[00:00] AkhiAbdullah has joined the channel
[00:00] diverdude has joined the channel
[00:00] _unary has joined the channel
[00:00] looopy has joined the channel
[00:01] diverdude: konobi_: hey, are you there?
[00:01] killing-joke: zomg, thanks.  i'm creating some metrics.  i was hoping to skip past the "yeah, we just totally make that shit up" part, to "some smart people thought about this, and i included their wisdom"  :)
[00:01] Joeysomo has joined the channel
[00:02] jacobolus has joined the channel
[00:02] navinpy has joined the channel
[00:03] kei has joined the channel
[00:03] rauchg has joined the channel
[00:03] average_drifter has joined the channel
[00:05] maru_cc has joined the channel
[00:06] riskish has joined the channel
[00:06] CIA-102: node: 03isaacs 07master * r0cdf85e 10/ (61 files in 6 dirs): Lint all the JavaScripts. - http://git.io/KBMk5Q
[00:06] CIA-102: node: 03isaacs 07master * r9633843 10/ Makefile : jslint src/ and lib/ on 'make test' - http://git.io/GJ-diA
[00:07] marcello3d has joined the channel
[00:07] lfppv has joined the channel
[00:07] jakehow has joined the channel
[00:08] lfppv: hi
[00:08] lfppv: why i cant use Iterator pattern of javascript 1.7 in node
[00:08] lfppv: ?
[00:09] isaacs: lfppv: because that's not something v8 implements
[00:09] LowValueTarget has joined the channel
[00:10] lfppv: isaacs: im using google closure tools, the goog.structs.Set object
[00:10] lfppv: and i need to iterate it
[00:10] isaacs: lfppv: ok.  node doesn't modify the language.  it just runs whatever v8 gives it
[00:11] sreeix has joined the channel
[00:11] sauerbraten: I'm using javascript for 2 days now, how can I add a key and value pair to an array?
[00:11] sauerbraten: *object
[00:11] torvalamo: obj.key = value
[00:12] isaacs: sauerbraten: obj[key] = val?
[00:12] torvalamo: or that, if it's a literal
[00:12] isaacs: obj.foo = bar is the same as obj['foo'] = bar
[00:12] sauerbraten: ah ok
[00:12] sauerbraten: I thought it woul be harder than that o.O
[00:13] isaacs: :)
[00:13] isaacs: sauerbraten: javascript's a pretty easy language.
[00:13] sauerbraten: somtimes it is confusingly easy, though :D
[00:13] arkh_ has left the channel
[00:14] lfppv: isaacs: ok, i just found it, google closure tools implement iterator pattern of javascript 1.7 but not nativelly
[00:14] torvalamo: or you can do Object.defineProperty(obj, key, {value: value}) and be totally ECMA 5!
[00:14] lfppv: i need to use goog.iter.toIterator
[00:14] torvalamo: who the fuck told them to make JS into Java
[00:15] torvalamo: i saw a presentation video yesterday where someone was extremely excited about this leap backwards in time
[00:16] sandfox has joined the channel
[00:17] isaacs: torvalamo: Object.defineProperty is silly to use in that case.  obj.key = value is 100% the same as Object.defineProperty(obj, 'key', {value:value, enumerable: true, writable: true, configurable: true })
[00:17] torvalamo: that's what i said
[00:17] jtsnow has joined the channel
[00:17] torvalamo: except i used cursewords
[00:18] isaacs: torvalamo: but obj.key=value is just as "ecmascript 5"
[00:18] torvalamo: it was a sarcastic remark by me
[00:19] torvalamo: just the fact that it has to be introduced
[00:19] ajackbot has joined the channel
[00:19] qsobad has joined the channel
[00:19] torvalamo: why don't they just make a new browser script langua... oh wait they are
[00:20] isaacs: well, it's kind of nice if you DO want to set something as read-only
[00:20] wink_ has joined the channel
[00:20] torvalamo: read-only is bollox. the only use i see for this is the enumerable thing, and even that probably is NEVER used.
[00:21] agnat_ has joined the channel
[00:22] torvalamo: and having a flag for deciding whether or not the other flags can be edited... that's so meta and a waste of time i have never seen.
[00:22] diverdude has joined the channel
[00:23] ohtogo has joined the channel
[00:24] tkahn6_ has joined the channel
[00:25] sandfox: read only is good if you want to stop other developers completely abusing your objects too easily
[00:25] torvalamo: but isn't that their choice?
[00:25] torvalamo: locking up your code isn't very nice at all
[00:25] jacobolus has joined the channel
[00:27] sandfox: it is there choice, but you can dissuade them, and it seems to work ok in the classical inheritance world
[00:27] markwubben has joined the channel
[00:27] hij1nx_ has joined the channel
[00:27] torvalamo: if setting some of these flags somehow optimizes the access time and/or sizes of the objects then sure that's a good idea... but somehow i doubt it does... at all
[00:28] samBiotic has joined the channel
[00:29] isaacs: torvalamo: enumerable:false is important.  and getter/setter properties are extremely useful in some situations.  (do you think we attach global.console in node if you never use it?)
[00:30] shanebo has joined the channel
[00:30] jtsnow has joined the channel
[00:30] ashishg has joined the channel
[00:30] eb4890 has joined the channel
[00:31] diverdude: anybody tried setting node.js up with haproxy?
[00:32] davidwalsh has joined the channel
[00:33] sandfox: yeah
[00:33] sandfox: diverdude: having problems?
[00:34] rauchg has joined the channel
[00:37] imsky_ has joined the channel
[00:41] criswell has joined the channel
[00:42] gr4yscale has joined the channel
[00:43] norviller has joined the channel
[00:45] wink___ has joined the channel
[00:46] rauchg has joined the channel
[00:46] bingomanatee: are there any decimal formatting hints for until.format?
[00:47] bingomanatee: util.format
[00:49] diverdude: sandfox: not anymore :D
[00:50] diverdude: i finally freakin fixed it hehe
[00:50] diverdude: sandfox: took me a helluva time i tell ya
[00:51] joshsmith has joined the channel
[00:51] jgaui has joined the channel
[00:52] sandfox: diverdude: first time can be a bit of pain
[00:52] diverdude: sandfox: oh yes...i found a config that helped me 
[00:52] diverdude: sandfox: this guy: https://github.com/mixu/sioconfig
[00:52] aristidesfl has joined the channel
[00:53] diverdude: sandfox: im gonna test now if my auth routine also works in this setup
[00:53] sreeix has joined the channel
[00:54] gr4yscale has joined the channel
[00:55] diverdude: sandfox: you using php+socket.io in same app?
[00:56] sandfox: divderdude: sort of… I've never had php directly handle socket-io though as it i personally think it's completely the wrong tool for job
[00:56] chjj has joined the channel
[00:57] diverdude: sandfox: oh thats not what i meant
[00:57] sandfox: diverdude: I've had nodejs front ends that handle socket.io and then if needed make restful calls to a php server
[00:57] diverdude: sandfox: i have socket.io running on node.js sharing session with php app running on apache
[00:58] jiboumans_ has joined the channel
[00:58] diverdude: sandfox: sharing session via mysql and authing using sessionID in cookie
[00:58] bnoordhuis: bingomanatee: no
[00:58] diverdude: sandfox: or at least  hope i am...i didnt test the auth with my haproxy setup yet
[00:58] diverdude: sandfox: thats my next task
[01:00] sandfox: diverdude: gotcha, i've only done session sharing using redis as the session store for both node and php (but should work the same way as your mysql setup) 
[01:01] diverdude: sandfox: yeah ok...redis is some shared memory right?
[01:01] ritch- has left the channel
[01:01] sandfox: diverdude: make sure your domains on the cookies are set correctly (this caused me much pain in testing when using random domains). yeah, redis is an in memory key->value store
[01:03] diverdude: sandfox: yeah...domains are really important....i already experienced some of that pain
[01:03] diverdude: sandfox: it hurts
[01:04] ryanfitz has joined the channel
[01:06] mAritz has joined the channel
[01:09] wink_ has joined the channel
[01:11] stef_ has joined the channel
[01:14] stagas has joined the channel
[01:14] johnhamelink has joined the channel
[01:15] rauchg has joined the channel
[01:16] Vespakoen has joined the channel
[01:16] k1ttty has joined the channel
[01:17] wink_ has joined the channel
[01:18] konobi_: diverdude: mmm?
[01:20] _th_n has joined the channel
[01:22] isaacs has joined the channel
[01:22] wink_ has joined the channel
[01:23] simenbrekken has joined the channel
[01:23] joshkehn has joined the channel
[01:23] diverdude: konobi_: nevermind...i think i got things figured out for now :) sry for disturbing you
[01:25] neoesque has joined the channel
[01:25] blueadept has joined the channel
[01:26] ryanfitz has joined the channel
[01:28] wink___ has joined the channel
[01:30] rauchg has joined the channel
[01:33] shinuza has joined the channel
[01:34] saikat has joined the channel
[01:36] trupppOFF has joined the channel
[01:36] RyanD has joined the channel
[01:41] jldbasa has joined the channel
[01:42] nerdfiles1 has joined the channel
[01:42] wink_ has joined the channel
[01:42] nerdfiles1 has left the channel
[01:42] thalll has joined the channel
[01:47] CIA-102: node: 03Mikeal Rogers 07master * rd530ee6 10/ (4 files in 3 dirs): Issue #2762. Add callback to close function. - http://git.io/mB5DSQ
[01:47] wink___ has joined the channel
[01:50] diverdude: is there a node.js debugger which may be recommended?
[01:54] wink_ has joined the channel
[01:57] subbyyy has joined the channel
[01:59] wink___ has joined the channel
[02:00] shinuza_ has joined the channel
[02:01] mikeal has joined the channel
[02:03] joshfinnie has joined the channel
[02:06] cagdas has joined the channel
[02:06] wink_ has joined the channel
[02:06] trupppOFF has joined the channel
[02:10] jblanche has joined the channel
[02:11] Danielss89 has joined the channel
[02:11] Danielss89: HI
[02:11] Danielss89: I have a module which have exports.foo and export.bar...
[02:11] Danielss89: how do i call foo from bar?
[02:13] _jzl has joined the channel
[02:14] isufy has joined the channel
[02:14] innociv has joined the channel
[02:16] SubStack: Danielss89: exports.foo()
[02:17] SubStack: or var foo = exports.foo = ...; foo()
[02:17] SubStack: or exports.foo = foo; function foo () { ... } then elsewhere foo()
[02:17] wink_ has joined the channel
[02:17] SubStack: (this is another thing about harmony modules that I hate)
[02:19] mandric has joined the channel
[02:19] mappum has joined the channel
[02:19] mappum: isnt there a way to make node scripts restart if they fail int he command line?
[02:19] SubStack: while true; do node beep.js; done
[02:20] SubStack: or forever or monit or whatever other tool
[02:20] mappum: isnt there something built in, thoough?
[02:20] SubStack: why should there be?
[02:20] mappum: i'm just wondering, i feel like i remember seeing that somewhere
[02:23] joshthecoder has joined the channel
[02:24] ningu: woohoo, I have implemented the most horrendous function in the world
[02:25] wizonesolutions has joined the channel
[02:25] ningu: http://pastebin.com/M1FsPWXx
[02:26] r04r has joined the channel
[02:27] ningu: I guess a 100 line function is actually not so bad.
[02:27] ningu: but the nesting....
[02:27] brianleroux has joined the channel
[02:29] joshkehn has joined the channel
[02:29] joshkehn has left the channel
[02:31] wink_ has joined the channel
[02:31] mAritz: 
ACTION slaps ningu with a large trout

[02:31] r04r has joined the channel
[02:31] r04r has joined the channel
[02:31] mAritz: you disgust me, sir!
[02:33] trupppOFF has joined the channel
[02:34] ningu: mAritz: there aren't many ways to make that function less ugly. believe me, I tried.
[02:34] ningu: I'll try again once I get some sleep ;)
[02:35] mAritz: every callback you define could easily be defined outside
[02:35] mAritz: making it simpler to maintain and possibly even simpler to test
[02:36] ningu: mAritz: actually, a lot of them need to access things in the context.
[02:36] mAritz: so?
[02:36] mAritz: wrap them
[02:36] ningu: so if they're defined outside, they don't have access to things I've declared in the calling function.
[02:36] mAritz: wrap them
[02:37] ningu: maybe I don't know what wrapping is.
[02:37] ningu: if you mean bind "this", it won't be enough...
[02:37] r04r has joined the channel
[02:38] mAritz: lambda
[02:38] ramitos has joined the channel
[02:39] ningu: I don't see how it solves the problem. var foo = bar; asyncThingy(function() { callback(stuff} })
[02:39] ningu: the callback still won't have access to foo.
[02:39] mAritz: you need x, so do someAsyncFunction(something, getSomeCallbackHandler(x)); where getSomeCallbackHandler returns your callback handler ;)
[02:39] tkahn6_ has joined the channel
[02:40] ningu: whoops, mixed up a brace with a paren there. anyway, yes, I've used that technique sometimes.
[02:40] mAritz: makes it much more readable imo
[02:41] ningu: I'll have to think about whether it makes sense to do it for this function. most of the stuff it does hangs together.
[02:41] mAritz: to be honest, i try to write functions that do as little as possible
[02:41] diogogmt has joined the channel
[02:42] sharkbird has joined the channel
[02:42] boccato has joined the channel
[02:42] mAritz: (though i fail miserably every time)
[02:42] ningu: well, this function is necessary due to what the site is supposed to do -- so the feature is not an option. maybe the implementation could be better. I hope so.
