21:29] ryah: jae: it's in the process of being rewritten
[21:29] jae: oh ok
[21:29] ryah: i'll publush my not-working https branch
[21:29] ryah: seems like it is not so hard to get a server running
[21:30] ryah: maybe mjr will be inspired
[21:30] mjr_: ha
[21:30] mjr_: Maybe ryah will be inspired that he needs it for botnet success.  
[21:30] CIA-121: node: 03Ryan Dahl 07https * rc8aee07 10/ (lib/http.js lib/tls.js https-test.js lib/https.js): ... - http://bit.ly/gH45GP
[21:30] ryah: http is for losers
[21:31] jamund has joined the channel
[21:31] matt_c_ has joined the channel
[21:31] ryah: https://github.com/ry/node/blob/c8aee07878294b4dae9cd50baacc1594dba1a509/lib/https.js
[21:31] ryah: it shouldn't be much more than that
[21:31] mraleph: ryah: I think it's not just reference count but also mirrors for all referencing objects.
[21:31] mjr_: yeah, it's a 2 hour fix, I'm sure.
[21:31] ryah: mjr_: i think so
[21:32] mjr_: oh hey, can I get some of those 2 hours back from your debugger then?
[21:32] ryah: mjr_: :)
[21:32] jae: ndb?
[21:32] mjr_: 
ACTION feeds the trolls

[21:33] PyroPeter has joined the channel
[21:33] jae: Hey Ryah, thanks much for everything
[21:33] ryah: if i get https server working in two hours can i have a free beer?
[21:33] mjr_: Hmmm, no
[21:33] mjr_: Still more bugs
[21:33] bmizerany has joined the channel
[21:33] jae: if you get https client working i'll buy you a beer
[21:33] wink_: man 2 hours of ryah's time for a beer, thats a pretty good deal :p
[21:33] ryah: 
ACTION goes back to fixing base64 on v0.2

[21:33] mjr_: Hmm, on second thought, you can have a free beer.
[21:34] mjr_: Beer is cheap
[21:34] jae: a keg?
[21:34] jae: an eighball?
[21:36] chapel: Ill buy you beer ryah
[21:37] ryah: okay. going to fix this base64 thing first
[21:37] mjr_: ryah is going to have too many beers
[21:37] mjr_: Then v8bot will have to cut him off.
[21:39] pyrotechnick: ryah has a drinking problem
[21:39] chapel: ryah: which version are you talking about, this isn't for 0.3.x?
[21:42] ryah: chapel: base64 is broken on v0.2
[21:42] sechrist has joined the channel
[21:42] chapel: oh, well still deserves a beer, but I dont use 0.2.x
[21:42] chapel: :(
[21:42] chapel: https client is still broken in 0.3.x :(
[21:43] ryah: soon
[21:43] zomgbie has joined the channel
[21:43] ryah: the client is harder, the server is possible in 2 hours
[21:43] ryah: maybe
[21:44] under___: for tls in general, is there going to be a way to upgrade from plaintext -> tls similar to the .2 net.Stream ?
[21:44] ryah: under___: there is already - it's just a rather low-level API
[21:44] ryah: under___: something cuter should be done
[21:45] under___: ryah: cool thx.  i guess i need to learn more JS :)  I poked in .3 tls code for a bit, but, uh, way over my head
[21:45] ryah: var cleartext = tls.update(socket); or something
[21:45] ryah: s/update/upgrade/
[21:45] under___: cool, sounds awesome
[21:46] liar has joined the channel
[21:46] under___: on the 2 hours, uh, take your time. TLS isn't something to rush
[21:49] ryah: aslakhellesoy: found the fix: https://github.com/ry/node/commit/cf1db4f3043fce0474c421dc547b8bd2f9616354
[21:50] Mikushi1 has joined the channel
[21:50] aslakhellesoy: ryah: sweet - and that commit is only on the 0.3.x branch?
[21:50] mikew3c_ has joined the channel
[21:51] ryah: yeah - 
[21:51] ryah: it's difficult to back port it
[21:51] ryah: but will do..
[21:52] jimt has joined the channel
[21:52] skohorn has joined the channel
[21:53] MrNibbles: anyone fancy having a quick peek at something im managing to bugger up?
[21:53] mjr_: I'm going to buy ryah two hours instead of a beer.
[21:53] MrNibbles: not directly node related actually :S but seems a bit weird
[21:53] mjr_: And then I'm going to buy him more minutes for his cellphone.
[21:53] creationix has joined the channel
[21:53] isaacs has joined the channel
[21:53] MrNibbles: in a require.js module, the setTimeout call in the loadIn function always fires instantly
[21:53] MrNibbles: http://pastebin.com/BvZABswP
[21:54] MrNibbles: cant fkin work out why
[21:54] mjr_: Maybe creationix will get ryah a node phone and he won't have to deal with prepaid minute plans anymore.
[21:55] mape: anyone know of a good module to deal with Date objects? convert to different formats, free text time settings and whatnot?
[21:55] MrNibbles: date.js!
[21:55] ryah: mjr_: :)
[21:56] Hello71 has joined the channel
[21:56] Hello71 has joined the channel
[21:56] mape: MrNibbles: Neat, thanks
[21:57] MrNibbles: mape: np!
[21:57] MrNibbles: now if you can work out my bloody timeout woes..
[21:57] MrNibbles: ;)
[21:58] pyrotechnick has left the channel
[21:58] mape: MrNibbles: remove the () on callback.call() ?
[21:58] mape: now you are waiting to run whatever returns from that execution?
[21:59] mape: setTimeout(function(){callback.call()}, 2000);
[22:00] MrNibbles: has to be called in the context its passed in as
[22:00] MrNibbles: and the anon function wrapper still gets fired instantly
[22:01] creationix has left the channel
[22:02] mape: that sounds strange
[22:02] MrNibbles: yeah, im thinking its a bug
[22:02] pyrotechnick has joined the channel
[22:02] pyrotechnick: ryah: what do i need to change in my bindings for 0.3.2 -> 0.3.3
[22:02] pyrotechnick: everythings returning an empty object
[22:03] pyrotechnick: before  i got a rigidBody [object RigidBody]
[22:03] pyrotechnick: now im getting
[22:03] pyrotechnick: rigidBody {}
[22:03] robotarmy has joined the channel
[22:04] mape: Bah, and Date.js seems broken
[22:05] tjholowaychuk: mape: I had some stuff in ext.js but the project is kinda defunct, I want to eventually separate them into different mods
[22:05] tjholowaychuk: but might want to grab some of that i dunno
[22:05] karboh has joined the channel
[22:05] tjholowaychuk: depends what you need
[22:05] pyrotechnick: tjholowaychuk: have you tried 0.3.3?
[22:05] pyrotechnick: my bindings are returning empty objects
[22:05] tjholowaychuk: pyrotechnick: nope
[22:06] mape: Right now the parse feature, Date.parse('today') gives Sun, 02 Jan 2011 23:00:00 GMT and date on the server is Mon Jan  3 23:05:40 CET 2011
[22:06] pyrotechnick: are there any bindings that work with 0.3.3?
[22:06] JusticeFries has joined the channel
[22:06] mape: Or perhaps it is just GMT+1 messing upp.. hmm
[22:06] pyrotechnick: what happened to hummingbird?
[22:07] robotarm_ has joined the channel
[22:07] ryah: pyrotechnick: hm - i dont think anything has changed...
