[00:01] eddanger has joined the channel
[00:01] onar has joined the channel
[00:01] taf2 has joined the channel
[00:01] keeto has joined the channel
[00:12] beppu has joined the channel
[00:12] aguynamedben has joined the channel
[00:12] kriszyp has joined the channel
[00:15] micheil_mbp has joined the channel
[00:16] technoweenie has joined the channel
[00:19] jacobat has joined the channel
[00:21] beppu has joined the channel
[00:21] aguynamedben has joined the channel
[00:21] kriszyp has joined the channel
[00:22] jacobat has joined the channel
[00:22] isaacs_ has joined the channel
[00:22] isaacs_ has joined the channel
[00:24] mattly has joined the channel
[00:26] stevestmartin has joined the channel
[00:28] gbot2 has joined the channel
[00:54] binary42 has joined the channel
[01:19] isaacs_mobile has joined the channel
[01:22] SubtleGradient has joined the channel
[01:22] SubtleGradient: is there anything like env.js that works in node.js?
[01:24] inimino: not unless you port it
[01:24] SubtleGradient: is there any sort of DOM-like API in anything that will run on node?
[01:27] inimino: not that I'm aware of
[01:27] SubtleGradient: thx
[01:28] inimino: but then I've never understood while people want it, so I haven't looked and might not be the best person to ask
[01:28] SubtleGradient: I was just asking the room in general
[01:59] mattly has joined the channel
[02:00] stephenlb: 
ACTION likes how fast node is and wants it to stay fast or get faster.  and to keep it that way with low memory foot by excluding the unnecessary.

[02:01] SubtleGradient: um… thanks for that elmo
[02:01] SubtleGradient: :P
[02:05] brandon_beacher has joined the channel
[02:10] ashb: inimino: the reason to want a DOM is to support client side code easier
[02:10] ashb: 
ACTION hands stephenlb loadable modules

[02:17] inimino: ashb: so I've heard
[02:17] inimino: ashb: it's not that I doubt people have compelling use cases, it's just that I've never seen or been able to imagine one
[02:17] ashb: but yeah, the DOM is a horrible API
[02:17] isaacs_mobile has joined the channel
[02:18] inimino: yes...
[02:21] un0mi has joined the channel
[02:28] charlenopires has joined the channel
[02:28] aguynamedben has joined the channel
[02:29] hassox has joined the channel
[02:37] rictic has joined the channel
[02:46] isaacs_mobile_ has joined the channel
[03:05] SubtleGradient has left the channel
[03:20] mattly has joined the channel
[03:30] mynyml has joined the channel
[03:39] un0mi has joined the channel
[03:56] JimBastard has joined the channel
[03:57] sstmartin has joined the channel
[03:58] JimBastard: sup inimino
[03:58] JimBastard: do you think you'd be able to help me track down my session.js issue at some point tonight? if you busy its cool
[04:08] binary42 has joined the channel
[04:12] allnickstaken has joined the channel
[04:13] micheil: JimBastard: hm.. what's the issue?
[04:13] micheil: 
ACTION also authored a http session library

[04:33] technowe_ has joined the channel
[04:35] unom1 has joined the channel
[04:43] JimBastard: is freenode being attacked?
[04:54] unomi has joined the channel
[04:57] lifo has joined the channel
[04:58] steadicat has joined the channel
[05:16] cloudhead has joined the channel
[05:17] JimBastard: well micheil i can't figure out why http://github.com/Marak/session.js/blob/master/session.js is working on one node instance and not on another
[05:17] JimBastard: its probaly something stupid on my end
[05:21] JimBastard: how goes the smtp
[05:21] JimBastard: im pretty excited about being able to spam people with node, i mean send blast emails
[05:21] JimBastard: ^_^
[05:30] micheil: it's almost ready.
[05:31] micheil: there's a bit still to do (I haven't begun implementing the auth segments yet)
[05:31] micheil: what are the two node versions?
[05:31] JimBastard: then all we need to do is coherse someone into building the .server
[05:32] JimBastard: v0.1.22-15-g8fd46a3 and v0.1.22-7-g3d24e11
[05:32] micheil: I've got meta code so that the server can be built
[05:32] JimBastard: but i think it might be because its on an IP and not a domain?
[05:32] micheil: try bringing them both up to HEAD
[05:32] micheil: maybe
[05:32] JimBastard: yaa you think
[05:32] micheil: try running "localhost"
[05:32] micheil:  / 127.0.0.1
[05:33] JimBastard: what you mean
[05:33] cdorn has joined the channel
[05:33] micheil: if one server is running on an IP and the other is on a domain, maybe there's an issue there
[05:33] JimBastard: thats what im thinking
[05:34] JimBastard: but where in the code
[05:34] micheil: ryah: did you see that Node is the github featured project?
[05:34] inimino: JimBastard: ah, I'll look
[05:34] JimBastard: woooooot
[05:34] JimBastard: sup inimino
[05:34] JimBastard: i can send you the site where its running at if that helps
[05:34] JimBastard: its probaly a stupid mistake, but it shouldnt be failing silently
[05:36] inimino: JimBastard: got a public URL?
[05:36] JimBastard: yeah
[05:36] JimBastard: ill pm
[05:41] micheil: JimBastard: aww.. not going to show me? :(
[05:42] micheil: hmm.. what testing frameworks do people use?