[02:43] joshsmith has joined the channel
[02:43] gr4yscale has joined the channel
[02:43] r04r has joined the channel
[02:43] mAritz: no, what i mean is that i rather have a couple of small functions with one small purpose than one big function with one big purpose and lots of steps.
[02:44] mAritz: (of course splitting it up too much is bad as well)
[02:44] mAritz: all about finding balance :)
[02:45] ningu: yes, but not everything is easy to break down into small purposes. you can break everything down into steps, but it's trickier if the steps depend on each other in a complex way.
[02:46] jtsnow has joined the channel
[02:47] skyler_brungardt has joined the channel
[02:47] mAritz: yes, but in most cases you'll then be able to simplify your logic :)
[02:48] SubStack: sometimes you can reframe the problem slightly to make it easier
[02:48] lee2012 has joined the channel
[02:48] SubStack: it can take a lot of time to refactor pieces into something more elegant
[02:49] r04r has joined the channel
[02:50] mAritz: which is why i almost never get anything done in my own projects :D
[02:51] devaholic: SubStack, https://gist.github.com/1861683
[02:51] mAritz: maybe i need a refactor counter for parts of my code and if it hits 10 i'll have 2 hours left until i'm completely blocked from changing these files for 2 months. :D
[02:55] SubStack: devaholic: people keep asking me for more tshirts too
[02:55] SubStack: what have you unleased‽
[02:55] devaholic: i saw that.
[02:55] devaholic: i still need to send the ones i have for you
[02:56] devaholic: eep
[02:56] SubStack: that campaign was crazy successful
[02:58] SubStack: devaholic: sweet gist, I've been doing something similar with seaport
[02:58] SubStack: still in the mad science phase though
[02:58] devaholic: ive got node-network-deploy and node-haproxy on deck to follow up with that
[02:59] devaholic: this might be a little bit, but ive been testing it all day
[02:59] devaholic: we'll see how it goes!
[02:59] SubStack: yep
[02:59] ph^ has joined the channel
[02:59] SubStack: I've been working on the experiment harness for my deploy stack today
[03:00] SubStack: although I haven't gotten much done
[03:00] r04r has joined the channel
[03:00] kurtzhong has joined the channel
[03:01] infynyxx has joined the channel
[03:02] devaholic: sometime i may work on that other idea you had when this is done
[03:02] jimmy has joined the channel
[03:02] devaholic: deploying just certain commits to certain machines and keeping track of it all
[03:06] wink___ has joined the channel
[03:07] SubStack: devaholic: I nearly have that working with propagit
[03:07] devaholic: i want it :D
[03:07] kurtzhong_ has joined the channel
[03:08] SubStack: I mean, propagit works now but the pieces to make it friendly and easy aren't finished yet
[03:08] devaholic: oh i didnt know about this
[03:08] intripoon has joined the channel
[03:08] r04r has joined the channel
[03:08] devaholic: you should make a module to search all of your modules
[03:08] ph^ has joined the channel
[03:08] ritch- has joined the channel
[03:09] joshsmith has joined the channel
[03:09] SubStack: stagecoach is all about tying together propagit to seaport
[03:10] SubStack: and managing the running instances from one place
[03:10] SubStack: I can't remember what I was hung up on finishing it
[03:10] SubStack: it's pretty close
[03:10] warz has joined the channel
[03:10] warz has joined the channel
[03:11] devaholic: sweet
[03:11] ningu: is something better than toolbox.no.de in the works?
[03:11] devaholic: i remember looking at stage coach, but didnt see a use case for me
[03:11] devaholic: i'll have a looksie 
[03:11] devaholic: coffee first!
[03:12] ningu: at least toolbox.no.de has heard of express, unlike search.npmjs.org
[03:12] r04r has joined the channel
[03:13] ningu: actually, it's there under "most depended on". but the search box is seriously broken.
[03:13] jtsnow has joined the channel
[03:14] Aria: I use npm search myself.
[03:15] ningu: if you search for express it turns up a lot of express modules but not express itself
[03:15] SubStack: coffee indeed!
[03:15] devaholic: duckduckgo.com
[03:15] devaholic: type !npm express
[03:16] devaholic: victory
[03:16] SubStack: devaholic: did you see https://gist.github.com/1734155 ?
[03:17] SubStack: the dependency part is very similar to your gist
[03:17] ningu: btw what are the use cases for redis over, say, mongodb?
[03:17] SubStack: even uses the same unicode chars ripped from npm ls :p
[03:17] ningu: devaholic: hehe, cool, thanks
[03:17] sdwrage has joined the channel
[03:18] devaholic: i want to make network.json as standard as i can
[03:18] devaholic: taking a look now.
[03:19] SubStack: I need something like a network.json for stagecoach too
[03:19] r04r has joined the channel
[03:19] wink_ has joined the channel
[03:19] SubStack: but just to map services in the current build to versions
[03:20] devaholic: if you npm install network-deploy and open readme you can see a full version
[03:20] Aria: ningu: npm search express returns express for me.
[03:20] SubStack: so I can feed those versions into seaport to go back through the git history to find an acceptable version
[03:20] devaholic: gotta get it on github
[03:20] SubStack: or rather, just the deployed versions
[03:21] SubStack: since I don't want intermediate commits to accidentally get deployed
[03:21] devaholic: hmm
[03:21] devaholic: yeah i was thinking about using tags
[03:21] webguynow has joined the channel
[03:25] SubStack: devaholic: the problem with tags is that it's only a single version
[03:25] SubStack: and perhaps you have multiple services in the same repo
[03:26] r04r has joined the channel
[03:26] r04r has joined the channel
[03:26] mansoor has joined the channel
[03:26] SubStack: putting that info into a json file makes the most sense I think
[03:27] ningu has joined the channel
[03:27] mansoor has joined the channel
[03:27] benvie has joined the channel
[03:28] rwaldron has joined the channel
[03:30] cburyta has joined the channel
[03:32] r04r has joined the channel
[03:33] mAritz has joined the channel
[03:34] redir has joined the channel
[03:36] jtsnow has joined the channel
[03:39] r04r has joined the channel
[03:39] jtsnow has joined the channel
[03:40] pyrotechnick has joined the channel
[03:41] joshkehn has joined the channel
[03:45] joshgillies has joined the channel
[03:45] r04r has joined the channel
[03:47] wink___ has joined the channel
[03:48] Vennril2 has joined the channel
[03:49] crutex has joined the channel
[03:49] crutex has joined the channel
[03:49] pendlepants has joined the channel
[03:50] blueadept has joined the channel
[03:51] arduix has joined the channel
[03:52] r04r has joined the channel
[03:52] r04r has joined the channel
[03:53] meso has joined the channel
[03:54] meso has joined the channel
[03:59] bbbb has joined the channel
[03:59] r04r has joined the channel
[04:01] wink_ has joined the channel
[04:04] tekky has joined the channel
[04:05] r04r has joined the channel
[04:06] wink___ has joined the channel
[04:07] sdwrage has joined the channel
[04:08] isaacs_mobile has joined the channel
[04:10] kurtzhong has joined the channel
[04:10] r04r has joined the channel
[04:10] isufy has joined the channel
[04:10] mAritz1 has joined the channel
[04:12] eukreign has joined the channel
[04:12] sdwrage has joined the channel
[04:14] qsobad has joined the channel
[04:14] qsobad has joined the channel
[04:15] r04r has joined the channel
[04:15] rurufufuss has joined the channel
[04:18] ryanfitz has joined the channel
[04:21] r04r has joined the channel
[04:26] munichlinux has joined the channel
[04:26] r04r has joined the channel
[04:27] richardr has joined the channel
[04:27] sreeix has joined the channel
[04:28] jaime_ has joined the channel
[04:31] wingie: what i have learned today: do not use CTRL+Z to kill a program
[04:31] Genius_ has joined the channel
[04:31] westg has joined the channel
[04:31] pyrotechnick has joined the channel
[04:32] eukreign: sometimes you have to
[04:32] eukreign: if CTRL-C doesn't do it
[04:32] r04r has joined the channel
[04:32] r04r has joined the channel
[04:32] Skaag has joined the channel
[04:32] eukreign: wingie: what's wrong with CTRL+Z?
[04:33] SomeoneWeird: is there a way to parse a string into an array?
[04:33] wingie: it suspends the program
[04:33] wingie: doesn't kill it
[04:33] wingie: eukreign: which gave me some issues with running vows since there were 50 vows processes still in the memory
[04:33] wingie: and it somehow collided with my current running vows script
[04:34] SomeoneWeird: killall node
[04:34] wingie: now when i killed them all with "kill -9" it works great
[04:34] wingie: oh
[04:34] wingie: stypid me
[04:34] wingie: i did kill -9 pid1 pid2 pid3 pid4 :/
[04:34] ningu: yeah, if Ctrl-C doesn't work you use Ctrl-\, not Ctrl-Z :)
[04:35] devaholic: you probably dont want to do killall node
[04:35] wink_ has joined the channel
[04:36] ningu: btw why is node-inspector apparently somewhat flaky?
[04:36] wingie: devaholic: why?
[04:36] ningu: sometimes I set breakpoints one place and they don't work but another they do, for the same call chain
[04:37] tristanz has joined the channel
[04:37] Genius_: anyone in here have some extra time to take on a project for a startup that is about to be funded in the 7 figures?
[04:38] devaholic: killall -i
[04:38] st_luke has joined the channel
[04:39] r04r has joined the channel
[04:39] devaholic: Genius_, so you have no money?
[04:40] Genius_: its not about th emoneyh, its about the opportunity
[04:40] Genius_: the money*
[04:40] wingie: Genius_: what's the status of the startup .. you have devs?
[04:40] wingie: the app will be created from scratch?
[04:41] nuba has joined the channel
[04:42] orospakr has joined the channel
[04:42] jerkelens has joined the channel
[04:44] spolu has joined the channel
[04:44] r04r has joined the channel
[04:46] harthur has joined the channel
[04:46] pyrotechnick has left the channel
[04:47] pquerna has joined the channel
[04:47] wink___ has joined the channel
[04:50] r04r has joined the channel
[04:55] sreeix has joined the channel
[04:55] wink_ has joined the channel
[04:56] lperrin1 has joined the channel
[04:56] OmidRaha has joined the channel
[04:57] r04r has joined the channel
[05:00] jergason has joined the channel
[05:00] wink___ has joined the channel
[05:03] r04r has joined the channel
[05:05] wink_ has joined the channel
[05:05] diogogmt has joined the channel
[05:06] ckknight has joined the channel
[05:07] ckknight has joined the channel
[05:08] kvey has joined the channel
[05:08] skyler_brungardt has joined the channel
[05:08] merlin83 has joined the channel
[05:08] merlin83_ has joined the channel
[05:09] papertigers has joined the channel
[05:09] Emmanuel has joined the channel
[05:10] r04r has joined the channel
[05:10] papertigers: anyone using jade? whats the proper way to include an iframe such as a youtube video?  I dont see iframs in the docs
[05:12] wink___ has joined the channel
[05:13] chjj has joined the channel
[05:14] brainproxy: I've got an binary executable (on linux) which prints lots of stuff when I invoke it manually in my shell; but for some reason the usual trick of ( spawn... ).stdout.pipe(process.stdout) results in nothing showing up
[05:14] ningu: I thought iframes were strongly discouraged nowadays, but I dunno
[05:14] brainproxy: tried stderr too, but not difference
[05:15] r04r has joined the channel
[05:15] ningu: did you try nonstderr? :P
[05:16] brainproxy: :p
[05:17] brainproxy: well, i've tinkered with the options hash -- cwd, env -- just to see if that has an effect
[05:17] josh-k has joined the channel
[05:18] brainproxy: .. it didn't; so i guess my question is if anyone has any suggestions for figuring out just how it might be outputting text into my console if it's not coming over stdout or stderr
[05:18] catb0t:  Exception: SyntaxError: Unexpected identifier
[05:18] bartt1 has joined the channel
[05:20] ningu: brainproxy: hrm, could be a buffering issue, not sure
[05:20] r04r has joined the channel
[05:21] boltR has joined the channel
[05:21] norviller has joined the channel
[05:23] sreeix has joined the channel
[05:23] wdbl: is there a good way to tell if a given function is a class constructor?
[05:25] brainproxy: wdbl: generally speaking any function can be used as a contructor
[05:25] a_suenami has joined the channel
[05:26] r04r has joined the channel
[05:26] wdbl: yeah
[05:27] wdbl: what if I inspect the functions prototype and find stuff that wouldn't be there for a non-class function?
[05:28] wdbl: a normal function probably wouldn't have any prototype additions, but a class function most likely would. but...not always
[05:28] sdwrage has joined the channel
[05:28] leif has joined the channel
[05:28] SubStack: wdbl: why do you want to do that?
[05:28] ningu: wdbl: not sure that distinction really exists in js...
[05:28] ningu: in practice it sort of does.