[22:08] ryah: pyrotechnick: except a new v8
[22:08] webr3 has joined the channel
[22:10] MrNibbles: mape: i had to hack it a while back for timezone bits
[22:10] MrNibbles: think i ended up forcing a certain timezone :S
[22:12] tjholowaychuk: mape: I know guillermo has issues with it here as well
[22:12] tjholowaychuk: I dont know the details
[22:12] tjholowaychuk: all I know is we are replacing it soonish
[22:12] desaiu: did i read this correctly: http://www.quirkey.com/blog/2009/09/15/sammy-js-couchdb-and-the-new-web-architecture/ frameworks aren't needed? 
[22:12] desaiu: and there's this quote: CouchDB makes Django look old- school in the same way that Django makes ASP look outdated.” — Jacob Kaplan-Moss, Django Developer
[22:12] desaiu: can someone clear this up for a noob like myself?
[22:13] pyrotechnick: desaiu: there is a trend for thick clients and thin servers
[22:13] mape: tjholowaychuk: Well if you open source it give me a shout :) Seems like there is an add but no remove.. strange
[22:13] pyrotechnick: sometimes even dropping the server alltogether and using couchdb
[22:13] tjholowaychuk: mape: will do
[22:13] pyrotechnick: so you use something like sammy or backbone and back it directly with couch
[22:14] pyrotechnick: thus having no application server and only having couch or mongo via a rest adapter or something like that
[22:14] herbySk74 has joined the channel
[22:15] desaiu: interesting
[22:16] pyrotechnick: it's just a trend, it doesnt make sense in some of situations, but most it does
[22:16] pyrotechnick: for something like twitter or facebook it makes alot of sense
[22:16] pyrotechnick: theyre purely data-driven clients
[22:16] chapel: there are a lot of instances where it makes sense, but it isn't one size fits all
[22:16] chapel: but it definitely is cool
[22:16] tjholowaychuk: i dont think FB is a good example
[22:16] chapel: and I see it being something that people go to in the future as an option
[22:17] pyrotechnick: tjholowaychuk: they still need services
[22:17] pyrotechnick: lie chat and stuff
[22:17] tjholowaychuk: their shit needs to run on old granny computers
[22:17] MrNibbles: mape: im being a dubass
[22:17] tjholowaychuk: im sure they would rather pay a bit more instead of crashing their computers
[22:17] mape: MrNibbles: Those are the worst bugs ;)
[22:17] MrNibbles: of course callback is a function, so just callback works fine >_<
[22:17] pyrotechnick: tjholowaychuk: right. there's the whole shitt windows and ie thing too
[22:17] tjholowaychuk: yeah haha
[22:17] pyrotechnick: tjholowaychuk: god bless their souls
[22:17] pyrotechnick: i wouldnt let my granny use ie6
[22:17] tjholowaychuk: eventually I will agree, when everyone has a reasonable machine
[22:18] pyrotechnick: tjholowaychuk: you make a good point, one i often overlook being a webgl developer
[22:18] mape: Must be a lot of fun to work on the IE team, getting tons of props from everyone all the time
[22:18] chapel: eh, chrome works well on all sorts of hardware
[22:18] tjholowaychuk: my grandma is rocken the ipad haha
[22:18] pyrotechnick: what a mad chick
[22:18] pyrotechnick: MrNibbles: my gf just walked past
[22:18] chapel: the ipad would make a great thick client :)
[22:18] pyrotechnick: < funny voice >"Mr Nibbles"< / funny voice >
[22:19] desaiu: :P
[22:19] gnrfan has joined the channel
[22:19] pyrotechnick: yeah my bindings are definitely broken
[22:19] bingomanatee_: What is Aria's git library at?
[22:19] MrNibbles: the ladies love it
[22:19] pyrotechnick: lol
[22:19] pyrotechnick: is your first name ear
[22:19] pyrotechnick: and your middle name lobe
[22:19] hober has joined the channel
[22:19] sitron has joined the channel
[22:20] wattz has joined the channel
[22:20] pyrotechnick: mr ear lobe nibbles
[22:20] adrienf has joined the channel
[22:20] WaterCooled has joined the channel
[22:20] desaiu: what situations does it not make sense to bypass server and just use couchdb... also i don't understand how it is bypassing server right now.. a server has to be on the other side right?
[22:20] MrNibbles: i intentionally leave those to the readers imagination
[22:20] pyrotechnick: desaiu: theyre not dropping servers, theyre dropping application servers
[22:20] desaiu: is sammy.js getting rid of the mvc framework i've come to know and love?
[22:20] pyrotechnick: like view servers
[22:20] pyrotechnick: no
[22:20] pyrotechnick: its moving it to the client
[22:20] gnrfan has left the channel
[22:20] pyrotechnick: where it kind of belongs
[22:21] pyrotechnick: check out the google app store
[22:21] desaiu: okay
[22:21] pyrotechnick: and html5 manifests
[22:21] pyrotechnick: and that kind of thing
[22:21] pyrotechnick: youll see how apps are going to be soon
[22:21] MrNibbles: pyrotechnick:  theres 8 bits in a byte, 4 in a nibble (or nybble) and i started out making 4 bit sound software :S
[22:21] pyrotechnick: i get it
[22:21] pyrotechnick: clever
[22:21] zemanel: http://preview.tinyurl.com/JSCodeStandardsEnforcer
[22:21] chapel: desaiu: instead of handling all the 'dynamic' stuff on the server and sending it to the client, you have the client request it
[22:21] MrNibbles: and besides, ColonelPanic was taken ;)
[22:22] pyrotechnick: haha
[22:22] pyrotechnick: mac user?
[22:22] MrNibbles: how did you guess :P
[22:22] mape: chapel: Then how do you deal with the foil hat people? :P
[22:22] chapel: what foil hat people?
[22:22] onar_ has joined the channel
[22:22] dthompson has joined the channel
[22:22] mape: The ones with tin foil that hate js because it gives you nice little cookies
[22:22] chapel: lol
[22:23] markwubben has joined the channel
[22:23] chapel: the amount of people that don't run js is very small
[22:23] desaiu: i'm guessing after i finish my initial read of some js and node.js documentation, i'm guessing reading the sammy.js and couchdb documentation will help me to understand this better
[22:23] MrNibbles: and omfg, code executing in a browser!
[22:23] chapel: so its a matter of wanting to support them or not
[22:23] langworthy has joined the channel
[22:23] chapel: my personal take on it, the sites I would make using thick clients wouldn't cater to those with tin foil hats
[22:24] pyrotechnick: lol
[22:24] isaacs has joined the channel
[22:24] chapel: but one thing that could assuage them, they can see all the code that is running (well most)
[22:24] pyrotechnick: i wish #node.js had a quote system
[22:24] chapel: make a bot
[22:24] chapel: :)
[22:24] pyrotechnick: i will
[22:25] dgathright has joined the channel
[22:25] mape: Someone should get the announcer back running, and white list people
[22:25] jakehow has joined the channel
[22:26] pyrotechnick: wats it do
[22:27] mape: it announces things
[22:27] chapel: twitter announcer right?