[05:43] micheil: (something that isn't the assert module.)
[05:50] erichocean has joined the channel
[05:52] JimBastard: micheil we figured it out, i wasnt calling the setcookie method in the header
[05:52] JimBastard: i assumed it was magic but its not
[05:55] inimino: I think someone's fork has the magic, but I was a purist and didn't want to mess with node's internals to make it work
[05:55] inimino: the Right Thing is some kind of middleware...
[05:56] JimBastard: middleware for http requests?
[05:56] inimino: Coming Soon™
[05:56] JimBastard: id would have to be baked into node to make sense
[05:58] inimino: JimBastard: maybe, or something layered on top that has some kind of ... layer, and abstracts away the request/response objects
[05:58] inimino: 
ACTION is off

[05:58] inimino: bbl
[06:02] erichocean has joined the channel
[06:12] keeto has joined the channel
[06:13] keeto has joined the channel
[06:13] keeto has joined the channel
[06:14] unom1 has joined the channel
[06:31] spoob has joined the channel
[06:36] scudco has joined the channel
[06:57] scudco has joined the channel
[07:06] piranha has joined the channel
[07:24] qFox has joined the channel
[07:51] mikeal has joined the channel
[08:07] jamiew has joined the channel
[08:23] mikeal has joined the channel
[08:31] tlrobinson has joined the channel
[08:33] tlrobinson has joined the channel
[08:37] jamiew_ has joined the channel
[08:38] tlrobinson_ has joined the channel
[08:39] tlrobinson1 has joined the channel
[08:59] jamiew has joined the channel
[09:39] teemow has joined the channel
[10:07] markwubben has joined the channel
[10:28] piranha has joined the channel
[10:36] micheil has joined the channel
[10:39] tlrobinson1 has left the channel
[10:53] steadicat has joined the channel
[10:56] webben has joined the channel
[11:00] micheil: hmm.. what was vendor stuff in the various frameworks and libraries about?
[11:00] micheil: eg, http://github.com/danwrong/restler/tree/master/lib/vendor/
[11:04] elliottcable has joined the channel
[11:32] kennethk_ has joined the channel
[11:54] felixge has joined the channel
[11:58] hassox has joined the channel
[12:16] rtomayko has joined the channel
[12:19] rtomayko has joined the channel
[12:25] piranha: micheil: that's dependecies (if you still need an answer :))
[12:25] piranha: *dependencies
[12:25] micheil: yeah, okay cool
[12:25] micheil: 
ACTION is moving some of the stuff that isn't directly interacting with node-smtp out of it's core sections

[12:31] kennethk_: micheil: i've started hacking away at a smtp server implementation
[12:35] micheil: kennethkalmer: why don't you fork http://github.com/miksago/node-smtp
[12:36] micheil: it has hooks for an SMTP Server
[12:36] micheil: (then I'll pull back in and add as contributor.)
[12:37] kennethkalmer: great idea
[12:37] kennethkalmer: although I have one concern
[12:37] micheil: have a look at /lib/smtp/server.js
[12:37] micheil: yeah?
[12:37] kennethkalmer: wouldn't it bloat the library too much ? bundling a client and server in one lib...
[12:37] micheil: not really
[12:37] kennethkalmer: thing is, the server needs the client (to relay)
[12:37] micheil: because it's a protocol implementation
[12:38] kennethkalmer: gotcha
[12:38] kennethkalmer: expecting all eyes on http://github.com/kennethkalmer/node-smtp :)
[12:38] micheil: and the way that I've structured lib/smtp.js is so that the server or client stuff is only included when you create either
[12:38] kennethkalmer: +1
[12:39] kennethkalmer: just one request: have patience with me as I get my non-DOM javascript-fu going
[12:39] micheil: I'm currently adding in the queue libs as a submodule, so it's a bit rough
[12:40] kennethkalmer: watching
[12:41] kennethkalmer: i'm curious, what is your plans for the smtp server component ?
[12:43] felixge_ has joined the channel
[12:55] micheil: kennethkalmer: I'm not sure yet
[12:55] micheil: I'm working on getting the client working first, although, it makes sense just to have one smtp module
[12:56] micheil: kennethkalmer: there's an upstream version as well, just made a few changes to the structure
[12:57] kennethkalmer: upstream ?
[12:58] hassox: hey kennethkalmer :D
[12:58] kennethkalmer: hey hassox :)
[12:58] kennethkalmer: off the train I hope ?
[12:58] hassox: yeah mate :D
[12:58] hassox: finally got to read your email ;)
[12:58] kennethkalmer: awesome
[12:59] kennethkalmer: hassox: summer killing you guys as well ? exceptionally hot over here :/
[12:59] hassox: yeah
[12:59] hassox: very hot
[13:00] micheil: where abouts are you two?
[13:00] hassox: 37ºC the other night
[13:00] hassox: Melbourne Australia
[13:00] micheil: ah
[13:00] micheil: I'm a bit north of that
[13:00] kennethkalmer: Johannesburg, South Africa
[13:00] kennethkalmer: 37, dammit
[13:00] mediacoder: Darmstadt, Germany :-) ..allright..your cities are better known ;-)
[13:01] hassox: mediacoder: I'd love to spend a bit of time over there
[13:01] kennethkalmer: we're around 32C, but humidity is 70+
[13:01] hassox: ouch
[13:01] hassox: that's horrible
[13:01] mediacoder: here its like -7°C
[13:01] kennethkalmer: ouch
[13:01] hassox: mediacoder: it's dry when it's that cold right?