[05:29] wdbl: I'm just exploring ways of modeling Js code and writing meta-data about it
[05:29] SubStack: wdbl: fun fact, if you use new on a function and it has a return value, the prototype gets discarded in favor of the return value
[05:29] russfrank has joined the channel
[05:30] wdbl: SubStack: hmmmm
[05:30] SubStack: BUT, it only works if the return value is a typeof 'object'
[05:30] SubStack: strange
[05:30] SubStack: .. function F () { return { a : 3 } } new F instanceof F
[05:30] catb0t:  false
[05:30] SubStack: .. function F () { } new F instanceof F
[05:30] catb0t:  true
[05:30] SubStack: dnode abuses this property for its constructor syntax
[05:31] SubStack: so you can assign into `this` or just return an object
[05:31] wdbl: well, I always use a capital letter for my constructor function names to indicate it's a class so I guess I could just require that or some sort of other comment-notation for my project to work.
[05:32] wdbl: conventions are good
[05:32] SubStack: what api are you trying to achieve?
[05:33] r04r has joined the channel
[05:33] wdbl: at the moment, I'm building a generator to spit out Intellisense meta-data for Visual Studio - https://github.com/waynebloss/vsNodeRef
[05:33] spion has joined the channel
[05:34] wdbl: I wrote all the m-data that's there so far by hand just to make sure the system works
[05:34] gavin_huang has joined the channel
[05:34] benvie: the ultra repl knows all about exploring prototype strees
[05:35] wdbl: ahh, yes ben thank you - I should have taken a closer look at that
[05:35] benvie: by default it shows [[prototos]]
[05:35] benvie: and has a toggle for hidden prototypes which prototypes usually are
[05:37] devaholic: benvie, do you know of any projects to inject a repl into deployed code?
[05:37] devaholic: modules rather
[05:37] SubStack: I've heard of that being done on applications
[05:37] SubStack: can't remember who did this though
[05:38] SubStack: wdbl: neat, inspecting the prototype will probably work well enough for that I would suppose
[05:38] r04r has joined the channel
[05:38] SubStack: at least for code written that way
[05:38] ryan1 has joined the channel
[05:39] benvie: I don't know of any no but the stuff I've done for my thing has all the tools for doing it
[05:39] benvie: I just need to set up the debugger protocol 
[05:39] devaholic: link?
[05:39] benvie: https://github.com/Benvie/Node.js-Ultra-REPL
[05:40] Genius_: so is benvie the authority on all things node.js?
[05:40] benvie: with debugger protocol it'll just be normal usage basically
[05:40] benvie: I'm planning on hooknig up esprima to more stuff
[05:40] benvie: and using that to do the whole function rewriting for entrance logging and whatnot, or maybe just Proxies
[05:41] SubStack: wdbl: you might also find this useful http://github.com/substack/node-burrito
[05:41] SubStack: makes it really easy to walk the AST
[05:41] devaholic:  benvie i think i saw when you first released this actually
[05:42] SubStack: so you could for instance grab all the vars in a particular block
[05:42] SubStack: or what-have-you
[05:42] markdaws has joined the channel
[05:43] brianseeders has joined the channel
[05:43] wdbl: SubStack: beautiful, thank you very much
[05:43] benvie: yeah brendan eich tweeted about it the first day I put it up when it was like all broken and the result of being awake for 4 days straight
[05:44] devaholic: it would be nice if there was one place where you could connect a bunch of servers and somehow switch between them
[05:44] devaholic: not sure about the second part of that
[05:45] benvie: yeah as it currently is it's built around being able to switch between multiple contexts easily
[05:45] cjm has joined the channel
[05:45] benvie: and have them all going
[05:45] r04r has joined the channel
[05:45] benvie: so it's a matter of just having the debugger protocol hooked up to do that with external stuff
[05:46] devaholic: i think this would be the repl to use. watched
[05:47] devaholic: im about to give it a try :)
[05:47] benvie: I keep getting distracted trying to make gui stuff work on windows
[05:47] devaholic: make remote-repl
[05:47] devaholic: you would win the game
[05:47] benvie: but I have a lot of thigns I want to do with the repl. It's a great platform to build debug/development stuff on
[05:48] benvie: yeah
[05:48] benvie: also it'd be awesome because it'd work with chromium too
[05:48] benvie: which would be quite awesome
[05:48] benvie: also ipad/iphone I think?
[05:49] devaholic: awh, sad
[05:49] pixel13 has joined the channel
[05:49] devaholic: colors are borked in konsole
[05:49] benvie: hmmm
[05:50] benvie: I've used it on windows and ubuntu
[05:50] Broolucks: fyi I'm making a browser-based terminal emulator with inline html
[05:50] Broolucks: if that can be useful
[05:50] r04r has joined the channel
[05:50] devaholic: http://konsole.kde.org/
[05:51] benvie: I had it half hooked up to tty.js
[05:51] Broolucks: too bad setWindowSize got broken
[05:51] benvie: but tty.js was missing a number of the escape sequences that I use and I didn't get it fixed up yet
[05:51] benvie: but the ability to hook up remotely to the repl is almost good to go
[05:51] benvie: just buggy
[05:52] benvie: yeah for that stuff it's going to involve implementing more of the protocol
[05:52] Broolucks: https://github.com/breuleux/terminus
[05:52] benvie: for negotating screen sizes and whatnot
[05:52] Broolucks: maybe I should make screenshots at some point
[05:53] benvie: but ultrarepl reimplements node's repl and readline entirely do it's not limited by whatever they decide to read
[05:53] take_cheeze has joined the channel
[05:55] redir has joined the channel
[05:55] CrypticSwarm has joined the channel
[05:56] yoy_xus has joined the channel
[05:56] dshaw_ has joined the channel
[05:56] YoY has joined the channel
[05:57] r04r has joined the channel
[06:00] wink_ has joined the channel
[06:03] r04r has joined the channel
[06:05] lhoffman has joined the channel
[06:05] wink___ has joined the channel
[06:06] k1ttty has joined the channel
[06:08] boehm has joined the channel
[06:09] r04r has joined the channel
[06:10] Tim has joined the channel
[06:11] looopy has joined the channel
[06:11] lee2012 has left the channel
[06:15] caolanm_ has joined the channel
[06:15] r04r has joined the channel
[06:19] wink_ has joined the channel
[06:21] objectiveous has joined the channel
[06:21] looopy has joined the channel
[06:21] copongcopong1 has joined the channel
[06:21] r04r has joined the channel
[06:22] deeprogram has joined the channel
[06:22] looopy has joined the channel
[06:26] r04r has joined the channel
[06:29] torvalamo has joined the channel
[06:30] TimTimTim has joined the channel
[06:33] r04r has joined the channel
[06:33] r04r has joined the channel
[06:35] odie5533 has joined the channel
[06:36] wink___ has joined the channel
[06:37] looopy has joined the channel
[06:38] EvRide has joined the channel
[06:39] r04r has joined the channel
[06:40] devaholic: benvie, how did i miss this http://nodejs.org/docs/latest/api/all.html#repl.start
[06:40] benvie: oh hah
[06:41] devaholic: like 2 scrolls down even
[06:41] Broolucks: anyone knows of a good js plotting library that's interactive and svg-based?
[06:42] devaholic: d3.js
[06:42] boltR has joined the channel
[06:43] dnjaramba has joined the channel
[06:44] r04r has joined the channel
[06:44] devaholic: hmm this is interesting
[06:46] chjj has joined the channel
[06:47] Broolucks: d3 seems nice, though I'd basically have to implement the plotting library myself
[06:49] r04r has joined the channel
[06:51] sreeix has joined the channel
[06:51] Nivka has joined the channel
[06:52] gavin_hu_ has joined the channel
[06:52] cjm has left the channel
[06:53] jiboumans_ has joined the channel
[06:53] cjm has joined the channel
[06:54] r04r has joined the channel
[06:55] dnjaramba has joined the channel
[06:56] disappearedng: So I have a rails webserver rendering html pages that connects to my nodejs server. I somehow need nodejs to communicate with my rails server that a particular action whether a particular action has started or not
[06:57] objectiveous has joined the channel
[06:59] r04r has joined the channel
[07:01] LarsSmit has joined the channel
[07:01] jergason has joined the channel
[07:02] zomgbie has joined the channel
[07:04] r04r has joined the channel
[07:04] r04r has joined the channel
[07:05] dnjaramba has joined the channel
[07:08] ohtogo has joined the channel
[07:08] r04r has joined the channel
[07:08] stagas has joined the channel
[07:09] `3rdEden has joined the channel
[07:11] disappearedng: is there a good way to get my nodejs to be able to talk to the db
[07:11] disappearedng: I mean the rails server, 1 obvious way is to share the db 
[07:12] Lingus: RPC or MQ are common
[07:12] wink_ has joined the channel
[07:13] Lingus: Alternatively, don't.
[07:13] r04r has joined the channel
[07:14] Joeysomo has joined the channel
[07:14] dnjaramba has joined the channel
[07:15] Joeysomo has joined the channel
[07:16] EhevuTov has joined the channel
[07:17] r04r has joined the channel
[07:18] objectiveous has joined the channel
[07:19] tonist has joined the channel
[07:20] dnjaramba has joined the channel
[07:22] r04r has joined the channel
[07:23] sdwrage has left the channel
[07:24] dnjaramba_ has joined the channel
[07:26] wink_ has joined the channel
[07:27] r04r has joined the channel
[07:31] dnjaramba has joined the channel
[07:31] wink___ has joined the channel
[07:31] r04r has joined the channel
[07:34] Skola has joined the channel
[07:35] dnjaramba_ has joined the channel
[07:36] r04r has joined the channel
[07:36] SubStack: aha, (req, res, rec)
[07:36] moreno has joined the channel
[07:36] SubStack: the rec is for record!
[07:37] simenbrekken has joined the channel
[07:39] wink_ has joined the channel
[07:39] dnjaramba has joined the channel
[07:40] harthur has joined the channel
[07:41] r04r has joined the channel
[07:42] take_cheeze has joined the channel
[07:42] qsobad has joined the channel
[07:43] vdemedes has joined the channel
[07:44] vdemedes: Hello!
[07:44] zomg has joined the channel
[07:45] doffm has joined the channel
[07:45] r04r has joined the channel
[07:47] patcito has joined the channel
[07:48] SQLServerIO: is there a way to over ride a function in the connect/express framework? like res.sendfile?
[07:48] SQLServerIO: proxy it maybe... hrum.
[07:49] maxogden: SQLServerIO: why would you want to monkeypatch like that
[07:50] r04r has joined the channel
[07:50] SQLServerIO: maxogden: what else would I do? I need to override some of the functionality or add to it in some way. 
[07:51] maxogden: a) fork the library and send a pull request b) use another module such as filed to read files and pipe() them to the response object
[07:52] wink_ has joined the channel
[07:52] mappum: how can I make a node vm script have a timeout?
[07:52] SQLServerIO: Or build a module. I have to do some fun stuff and piping to res isn't working due to a couple of things that res.send is doing. It wouldn't be useful for most folks so I doubt it would be merged in.
[07:54] maxogden: its important to not break streaming
[07:54] maxogden: imo
[07:55] r04r has joined the channel
[07:56] SQLServerIO: I agree, I'm trying not to, but I'm at the point where I ether monkeypatch express or fork it and handle merges and the like. I don't have the time to write a framework from scratch.
[07:56] margle has joined the channel
[07:56] Morkel has joined the channel
[07:57] wink___ has joined the channel
[07:57] westg has joined the channel
[07:57] erikzaadi has joined the channel
[07:59] r04r has joined the channel
[08:00] dnjaramba has joined the channel
[08:01] aliem has joined the channel
[08:04] r04r has joined the channel
[08:08] asimo has joined the channel
[08:08] wink_ has joined the channel
[08:08] dreamdust has joined the channel
[08:09] `3rdEden has joined the channel
[08:09] Skola has joined the channel
[08:09] k1ttty has joined the channel
[08:09] isaacs has joined the channel
[08:11] dnjaramba has joined the channel
[08:12] r04r has joined the channel
[08:12] r04r has joined the channel
[08:12] Wizek has joined the channel
[08:13] kirbysayshi has joined the channel
[08:15] wink_ has joined the channel
[08:16] dnjaramba has joined the channel
[08:16] lzskiss has joined the channel
[08:17] r04r has joined the channel
[08:18] lzskiss has joined the channel
[08:20] illourr: gnight node
[08:20] markq has joined the channel
[08:20] wink___ has joined the channel
[08:21] r04r has joined the channel
[08:21] dnjaramba has joined the channel
[08:23] westg has joined the channel
[08:23] lzskiss has joined the channel
[08:25] k1ttty has joined the channel
[08:25] wink_ has joined the channel
[08:26] diverdude: Is it possible to share variables between connected sockets in node.js?
[08:26] r04r has joined the channel
[08:27] tonist has joined the channel
[08:27] SubStack: diverdude: I wrote https://github.com/substack/node-replicant to do that
[08:27] SubStack: or a lot of people use redis
[08:28] benvie: you sure can'
[08:28] benvie: can't do it with v8 handles though, not without wrapping them or something
[08:29] mappum: What does an "oob" error mean when reading from Buffers?
[08:29] dreamdust has joined the channel
[08:29] benvie: almost boob
[08:29] mappum: :P
[08:29] benvie: out of bounds
[08:29] diverdude: i am making a chat and i need to somehow keep an array of connected sockets so that when one client sends a message i can forward it to another
[08:29] benvie: read beyond buffer size
[08:29] mappum: but it usually gives me a more specific message
[08:29] benvie: buffers are allocated with a specific size and can't increase it
[08:30] mappum: i know
[08:30] mappum: oh, it must be on my slice
[08:30] benvie: ah yeah
[08:30] LordTrev has joined the channel
[08:30] r04r has joined the channel
[08:30] LordTrev: 
VERSION

[08:32] Druid_ has joined the channel
[08:33] k1ttty has joined the channel
[08:33] diverdude: SubStack: is your node-repliant for forwarding messages between sockets?