[22:28] pyrotechnick: ryah: do i need to use that NODE version macro thing
[22:28] mape: Yeah
[22:28] pyrotechnick: tjholowaychuk: i think you need some macro
[22:28] pyrotechnick: to make your bindings work in 0.3.3
[22:29] markwubben1 has joined the channel
[22:29] tjholowaychuk: pyrotechnick: hmm, Buffer has changed,  but I think init() is still supported
[22:29] tjholowaychuk: maybe not
[22:29] pyrotechnick: NODE_MODULE(helloworld, init);
[22:29] pyrotechnick: that
[22:30] tjholowaychuk: Buffer data access requires a version check for compat though
[22:30] pyrotechnick: im not using buffers
[22:30] tjholowaychuk: ah
[22:30] pyrotechnick: i think u need that
[22:30] pyrotechnick: ill try it
[22:30] pyrotechnick: Thehttp://github.com/ry/node/blob/v0.2.0/src/node.h#L101, expands into a struct that node uses to store information about your module, like what version of the API it is using, and in the future will be used to help prevent crashes of modules built against incompatible APIs than the current Node.js version.
[22:30] tim_smart has joined the channel
[22:31] tjholowaychuk: ya i just checked out the macro
[22:31] pyrotechnick: i dont know how to put it in mine
[22:31] strmpnk_ has joined the channel
[22:31] pyrotechnick: https://www.cloudkick.com/blog/2010/aug/23/writing-nodejs-native-extensions/
[22:31] pyrotechnick: halfway down
[22:31] pyrotechnick: he does extern "C" {
[22:31] pyrotechnick: i do
[22:31] pyrotechnick: extern "C" void
[22:31] pyrotechnick: init (Handle<Object> target) {
[22:31] pyrotechnick: which is correct?
[22:31] pyrotechnick: i guess u need his way if u wanna do that
[22:32] tjholowaychuk: you can use it per statement or a block
[22:32] pyrotechnick: ryah: do you have bindings not in node.js that work with 0.3.3
[22:32] pHcF has joined the channel
[22:32] pyrotechnick: like an example
[22:32] pyrotechnick: or you tjholowaychuk
[22:33] pyrotechnick: i guess you havent tried it yet
[22:33] pyrotechnick: argh im going back to 0.3.2 i think lol
[22:33] sudoer has joined the channel
[22:33] pyrotechnick: targeting edge node is like chasing the dragon
[22:33] pyrotechnick: you never quite catch it
[22:34] ryah: pyrotechnick: are you recompiling them?
[22:34] pyrotechnick: yes
[22:34] ryah: nothing really has changed...
[22:34] jae has joined the channel
[22:35] jae has left the channel
[22:35] adambeynon_ has joined the channel
[22:35] pyrotechnick: right. im not saying 0.3.3 is broken im just saying im not adhering to some standard 0.3.3 is now enforcing
[22:35] pyrotechnick: and id be interested to know what that is
[22:35] pyrotechnick: whether its this NODE_MODULE macro or what
[22:35] keeran_ has joined the channel
[22:35] zemanel_ has joined the channel
[22:35] teddy_ has joined the channel
[22:36] ryah: well - that's optional
[22:36] HAITI_ has joined the channel
[22:36] jae has joined the channel
[22:36] ryah: you need to define a init symbol
[22:37] pyrotechnick: what the hell
[22:37] pyrotechnick: is with freenode lately
[22:37] MrNibbles: 
ACTION rides the splits

[22:37] MrNibbles: surfs up gents!
[22:37] beta_ has joined the channel
[22:37] zentooo has joined the channel
[22:38] jae: 
ACTION washout

[22:38] chapel: I dont see the splits
[22:39] chapel: http://imgur.com/rK7OI.png
[22:39] rauchg_ has joined the channel
[22:39] pyrotechnick: do u see when people leave?
[22:39] MrNibbles: chapel: you dont have parts/joins listed, it will show the reason on the part message
[22:40] matt_c has joined the channel
[22:40] pyrotechnick: yeah
[22:40] rednul has joined the channel
[22:40] chapel: I know I dont see them cause I have them disabled on freenode (join/parts)
[22:40] mraleph: ryah: one of the guys from gdb mailing list devised a patch for gdb which apparently makes gdb jit interface 100x faster. *sigh* I have not tried it yet myself though
[22:41] tjholowaychuk has joined the channel
[22:41] bvleur has joined the channel
[22:41] jae: what irc client is that chapel?
[22:41] chapel: Linkinus
[22:41] jae: cool. better than my term
[22:41] pyrotechnick: adiums not bad
[22:42] rubydiamond has joined the channel
[22:42] olegp has joined the channel
[22:42] MrNibbles: I tried linkinus, found it a bit too glitzy and gimmicky
[22:42] MrNibbles: although, i use colloquy so i dont really have a say in this :P
[22:42] pyrotechnick: i just telnet into irc
[22:42] chapel: MrNibbles: well the new beta is super fast, its asynchronous :) well as much as objc can be
[22:42] pyrotechnick: i have to handle all of your messages with telnet commands
[22:42] pyrotechnick: so please try not spam
[22:43] chapel: heh
[22:43] chapel: MSG #node.js blah blah blah
[22:43] Mikushi1 has left the channel
[22:43] pyrotechnick: no chapel i use binary
[22:43] chapel: even better
[22:43] MrNibbles: chapel: might have to give that a try
[22:44] jae: i use straignt up `nc`
[22:44] MrNibbles: colloquy basically freezes whenever i get more than about 400 messages out of the buffer from znc
[22:44] MrNibbles: its a pain
[22:44] ryah: mraleph: !
[22:44] ryah: mraleph: what did they do?
[22:44] chapel: I haven't tried large buffers, maybe I should to test
[22:45] drudge: MrNibbles: how is it more glitzy and gimmicky than colloquy
[22:45] chapel: :)
[22:45] chapel: da man
[22:45] chapel: ryah++
[22:45] v8bot:  chapel has given a beer to ryah. ryah now has 1 beers.
[22:45] chapel: see I deliver ryah
[22:45] MrNibbles: drudge: hence the comment after that ;)
[22:45] chapel: Someone needs to make an automated beer delivery service
[22:45] pyrotechnick: tjholowaychuk: macro doesnt fix them
[22:45] MrNibbles: just thought it had an over developed gui
[22:46] MrNibbles: and it wasnt free
[22:46] mraleph: ryah: some darkish gdb magic. apparently it was scanning each added elf-object for some predefined builtin symbols over and over again... or something like that.
[22:46] drudge: free i understand, overdeveloped gui not so much :)
[22:46] chapel: MrNibbles: I feel the gui is a good mix of simplicity with form
[22:46] chapel: that and you can make your own themes if you want
[22:46] tjholowaychuk has joined the channel
[22:46] ryah: mraleph: sweet. i can't wait to try it
[22:46] MrNibbles: it was a loong time ago i checked it out, might be worth another revisit
[22:46] mraleph: ryah: http://permalink.gmane.org/gmane.comp.gdb.devel/29725
[22:46] chapel: so irc can look like irssi, or some strange alien world
[22:46] mraleph: ryah: oh. he works for Google... just noticed that.
[22:47] drudge: MrNibbles: http://cloud.penree.com/3r1u
[22:47] chapel: show off drudge
[22:47] MrNibbles: drudge: that looks almost the same as my colloquy :D
[22:47] mikeal: ryah: did you see the numbers when serving from cache?
[22:47] mikeal: ab -n 1000 -c 10 http://127.0.0.1:8888/perf/cached.html
[22:47] mikeal: Requests per second:    3753.70 [#/sec] (mean)
[22:48] drudge: MrNibbles: http://cloud.penree.com/3r67
[22:48] mraleph: ryah: I should try it myself but I do want to build gdb from source :-)
[22:48] mraleph: do *not* want
[22:48] pyrotechnick: oh god
[22:48] pyrotechnick: that client looks sick
[22:48] drudge: pyrotechnick: thanks
[22:48] mikeal: what IRC client is that?