[13:02] mediacoder: hassox: yea, at the moment its pretty dry
[13:02] mediacoder: (humidity) ..much snow tho
[13:03] micheil: mediacoder: hmm, yeah, I'm not sure what the weather is at the moment in Dusseldorf where my colleagues are based
[13:04] micheil: probably something near -7ªC
[13:06] mediacoder: micheil: im a bit more south, but there was much snow this year allover germany, i guess (not as much as UK tho)
[13:07] micheil: hmm.. anyone know all the commands to working with git forks?
[13:07] kennethkalmer: which are you looking for ?
[13:08] micheil: the stuff for pull and pushing and managing the forks
[13:08] micheil: kennethkalmer: did you get your repo up to date?
[13:08] kennethkalmer: yep
[13:08] kennethkalmer: submodules et all
[13:09] kennethkalmer: first thing is to add the forks as remotes
[13:09] kennethkalmer: git remote add kennethkalmer git://github.com/kennethkalmer/node-smtp.git
[13:09] kennethkalmer: after adding remotes, let git update itself : git remote update
[13:09] kennethkalmer: then you can create 'tracking branches'
[13:10] kennethkalmer: git branch --track kennethkalmer-master kennethkalmer/master
[13:10] kennethkalmer: where kk-master is a local branch by you, tracking kennethkalmer (remote) / master (branch)
[13:10] kennethkalmer: then from there you just checkout and pull and merge as usual
[13:10] kennethkalmer: does that help ?
[13:11] micheil: yeah
[13:12] kennethkalmer: or you can try these tools ( i haven't yet )
[13:12] kennethkalmer: http://github.com/nowells/git-goggles
[13:12] kennethkalmer: http://github.com/defunkt/hub
[13:12] micheil: so, git branch --track [branch name] [remote repo]
[13:12] micheil: yeah?
[13:13] kennethkalmer: git branch --track [branch name] [refspec]
[13:13] micheil: yeah'
[13:13] rolfb has joined the channel
[13:13] kennethkalmer: yip
[13:13] inimino has joined the channel
[13:13] micheil: okay..
[13:13] micheil: have you made any changes?
[13:13] kennethkalmer: not yet, studying your code
[13:14] kennethkalmer: will send you pull requests
[13:14] micheil: okay
[13:14] kennethkalmer: let me gist what I had
[13:14] grantmichaels has joined the channel
[13:14] micheil: hopefully the filesystem's easy enough to understand
[13:14] micheil: (the only one which might be confusing is the ../vendor/Queue/lib/Queue)
[13:15] kennethkalmer: but that is submodules, it's fine
[13:15] micheil: yeah
[13:15] micheil: I might redo Queue to be system independent later, but I'm not sure for now
[13:15] micheil: also, for testing I've been working on /mock/Server
[13:17] joshthecoder has joined the channel
[13:17] kennethkalmer: http://gist.github.com/278814
[13:17] kennethkalmer: that was my last working copy before refactoring
[13:18] micheil: hey, you know, I never thought of using regex..
[13:18] kennethkalmer: got that from EventMachine's smtpd implementation
[13:19] kennethkalmer: combined it with jan____'s awesome-style callbacks
[13:19] micheil: I've been working off ruby's net::smtp for now
[13:20] kennethkalmer: i saw
[13:20] micheil: it seemed a bit sane, although, I'm only using it as a guide
[13:21] kennethkalmer: i'm going to lug my existing refactor into lib/smtp/server.js and finish that up
[13:21] micheil: okay
[13:23] micheil: I'm yet to read through the various RFC's for smtp servers, hopefully there'll be uses for things like the queue manager and stuff like that
[13:23] kennethkalmer: definitely
[13:23] kennethkalmer: smtp server will need a spool
[13:24] kennethkalmer: it will need the client when relaying mesages
[13:24] micheil: hmm..
[13:25] micheil: haha, I'm reading through your code, and normally I'll write things in the order of which they'll happen, eg, connect -> commands -> quit
[13:25] micheil: So I miss took the quit callback as a connect on :P
[13:25] kennethkalmer: :)
[13:26] kennethkalmer: i tackle low hanging fruit first, usually :)
[13:27] micheil: btw, with the EHLO reply, maybe it'd work if we had a set of options, to control what the server should implement
[13:27] micheil: or do you think it'd be better to implement everything we have?
[13:28] kennethkalmer: i think we should offer the bare minimum and allow implementers to easily extend
[13:28] kennethkalmer: bare minimum is still a lot though
[13:28] micheil: hmm..
[13:29] micheil: http://tools.ietf.org/html/rfc5248
[13:29] markwubben has joined the channel
[13:30] kennethkalmer: sigh, I'm only busy with rfc2821 atm
[13:31] kennethkalmer: i'm already seeing an overlap in my refactored code
[13:31] kennethkalmer: i've got a SMTPSession object going
[13:33] micheil: so what's SMTPSession do?