[08:34] SubStack: devaholic: it's for doing synchronization of objects
[08:34] SubStack: diverdude: what are you trying to build?
[08:35] diverdude: SubStack: its a chat system where 2 users can chat privately
[08:35] r04r has joined the channel
[08:35] wink___ has joined the channel
[08:36] SubStack: diverdude: there are tons of libraries to make that easy
[08:36] SubStack: socket.io is pretty much built around that exact use case
[08:36] bradhe has joined the channel
[08:36] diverdude: SubStack: i tried creating a global array where each users socket is saved on connect...but this variable does not seem to be shared 
[08:36] diverdude: SubStack: i am using socket.io already
[08:37] SubStack: don't use globals
[08:37] SubStack: ever
[08:37] SubStack: unless you mean a module-level variable which is fine
[08:37] diverdude: SubStack: how can i then convey a message from one socket to another?
[08:37] skyler_brungardt has joined the channel
[08:38] diverdude: SubStack: broadcast is no good because it sends to all connected sockets
[08:38] SubStack: diverdude: with programming?
[08:38] SubStack: don't use broadcast
[08:38] diverdude: thats what im saying
[08:39] SubStack: just write to the socket that matters
[08:39] diverdude: SubStack: sure, but how can i access that socket?
[08:39] SubStack: and listen from the socket that matters
[08:39] phidah has joined the channel
[08:40] r04r has joined the channel
[08:40] r04r has joined the channel
[08:40] diverdude: SubStack: this seems to now work: http://bin.cakephp.org/view/61110497
[08:40] diverdude: now=not
[08:40] dshaw_ has joined the channel
[08:41] k1ttty has joined the channel
[08:44] SubStack: diverdude: why all the handshake userdata noise?
[08:44] r04r has joined the channel
[08:45] SubStack: var id = Math.random().toString(16).slice(2); // or whatever
[08:45] SubStack: then clients[id] = socket
[08:45] michaelhartau has joined the channel
[08:46] mac^ has joined the channel
[08:46] SubStack: then emit('chatmessage', { from : id, to : msg.to, message : msg.message })
[08:46] stef has joined the channel
[08:47] Dotan has joined the channel
[08:47] SubStack: diverdude: where did you get that handshake stuff from, I don't see that documented
[08:48] ph^ has joined the channel
[08:49] r04r has joined the channel
[08:49] marlun has joined the channel
[08:49] darkixion has joined the channel
[08:49] elnn has joined the channel
[08:50] diverdude: SubStack: i have a routine which takes care of that: http://bin.cakephp.org/view/493728144
[08:50] JakeSays has joined the channel
[08:51] wink_ has joined the channel
[08:51] arduix has joined the channel
[08:51] dreamdust has joined the channel
[08:52] mappum: how can i make node deallocate buffers?
[08:53] SubStack: diverdude: so why didn't you post that originally >_<
[08:53] Illourr| has joined the channel
[08:53] Qzen: do you guys know a simple chat that i can use for my website?
[08:53] SubStack: going to bet that the error is in that ugly thing somewhere
[08:53] r04r has joined the channel
[08:53] r04r has joined the channel
[08:53] Qzen: i am trying to implement only open-source scripts
[08:54] Qzen: and therefor i dont feel like doing it myself :P
[08:54] diverdude: SubStack: because the relevant part is how i connect 2 clients
[08:54] padd173 has joined the channel
[08:55] Qzen: well the clients connect to the server and the server just seperates those connections with some nodejs magic
[08:55] Qzen: its simple
[08:57] SubStack: diverdude: why not generate a unique id and then attach data onto that?
[08:57] diverdude: Qzen: https://github.com/Fabryz/support-chat
[08:57] padd173: Is there a db driver that does connection pooling?
[08:57] SubStack: var users = {}; then users[id] = { socket : sock, name : ... }
[08:57] diverdude: SubStack: thats actually what im doing
[08:57] Dotan__ has joined the channel
[08:58] SubStack: except you're doing it really confusingly by putting everything on the sock.handshake
[08:58] Illourr| has joined the channel
[08:58] stagas: diverdude: no need for channels if you only pair 2, do it like this: socket_a.partner = socket_b; socket_b.partner = socket_a;
[08:58] diverdude: SubStack: im putting validation on the handshake
[08:58] r04r has joined the channel
[08:59] stagas: the on message, socket.partner.send('blah')
[08:59] stagas: then
[08:59] SubStack: diverdude: well it doesn't work so one or more of your assumptions must be wrong
[09:00] diverdude: SubStack: hehe that i cannot argue ;)
[09:00] SubStack: anyways, don't put data on the handshake that's ugly
[09:00] SubStack: just var users = {}
[09:00] SubStack: then it's easier to inspect
[09:00] SubStack: and less ridiculous
[09:00] jomoho has joined the channel
[09:01] raja: In a for loop in node, what is the equiv of continue
[09:01] raja: continue seems to break
[09:01] rendar has joined the channel
[09:02] konobi: raja: let me google that for you...
[09:02] SubStack: raja: node is javascript
[09:02] wink_ has joined the channel
[09:02] SubStack: full stop
[09:02] SubStack: nothing magic going on
[09:03] raja: yep i know.. 
[09:03] adambeynon has joined the channel
[09:03] raja: i was originally iterating with forEach but that is functional and doesn't allow you to break or continue
[09:03] raja: so i changed to a for loop
[09:03] Illourr| has joined the channel
[09:03] SubStack: forEach doesn't break or continue
[09:03] raja: and now break works but continue works like a break
[09:03] SubStack: use a different combinator
[09:03] raja: SubStack: thats what i wrote
[09:03] r04r has joined the channel
[09:03] r04r has joined the channel
[09:04] SubStack: raja: paste your code
[09:04] SubStack: I hate speculating
[09:05] raja: https://gist.github.com/45c7086a6e0fd315754f
[09:06] raja: i'll simplify the example too
[09:06] raja: so the context doesn't matter
[09:06] huggies has joined the channel
[09:07] SubStack: don't do that
[09:07] SubStack: I hate it when people take out code
[09:07] SubStack: because the part that they took out is overwhelmingly where the bug is
[09:07] raja: true
[09:08] Edy has joined the channel
[09:08] Edy has joined the channel
[09:08] raja: standard for loop works fun and continues fine.. just iterating over numbers
[09:08] r04r has joined the channel
[09:08] Illourr| has joined the channel
[09:08] raja: so not sure why this doesn't work
[09:09] SubStack: raja: are you sure that tweet.match(...) returns a real array?
[09:09] SubStack: match objects are strange
[09:09] konobi: your match will always be 0
[09:09] konobi: (you're not capturing)
[09:09] SubStack: .. 'beep boop'.match(new RegExp('(\\S+)', 'g'))
[09:09] catb0t:  [ "beep", "boop" ]
[09:10] Dotan has joined the channel
[09:10] raja: SubStack: yeah it returns a real array
[09:10] markq has joined the channel
[09:10] kwy__ has joined the channel
[09:11] raja: konobi: capturing where?
[09:11] SubStack: just console.dir(negMatch)
[09:11] Lorentz has joined the channel
[09:11] SubStack: I would do that before I conclude that continue is broken
[09:11] raja: negMatch.length is not always zero..
[09:12] vincentcr has joined the channel
[09:12] r04r has joined the channel
[09:13] Skola has joined the channel
[09:13] chilts: raja: as an aside, all of your 'var' definitions in the loop are hoisted to the top of the function ... not that it matters in this case, but something to be aware of
[09:13] chilts: I never like it catching me out :)
[09:13] westg has joined the channel
[09:13] Illourr| has joined the channel
[09:14] niallo has joined the channel
[09:14] raja: console.dir inspects array and stdio's it fine
[09:15] raja: chilts: thanks, will fix
[09:15] qsobad has joined the channel
[09:16] Dotan__ has joined the channel
[09:16] megalomix has joined the channel
[09:16] megalomix: hello
[09:17] megalomix: i added PATH=/home/node/0.6.11/bin:$PATH in .bashrc of my home (home/vhosts/user1/.bashrc)
[09:17] dlg: thats a great story
[09:17] megalomix: but if i write node it does not start
[09:17] r04r has joined the channel
[09:17] dlg: whats it look like if you go echo $PATH?
[09:17] megalomix: i need to go to /home/node/0.6.11/bin
[09:18] konobi: `hash -r`
[09:18] megalomix: and write ./node
[09:18] dlg: maybe you need export PATH after you set it in .bashrc
[09:18] avih has joined the channel
[09:18] dlg: also, you might have to have something like this in .bash_profile:
[09:18] megalomix: $ echo $PATH
[09:18] megalomix: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
[09:18] dlg: if [ -f ~/.bashrc ]; then
[09:18] dlg: 	source ~/.bashrc
[09:18] dlg: fi
[09:19] raincole has joined the channel
[09:19] methodt: i second: source ~/.bashrc
[09:19] megalomix: yes i have .profile
[09:19] Dotan_ has joined the channel
[09:19] dlg: does it source the rc?
[09:20] megalomix: uh? i have PATH=/home/node/0.6.11/bin:$PATH in .bashrc
[09:20] megalomix: at the end
[09:20] herbySk has joined the channel
[09:20] dlg: thats not what i asked
[09:21] megalomix: what did you ask?
[09:21] markq has joined the channel
[09:22] r04r has joined the channel
[09:22] r04r has joined the channel
[09:22] dnjaramba has joined the channel
[09:23] megalomix: dlg, this is my .profile https://gist.github.com/1862722
[09:23] stayarrr has joined the channel
[09:23] Illourr| has joined the channel
[09:24] sigurding has joined the channel
[09:25] megalomix: dlg, still there ? :)
[09:26] insin has joined the channel
[09:27] r04r has joined the channel
[09:28] Illourr| has joined the channel
[09:29] RLa has joined the channel
[09:29] aranw has joined the channel
[09:30] aranw has joined the channel
[09:30] megalomix: methodt, you?
[09:31] methodt: megalomix, yeah I was just asking if you sourced it after you added that line
[09:31] methodt: or close the terminal
[09:31] methodt: and reopen it
[09:31] megalomix: wait methodt, excuse me what to you mean with "sourced" ?
[09:31] megalomix: yes
[09:31] megalomix: i done it
[09:31] TheFuzzball has joined the channel
[09:31] andrew12 has joined the channel
[09:31] r04r has joined the channel
[09:31] methodt: sourced meaning source .bashrc
[09:31] megalomix: ok
[09:32] methodt: well hmm
[09:32] megalomix: yes i add the line there....saved
[09:32] megalomix: close and reopen the shell
[09:32] megalomix: (i connect to this VPS) with ssh
[09:32] megalomix: closed and reopened
[09:32] megalomix: but, why if i do echo $PATH
[09:32] megalomix: i don't see it ?
[09:33] megalomix: $ echo $PATH
[09:33] megalomix: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
[09:33] methodt: try 'export PATH=$PATH:/home/node/0.6.11/bin' 
[09:34] Illourr| has joined the channel
[09:34] megalomix: i try
[09:34] chakrit has joined the channel
[09:34] megalomix: $ echo $PATH
[09:34] megalomix: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/home/node/0.6.11/bin
[09:35] megalomix: hmmmmm now i have it :-/ why ?
[09:35] megalomix: why .bashrc is not enough to do it ?
[09:35] methodt: bc you need to add 'export' to your .bashrc line
[09:36] megalomix: methodt, i use: PATH=/home/node/0.6.11/bin:$PATH   also in /root/.bashrc
[09:36] dnjaramba has joined the channel
[09:36] megalomix: and it works
[09:36] r04r has joined the channel
[09:37] Industrial has joined the channel
[09:37] megalomix: methodt, so, is this wrong?
[09:38] megalomix: or better do i have to use it in /root/.bashrc but for the users i need to export it ?
[09:39] Illourr| has joined the channel
[09:39] methodt: export takes a local variable and makes it global
[09:39] methodt: so yes
[09:39] methodt: you dont need it in root
[09:39] methodt: but you need 'export' for users
[09:41] r04r has joined the channel
[09:41] dnjaramba has joined the channel
[09:42] megalomix: methodt, do i have to add the same line in .bashrc ?
[09:42] megalomix: i mean: export PATH=$PATH:/home/node/0.6.11/bin
[09:43] megalomix: exactly this line ?
[09:43] methodt: megalomix, if you have it in root
[09:43] methodt: then no
[09:43] megalomix: aaah ok
[09:43] megalomix: do you mean that i have to add it ONE TIME In root
[09:43] megalomix: and stop ?
[09:44] megalomix: so i have to change that line in my /root/.bashrc
[09:44] methodt: .bashrc is run everytime you open a terminal, so you leave it there
[09:44] methodt: you can have it in root or user, it's your choice
[09:45] megalomix: methodt, yes but what i mean is that NOW i have PATH=/home/node/0.6.11/bin:$PATH  in /root/.bashrc
[09:46] megalomix: to leave the user use node everywhere
[09:46] megalomix: i need to change it
[09:46] r04r has joined the channel
[09:46] r04r has joined the channel
[09:46] megalomix: and add export to it
[09:46] megalomix: so ONE TIME In /root/.bashrc
[09:46] megalomix: i login with users in the VPS
[09:47] Vennril has joined the channel
[09:47] methodt: You root will apply to all users, if you have it in user.. it will only apply to that user. You don't need it in both .bashrc files.