[22:48] chapel: I use the dark version of that theme
[22:48] chapel: damn drudge
[22:48] chapel: I used to hate whisper
[22:49] drudge: mikeal: http://linkinus.com
[22:49] Aria has joined the channel
[22:49] ryah: mraleph: i did it- it's pretty easy
[22:49] MrNibbles: drudge: http://www.webnibbles.co.uk/dropbox/Screen%20shot%202011-01-03%20at%2022.46.37.png
[22:49] tjholowaychuk: chapel: haha thats what im using
[22:49] tjholowaychuk: I haits annoying having the sidebars
[22:49] tjholowaychuk: its*
[22:49] tjholowaychuk: wtf
[22:49] pyrotechnick: but it cant do pictures
[22:49] matt_c has joined the channel
[22:49] tjholowaychuk: but its annoying not having them too
[22:49] chapel: tjholowaychuk: simplified used to be my favorite theme, but after the beta, it just wasn't fast like it used to be
[22:49] PyroPeter has joined the channel
[22:50] tjholowaychuk: pyrotechnick: it inlines images and (parts) of gists etc
[22:50] MrNibbles: drudge: does look nice i agree
[22:50] pyrotechnick: tjholowaychuk: doesnt look like it
[22:50] MrNibbles: your preferred ui anyway, dont like the second version
[22:50] pyrotechnick: theyre just links not images
[22:50] tjholowaychuk: chapel: it used to have memory issues, much snappier now
[22:50] mraleph: ryah: no doubt. but it will take some time and my laptop will start burning my lap :-)
[22:50] drudge: mikeal:, MrNibbles, pyrotechnick: disclaimer: this is a product of mine
[22:50] tjholowaychuk: pyrotechnick: probably a pref
[22:50] jae1 has joined the channel
[22:50] mraleph: ryah: I do not have good old workstation at home since I moved to Denmark.
[22:50] jae1: testing
[22:51] MrNibbles: drudge: congrats matey, its a good one
[22:51] jae1: i don't think i'm in kansas anymore, toto
[22:51] jae: no you aint
[22:51] chapel: tjholowaychuk: yeah, much nicer, btw are you going to help get node.js in linkinus?
[22:51] gf3 has joined the channel
[22:51] mikeal: what would getting node.js in linkinus entail?
[22:52] tjholowaychuk: chapel: not sure what you mean
[22:52] drudge: for plugin api
[22:52] tjholowaychuk: ohhh
[22:52] tjholowaychuk: that would be great
[22:52] chapel: :)
[22:52] tjholowaychuk: I just noticed a few seconds ago that it was webkit
[22:52] tjholowaychuk: haha
[22:52] pyrotechnick: so its kod
[22:52] mikeal: i have high hopes for Kod
[22:52] mikeal: i can't use it yet
[22:52] pyrotechnick: i wish kod used webkit
[22:52] mikeal: but I'm optimistic
[22:52] drudge: i'm still using vim
[22:52] chapel: it doesn't use webkit?
[22:52] mikeal: i'm using TextMate
[22:52] MrNibbles: ooh has anyone been hacking on kod yet? it looks reasonably promising
[22:52] pyrotechnick: no it uses native text component
[22:52] chapel: textmate here
[22:52] MrNibbles: im getting a bit bored of textmate
[22:52] chapel: hmm
[22:52] pyrotechnick: yeah im still on textmate
[22:52] chapel: native text with css?
[22:52] drudge: vim for life
[22:53] pyrotechnick: sometimes i use intellij, kate and aptana
[22:53] pyrotechnick: anyone used kate?
[22:53] mraleph: i am using emacs and have "develop ideal editor" in my todo list
[22:53] tjholowaychuk: drudge should make a webkit one :)
[22:53] tjholowaychuk: haha
[22:53] chapel: I love chrome tabs, so kod is interesting, but its missing a lot of must have features for regular use
[22:53] pyrotechnick: like coffeescript syntax highlighting lol
[22:53] drudge: i've yet to find an editor i like as much as vim
[22:53] chapel: I think one issue with using webkit would be speed/performance
[22:53] MrNibbles: drudge: whats the upload service you have mapped to cloud.penree.com
[22:53] mikeal: ryah: node pins the core during the uncached test
[22:54] drudge: MrNibbles: cloudapp
[22:54] mikeal: jsdom hits a lot of slow paths in v8
[22:54] mikeal: :(
[22:54] zemanel: chapel: true, but its a good start. also has some bugs
[22:54] tjholowaychuk: 
ACTION hates textmate grammars

[22:54] MrNibbles: drudge: thanks, its pretty neat
[22:54] mikeal: i hate everything about TextMate except using it
[22:54] drudge: MrNibbles: it's made by a friend of mine
[22:54] isaacs has joined the channel
[22:54] mraleph: mikeal: benchmark?
[22:54] dmcquay has joined the channel
[22:55] mikeal: http://www.mikealrogers.com/post/jspp-initial-performance
[22:55] jae1 has left the channel
[22:55] hunterloftis has joined the channel
[22:55] jae1 has joined the channel
[22:55] blueadept has joined the channel
[22:55] drudge: tjholowaychuk: friend of mine is making a really nice text editor for mac, i think we talked him into using node for the plugin interface this evening :P
[22:56] tjholowaychuk: drudge: rad
[22:56] chapel: nice
[22:56] tjholowaychuk: js plugins would be great
[22:56] mraleph: mikeal: mmm. standalone one? when you say that it hits slow paths you probably have something particular in mind? 
[22:56] mikeal: jsdom is all getters and setters
[22:56] mikeal: which aren't optimized in v8
[22:57] drudge: macvim with taglist + supertab + command-t + jslint + conque is really nice for node dev though
[22:57] tjholowaychuk: mikeal: I noticed when I did a few benchmarks the other day, even a single getter took a massive chunk out
[22:57] tjholowaychuk: :s
[22:57] tjholowaychuk: scared me
[22:57] tjholowaychuk: lol
[22:57] drudge: oh oh and snipmate
[22:57] springmeyer has joined the channel
[22:57] drudge: and jamescarr's snipmate-nodejs
[22:57] chapel: mikeal: I am surprised no one has made a node optimized version of jquery, not jquery itself but the language it uses (blah blah blah, I talk to much)
[22:58] mraleph: mikeal: afaik there should be optimized ICs for getters and setters.
[22:58] mraleph: let me check
[22:58] mikeal: tjholowaychuk: yeah, i've done that too
[22:58] mikeal: the thing is, jspp has jQuery OOTB but i want it to be pluggable
[22:59] mikeal: cause some people want YUI or Dojo or <insert my framework here>
[22:59] mikeal: and for that you need a working DOM
[22:59] chapel: drudge: you should do a quick screencast (just video) of you working in macvim so I can see what it looks like with all your fancy doodads
[22:59] mikeal: most of the time spent in jsdom is creating the intial DOM tree
[22:59] rauchg_: im sure you can cache it 
[22:59] rauchg_: per template 
[23:00] rauchg_: a "compilation" in a way
[23:00] mikeal: not that I can see
[23:00] mikeal: cause I REALLY wanted that
[23:00] Ond has joined the channel
[23:00] mikeal: that is my use case
[23:00] rauchg_: how come
[23:00] Aria: Which parser are you using, mikeal?
[23:00] mikeal: i need to basically "copy" a pre-parsed DOM
[23:00] mikeal: jsdom
[23:01] Aria: jsdom isn't a parser.