[13:34] kennethkalmer: beefed up PacketHandler it seems
[13:34] kennethkalmer: actually, it handles everything
[13:35] micheil: okay
[13:35] micheil: I'd be interested to see that part of the code
[13:35] micheil: because I'm not sure on the packetHandler just yet
[13:36] micheil: it's one of my newer pieces of code
[13:36] kennethkalmer: ah
[13:36] kennethkalmer: great, let me just get it working properly
[13:36] kennethkalmer: i'm very new to prototypical js
[13:36] kennethkalmer: so this quite messy atm
[13:37] micheil: yeah, we can clean it up and all that later..
[13:37] micheil: We just need to make sure we tag it before making clean ups
[13:37] kennethkalmer: +1
[13:37] kennethkalmer: or keep a history file
[13:38] micheil: hmm.. yeah
[13:38] micheil: I tagged v0.5, which was before my latest refactor
[13:38] micheil: the difference between the code now and then is fairly large
[13:40] micheil: Once I have the client protocol working, I'm going to work on adding in a wrapper library (which is mailer.js)
[13:40] kennethkalmer: aaah
[13:40] kennethkalmer: that makes sense
[13:40] kennethkalmer: thought it was very low level
[13:41] micheil: yeah, still trying to work out the best way to handle things..
[13:42] micheil: I'm not too keen on using events for absolutely everything, I'm thinking callbacks might be the way to go
[13:44] kennethkalmer: i hate that node just quits without any output to the console :/
[13:44] micheil: heh, I actually used extra timers with empty functions just to hold node open
[13:45] micheil: (just for debugging, of course ;P)
[13:48] emilis_info has joined the channel
[13:49] piranha has joined the channel
[13:50] micheil: kennethkalmer: I don't suppose you have any preference as to testing frameworks?
[13:50] kennethkalmer: not for node (or js in general) yet
[13:50] micheil: okay
[13:50] kennethkalmer: but I do like rspec-style testing
[13:51] micheil: hmm..
[13:53] micheil: How about using: http://github.com/ecavazos/speks
[13:55] kennethkalmer: looks good
[13:55] micheil: okay, I'll work on setting that up in the repo
[13:57] kennethkalmer: great
[13:59] micheil: although, it does seem to have a few issues, I'm looking at the source to try and resolve them
[14:01] kennethkalmer: hassox: what testing library do you recommend ?
[14:01] hassox: spectacular :)
[14:02] hassox: http://github.com/jcrosby/spectacular
[14:02] hassox: the async testing support is really simple :)
[14:02] kennethkalmer: showoff ;)
[14:02] hassox: http://github.com/jcrosby/spectacular/blob/master/spec_example.js
[14:02] hassox: ;)
[14:06] micheil: hassox: I looked at it, but it looked maybe too thin?
[14:06] hassox: micheil: how so?
[14:06] hassox: not enough assertion types?
[14:06] micheil: well, because it's just three files
[14:06] hassox: it's really only 1 ;)
[14:06] micheil: I look at others and there's a lot more
[14:06] hassox: the other is an example, and a .gitignore ;)
[14:06] micheil: that's what I mean
[14:07] hassox: micheil: does that make the others more useful?
[14:07] micheil: not so, but the others do have tests and a consistent file structure
[14:07] micheil: eg the /lib, /tests, doc
[14:08] hassox: micheil: dunno mate...
[14:08] hassox: it's simple, and it does async
[14:08] hassox: that's 2 big ticks from me ;0
[14:08] hassox: ;)
[14:09] hassox: micheil: which do you recommend?
[14:09] micheil: okay, wins my vote..what's the command line usage like?
[14:09] hassox: node tacular.js
[14:09] micheil: hassox: I always use a specific directory structure
[14:09] hassox: micheil: it's not perfect
[14:10] hassox: micheil: do you have a preferred test runner / setup?
[14:10] micheil: umm.. not really, recently I've done a little testing on dojo's DOH
[14:10] micheil: but it's an odd test suite
[14:11] technoweenie_: http://github.com/technoweenie/ntest
[14:11] technoweenie_: is my testing lib
[14:11] hassox: initially I looked at jquerys test suite, but it's pretty DOM centric
[14:11] hassox: ohh
[14:11] hassox: looking
[14:11] hassox: technoweenie: how does it do async?
[14:11] technoweenie_: i really liked rtomayko's test suite for his bert lib, which was just a test(..., func) method
[14:11] micheil: technoweenie: checkin' it out, also, if it's a node project, add it to http://github.com/ry/node/wiki/modules
[14:12] hassox: technoweenie: it looks pretty nice
[14:12] technoweenie_: i use promise.wait() a lot
[14:12] technoweenie_: http://github.com/technoweenie/wheres-waldo/blob/master/test/api_test.js
[14:12] hassox: technoweenie: can it do nested describes with before / after
[14:12] hassox: ?
[14:12] micheil: hmm.. would it be wise to start defining a bit of a module structure?
[14:12] technoweenie_: no nesting
[14:13] technoweenie_: i like the syntax in jcrosby's
[14:13] technoweenie_: i wanted to get away from something like 
[14:14] technoweenie_: describe('foo', function() { this.test(..., function() ...