[09:47] methodt: Your root .bashrc file will apply to all users*
[09:47] konobi: o.O
[09:49] megalomix: yes i have understand :D ahaha i mean I DON'T use it in user/.bashrc but ONLY In /root/.bashrc
[09:49] Illourr| has joined the channel
[09:49] megalomix: the thing i don't understand is
[09:49] megalomix: what line do i have to add to /root/.bashrc
[09:49] megalomix: because only PATH=/home/node/0.6.11/bin:$PATH does't works for users
[09:49] megalomix: so do i need to add export there
[09:49] megalomix: right?
[09:49] methodt: correct
[09:50] megalomix: ook
[09:50] konobi: methodt: "root" is just another user... it doesn't get sourced for everyone
[09:50] megalomix: hmm
[09:50] megalomix: konobi, so what can i do?
[09:51] r04r has joined the channel
[09:51] konobi: megalomix: try reading up on bash and profile scripts in general
[09:52] jblanche has joined the channel
[09:52] dreamdust has joined the channel
[09:52] megalomix: konobi, something more specific? :)
[09:53] EhevuTov: cat /etc/profile; cat /etc/bashrc
[09:53] megalomix: EhevuTov, ?
[09:53] megalomix: i have PATH=/home/node/0.6.11/bin:$PATH in .bashrc
[09:53] EhevuTov: were you looking for system wide sh settings?
[09:54] megalomix: i have the same in /root/.bashrc
[09:54] EhevuTov: megalomix, are you trying to add paths globally?
[09:55] megalomix: EhevuTov, exactly
[09:55] daglees has joined the channel
[09:55] daglees has joined the channel
[09:55] megalomix: i would like to add it ONE time and all the users can launch node everywhere
[09:55] EhevuTov: I find that odd, but if you do, you'd modify /etc most likely. Check out the /etc/paths file
[09:55] dnjaramba has joined the channel
[09:55] xSmurf has joined the channel
[09:55] megalomix: why odd ?
[09:55] r04r has joined the channel
[09:55] EhevuTov: megalomix, it's not managed
[09:56] EhevuTov: you should be adding node via package or something
[09:56] raja: SubStack, konobi: Found my problem.. the match length was always >= 1 because the regex was often matching on whitespace 
[09:56] EhevuTov: this allows for better maintainability, particularly since Node is in such heavy development
[09:56] raja: thanks for the help
[09:56] EhevuTov: megalomix, you don't want to share files from your /root
[09:57] megalomix: EhevuTov, hmmm i don't getting you, if i change the node version i will modify to the new path, what is odd? i mean....if i donwload the last stable i would like to use it everywhere
[09:57] joshontheweb has joined the channel
[09:58] megalomix: EhevuTov, i don't have /etc/paths
[09:58] EhevuTov: megalomix, node is a binary similar to other scripting languages. WHere is your perl, python, ruby bins?
[09:58] megalomix: i'm using ubuntu 11.10
[09:58] EhevuTov: you should be putting it in something like /usr/local/bin
[09:58] megalomix: /usr/local/bin
[09:58] lmorchard has joined the channel
[09:58] megalomix: aaah
[09:58] megalomix: understand
[09:58] megalomix: ok
[09:58] megalomix: so wait i can add a sym link to node
[09:58] megalomix: right ?
[09:59] EhevuTov: and with ubuntu, there is probably a package for node. Just install it that way. You're wanting to use Node in a normal way, right?
[10:00] EhevuTov: megalomix, you shouldn't need a symlink. Your system should already check the /usr/local/bin as it's most likely already in the systemwide PATHS environment var
[10:00] megalomix: with "package" do you mean install it using apt-get ? i'm installing it from source but i use --prefix= during the configuration
[10:00] r04r has joined the channel
[10:00] EhevuTov: yes, but why are you modifying the default with the --prefix= ? What is your reasoning?
[10:01] dnjaramba has joined the channel
[10:01] megalomix: because i have a directory with all the node versions inside
[10:01] megalomix: so i switch to versions quickly
[10:01] EhevuTov: megalomix, so you're trying to create a multiversioned environment
[10:01] megalomix: *i change version if i want to go back and not using the last
[10:01] megalomix: exactly
[10:02] megalomix: now i have:   /home/node/0.6.*
[10:02] megalomix: inside node i have all the versions
[10:02] methodt has joined the channel
[10:02] megalomix: *inside /home/node
[10:03] trupppOFF has joined the channel
[10:03] plutoniiix has joined the channel
[10:03] [[zzz]] has joined the channel
[10:04] megalomix: EhevuTov, what do you think creating a sym link is odd ? i think is very easy to do and i can switch the version (change the dir) easyly
[10:05] r04r has joined the channel
[10:05] r04r has joined the channel
[10:05] EhevuTov: megalomix, I don't know of the symlink will resolve. you could try it
[10:05] megalomix: EhevuTov, i can add a link from /usr/bin/node to /home7node/0.6.11/bin/node
[10:06] megalomix: no?
[10:06] EhevuTov: right
[10:06] EhevuTov: megalomix, you can do that
[10:06] megalomix: and when i will change it to 0.6.12 i will change that sym link
[10:06] megalomix: without doing nothing else
[10:06] dreamdust has joined the channel
[10:07] megalomix: but i would like to undetsand why this is not a good solution for you
[10:07] EhevuTov: right, but you'll have to change your global paths for every /home/node/<version>/bin
[10:07] megalomix: global paths?
[10:07] megalomix: why?
[10:07] megalomix: i add the link inside /usr/bin
[10:07] EhevuTov: so that your users can access the previous versions
[10:07] megalomix: it will be accessible everywhere as the other application on the system no?
[10:07] megalomix: aah
[10:08] megalomix: yes i need to switch it if i nmeed to change
[10:08] EhevuTov: you said earlier you wanted your users to have access to previous version
[10:08] megalomix: right
[10:08] megalomix: yes
[10:08] megalomix: but if i want to switch to a previous version i iwll change the sym
[10:08] megalomix: if i want to use 0.6.10 i will use that version for ALL
[10:08] megalomix: the user
[10:08] megalomix: *users
[10:09] EhevuTov: Are you running some type of testing system for a lot of users?
[10:09] megalomix: i have 10 users
[10:09] EhevuTov: ok
[10:09] megalomix: yes i would like to test it separately
[10:10] megalomix: EhevuTov, maybe i need to add --> n
[10:10] megalomix: i have seen that is a tool to manage node versions
[10:10] megalomix: *i need to install
[10:10] EhevuTov: megalomix, I'm not familiar with that
[10:10] bradleyg has joined the channel
[10:10] r04r has joined the channel
[10:11] EhevuTov: megalomix, have you looked into something like this: https://github.com/creationix/nvm
[10:11] megalomix: yes exactly n is something line nvm and nave
[10:11] megalomix: created by TJ Holowaychuk
[10:11] EhevuTov: yeah, check that out. That's beyond me though. i wish you the best. I have to sleep :-)
[10:12] tjmehta has left the channel
[10:12] megalomix: :)
[10:12] megalomix: gnight
[10:12] megalomix: bye
[10:12] megalomix: thank you!
[10:12] EhevuTov: you're welcome
[10:14] Illourr| has joined the channel
[10:14] spolu has joined the channel
[10:15] r04r has joined the channel
[10:15] r04r has joined the channel
[10:15] lmorchard has joined the channel
[10:17] johnhamelink has joined the channel
[10:17] SvenDowideit has joined the channel
[10:18] zomgbie has joined the channel
[10:19] Danielss89 has joined the channel
[10:19] r04r has joined the channel
[10:20] johnnychimpo has joined the channel
[10:20] thalll has joined the channel
[10:21] wink___ has joined the channel
[10:24] r04r has joined the channel
[10:24] Illourr| has joined the channel
[10:24] d0k has joined the channel
[10:27] wink_ has joined the channel
[10:29] r04r has joined the channel
[10:29] Illourr| has joined the channel
[10:30] `3rdEden has joined the channel
[10:30] spolu: AvianFlu jesusabdullah mmalecki[zzz]: hmmm you know it's winter right? http://nodejitsu.com/
[10:31] wink_ has joined the channel
[10:32] __doc__ has joined the channel
[10:33] megalomix: guys i'm using nginx as proxy of node
[10:33] joshgillies has joined the channel
[10:33] megalomix: i have seen on the log of express.js that the ip is ALWAYS 127.0.0.1
[10:33] megalomix: why?
[10:33] simenbrekken has joined the channel
[10:33] r04r has joined the channel
[10:33] r04r has joined the channel
[10:34] Illourr| has joined the channel
[10:36] JakeSays has joined the channel
[10:37] wink___ has joined the channel
[10:37] shaysm has joined the channel
[10:37] dnjaramba has joined the channel
[10:38] r04r has joined the channel
[10:39] kwmiebach has joined the channel
[10:39] Illourr| has joined the channel
[10:40] xSmurf has joined the channel
[10:41] robotmay has joined the channel
[10:41] dnjaramba has joined the channel
[10:42] hkjels has joined the channel
[10:42] mansoor has joined the channel
[10:42] r04r has joined the channel
[10:45] Illourr| has joined the channel
[10:45] wink_ has joined the channel
[10:46] wereHamster: megalomix: how are you accessing the server?
[10:46] wereHamster: megalomix: ah, nginx. that explains it
[10:46] eddyb has joined the channel
[10:47] dnjaramba has joined the channel
[10:47] r04r has joined the channel
[10:48] eddyb: I need some JS trickery and I'm not sure how to do it, or if there's a complete solution out there
[10:49] eddyb: (I like the topic, btw)
[10:49] eddyb: so, I have a JSON file with some data in it
[10:49] eddyb: I want to read that JSON file into a JS object
[10:49] AndreasMadsen has joined the channel
[10:50] Illourr| has joined the channel
[10:50] eddyb: and every time something changes in that object, I want to convert it to JSON and save it into the file
[10:50] RLa has joined the channel
[10:50] eddyb: I know that proxies might help me, but I think this trick is used in some other places and there could be an easier way of doing it
[10:51] eddyb: *question mark*
[10:52] boltR: ls
[10:52] boltR: oops
[10:52] r04r has joined the channel
[10:52] mAritz has joined the channel
[10:52] chjj has joined the channel
[10:55] dnjaramba has joined the channel
[10:55] Illourr| has joined the channel
[10:55] wereHamster: read the file contents, use JSON.parse() to convert it to a JS object, and JSON.stringify() to convert it to a string so you can write it into the file
[10:55] zomgbie has joined the channel
[10:55] eddyb: wereHamster: except
[10:56] eddyb: json.someProperty = 5; won't do anything
[10:56] eddyb: I want save-on-change
[10:56] r04r has joined the channel
[10:57] wereHamster: make a api around it. smth.set('someProperty', 1)
[10:57] eddyb: I know what I'm saying is possible
[10:58] eddyb: lol "I'll create a GUI in Visual Basic to track his IP address" - I've read about that line in CSI:NY, now I get to watch it
[10:59] SubStack: are there any non-ridiculous cookie modules on npm?
[10:59] wink___ has joined the channel
[10:59] SubStack: because all the ones I can find either have no documentation or monkey punch ServerRequest or have terrible needlessly indirect APIs
[11:00] Illourr| has joined the channel
[11:00] snuggl: eddyb: __defineGetter/Setter__ can do that if you encapsulate it 
[11:00] eddyb: what about creating new properties?
[11:00] danielss89_ has joined the channel
[11:01] stonebranch has joined the channel
[11:01] SubStack: I just want a thing that parses and generates cookie values as a pure function with parse :: string -> object and deparse :: object -> string
[11:01] snuggl: eddyb: but tbh, do the re-save to JSON when the JSON is used instead
[11:01] wereHamster: SubStack: the one from connect
[11:02] SubStack: wereHamster: is that seperable from connect?
[11:02] eddyb: SubStack: Haskell much?
[11:02] r04r has joined the channel
[11:02] wereHamster: SubStack: no, but you can easily extract it
[11:03] QaDeS has joined the channel
[11:03] eddyb: snuggl: not sure if I follow
[11:04] wink_ has joined the channel
[11:04] eddyb: I want every change in that object to cause a save
[11:04] ph^ has joined the channel
[11:05] arduix has joined the channel
[11:06] konobi: eddyb: you could have a setter that marks an object internally as "dirty" and then have a .save() method that takes a callback
[11:07] ppcano has joined the channel
[11:07] r04r has joined the channel
[11:07] r04r has joined the channel
[11:07] ajackbot has joined the channel
[11:08] mAritz has joined the channel
[11:09] Sahua has joined the channel
[11:10] dnjaramba has joined the channel
[11:10] Illourr| has joined the channel
[11:12] r04r has joined the channel
[11:12] r04r has joined the channel
[11:13] dnjaramba has joined the channel
[11:13] snearch has joined the channel
[11:15] Illourr| has joined the channel
[11:16] arduix has joined the channel
[11:16] r04r has joined the channel
[11:16] wink___ has joined the channel
[11:20] kalms has joined the channel
[11:20] Illourr| has joined the channel
[11:21] r04r has joined the channel
[11:21] r04r has joined the channel
[11:21] wink_ has joined the channel
[11:22] mAritz has joined the channel
[11:23] benvie: don't use defineGetter/Setter
[11:23] benvie: use Object.defineProperty or use literal getters and setters
[11:23] jetienne has joined the channel
[11:23] eddyb: I think I'll just make a contraption with the Harmony proxies
[11:23] koxa has joined the channel
[11:24] benvie: .. { get prop(){ return 'p' }, set prop(v){} }
[11:24] catb0t:  Exception: SyntaxError: Unexpected identifier
[11:24] benvie: .. ({ get prop(){ return 'p' }, set prop(v){} })
[11:24] catb0t:  { prop: "p" }
[11:24] benvie: catb0t you resolved the getter
[11:24] catb0t: resolved deltas faster than my server clock out getter
[11:25] johnnus_ has joined the channel
[11:25] eddyb: lol, catb0t can speak
[11:25] catb0t: speak English
[11:25] benvie: lol
[11:25] Illourr| has joined the channel
[11:25] eddyb: catb0t: I do
[11:25] ullmark has joined the channel
[11:25] catb0t: do deploys to our administration site, but it's still not scriptkidproof because not all obj's in I
[11:26] r04r has joined the channel
[11:27] ullmark: hi guys, i upgraded node to 0.6.11.. isn't forever compatible with that version?