[23:01] mikeal: rauchg_: ask tmpvar
[23:01] ryah: wir machen ein stunde ein jahr
[23:01] tjholowaychuk: htmlparser
[23:01] mikeal: jsdom includes a parser
[23:01] Aria: Ah, okay. 
[23:01] Aria: Yeah, kinda. References anyway.
[23:01] mikeal: didn't even realize it was a different library :)
[23:01] jpick has joined the channel
[23:01] mikeal: where is davglass?
[23:01] Aria: (I know, because I wrote one of the other parsers that works with it ;-))
[23:02] tjholowaychuk: yuppers, I think they are both pretty slow
[23:02] mikeal: he's spent more time on jsdom perf than anyone else
[23:02] Aria: Yeah. I've hit some majorly slow interactions with v8.
[23:02] steffkes: ryah, what? ;o
[23:02] Aria: still trying to find a great path forward, but the speed of creating strings is painful.
[23:02] mikeal: we need better benchmarks for js
[23:02] mikeal: because all performance work is benchmark driven
[23:03] mikeal: and sunspider is 10 years old
[23:03] c4milo: mikeal++ 
[23:03] v8bot:  c4milo has given a beer to mikeal. mikeal now has 4 beers.
[23:03] Atmoz has joined the channel
[23:03] mikeal: the v8 guys dominate benchmarks
[23:03] mikeal: they just need some better ones :)
[23:04] patrick has joined the channel
[23:04] mraleph: if you see some really bad performance case --- just create a pure js example illustrating it and send it to v8-users. we are usually happy to accept and investigate.
[23:05] mikeal: will do :)
[23:05] tjholowaychuk: mikeal: let me know how that goes, when we made tobi we realized jsdom/htmlparser were crazy slow as well
[23:05] mikeal: mraleph: I don't know if this is still the case, but at some point if your arguments length when callings != the definition length you opted out of inlining 
[23:06] tjholowaychuk: so definitely interested there
[23:06] ryah: aslakhellesoy: okay fixed the base64 thing
[23:06] mikeal: is that still the case
[23:06] mikeal: tjholowaychuk: we need to spur some competition 
[23:06] sonnym has joined the channel
[23:06] tjholowaychuk: gahh
[23:06] aslakhellesoy: ryah: awesome. thanks a lot!
[23:06] tjholowaychuk: lol
[23:06] CIA-121: node: 03Ryan Dahl 07master * r282cce1 10/ lib/_debugger.js : debugger: don't display whole path of scripts - http://bit.ly/eF9OsB
[23:06] CIA-121: node: 03Ryan Dahl 07master * rd040f1d 10/ lib/_debugger.js : debugger: use correct handle handle - http://bit.ly/f1hben
[23:06] mikeal: polotek was working on libxml2 bindings
[23:07] mikeal: that could also pan out
[23:07] Aria: Indeed, mraleph. I keep trying to figure out what the actual slow parts are. I'll send benchmark suggestions when I find 'em.
[23:07] tjholowaychuk: mikeal: yeah I was going to use that actually but it was exploding in a variety of ways
[23:07] tjholowaychuk: so I gave up
[23:07] ryah: aslakhellesoy: https://github.com/ry/node/commit/f11e24b864a3ca3788bb8d0e5a05390ce676a9a3
[23:07] tjholowaychuk: but would be great
[23:07] technoweenie has joined the channel
[23:07] mraleph: mikeal: well. it is by design and it will remain this way for some time.
[23:07] mikeal: it's not ready yet
[23:07] mikeal: mraleph: ok
[23:07] pyrotechnick: static Handle<Value>
[23:07] pyrotechnick:   New (const Arguments& args)
[23:07] pyrotechnick:   {
[23:07] pyrotechnick:     HandleScope scope;
[23:07] pyrotechnick:     return args.This();
[23:07] pyrotechnick:   }
[23:08] pyrotechnick: should that be scope.Close(args.This())
[23:08] ryah: pyrotechnick: i don't think it matters becuase args.This() isn't a local handle
[23:08] mraleph: mikeal: there is a reason behind that. if you have arguments count mismatch that means that you are probably using arguments right?
[23:08] pyrotechnick: ok
[23:08] mikeal: no
[23:08] mikeal: not necessarily
[23:09] CIA-121: node: 03Ryan Dahl 07v0.2 * rf11e24b 10/ (src/node_buffer.cc test/simple/test-buffer.js): 
[23:09] CIA-121: node: base64 decode should handle whitespace
[23:09] CIA-121: node: painfully backported from v0.3 cf1db4f3043fce0474c421dc547b8bd2f9616354 - http://bit.ly/fEeGGj
[23:09] mraleph: ok
[23:09] mikeal: a lot of time a few arguments are just optional
[23:09] mikeal: and if you do something like
[23:09] mikeal: arg2 == arg2 || "default"
[23:09] pyrotechnick: ryah: can you force a GC?
[23:09] mikeal: using arguments causes pain :)
[23:09] ryah: pyrotechnick: --expose_gc and call gc()
[23:10] pyrotechnick: from
[23:10] ryah: your code
[23:10] pyrotechnick: js
[23:10] ryah: yes
[23:10] pyrotechnick: ok
[23:10] pyrotechnick: is that guaranteed to clean up all loose objects
[23:10] pyrotechnick: or will it still only do it if it needs the memory
[23:10] ryah: i dont know
[23:10] mraleph: mikeal: actually I gave it a second thought. bailing out from inlining with mismatched number of arguments was probably just a shortcut.
[23:10] richcollins has joined the channel
[23:11] mikeal: mraleph: i mean, you can just fill it in under the hood with undefined
[23:11] patrickr: is this channel only  for ppl working on the node.js project?
[23:11] mikeal: patrickr: it's also for people who wrestle bears
[23:11] ryah: patrickr: no
[23:11] ryah: patrickr: it's for people using node.js
[23:12] patrickr: k
[23:12] mikeal: not me, i'm mostly a bear wrestler
[23:12] tjholowaychuk: patrickr: mostly bear wrestling
[23:12] pyrotechnick: ryah: thanks that works
[23:12] b_erb has left the channel
[23:12] pyrotechnick: cool now i dont have to wait 10 minutes for it to crash lol
[23:12] mraleph: mikeal: it's not that easy. you also have to think about deoptimization from inlined function. that would require to reconstruct arguments adaptor frame... nasty. not difficult, but nasty... would require some changes in existing deoptimization framework. 
[23:13] mikeal: well, now you have tracing
[23:13] eee_c has joined the channel
[23:14] mikeal: and if you're calling it with similar arguments that aren't the defined length at least you might get those optimizations
[23:14] mikeal: haven't really tested
[23:14] mraleph: it is not tracing. *sigh* and before crankshaft we did not have inlining of js function at all. 
[23:15] mikeal: i was under the impression it was basically tracing but a little different
[23:15] mraleph: no. it is not tracing at all. 
[23:15] mikeal: damn
[23:15] mraleph: tracing is way more fine grained.
[23:15] mikeal: get on that ;)
[23:16] mjr_: mraleph: do you ever make it to the SF area to visit the googleplex or whatever?  It'd be great for node to hear you talk about some of these V8 internal issues, how to make things fast, things to watch out for, etc.
[23:16] patrickr: so this might be a more server noob question, but how do i keep users in their own private instance of my node app? i don't want them to have to log in, but they will be authenticating to flickr and twitter. do i use a session to store their info in a db? 
[23:16] mikeal: yeah, there is much rediculous souders work on bad performance situations but basically nothing on how to write to optimizations 
[23:16] patrickr: and then use their session to create their own object? var user1234 = new flickr(blah, blah);
[23:17] mraleph: mjr_: *sigh* I am Russian. That means going to US requires visa which is pain in the brain.