[14:14] hassox: i like assert.equal
[14:14] technoweenie_: thats part of node.js
[14:14] technoweenie_: ntest has no assertions
[14:14] hassox: oh you're using nodes assertions :D
[14:14] hassox: sweet
[14:14] micheil: I don't like assert because it kills the thread
[14:14] hassox: micheil: ?
[14:14] hassox: what do you mean?
[14:14] micheil: it throws errors, which, unless caught explicitly will end the test suite
[14:15] hassox: technoweenie: what about the case where in the test you need to fire off somethign async, and then callback for the assertion?
[14:15] technoweenie_: so, catch them :)
[14:15] technoweenie_: hassox: i just call wait() on the promises
[14:15] hassox: technoweenie_: that assumes a promise is available though
[14:16] technoweenie_: what am i, a peasant?
[14:16] technoweenie_: no, i'm a coder damnit
[14:16] technoweenie_: http://github.com/technoweenie/wheres-waldo/blob/master/test/test_client.js
[14:16] technoweenie_: http client wrapper that returns a promise
[14:17] hassox: heh
[14:17] technoweenie_: micheil: caught exception: http://img.skitch.com/20100116-nx53gfpqn8fxii5ns26fj4gti9.png
[14:18] technoweenie_: well, the tests run immediately as they're defined
[14:18] micheil: will it continue to test though?
[14:18] technoweenie_: yea
[14:18] micheil: If I have an error, I want to know about it, but I want to know that the rest of the tests still run
[14:19] technoweenie_: http://img.skitch.com/20100116-1ybqbtr25ihmegbbr49e7p6n8i.png
[14:19] hassox: nice
[14:20] hassox: can you add a promise thing like in jcrosby's one?
[14:20] technoweenie_: yea...
[14:20] technoweenie_: i want to toy with the idea of having your tests run async too
[14:20] hassox: hawt
[14:21] hassox: technoweenie_: that could play havoc with the state of what you're testing though yeah?
[14:21] technoweenie_: yea perhaps
[14:21] technoweenie_: it's easy to wrap stuff in a promise though
[14:22] technoweenie_: did you see my blog post that explored async coding
[14:22] hassox: no
[14:22] technoweenie_: i'm pretty new to it
[14:22] hassox: please to be pointing
[14:22] hassox: me too
[14:22] technoweenie_: http://techno-weenie.net/2010/1/15/node-js-for-my-tiny-ruby-brain
[14:22] technoweenie_: basically my original lib code used wait() everywhere, until i got passing tests
[14:22] technoweenie_: then i added async and essentially moved the wait() calls to the tests
[14:23] hassox: async is a blast
[14:23] hassox: have you seen my twitter streamer?
[14:23] technoweenie_: no
[14:23] technoweenie_: nested describes might be cool, i just hate all the js syntax that it adds
[14:24] micheil: oh man.. dailup internet is so painful...
[14:24] hassox: http://github.com/hassox/my_node_playground/tree/master/tweet_node/
[14:24] technoweenie_: my tests have this pseudo indentation
[14:24] micheil: technoweenie: I thought I knew your name from somewhere, ENTP
[14:24] technoweenie_: micheil: yep
[14:24] hassox: technoweenie_: it reads from the twitter stream API and for each tweet it recieevs it emits a "newTweet" event
[14:24] technoweenie_: hassox: thats cool
[14:24] technoweenie_: we're probably going to do something like that, but have it emit webhooks
[14:25] hassox: each request has a listener setup on the streamer such that when it recieves the event, it pushes it out to the client
[14:25] hassox: technoweenie_: exactly :D
[14:25] hassox: technoweenie_: somethign that I've been thinking of that kennethkalmer is working on that you might like
[14:25] hassox: kennethkalmer: is looking at making something like the EM smtp server (a relay)
[14:26] kennethkalmer: not looking, is busy :)
[14:26] micheil: hassox: we've merged it into node-smtp
[14:26] hassox: so you could setup postfix to relay email to a smtp server (relay) inside the node instance on port 25
[14:26] hassox: micheil: awesome
[14:26] hassox: so
[14:26] un0mi has joined the channel
[14:26] hassox: have it emit an event "mailReceived" 
[14:26] hassox: then the app can add a listeenr to it, and process the email as it arrives
[14:27] micheil: hassox: We're welcome to suggestions
[14:27] hassox: micheil: that's my use case ^^ 
[14:27] hassox: that's what I'd love to see
[14:28] micheil: hassox: I've been experimenting with different api's between both client smtp and server smtp
[14:28] micheil: although, last I knew, the stuff that did the receiving of emails was imap / pop
[14:28] hassox: micheil: have you seen the Event machien one?
[14:28] technoweenie_: i'd rather have the email just dumped to a queue somewhere than rely on an active service that has to be running to receive events
[14:28] micheil: sort of
[14:28] hassox: micheil:  think of it as a relay instead of a mailbox
[14:29] technoweenie_: we process tons of emails a day, exim just dumps them to a maildir on the server, and a ruby script churns through them
[14:29] michaelk^ has joined the channel
[14:29] micheil: technoweenie: in that case, my Queue runner might interest you, it uses an evented queue / array to process data
[14:30] technoweenie_: i currently use a db table for a queue actually
[14:30] technoweenie_: well, the email server uses teh filesystem as a queue for new emails
[14:30] technoweenie_: ive been wanting to move to redis, just havent had an urgent need yet
[14:31] technoweenie_: hey hassox can i pick your brain about the twitter stream api real quick
[14:31] hassox: technoweenie_: has ezra made his distributed ring thing for redis yet?