[11:28] k1ttty has joined the channel
[11:29] eddyb: .. Proxy.create
[11:29] catb0t:  Exception: ReferenceError: Proxy is not defined
[11:29] eddyb: it works with --harmony
[11:30] benvie: I has some fun with ecmabot and with(Proxy.create(...))
[11:30] Illourr| has joined the channel
[11:31] r04r has joined the channel
[11:31] r04r has joined the channel
[11:32] artusrocha has joined the channel
[11:32] jetienne: q. with requirejs, when i do define(['./relative/path/file.js'], function(){}), the relative path is relative to what ?
[11:34] eddyb: catb0t: true
[11:34] catb0t: true too
[11:34] eddyb: damn
[11:34] eddyb: catb0t -e true
[11:34] catb0t: true , I hope it can use the data structure simple
[11:34] eddyb: .. --harmony Proxy
[11:34] catb0t:  Exception: SyntaxError: Unexpected identifier
[11:34] vdemedes has joined the channel
[11:34] tdegrunt has joined the channel
[11:35] benvie: .. function catb0t(){}; catb0t.prototype = catb0t; new catb0t
[11:35] Illourr| has joined the channel
[11:35] catb0t:  Timeout Error
[11:36] Skola has joined the channel
[11:36] jldbasa has joined the channel
[11:36] r04r has joined the channel
[11:38] pthm has joined the channel
[11:38] wink___ has joined the channel
[11:38] benvie: .. (function(a,b){for(b=a='';a++<36;b+=a*51&52?(a^15?8^Math.random()*(a^20?16:4):4).toString(16):'-');return b})()
[11:38] catb0t:  "ee98c4ee-deb0-45ee-a617-9cd815642464"
[11:38] DrPheltRight has joined the channel
[11:38] sandfox has joined the channel
[11:39] whaley has joined the channel
[11:40] pthm has joined the channel
[11:40] i42n has joined the channel
[11:40] Skola_ has joined the channel
[11:40] Illourr| has joined the channel
[11:41] r04r has joined the channel
[11:41] r04r has joined the channel
[11:41] mAritz has joined the channel
[11:42] eddyb: benvie: 140byt.es UUID?
[11:42] megalomix: guys, i'm using nginx as proxy of node.js i don't undersatnd why the logger of Express.js save that all the request are coming from 127.0.0.1 the real user ip is not forwarded
[11:42] benvie: yeah
[11:43] benvie: a newish one
[11:43] megalomix: oh hello benvie!
[11:43] megalomix: :)
[11:44] benvie: hello =D
[11:45] megalomix: how are you?
[11:45] benvie: braindead
[11:45] megalomix: :)
[11:45] r04r has joined the channel
[11:45] r04r has joined the channel
[11:46] Illourr| has joined the channel
[11:46] sandfox: megalomix: is nginx set to added in the 'x-forwaded-for' header?
[11:46] arduix has joined the channel
[11:46] hipsterslapfight has joined the channel
[11:47] diverdude: sandfox: hey, what was it again you use socket.io for?
[11:47] bas_ has joined the channel
[11:47] dnjaramba has joined the channel
[11:48] pickels has joined the channel
[11:49] megalomix: sandfox, yes
[11:49] megalomix: sandfox, https://gist.github.com/1863407
[11:50] wink_ has joined the channel
[11:50] sandfox: bunch of stuff, (some of it completely pointless) like sharing lots of mouse cursors on single page, sending iPhone accelerometer data back so it updates a 3d renderer on clients web pages, also for logging data, etc etc
[11:50] r04r has joined the channel
[11:50] r04r has joined the channel
[11:51] megalomix: sandfox, are you talling it to me ? :)
[11:51] megalomix: is this wrong ?
[11:51] satyr has joined the channel
[11:51] sandfox: megalomix: are $remote_addr and $proxy_add_x_forwarded_for both empty?
[11:52] sandfox: megalomix: (sorry first message was diverdude:)
[11:53] megalomix: sandfox, how can i control it? becase as i told the request are passed to NODE.js node send back the responses
[11:53] diverdude: sandfox: ohh that sounds nice
[11:53] megalomix: but i always see 127.0.0.1 so it is using nginx "ip"
[11:53] diverdude: sandfox: you use it with nginx right?
[11:53] boccato has joined the channel
[11:54] sandfox: megalomix: let me dig out one of configs somewhere...
[11:54] megalomix: ok
[11:54] megalomix: sandfox, thanks
[11:54] adrianF has joined the channel
[11:54] sandfox: diverdude: using nginx to proxy any http requests but it doesn't handle the web sockets stuff
[11:54] r04r has joined the channel
[11:56] diverdude: sandfox: right ok...so you just connect directly to nodejs when communicating websockets?
[11:56] Illourr| has joined the channel
[11:56] konobi: http://blog.mixu.net/2011/08/13/nginx-websockets-ssl-and-socket-io-deployment/
[11:56] dingomanatee has joined the channel
[11:57] mehlah has joined the channel
[11:58] jinmo123 has joined the channel
[11:58] jinmo123: does anyone know why tcp socket closes after connect event?
[11:58] jinmo123: https://gist.github.com/850c6f257a91a1c39f4f
[11:58] sandfox: megalomix: that seems like it should be ok to me, had you tried putting "$remote_addr and $proxy_add_x_forwarded_for" in your nginx logs so you can see what nginx thinks it doing? and can you console.log() the request headers? 
[11:58] sandfox: diverdude: correct, nginx has no ability to support web sockets (yet) (patches are in the works)
[11:59] alFReD-NSH has joined the channel
[11:59] alFReD-NSH has left the channel
[11:59] diverdude: sandfox: yeah ok. I actually managed to route my websockets through HAProxy
[11:59] r04r has joined the channel
[11:59] r04r has joined the channel
[11:59] eddyb: I got something: https://gist.github.com/1863435
[12:00] quijote has joined the channel
[12:00] megalomix: sandfox, i should try it
[12:00] konobi: jinmo123: try an .on('data'... in there too
[12:00] megalomix: do you think those vars are empty ?
[12:00] jinmo123: yep
[12:01] Illourr| has joined the channel
[12:01] wink_ has joined the channel
[12:01] jinmo123: hmm
[12:01] jinmo123: nothing happens
[12:02] jinmo123: no data event
[12:02] megalomix: sandfox, hmm i don't undersatnd how can they be empty
[12:02] megalomix: it is strange
[12:03] blup has joined the channel
[12:04] rickibalboa has joined the channel
[12:04] Emmanuel__ has joined the channel
[12:04] dnjaramba has joined the channel
[12:04] sandfox: are they empty on both nginx logs and node?
[12:04] r04r has joined the channel
[12:04] konobi: jinmo123: time to get wireshark out
[12:05] sandfox: megalomix: it could be express isn't picking them up? (not sure if express is meant to by default or needs logging)
[12:05] arduix has joined the channel
[12:06] jinmo123: no difference, konobi sir!
[12:06] Illourr| has joined the channel
[12:06] jinmo123: i tested this in C
[12:06] jinmo123: works
[12:06] jinmo123: and wireshark didn't show any difference
[12:06] megalomix: i need to do some tests
[12:07] megalomix: ok
[12:07] booo has joined the channel
[12:07] megalomix: sandfox, btw the nginx configuration is good ?
[12:08] sandfox: megalomix: I can't see any reason why it wouldn't work, unless there is something else whacky going on in your main config
[12:08] jinmo123: ㅗㅕㅗ
[12:08] jinmo123: huh
[12:09] sandfox: megalomix: the only think i can think of is the "$proxy_add_x_forwarded_for" is not being set properly and therefore is neither correct, nor defaulting to being "$remote_addr" either
[12:09] jinmo123: konobi: thanks i found a problem
[12:09] r04r has joined the channel
[12:09] konobi: jinmo123: mmm?
[12:10] jinmo123: that code doesn't have any problem. my mistake
[12:10] jinmo123: setEncoding problem.
[12:11] dnjaramba has joined the channel
[12:11] jinmo123: hmm
[12:11] Illourr| has joined the channel
[12:12] elnn has joined the channel
[12:13] bingomanatee_ has joined the channel
[12:13] r04r has joined the channel
[12:15] dnjaramba has joined the channel
[12:16] Illourr| has joined the channel
[12:17] xaq_ has joined the channel
[12:17] megalomix: ok
[12:17] megalomix: i try
[12:17] megalomix: thank you so much
[12:17] megalomix: goodbye
[12:17] xaq has joined the channel
[12:18] Ezku has joined the channel
[12:18] Skaag has joined the channel
[12:18] r04r has joined the channel
[12:18] r04r has joined the channel
[12:20] wink___ has joined the channel
[12:20] xaq has joined the channel
[12:20] robhawkes has joined the channel
[12:20] xaq has joined the channel
[12:20] Joeysomo_ has joined the channel
[12:22] Country has joined the channel
[12:22] Dreamer3 has joined the channel
[12:23] mAritz: hm, if i restart my socket.io server and don't refresh the client page, the client will receive all socket.io emits twice (or more if i restart again). is this normal and what's the easiest way to prevent this?
[12:23] r04r has joined the channel
[12:26] Joeysomo has joined the channel
[12:27] wink_ has joined the channel
[12:28] r04r has joined the channel
[12:28] bradleyg has joined the channel
[12:29] Danielss89 has joined the channel
[12:29] diverdude: sandfox: do you know what this debug output means which comes periodically quite often? http://bin.cakephp.org/view/1779071068  Is that the heartbeat or is the connection reestablished or what is it?
[12:29] dnjaramba has joined the channel
[12:30] sandfox: diverdude: thats happening because the client doesn't support web-sockets so socket.io is falling back to using xhr-long poliing. and the stuff your seeing there is the heartbeat of sorts.
[12:31] Illourr| has joined the channel
[12:33] wink___ has joined the channel
[12:33] r04r has joined the channel
[12:34] krnl has joined the channel
[12:36] Illourr| has joined the channel
[12:37] bingomanatee__ has joined the channel
[12:38] timoxley has joined the channel
[12:39] BAK_ has joined the channel
[12:40] erikzaadi has joined the channel
[12:40] Ezku has joined the channel
[12:42] samBiotic has joined the channel
[12:46] Illourr| has joined the channel
[12:47] josh-k has joined the channel
[12:48] xaq_ has joined the channel
[12:49] diverdude: sandfox: i see...do you know what could be the reason why i can only emit messages from client to server...but not from server to client?
[12:49] dnjaramba has joined the channel
[12:49] sandfox: divderdude: when you say you can't, what exactly goes wrong?
[12:49] erichynds has joined the channel
[12:51] xaq has joined the channel
[12:51] diverdude: sandfox: well nothing happens when i do socket.emit('chatmessage',"pooooong"); on the server
[12:52] Illourr| has joined the channel
[12:53] vdemedes has joined the channel
[12:53] larsschenk has joined the channel
[12:53] vdemedes: Hello
[12:53] sandfox: are you seeing anything at all in the server logs?
[12:53] vdemedes: I'm wondering, how much does it cost to maintain npm?
[12:54] Skyjoker has joined the channel
[12:54] diverdude: sandfox: this is my server and my client: http://bin.cakephp.org/view/426942132
[12:54] diverdude: sandfox: no..nothing in the serverlogs either....at least not in the debug output which is printed in the terminal
[12:57] Illourr| has joined the channel
[12:57] diverdude: sandfox: sorry that was not true.,...i do see something iin the log actually
[12:58] diverdude: sandfox: this is debug output: http://bin.cakephp.org/view/1510798279
[12:58] ramitos has joined the channel
[12:58] mAritz: vdemedes: what?
[12:58] diverdude: sandfox: it actually writes; debug - xhr-polling writing 5:::{"name":"chatmessage","args":["pooooong"]}
[12:58] vdemedes: mAritz: what?
[12:58] diverdude: sandfox: ahh nevermind i solved it :)
[12:59] mAritz: vdemedes: ok
[12:59] sandfox: diverdude: hehe
[12:59] larsschenk has left the channel
[12:59] diverdude: sandfox: i did chatsock.socket.on <--- wrong :) should be ofc. chatsock.on
[13:00] sandfox: diverdude: yeah i was just tracing it back to see if was actually an object you were setting the listener on
[13:01] diverdude: sandfox: but i find it strange that my ff an chrome which are both quite new versions do not support websockets
[13:02] sandfox: chrome should…. (it does on my mac anyway), what version and OS are you running?