[23:17] mjr_: our immigration policy is so annoying
[23:17] mikeal: you should get one to come to NodeConf
[23:17] mikeal: :)
[23:17] mikeal: http://www.nodeconf.com/
[23:18] tjholowaychuk: mraleph: I think I need one too, but I just go
[23:18] tjholowaychuk: not sure
[23:18] bmizeran_ has joined the channel
[23:18] mraleph: I will wait for nodeconf.eu :-)
[23:18] mjr_: Ouch, the information keeps piling up on that site.  I don't have time to digest it all.
[23:18] mikeal: mjr_: call for speakers is coming soon
[23:18] mikeal: i finished the app
[23:18] mikeal: it's just a matter of when we announce
[23:18] mikeal: mraleph: no plans for .eu this year :(
[23:19] mikeal: maybe next year
[23:19] PyroPeter has joined the channel
[23:19] mraleph: pity. 
[23:19] mikeal: we have our hands full this year
[23:19] mikeal: 4 US conferences and one EU conference (JSConf.eu) 
[23:20] mjr_: mraleph: surely the corporate mega-machine under which you work has the legal muscle to get you into the US with the minimum of hassle, no?
[23:20] MrNibbles: drudge: trying linkinus again ;) seems quite a bit better than i remember! It handled the deluge from my bnc quite successfully
[23:20] drudge: MrNibbles: glad to hear it :)
[23:21] chapel: are you using the beta MrNibbles ?
[23:21] drudge: chapel: 2.2 was released today
[23:21] chapel: oh
[23:21] chapel: :)
[23:21] chapel: kk
[23:21] drudge: way to read the blog yo
[23:21] xSmurf: how can I list events listeners of an object??
[23:21] drudge: or our twitters
[23:21] mraleph: mjr_: there are certain procedures I still have to do myself. like collecting some paperwork and going to embassy in person (embassy is in Copenhagen, 3 hours by train from Aarhus) and having an interview there... too much pain for me. 
[23:21] chapel: hey, why should I read the blog :P
[23:21] MrNibbles: chapel
[23:22] mjr_: mraleph: well, that sucks.
[23:22] MrNibbles: oops
[23:22] MrNibbles: chapel: im not, only on the demo
[23:22] xSmurf: MrNibbles: did you try Textual? it's free and much quicker if you have a bouncer behind
[23:22] ryah: DHS: not helping
[23:22] chapel: well sounds like you should have the newest version
[23:22] xSmurf: much nicer for me than Linkinus
[23:22] mraleph: mjr_: yep. Iron Curtain is still in place
[23:22] drudge: xSmurf: have you tried 2.2?
[23:22] mape: mikeal: did you see that request issue I posted?
[23:23] xSmurf: when was that released?
[23:23] MrNibbles: xSmurf: nice will give that a go too
[23:23] mjr_: mraleph: perhaps just converted to a Red Tape Curtain.
[23:23] drudge: xSmurf: today
[23:23] xSmurf: no, I ditched it a couple months ago
[23:23] drudge: xSmurf: give it another shot, we put a lot of work into 2.2. much faster
[23:23] chapel: xSmurf: it is much faster
[23:23] xSmurf: it just couldn't handle it for me (small machine, dell mini 9)
[23:23] mikeal: mape: i don't know, what was it?
[23:23] chapel: and MrNibbles you have the newest version as the beta :)
[23:24] MrNibbles: yep, just checked it! assumed 'beta' meant not the demo ;)
[23:24] mape: mikeal: 16:47 < mape> mikeal: Any know reason this doesn't work? require('request')({'uri': 'https://www.google.com/accounts/AuthSubSessionToken'}, function(){console.log(arguments)});
[23:24] xSmurf: unless resources went down an order of magnitude or two then it's not gonna cut it for me
[23:24] dgathright has joined the channel
[23:24] xSmurf: it's too bad, I really liked it
[23:24] isaacs has joined the channel
[23:25] xSmurf: but 2gb ram 1.62ghz system... tight resources
[23:25] xSmurf: 32bit
[23:26] chapel: xSmurf
[23:26] chapel: I use it on my 2006 macbook, 1.83ghz cpu 2gb ram
[23:26] drudge: xSmurf: 2.2 cut down resource usage a ton
[23:26] chapel: and it still is snappy as usual
[23:26] drudge: it's a complete backend rewrite
[23:26] mape: mikeal: seems to work on v0.3.0-pre (had old version on laptop) but not on the v0.3.2-pre i had on my server
[23:27] galaxywatcher_ has joined the channel
[23:27] xSmurf: chapel: do you often load relatively big backlogs from your bouncer?
[23:27] chapel: well atm no, just 50
[23:27] chapel: but it handles tons of super active channels just fine
[23:28] chapel: and memory usage is way down
[23:28] MrNibbles: how do i disable smilies!?
[23:28] MrNibbles: those things drive me nuts
[23:28] drudge: under preferences
[23:28] drudge: styles
[23:28] mikeal: mape: is it just broken because https is broken?
[23:28] MrNibbles: ah, somehow looked at that page about 5 times and missed it each one :S
[23:29] MrNibbles: thanks :D
[23:29] drudge: they pop when you click them too btw, if they were in code bothering you
[23:29] MrNibbles: ooh thats quite a near feature
[23:29] mape: mikeal: that would be a logical explanation.. https is all broken right now?
[23:29] MrNibbles: neat*
[23:29] m64253 has joined the channel
[23:29] mikeal: it's broken in very unpredictable ways
[23:29] mape: ah :/
[23:29] mikeal: and the underlying TLS code has changed a lot in 0.3.x releases
[23:29] mikeal: mostly for the better
[23:29] mape: Serves me for not using stable
[23:30] mikeal: ryah started a branch today to gather fixes
[23:30] mikeal: at least that's what I'm pretty sure the branch is for :)
[23:30] mape: Guess this could be a good thing to test then 
[23:30] mikeal: yup
[23:30] mikeal: try it without request and if it works then I'll try to figure it out
[23:31] pyrotechnick: okay i've got the problem down to this one file
[23:31] pyrotechnick: if you can help please do https://github.com/feisty/node-bullet/blob/master/src/RigidBody.cc
[23:31] pyrotechnick: can anyone who knows bindings take a quick look over it and tell me why that would be being gc'd whilst still in scope
[23:32] tbeseda has joined the channel
[23:33] mape: mikeal: Btw, what is the use case for jspp?
[23:33] mikeal: i build like a little app a week
[23:33] mikeal: and sometimes there is stuff that I can't do easily with a couchapp
[23:34] mikeal: and it's easier to write something using the filessystem as the api
[23:34] mape: k
[23:34] mikeal: and uses server side jquery
[23:34] mikeal: which is a better environment to create pages in than any template system
[23:34] thebigbad has joined the channel
[23:34] mape: Yeah, saw you mentioned the speed of jsdom
[23:35] mikeal: then add shit loads of caching because jsdom is slow :)
[23:35] mape: Turns out to be quite an issue with my scraper :/
[23:35] tjholowaychuk: mape: same with tobi
[23:35] tjholowaychuk: did you bench it?