[14:31] hassox: yaeh..
[14:31] hassox: I don't know heaps about it
[14:31] technoweenie_: "The default access level allows up to 200 track keywords and 400 follow userids"
[14:31] hassox: but hit me
[14:31] technoweenie_: does that mean you can have 1 connection that tracks 200 keywords and follows 400 users?
[14:31] hassox: yeah I believe so
[14:32] technoweenie_: hmm ok thats uh pretty sweet :)
[14:32] hassox: each user id can only have 1 active connection to the api
[14:32] technoweenie_: well, we're kicking around the idea of integrating twitter into Tender Support
[14:32] technoweenie_: and i was imagining having a tweet stream connection for each company acct
[14:32] hassox: my idea of usage was that one node could be connected, and my "app" cluster could hold open a connection to it to receive
[14:32] hassox: oh nice :)
[14:33] technoweenie_: well, do keep thsi node tweet streamer oss :)
[14:33] technoweenie_: i may just run a simple one that fires off webhooks and do the processing in the rubies
[14:34] hassox: yeah
[14:34] hassox: that app is just an exp btw
[14:34] technoweenie_: sure
[14:35] hassox: soz
[14:35] hassox: baby just woke up
[14:35] hassox: brb
[14:35] technoweenie_: the only issue is it only returns public tweets
[14:35] technoweenie_: which makes sense i guess, but if i'm connecting w/ my acct is hould see private tweets of my friends? hm
[14:37] michaelk^: question about the http server / client and this snippet: http://gist.github.com/278836 - if i have a client coming in with "Connection: close", why is the connection closed before i ever call res.finish?
[14:38] bry has joined the channel
[14:39] hassox: bak
[14:39] hassox: technoweenie_: I think that's the case yeah
[14:39] hassox: but not sure... I just played with the public stuff
[14:40] hassox: was more an experiment to have 1 stream going to n client connections from a single node
[14:40] hassox: concurrent connections and all that stuff
[14:40] kriszyp has joined the channel
[14:40] technoweenie_: yea cool
[14:40] technoweenie_: how do you find your twitter user id
[14:41] hassox: I have a json configuration object example there
[14:41] hassox: is that what you mean?
[14:42] technoweenie_: the follower param needs a numerical twitter id
[14:42] hassox: oh right
[14:42] technoweenie_: maybe i need to make an api call to see it
[14:43] piranha has joined the channel
[14:43] technoweenie_: oh its on the rss feed link i think
[14:43] hornbeck has joined the channel
[14:44] hassox: yeah tehre it is
[14:45] hassox: k guys
[14:45] hassox: I'm off for the night
[14:45] technoweenie_: later
[14:45] hassox: technoweenie_: I'd be interested to hear how you go :)
[14:45] hassox: cyas
[14:47] micheil: bye hassox 
[14:48] micheil: kennethkalmer: do you intend on implementing smtp-auth extension?
[14:48] kennethkalmer: ciao hassox !
[14:48] kennethkalmer: micheil: yep
[14:49] micheil: okay, I've got a base64 lib in vendor, just checking out jsdigest for md5 and the others
[14:50] micheil_mbp has joined the channel
[14:53] micheil_mbp: sorry about that, net maintainence
[14:54] micheil_mbp has joined the channel
[14:54] un0mi has joined the channel
[14:55] micheil_mbp: mm..
[14:55] micheil: that's better.
[14:57] micheil: kennethkalmer: it looks like smtp-auth will only work on server for now, until we have ssl connections on the clients in tcp
[14:58] kennethkalmer: is that on ryah's todo list ?
[14:58] micheil: node-origin/net2
[14:58] micheil: hopefully
[14:58] kennethkalmer: cool
[14:59] micheil: also..
[14:59] micheil: Note: A server implementation MUST implement a configuration in which it does NOT permit any plaintext password mechanisms, unless either the STARTTLS [SMTP-TLS] command has been negotiated or some other mechanism that protects the session from password snooping has been provided. Server sites SHOULD NOT use any configuration which permits a plaintext password mechanism without such a protection mechanism against password snooping.
[14:59] micheil: from RFC4954
[15:02] micheil: hmm..
[15:03] micheil: I think I'll keep the idle callback in smtp-client, although, I'll change client.mail(to_addr), to client.mail(to_addr, callback)
[15:03] micheil: or something
[15:04] micheil: client.mail returns promise?
[15:05] kennethkalmer: think so
[15:05] kennethkalmer: then the promise can fail or succeed
[15:05] kennethkalmer: i still have plenty to learn around promises
[15:05] kennethkalmer: one last smoke then I have to call it a day
[15:06] kennethkalmer: almost done with the prototypical refactoring
[15:06] kennethkalmer: haven't cursed at js-scoping issues quite like this before
[15:07] micheil: prototypical refactoring?
[15:07] micheil: gotta love long words.. :P
[15:08] kennethkalmer: hehe
[15:08] micheil: brb
[15:08] JoePeck has joined the channel
[15:15] micheil has joined the channel
[15:15] d3w has joined the channel
[15:15] r11t has joined the channel
[15:15] micheil: righteo..