[13:02] Illourr| has joined the channel
[13:03] diverdude: sandfox: 16.0.912.77 (Developer Build 118311 Linux) Built on Ubuntu 11.10, running on LinuxMint 12
[13:04] sandfox: hmm, I think web sockets were in 16, I'm on 17.0.963.56 
[13:06] diverdude: sandfox: yeah ok....ohh well at least in 1 year or so they will be much more common
[13:06] sandfox: diverdude: that is strange, web sockets were supported from 14 upwards,
[13:06] diverdude: sandfox: and i will be prepared hehehe :D
[13:06] diverdude: sandfox: hmmm yes that is trange
[13:06] fairwinds has joined the channel
[13:07] diverdude: sandfox: does it have anything to do with the fact that im doing this: io.set('transports', ['xhr-polling','htmlfile','jsonp-polling']); on the server?
[13:07] Illourr| has joined the channel
[13:07] sandfox: diverdude: hehe, yes, 
[13:07] diverdude: sandfox: im not setting websockets as an option i think no?
[13:08] diverdude: sandfox: hehe sry im a bit ignorant sometimes :)
[13:08] sandfox: diverdude: if no transport is specified then all are available, otherwise as soon as you specify one, then only the explicitly stated protocols will be used
[13:09] diverdude: sandfox: ahh yes i commented out...much better :D
[13:09] mAritz1 has joined the channel
[13:09] megalomix has joined the channel
[13:09] diverdude: sandfox: now its doing: emitting heartbeat for client 1615203448304870986 instead
[13:09] megalomix: sandfox, 
[13:09] megalomix: hello everybody
[13:10] megalomix: sandfox, i saw log and there is remode addr :(
[13:10] mansoor: ohhh wow
[13:10] mansoor: 3d rigid body physics engine in JS? :S  its a brave new world indeed
[13:11] diverdude: sandfox: thx for your help and insight :) I gotta go for family dinner now hehe.
[13:11] megalomix: mansoor, 3D engine for JS ?
[13:11] mansoor: megalomix, yes
[13:11] megalomix: name?
[13:11] mansoor: Webgl? :p
[13:11] megalomix: aah webgl
[13:11] megalomix: ok
[13:12] mansoor: Here is a sweet engine for it https://github.com/mrdoob/three.js
[13:12] SubStack: hey what, request seems to be ignoring my { end : false } options in .pipe()
[13:12] rickibalboa has joined the channel
[13:12] Illourr| has joined the channel
[13:13] SubStack: suspicions confirmed https://github.com/mikeal/request/blob/master/main.js#L620
[13:14] mmalecki: SubStack: no worries, I'll fix that soon :)
[13:14] SubStack: it's really trivial to fix
[13:14] mmalecki: (fixing problems by releasing new modules ftw)
[13:15] SubStack: mmalecki: you're building something?
[13:15] mmalecki: SubStack: yeah! something like request, but closer to the metal
[13:16] dnjaramba has joined the channel
[13:17] Illourr| has joined the channel
[13:18] mmalecki: SubStack: any API tips, btw?
[13:18] megalomix: guys i proxy the requests from nginx to node.js
[13:18] megalomix: remote_addr what not set
[13:18] megalomix: or better
[13:18] megalomix: express.js does not store it as the real IP
[13:18] megalomix: but it always use 127.0.0.1 :(
[13:19] megalomix: https://gist.github.com/1863407
[13:21] sandfox: megalomix: http://stackoverflow.com/questions/7139369/remote-ip-address-with-node-js-behind-amazon-elb
[13:21] Edy: megalomix: what ist you set the remote_addr in yout nginx config?
[13:21] SubStack: mmalecki: what will your lib do that request doesn't do very well presently?
[13:21] sandfox: megalomix: might help (I don't know much about express)
[13:22] megalomix: Edy, https://gist.github.com/1863407
[13:22] megalomix: sandfox, i see
[13:22] SubStack: mmalecki: yourlib() should return a res object that I can call .setHeader() on for one
[13:22] SubStack: I don't like stuffing everything into an options hash
[13:22] Edy: megalomix: proxy_set_header   REMOTE_ADDR        $remote_addr;
[13:22] Illourr| has joined the channel
[13:22] Edy: could work :)
[13:23] SubStack: both ways should work
[13:23] megalomix: wait
[13:23] megalomix: do i have to add it
[13:23] megalomix: and remove real-ip ?
[13:23] dnjaramba has joined the channel
[13:23] megalomix: X-Real-IP 
[13:23] megalomix: ?
[13:24] Edy: leave x-real-ip and add remote_addr
[13:24] megalomix: what is the difference ?
[13:24] SubStack: mmalecki: it's frustrating to do var res = request.put({ uri : uri, headers : { 'content-type' : 'application/json' } }); res.end(JSON.stringify(obj))
[13:25] Edy: megalomix: just see what happens
[13:25] SubStack: mmalecki: request.put(uri).setHeader('content-type', 'application/json').end(obj) would be boss
[13:25] conntrack has joined the channel
[13:25] megalomix: ok
[13:25] megalomix: Edy, case sensitive ?
[13:25] SubStack: or saving to an intermediate res would be fine
[13:25] megalomix: REMOTE_ADDR ? Remote_Addr ?
[13:25] Edy: REMOTE_ADDR
[13:26] megalomix: ok
[13:26] megalomix: i try now
[13:27] wink_ has joined the channel
[13:27] dnjaramba has joined the channel
[13:27] Illourr| has joined the channel
[13:30] megalomix: Edy, restarted nginx restarted node
[13:30] megalomix: i do some requests
[13:31] megalomix: Edy, nothing �_�
[13:31] megalomix: 127.0.0.1 [Sun, 19 Feb 2012 13:33:00 GMT]  <---------- 
[13:31] megalomix: :/
[13:32] Edy: well then http://stackoverflow.com/questions/7139369/remote-ip-address-with-node-js-behind-amazon-elb
[13:33] Illourr| has joined the channel
[13:33] garrensmith has joined the channel
[13:34] wink___ has joined the channel
[13:35] zomgbie has joined the channel
[13:36] vguerra has joined the channel
[13:36] mmalecki: SubStack: I see. so how'd request().pipe(stream) work?
[13:37] mmalecki: SubStack: the way it does with request?
[13:37] dnjaramba has joined the channel
[13:37] SubStack: it works fine, it's just that the optional second arg opts is missing from the api
[13:38] Illourr| has joined the channel
[13:38] spolu has joined the channel
[13:39] megalomix: Edy, unfortunately i use express logger
[13:39] mmalecki: SubStack: ah ok. yeah, I kinda like .setHeader API
[13:39] megalomix: so i can do a change inside it
[13:39] wink_ has joined the channel
[13:39] mAritz1: any suggestions on ACLs  or other role management in node.js? the packages i've found through search.npm.js all seam either unmaintained or sub-optimal for me.
[13:40] sandfox: member:identifier:maritz1 roll your own?
[13:40] shanem_ has joined the channel
[13:41] mAritz1: sandfox: yeah, i guess i'll have to :(
[13:41] SubStack: mmalecki: yep and if .end() could detect for typeof obj and write the JSON.stringify that would also be super convenient
[13:42] SubStack: not sure how well that fits into your "close to the metal" vision though
[13:42] mmalecki: SubStack: yeah, that's kinda the thing I'm missing in node
[13:42] sandfox: mAritz1: you know it's the node way, on the plus side you could make a decent one and open source it? (i  have the same problem frequently)
[13:42] mmalecki: SubStack: I even pull requested that, but it got rejected
[13:43] Illourr| has joined the channel
[13:43] mAritz1: sandfox: i don't really have the patience right now to create a module that is of sufficient quality to open source it i'll likely take some existing module and force it into my app :P
[13:45] SubStack: mmalecki: it would be super handy for interacting with couch and rest apis
[13:45] SubStack: oh bonus!
[13:46] SubStack: if end() gets called with an object, set the content-type to application/json if it hasn't been set already
[13:46] ullmark has joined the channel
[13:46] SubStack: request.put(uri).end(obj)
[13:46] SubStack: MAGIC
[13:46] mmalecki: HELL YEAH
[13:48] wink___ has joined the channel
[13:48] SubStack: it should be a readable/writable stream too of course
[13:48] Illourr| has joined the channel
[13:48] sandfox: mAritz1: I sympathise… There is a world of half finished projects and dozens of 'good enough for my needs' git repos's on littered around the war zone that is my disk drive
[13:48] stagas has joined the channel
[13:48] mAritz1: :D
[13:51] hellp has joined the channel
[13:53] stonebranch has joined the channel
[13:53] SubStack: mmalecki: maybe instead of how request has a callback body parameter you could listen for a 'body' event instead
[13:53] warz has joined the channel
[13:53] warz has joined the channel
[13:53] johnhamelink has joined the channel
[13:53] SubStack: and if there is no 'body' listener on the first tick then the request doesn't get buffered
[13:54] WeeJeWel has joined the channel
[13:54] munichlinux has joined the channel
[13:54] conntrack has joined the channel
[13:55] dnjaramba has joined the channel
[13:56] mmalecki: SubStack: so, mylib(url).on('body') <= request doesn't get buffered?
[13:56] philipd has joined the channel
[13:57] SubStack: if you listen for a 'body' then it does get buffered
[13:57] mmalecki: ah ok
[13:57] wink_ has joined the channel
[13:57] SubStack: you can just check self.listeners('body').length on the first tick
[13:57] mmalecki: so you get whole body in this event?
[13:58] SubStack: I think that would be handy yes
[13:58] SubStack: but if you don't listen for it then the request doesn't get buffered at all
[13:58] rickibalboa: So I have a tls server listening for connections/data. I have this code in it: s.on('close', function() { console.log(s.remoteAddress); }); it prints undefined, any reason for this? It doesn't in the data event.
[13:58] Illourr| has joined the channel
[13:59] dnjaramba has joined the channel
[13:59] mmalecki: SubStack: hm, so I'd have to perform the actual request in the next tick?
[14:00] Cromulent has joined the channel
[14:00] mlangenberg: Argh, on the client a Message.create triggers and AJAX post. On the server this does a push to all Faye clients, they all do Message.create and again also execute the AJAX post. Catch22!
[14:00] mlangenberg: this is do much fun :P
[14:00] SubStack: mmalecki: but that is already the case to make piping work
[14:01] SubStack: and to make the .end(obj) api support setting the application/json content type
[14:02] ningu has joined the channel
[14:02] mmalecki: SubStack: yeah
[14:03] SubStack: oh request supposedly already has a chainable .setHeader()
[14:03] SubStack: looking at the source
[14:03] Ezku has joined the channel
[14:03] Illourr| has joined the channel
[14:04] ideado has joined the channel
[14:05] Topcat has joined the channel
[14:05] tpalohei has joined the channel
[14:06] wink___ has joined the channel
[14:06] kurtzhong has joined the channel
[14:07] AndreasMadsen has joined the channel
[14:07] dannyamey has joined the channel
[14:08] Illourr| has joined the channel
[14:09] mmalecki: SubStack: I actually dislike the idea of using nextTick. wondering about releasing close to metal library and some sweet API wrapper around it
[14:10] aristidesfl has joined the channel
[14:10] sreeix has joined the channel
[14:11] mAritz has joined the channel
[14:12] erikzaadi has joined the channel
[14:12] SubStack: mmalecki: what would the metal library do that http.Client doesn't do?
[14:13] mmalecki: SubStack: piping in and out, parsing the uri
[14:13] Illourr| has joined the channel
[14:14] adambeynon has joined the channel
[14:15] bradleyg_ has joined the channel
[14:15] wink_ has joined the channel
[14:15] hgg has joined the channel
[14:15] dnjaramba_ has joined the channel
[14:15] igl1 has joined the channel
[14:15] indistylo has joined the channel
[14:16] __class__ has joined the channel
[14:16] bradleymeck has joined the channel
[14:17] take_cheeze has joined the channel
[14:18] ph^ has joined the channel
[14:18] owenb has joined the channel
[14:18] Cromulen_ has joined the channel
[14:18] crutex has joined the channel
[14:18] Joeysomo has joined the channel
[14:19] Internet13 has joined the channel
[14:21] innoying has joined the channel
[14:21] nuba_ has joined the channel
[14:21] wink_ has joined the channel
[14:21] garrensm_ has joined the channel
[14:22] eukreign has joined the channel
[14:23] ramitos has joined the channel
[14:23] jldbasa has joined the channel
[14:23] SubStack: whoa bug alert
[14:24] SubStack: mmalecki: found another bug in request
[14:24] hipsterslapfight has joined the channel
[14:24] zed0 has joined the channel
[14:24] EvRide has joined the channel
[14:24] munichlinux: SubStack, what is it?
[14:24] SubStack: gisting
[14:25] Vennril has joined the channel
[14:25] roger_raymond has joined the channel
[14:26] GooMoo has joined the channel
[14:27] tkahn6_ has joined the channel
[14:27] ramitos has joined the channel
[14:28] SubStack: what the craps, the latest version of request on npm can't even req.setHeader()
[14:28] SubStack: but the one in git can
[14:28] kurtzhong has joined the channel
[14:28] SubStack: mmalecki: https://gist.github.com/1864060
[14:29] mmalecki: SubStack: wat
[14:29] SubStack: it does it for .put also
[14:29] looopy has joined the channel
[14:29] SubStack: mmalecki: the json message is repeated twice
[14:30] mmalecki: SubStack: yeah, I mean, wat.