[23:35] tjholowaychuk: I didnt get around to that
[23:35] mikeal: yeah, we should put together a good SSJS DOM benchmark
[23:35] mikeal: and put out some numbers and try to get some movement on this
[23:36] mape: tjholowaychuk: Not really, just that it breaks down. Was looking at using workers but the api would be so wonky I never went through with it
[23:36] mikeal: it would be great if someone at Mozilla decided to tease apart their DOM to remove rendering :)
[23:36] rphillips has joined the channel
[23:36] pyrotechnick: yeah that would be awesome
[23:36] estrathmeyer has left the channel
[23:37] pyrotechnick: that's what i thought about when you started node-canvas TJ
[23:37] dgathright has left the channel
[23:37] tjholowaychuk: pyrotechnick: yeah
[23:37] tjholowaychuk: it is really annoying
[23:37] boaz_ has joined the channel
[23:37] pyrotechnick: it would be so awesome if mozilla could decouple their canvas2d from the rest
[23:37] tjholowaychuk: you cant get at dom exceptions
[23:37] pyrotechnick: but i think
[23:37] pyrotechnick: ATM you just have to roll your own
[23:37] tjholowaychuk: and its not easy to extend from c++
[23:37] pyrotechnick: the browser components are too coupled
[23:37] tjholowaychuk: if at all possible without hacking v8
[23:38] mape: tjholowaychuk: So when will I be able to use jsdom+node-canvas to create screenshots? or "degrade gracefully"? ;)
[23:38] pyrotechnick: and especially now with 3d acceleration and sandboxing 
[23:38] dgathright has joined the channel
[23:38] tjholowaychuk: mape: haha :)
[23:38] tjholowaychuk: 5 years
[23:38] tjholowaychuk: haha
[23:38] bingomanatee_: why not create a virtual V8 in V8?
[23:38] bingomanatee_: The 13th floor solution
[23:38] pyrotechnick: lol
[23:38] tjholowaychuk: and in that v8 we can have another v8 vm
[23:38] tjholowaychuk: written with the v8 v8
[23:38] mape: I actually need a screenshot service, now I'm running a node script wrapping a headless x server but that isn't optimal :/
[23:38] pyrotechnick: *provides the backing music*
[23:39] ryah: https working
[23:39] ryah: 
ACTION times it

[23:39] pyrotechnick: http://www.youtube.com/watch?v=BlcJ-fDmDhA
[23:39] ryah: 32 minutes
[23:40] tjholowaychuk: 
ACTION wishes narcissus would actually run nicely on node :(

[23:40] pyrotechnick: v8 in v8 in v8 theme music ^^
[23:40] tjholowaychuk: recursive v8!!
[23:40] chapel: ryah++
[23:40] v8bot:  chapel has given a beer to ryah. ryah now has 2 beers.
[23:40] mape: ryah: pushed to unstable master?
[23:40] tjholowaychuk: its just like inception, the deeper you go the slower it is haha
[23:40] mape: hehe
[23:40] pyrotechnick: hahaha
[23:40] chapel: mape: 0.2.x
[23:40] drudge: twss
[23:41] sechrist: ryah++
[23:41] v8bot:  ryah is getting too many beers.  Don't let ryah get drunk!
[23:41] mraleph: pyrotechnick:  what is being collected?
[23:41] pyrotechnick: ryah+=1/0
[23:41] pyrotechnick: in my bindings?
[23:41] sechrist: v8bot++
[23:41] v8bot:  sechrist has given a beer to v8bot. v8bot now has 11 beers.
[23:41] sechrist: ryah++
[23:41] v8bot:  ryah is getting too many beers.  Don't let ryah get drunk!
[23:41] pyrotechnick: v8: Infinity
[23:41] v8bot: pyrotechnick: Infinity
[23:41] pyrotechnick: ryah+=Infinity
[23:42] pyrotechnick: mraleph: you mean in my bindings?
[23:42] ryah: mape: just making a test
[23:42] ryah: mape: then i'll push
[23:42] pyrotechnick: can you turn the gc off?
[23:43] mraleph: mikeal: yeah. getters and setters are not as optimized as they can be. no support in Crankshaft at all --- if someone adds proper support CS should be able to inline them... 
[23:43] mraleph: pyrotechnick: yep. in your bindings. no you cannot
[23:43] mikeal: who is "someone" and where do I find them :)
[23:43] pyrotechnick: mraleph: the rigidBodies
[23:43] pyrotechnick: mraleph: look in test/master.coffee
[23:44] pyrotechnick: all i am doing is instantiating 1 of them and then gc'ing
[23:44] pyrotechnick: …i suck at node
[23:44] rockstar has joined the channel
[23:44] pyrotechnick: it's something so simple and i still cant do it
[23:44] pyrotechnick: otherwise the bindings work
[23:44] pyrotechnick: if you dont force a gc they run for hours
[23:44] rockstar has joined the channel
[23:44] mape: Someone mind testing this on 0.3.4-pre https://gist.github.com/764187 ?
[23:44] pyrotechnick: but they eventually crash when it gc's
[23:44] mraleph: pyrotechnick: you instantiate two as far as I see
[23:44] dothebart: ryah: some tiny suggestion for test/simple/test-c-ares.js
[23:44] dothebart: some line like 
[23:44] dothebart: dns.resolveMx('nodejs.org', function(error, data)	      { console.log('Done'); console.dir(data);}); 
[23:44] mape: oops, 1sec
[23:45] dothebart: to demonstrate how mx resolving works.
[23:45] pyrotechnick: mraleph: for i in [0...1000]
[23:45] pyrotechnick:   rigidBody = new Bullet.RigidBody
[23:45] pyrotechnick:   console.log 'rigidBody', rigidBody
[23:45] pyrotechnick:   
[23:45] pyrotechnick:   gc()
[23:45] pyrotechnick: crashes after 1 or 2
[23:45] pyrotechnick: after 2
[23:45] m64253 has joined the channel
[23:46] chapel: https://gist.github.com/raw/764187/3dc4508f72167101df40e6945e8149e0cc2ca022/broken-https.js
[23:46] dothebart: 
ACTION is integrating this into citadel.org s smtp client.

[23:46] rockstar has left the channel
[23:46] briznad has joined the channel
[23:46] mraleph: pyrotechnick: gdb backtrace? 	
[23:46] dothebart: tnx for the nice mx-query parser ;-)
[23:46] MrNibbles: has anyone had a look at the code for the webkit inspector?
[23:46] MrNibbles: its pretty nuts
[23:46] pyrotechnick: its on github
[23:46] pyrotechnick: sec
[23:46] mape: mikeal: Hmm that gist seems broken on both v0.3.0-pre and v0.3.4-pre
[23:47] pyrotechnick: mraleph: https://gist.github.com/a30153ea9e367f004ddb
[23:47] MrNibbles: you can inspect the inspector which i find deliciously recursive
[23:47] pyrotechnick: lol
[23:47] pyrotechnick: it fucks it up though
[23:47] pyrotechnick: at least it used to
[23:47] MrNibbles: it seems pretty stable for me at the mo
[23:47] MrNibbles: im using the nightly builds
[23:48] mape: Or rather, on v0.3.0-pre I get "ECONNRESET, Connection reset by peer" every now and then, most of the time I don't though
[23:48] pyrotechnick: same
[23:48] mraleph: pyrotechnick: well it clearly says that pointer being freed was not allocated :-)
[23:48] possibilities has joined the channel
[23:48] pyrotechnick: because it's cleaning it up twice
[23:48] pyrotechnick: because the bindings are wrong
[23:48] pyrotechnick: the scope.Close() stuff or something
[23:49] pyrotechnick: the file is only like 3 methods 50 lines or something
[23:49] pyrotechnick: and theres a problem in it
[23:49] pyrotechnick: but unfortunately i just dont know enough about v8 bindings to solve it
[23:49] pyrotechnick: been trying to fix this for weeks TBH
[23:49] mraleph: pyrotechnick: well. you allocate with new-operator but freeing it with free() --- that is not going to work in C++
[23:49] pyrotechnick: no
[23:49] pyrotechnick: its not freeing a pointer its freeing a member in a bullet class
[23:49] pyrotechnick: i didnt write bullet
[23:49] pyrotechnick: im only making bindings
[23:50] pyrotechnick: and bullet is fine
[23:50] mraleph: pyrotechnick: lines 43 and 48 in your https://github.com/feisty/node-bullet/blob/master/src/RigidBody.cc
[23:50] pyrotechnick: its the scope thing in the bindings
[23:50] pyrotechnick: 48 is commented out
[23:50] pyrotechnick: 43 just makes one
[23:50] mjr_: ryah: 32 minutes?  What'll you do with the remaining 88 minutes?