[15:15] stevestmartin has joined the channel
[15:15] alex-desktop has joined the channel
[15:17] micheil: kennethkalmer: do you think it'd be beneficial to write a packet formatter?
[15:17] kennethkalmer: explain a bit more please
[15:18] un0mi has joined the channel
[15:18] micheil: well, something to create the packets, rather then just using strings
[15:19] kennethkalmer: ah
[15:19] micheil: this.get("MAIL FROM:<"+from_addr+">", function(packet){ => something more readable
[15:19] kennethkalmer: maybe
[15:19] kennethkalmer: but i don't think just yet
[15:20] micheil: this.get(packet("MAIL FROM:<%1>", "email@test.com"
[15:20] micheil: or something
[15:20] micheil: not really a packet formatter but sprintf
[15:22] kennethkalmer: maybe it is my inner rubyist that resists the formatter :)
[15:23] micheil: hmm..
[15:23] micheil: because the current style isn't the nicest
[15:25] micheil: hmm.. with the promise for all client commands, if I go with passing errback the packet, that should be fine..
[15:26] micheil: it'd probably also mean we could use .wait() in tests
[15:27] micheil: (yeah, reading technoweenie's article on node now..)
[15:28] mies has joined the channel
[15:28] kennethkalmer: still have to read that article
[15:29] micheil: kennethkalmer: the most useful part in the whole smtp docs has to be section 4.3.2
[15:29] micheil: with the chart to display the flow of status codes
[15:30] kennethkalmer: thanks, will digest that a bit later
[15:31] kennethkalmer: i've just pushed an smtp-server branch to github
[15:31] micheil: okay
[15:31] micheil: I'm just working on a promise pattern:
[15:31] kennethkalmer: still needs a lot of work, but will pick it up later
[15:31] kennethkalmer: http://github.com/kennethkalmer/node-smtp/commit/4ffb29bf9762d314c881e71da06022053d628c3f
[15:32] micheil: http://gist.github.com/278856
[15:32] micheil: oh, yeah, what os / editor are you using?
[15:33] kennethkalmer: osx/gvim
[15:33] kennethkalmer: erm, macvim
[15:34] micheil: http://gist.github.com/278858
[15:34] micheil: are the two snippets I'm using in textmate, that's for the headers
[15:34] kennethkalmer: i'll have to see how I can tweak vim to match
[15:34] micheil: I'm not sure if it's possible to port those to vim, but yeah
[15:35] micheil: (of course, change the file_info one to default to your name)
[15:35] micheil: I've really gotta figure out the tabstops in textmate..
[15:38] micheil: kennethkalmer: are you using 4 spaces = 1 tab?
[15:40] micheil: kennethkalmer: we should add modelines into the files
[15:40] micheil: (I've just added modeline support to my copy of textmate.. so, yeah)
[15:41] inimino: tabs are evil, just use spaces
[15:43] kennethkalmer: i default to 2 spaces per tab (ruby)
[15:43] kennethkalmer: micheil: I have to run
[15:43] micheil: okay
[15:43] kennethkalmer: thanks for the assistance and eagerness :)
[15:43] kennethkalmer: will catch you later again
[15:44] micheil: okay, and thanks for the contribution and work on the server
[15:49] un0mi has joined the channel
[15:51] technoweenie_: is there a node.js debugger?
[15:52] micheil: a few
[15:52] micheil: node_debug I think is the favoured
[15:54] micheil: kennethkalmer: #2eec492
[16:18] technoweenie_: damnit, node changed enough in the last week to break my shit
[16:19] un0mi: yeah, a few things broke 
[16:20] technoweenie_: dang i should've kept track of what commit i was on
[16:21] technoweenie_: running "./configure ; make ; make install" every time i update the node.js repo should install it properly, right?
[16:21] un0mi: 1.23 still had most working
[16:21] un0mi: probably uninstall first is a fine idea
[16:21] un0mi: make uninstall
[16:22] technoweenie_: ah ok
[16:22] un0mi: preferably *before* you update :)
[16:22] un0mi: although I guess the worst that can happen is some dangling unused files
[16:22] micheil: yeah
[16:22] micheil: no need to uninstall generally
[16:23] technoweenie_: yea it seems to know what it needs to compile
[16:23] micheil: I mean, I haven't noticed any side affects, but that's just me.
[16:23] technoweenie_: yea but how would you know???