[23:50] tjholowaychuk: is there any reason we use ObjectWrap and not v8's External stuff? I havent looked at External so maybe it is something completely different
[23:51] kylefox has joined the channel
[23:51] mraleph: pyrotechnick:  my glasses are fine. well. I can clearly see that it was _not_ commented out at the moment of this >>> https://gist.github.com/a30153ea9e367f004ddb <<< crash
[23:51] pyrotechnick: well give me a sec then
[23:51] CIA-121: node: 03Ryan Dahl 07master * re4dd5cd 10/ (lib/http.js lib/net.js lib/tls.js src/node.js): NODE_DEBUG uses strings instead of bitflags - http://bit.ly/hjFd5a
[23:51] CIA-121: node: 03Ryan Dahl 07master * r94f8368 10/ (4 files in 2 dirs): First pass at new https server - http://bit.ly/fPrO5y
[23:51] mraleph: pyrotechnick: also you do not need scope.Close(args.This())
[23:52] mraleph: pyrotechnick: but that should not cause any real problem
[23:52] pyrotechnick: ive done various things in that file based on various peoples opinions
[23:52] pyrotechnick: if you have an alternate i am willing to try it trust me
[23:52] pyrotechnick: some say u dont need them but it helps
[23:52] pyrotechnick: you say u dont need it
[23:52] ryah: mjr_, chapel, mape: https://github.com/ry/node/commit/94f8368cf9e5077050525e32a24188402f077074
[23:52] pyrotechnick: but thats why its there
[23:52] pyrotechnick: because others have said u can do it if u want
[23:52] mape: ryah: testing
[23:52] mraleph: yeah. Handles are dark magic. 
[23:52] pyrotechnick: "put it in for good measure"
[23:52] pyrotechnick: so i did
[23:53] pyrotechnick: weeks of wasted time dark magic
[23:53] mraleph: rule of thumb is simple
[23:53] mraleph: scope.Close(foo) is required if foo belongs to scope. 
[23:54] mraleph: is foo was passed from outside it clearly does not belong to scope.
[23:54] mraleph: like args.This()
[23:54] mjr_: ryah: I'll point some iOS clients at it tomorrow. 
[23:55] ryah: great
[23:55] kylefox: Anyone know if there's something like liquidmarkup.org for nodejs?
[23:55] pyrotechnick: thats how i had it all mraleph
[23:55] isaacs has joined the channel
[23:55] pyrotechnick: have i been led astray?
[23:55] tjholowaychuk: kylefox: I dont believe so
[23:55] pyrotechnick: i have a new bt
[23:55] dothebart: ryah: got that mx sample js?
[23:55] tjholowaychuk: I started a port months and months ago
[23:55] pyrotechnick: but you're not going to beleive me
[23:55] ryah: dothebart: ?
[23:55] sechrist: woo no more stunnel
[23:55] kylefox: tjholowaychuk: cool, any luck?
[23:55] mraleph: pyrotechnick: another C/C++ rule of thumb is: if you allocate with new you have to deallocate with delete, new[] --- delete[], malloc/realloc/calloc --- free.
[23:55] dothebart: 00:44  dothebart: ryah: some tiny suggestion for test/simple/test-c-ares.js
[23:55] pyrotechnick: i know that
[23:55] dothebart: 00:44  dothebart: dns.resolveMx('nodejs.org', function(error, data)I      { console.log('Done'); console.dir(data);}); 
[23:55] pyrotechnick: but again
[23:56] pyrotechnick: others said to try and use free
[23:56] mraleph: pyrotechnick: show me your backtrace =)
[23:56] tjholowaychuk: kylefox: yeah but I wrote http://jade-lang.com instead
[23:56] pyrotechnick: https://gist.github.com/a30153ea9e367f004ddb
[23:56] romainhuet has joined the channel
[23:56] sechrist: I would if a node process could be anywhere near as high functioning as haproxy + stunnel
[23:56] mjr_: IRC is burning up my internets today
[23:56] mape: ryah: But yeah, if you are interested you might want to take a look at https://gist.github.com/764187 when you have the time, no errors, nothing. Just exits node silently.
[23:56] sechrist: doing similar tasks
[23:56] sveimac has joined the channel
[23:56] pyrotechnick: it's seriously commented out mraleph
[23:56] pyrotechnick: i know you dont think it is but it is
[23:56] mape: Hah, http://wargamez.mape.me/ is still running
[23:56] mape: Love when I forget things and they don't break
[23:57] sechrist: dat nodejs
[23:57] kylefox: tjholowaychuk: hah, cool. That's the other template lang we're looking at using ...
[23:57] chapel: mape, thats hilarious
[23:57] ryah: mape: client isn't done yet
[23:57] mraleph: pyrotechnick: can you do: f 3, then dissasemble 0x000000012580483c
[23:57] ryah: mape: just server
[23:57] pyrotechnick: sure
[23:57] ryah: client is going to take some work
[23:57] mape: ryah: Ah k
[23:57] sechrist: tls client is more involved than server?
[23:57] chapel: ryah: are you going to make it similar to server, and separate http/s?
[23:57] sechrist: O.o
[23:58] mikew3c has joined the channel
[23:58] mape: ryah: Well at least if you didn't have an easy thing to test on, now you do :)
[23:58] ryah: chapel: yeah
[23:58] pyrotechnick: like "f 3" then another command "dissasemble 0x000000012580483c"
[23:58] mraleph: yep
[23:58] pyrotechnick: (gdb) f 3
[23:58] pyrotechnick: #3  0x000000012580483c in RigidBody::~RigidBody ()
[23:58] ryah: chapel: do you like that?
[23:58] pyrotechnick: do u want the actual pointer u sent
[23:58] pyrotechnick: or that one
[23:58] pyrotechnick: oh theyre the same lol
[23:58] adambeynon has joined the channel
[23:58] kylefox: tjholowaychuk: Any particular reason you gave up on a liquid port?
[23:59] mraleph: pyrotechnick: i mistype disassemble
[23:59] chapel: ryah: I think it is fine, can be easier in some ways, will make proxy/routers have a bit different logic, but no issues really
[23:59] tjholowaychuk: kylefox: not really, but jade is more flexible. you can check out my ejs implementation
[23:59] tjholowaychuk: you can switch <? and ?> to {% %} via options
[23:59] tjholowaychuk: and it has filter support
[23:59] ryah: chapel: you mean for upgrading from http to https? 
[23:59] pyrotechnick: mraleph: https://gist.github.com/a30153ea9e367f004ddb
[23:59] chapel: sure, if you are handling both