[16:23] micheil: make test ftw
[16:23] technoweenie_: oh right
[16:23] micheil: I run the test suites compulsively
[16:24] un0mi: make test only checks the files that in the build dir afaik
[16:24] un0mi: not the ones which are actually installed in your sys
[16:24] micheil: oh, yeah, true
[16:24] technoweenie_: crap, i think i left my redis server off, thats why tests failed
[16:24] technoweenie_: jeez
[16:24] micheil: but either way, if I file isn't used, it's not used
[16:24] micheil: no harm done
[16:24] micheil: anything that is used gets overwritten
[16:28] technoweenie_: ok well its fine, i should have something that fails early if redis is down :)
[16:28] technoweenie_: though my test is hanging, hmm
[16:43] micheil: 
ACTION is implementing promises into node-smtp client

[16:46] ibolmo has joined the channel
[16:47] JoePeck has joined the channel
[16:48] ibolmo: getting an error for the `make` 
[16:48] ibolmo: http://paste.mootools.net/d746dd702
[16:49] ibolmo: this is a PPC mac
[16:49] ibolmo: maybe that'd be problem
[16:49] dew: in deps/v8/SConstruct, find this: "ARCH_GUESS = utils.GuessArchitecture()" (line 574 on my copy)
[16:50] dew: and replace it with: ARCH_GUESS = utils.GuessArchitecture() or ''
[16:50] dew: then try ;)
[16:52] ibolmo: [ 8/22] libv8.a: deps/v8/SConstruct -> build/default/libv8.a
[16:52] ibolmo: scons: Reading SConscript files ...
[16:52] ibolmo: A value for option arch must be specified (arm, ia32, x64).
[16:52] ibolmo: so i guess the ARCH_GUESS should be returning ppc (or similar)
[16:52] dew: ah, *that*'ll be your ppc chip scuppering you ;)
[16:55] ibolmo: any ideas?
[16:55] ibolmo: looks like i could pass arch=ppc directly
[16:55] dew: i don't have much to offer i'm afraid mate, the above hack was from the mailing list
[16:56] Yuffster has joined the channel
[17:00] ibolmo: yeah looks like i'm sol
[17:00] ibolmo: if i remember, v8 does not support ppc
[17:02] micheil: kennethkalmer: hmm.. the packetHandler seemed to have a massive error in it, it's emptying the data array ever loop
[17:19] cloudhead has joined the channel
[17:20] cloudhead has joined the channel
[17:29] steadicat has joined the channel
[17:34] steadicat has joined the channel
[17:46] ibolmo has left the channel
[17:47] micheil: hmm.. hopefully I'll have mail...
[17:52] DamZ has joined the channel
[17:57] rictic has joined the channel
[17:59] micheil: JimBastard: Okay, the current HEAD of node-smtp can functionally send emails. (I'll confirm 100% when I receive the email I just sent with it, the server said okay, so now it's just a waiting game.)
[18:00] micheil: hash: f69d18e
[18:00] micheil: G'night chaps.
[18:14] aho has joined the channel
[18:24] jamiew has joined the channel
[18:24] cloudhead_ has joined the channel
[18:43] joshthecoder_ has joined the channel
[18:49] DamZ has joined the channel
[18:49] alex-desktop has joined the channel
[18:49] technoweenie_ has joined the channel
[18:49] JimBastard has joined the channel
[18:49] mattly has joined the channel
[18:49] technoweenie has joined the channel
[18:49] onar has joined the channel
[18:49] taf2 has joined the channel
[18:51] mattly has joined the channel
[18:51] taf2 has joined the channel
[18:51] technoweenie has joined the channel
[18:51] technowe_ has joined the channel
[18:52] steadicat has joined the channel
[18:52] DamZ has joined the channel
[18:54] technoweenie: is there a way to tell if a promise has fired other than if(!promise.hasFired && promise._blocking)
[18:55] technoweenie: or maybe just have wait() not blow up if its already been fired
[18:56] grantmichaels has joined the channel
[19:01] Booster has joined the channel
[19:03] Booster has joined the channel
[19:09] davidsklar has joined the channel
[19:13] technoweenie: async twitter stream lib: http://github.com/technoweenie/twitter-node
[19:20] cadorn has joined the channel
[19:22] binary42 has joined the channel
[19:27] piranha has joined the channel
[19:29] ako has joined the channel
[20:02] eviltwin has joined the channel
[20:03] micheil has joined the channel
[20:10] mies_ has joined the channel
[20:14] mies_ has joined the channel
[20:24] erichocean has joined the channel
[20:30] isaacs has joined the channel
[20:50] binary42 has joined the channel
[20:57] charlenopires has joined the channel
[21:01] inkubus08 has joined the channel
[21:02] inkubus08 has left the channel
[21:02] bry has joined the channel
[21:25] teemow has joined the channel
[21:39] DamZ has joined the channel
[21:50] technowe_: hey has anyone tried jack with node.js
[21:55] JimBastard has joined the channel
[21:58] felixge has joined the channel
[22:08] charlenopires has joined the channel
[22:18] un0mi has joined the channel
[22:25] felixge: ryah: does node only check or eio callbacks on ev_idle?
[22:28] steadicat has joined the channel
[22:32] piranha has joined the channel
[22:50] pdelgallego has joined the channel
[23:01] un0mi has joined the channel
[23:18] JimBastard: does anyone know what channels are being spammed?
[23:23] felixge: ryah: you around?
[23:23] felixge: JimBastard: why spammed?
[23:33] micheil: moins all
[23:35] DamZ has joined the channel
[23:38] JimBastard: felixge did you see the global notice?
[23:39] felixge: JimBastard: nope
[23:39] JimBastard: o
[23:39] un0mi has joined the channel
[23:43] micheil has joined the channel
[23:44] rektide has joined the channel
[23:45] aho: JimBastard, i heard that ##opengl was spammed quite a lot
[23:46] aho: probably happend to all rather big channels
[23:46] JimBastard: must have been those directx guys again