[14:30] SubStack: couch was complaining at me about invalid json
[14:31] SubStack: and when I put req.end() in a nextTick it adds content-length: 0
[14:31] SubStack: and then puts the message down!
[14:32] SubStack: why isn't it still multipart?
[14:33] SubStack: mmalecki: can you try running that code? I'm on 0.4 still and it could be a 0.4 issue
[14:33] SubStack: I need to set up nave
[14:33] mmalecki: SubStack: sure
[14:33] mmalecki: SubStack: and use give :)
[14:34] mmalecki: SubStack: https://github.com/mmalecki/give
[14:34] deeprogram has joined the channel
[14:35] joeytwiddle has joined the channel
[14:37] SubStack: aha!
[14:37] wink___ has joined the channel
[14:37] SubStack: the problem goes away when I use the latest version from npm
[14:38] satyr has joined the channel
[14:38] SubStack: that is more forgivable
[14:38] mmalecki: SubStack: seeing the same thing with git
[14:38] cha0s has joined the channel
[14:38] cha0s has joined the channel
[14:40] SubStack: mmalecki: probably mikeal already knows about the issue and that's why he hasn't published a new version to npm yet
[14:40] insin has joined the channel
[14:40] silky has joined the channel
[14:40] mmalecki: SubStack: yeah, possible
[14:40] dscape: SubStack: there where some issues with http and couchdb prior to 0.6
[14:41] dscape: i filed some
[14:41] dscape: and at least one was related to content-length
[14:41] dscape: in nano i was actually removing the header cause it was screwing my pipes
[14:41] SubStack: dscape: https://gist.github.com/1864060
[14:41] SubStack: this is just a request bug, couch was doing the right thing
[14:42] zivester has joined the channel
[14:42] mmalecki: dscape: oh, btw, that request issue you hacked around in secret-sauce
[14:42] wink_ has joined the channel
[14:42] eukreign has left the channel
[14:42] mmalecki: dscape: how'd you like to do that in my lib?
[14:43] jxie has joined the channel
[14:44] ohtogo has joined the channel
[14:44] dscape: mmalecki: https://github.com/mikeal/request/issues/171
[14:44] dscape: i love it that its like the only thing that is not tagged easyfix
[14:44] dscape: lol
[14:44] mmalecki: dscape: no, that signature thing
[14:45] dscape: SubStack: the error i found was https://github.com/joyent/node/issues/1569
[14:45] VladGh has joined the channel
[14:45] dscape: it seems unrelated
[14:45] artusrocha has joined the channel
[14:46] dscape: all the http code got changed in 0.6 and plenty of fixes went in
[14:46] dscape: i wish there was a new 0.4.13 release with patches but no api changes
[14:46] dscape: but i guess that would be an immense amount of work
[14:48] mmalecki: no, that's just retarded
[14:48] wink___ has joined the channel
[14:49] dscape: SubStack: what's ur node/request combo?
[14:50] infynyxx has joined the channel
[14:52] infynyxx has joined the channel
[14:53] Industrial: What's the best Node.JS ORM/ODM out there now? Not Mongoose (or JugglingDB, which uses Mongoose for MongoDB).
[14:54] mmalecki: Industrial: I like resourceful
[14:54] magnetik has joined the channel
[14:55] Industrial: https://github.com/pmlopes/p-odm looks promising. I know it is very fast and does submodels/subdocuments and references correctly
[14:56] SubStack: dscape: that bug is just for the request on git, the one on npm is fine
[14:56] daglees has joined the channel
[14:56] adrianF has joined the channel
[14:56] copongcopong has joined the channel
[14:58] dscape: Industrial: quite frankly if you are thinking of using an odm you probably should be using a relational database
[14:58] dscape: SubStack: lots of pull requests from a bunch of people does that :\
[14:58] dscape: plus request doesnt really have a good test suite
[15:00] dscape: SubStack: what is it that you are trying to do?
[15:00] looopy has joined the channel
[15:00] robhawkes has joined the channel
[15:00] dscape: oh, nvm its just a bug in request anyway
[15:01] jetienne has joined the channel
[15:02] ramitos has joined the channel
[15:03] Morkel has joined the channel
[15:03] Danielss89 has joined the channel
[15:03] SubStack: dscape: I'm just building a program that talks to couch and I wanted to req.pipe(process.stdout, { end : false })
[15:03] dscape: SubStack: but it works on request@npm right?
[15:04] dscape: SubStack: request.post({url: url, body: body).pipe(process.stdout) doesnt work?
[15:05] SubStack: .pipe(process.stdout) closes the stream
[15:05] SubStack: so I can't ever print anything else ever
[15:06] Industrial: dscape: I disagree. MongoDB lends itself fine for denormalized structures
[15:06] dscape: Industrial: exactly
[15:06] Industrial: with an odm you can capture part of your denormalized structure in the document as a model, use model methods on the subdocument just for that part
[15:06] Industrial: instead of e.g. a table
[15:06] Industrial: make references where needed, but that's true for using a nosql without an odm aswell
[15:07] kbhit has joined the channel
[15:07] Industrial: you just need to work with that those subdocument models may not have identifiers
[15:08] draginx has joined the channel
[15:08] SubStack: I don't like models
[15:08] objectiveous has joined the channel
[15:08] Industrial: but they do have indexes :)
[15:08] pandeiro has joined the channel
[15:09] dscape: Industrial: so do relational databases
[15:09] dscape: postgres even supports json
[15:09] bnoordhuis has joined the channel
[15:09] dscape: fact is that the advantage of document store is denormalization
[15:11] piscisaureus_ has joined the channel
[15:12] diva has joined the channel
[15:12] kbhit: i'm trying to figure out why webgl_terrain_dynamic makes uses of two scenes { scene, and sceneRenderTarget} .  seems like sceneRenderTarget is for the landscape mesh.. any idea why the benefits of having 2 scenes for this are?
[15:13] kbhit: sorry, wrong channel
[15:14] jetienne has joined the channel
[15:14] ajjajaja has joined the channel
[15:14] conntrack has joined the channel
[15:16] mfunkie has joined the channel
[15:16] Lingus: dscape: I'm not able to find any information on pgsql's JSON support.
[15:16] Lingus: Do you have a link?
[15:18] Masds has joined the channel
[15:18] Masds: hi
[15:19] Masds: i have a question, can you scrape web pages with javascript generated content using node.io?
[15:19] dscape: Lingus: http://people.planetpostgresql.org/andrew/index.php?/archives/255-JSON-for-PG-9.2-...-and-now-for-9.1!.html
[15:19] dscape: Masds: node.io ?
[15:19] dscape: what's that?
[15:19] ryanfitz has joined the channel
[15:20] dscape: oh that is something
[15:20] dscape: lol
[15:20] dscape: im not familiar
[15:20] level09 has joined the channel
[15:20] ajjajaja: anyone know a good node.js host?
[15:20] Masds: it's a framework for scraping
[15:21] Masds: so do you think it is possible just with node.js and jquery?
[15:21] dscape: Masds: well node.io is node right? 
[15:21] dscape: everything is possible doesnt mean you should use tools that already exist and help you achieve the result
[15:21] Masds: yes
[15:21] dthompso99 has joined the channel
[15:21] Lingus: Thanks
[15:21] dscape: but im not familiar with node.io, we use banzai instead but dont use it for scrapping
[15:22] Masds: oh
[15:22] dscape: ajjajaja: if you are interested in trying nodejitsu register and then gimmesomebeta.jit.su
[15:22] Masds: well, for scraping regular php pages it's great
[15:22] ajjajaja: is it instant or do u have to wait 1 week for invite?
[15:22] arcanis has joined the channel
[15:24] wink_ has joined the channel
[15:27] aranw has joined the channel
[15:27] ningu has joined the channel
[15:28] aranw has joined the channel
[15:31] wink___ has joined the channel
[15:31] dscape: ajjajaja: go to  gimmesomebeta.jit.su to get an invite
[15:32] enmand has joined the channel
[15:33] whitman has joined the channel
[15:33] slaskis has joined the channel
[15:36] saikat has joined the channel
[15:37] mAritz has joined the channel
[15:37] kwmiebach_ has joined the channel
[15:37] dob_ has joined the channel
[15:39] johnhamelink has joined the channel
[15:41] criswell has joined the channel
[15:43] ohtogo has joined the channel
[15:43] wink_ has joined the channel
[15:46] SubStack: mmalecki: I'm giving give a spin right now
[15:46] AndreasMadsen has joined the channel
[15:46] mmalecki: SubStack: awesome! tell me how do you like it :)
[15:46] SubStack: I like how simple the directions are so far
[15:47] subbyyy has joined the channel
[15:47] SubStack: mmalecki: does this handle npm versions too?
[15:47] SubStack: since 0.6 bundles npm but 0.4 doesn't 
[15:48] mmalecki: SubStack: no, it doesn't handle that - it was pissing me off in nvm
[15:48] mmalecki: SubStack: I can make it a switch tho
[15:49] SubStack: I'm thinking it would be really useful to integrate give with stagecoach now
[15:49] SubStack: so you can bump the platform version completely in isolation
[15:49] SubStack: so you can test against a few version separately and quickly roll back if it's broken in some way
[15:50] mmalecki: SubStack: sounds neat
[15:50] SubStack: I want to make it awesome to manage a whole cluster spanning multiple machines without distributing vms
[15:51] SubStack: since vm-based rollouts take too long to spin up and push out
[15:51] SubStack: and I pretty much want to deploy new code and see it live immediately
[15:51] mmalecki: SubStack: agreed. it's pretty easy in haibu, actually
[15:52] mmalecki: SubStack: unless you have your own thing to do that :D
[15:52] SubStack: yeah I was looking at haibu some
[15:52] geojeff has joined the channel
[15:52] SubStack: I still might switch if propagit doesn't work out for what I'm trying to accomplish
[15:54] mmalecki: SubStack: what do you need in haibu to use it? we can totally work something out.
[15:56] esundahl has joined the channel
[15:56] wink___ has joined the channel
[15:56] DarkGlass has joined the channel
[15:57] mandric has joined the channel
[15:57] hij1nx_ has joined the channel
[15:58] SubStack: mmalecki: one big thing is that there isn't a 1:1 correspondence between apps and repos
[15:58] SubStack: one repo might span several processes
[15:59] SubStack: I might be able to just overload the name attribute with that however
[15:59] Vespakoen has joined the channel
[16:00] aristidesfl has joined the channel
[16:01] vereteran has joined the channel
[16:02] mmalecki: SubStack: ah, I see what you mean. that's never been a problem tho - you can spawn subprocesses from your app
[16:02] AndreasMadsen: 
ACTION watching a very interesting discussion here

[16:03] Qzen: 
ACTION agrees with AndreasMadsen

[16:03] xbayrockx has joined the channel
[16:03] mmalecki: SubStack: controlling processes gets harder tho. I think what you'd like is an array as start script?
[16:04] SubStack: it's complicated
[16:05] SubStack: because I want to spin up just those services necessary that aren't already running to satisfy the service dependency graph
[16:06] satyr has joined the channel
[16:06] DarkGlass has left the channel
[16:06] SubStack: then seaport takes care of the port allocation and routing
[16:07] mmalecki: SubStack: so, the way I'd do it: have some hub you can deploy specific services to. this hub should talk to haibus deploy services which aren't running/need deployment
[16:07] mmalecki: SubStack: that way haibu becomes a dumb slave, pretty much what it was meant to be :)
[16:08] lennyx has joined the channel
[16:10] skylamer` has joined the channel
[16:10] sandfox has joined the channel
[16:11] lennyx: so I'm building a website that needs to do some fairly intense parsing of a website. Basically POST a login form, visit a few pages, scrape data with server-side jquery and JSDOM. Problem is that's pretty heavily bounded by the CPU, which a one-thread async-only environment such as node wouldn't really jive in. Should I use some sort of async RPC system to export the parsing jobs, or should I somehow make node do it?
[16:12] SubStack: jsdom is crazy slow
[16:12] SubStack: lennyx: you could use phantom
[16:12] SubStack: or a real browser
[16:13] lennyx: SubStack, phantom? And how would I use a real browser?
[16:13] mmalecki: SubStack: actually, that way you could even run automated test suite after deploying
[16:13] stagas has joined the channel
[16:13] SubStack: mmalecki: yes, once the services are up
[16:14] SubStack: bound to a custom subdomain based on the commit hash automatically
[16:14] ajjajaja: dscape what the hell
[16:14] ajjajaja: that jitsu site
[16:15] ajjajaja: dscape i wanted a node.js host.
[16:15] lennyx: oh I found PhantomJS. But I still don't see how I would use a real browser, unless you mean using a real browser's engine
[16:15] mmalecki: SubStack: yeah. also, if you add some balancer to it, you can easily scale it up (spawning new drones in haibus)
[16:16] SubStack: mmalecki: yes and writing those is easy
[16:17] SubStack: basically stagecoach hub 7000 on the deploy server then stagecoach drone --hub hubhost:7000 on each of the workers
[16:17] SubStack: and stagecoach router somewhere
[16:18] kylebt has joined the channel
[16:18] sirwan has joined the channel
[16:18] danielss89 has joined the channel
[16:18] SubStack: then you git push to the hub to deploy into a directory and stagecoach deploy to serve up that commit
[16:19] SubStack: with some convenience command to do both of those at once for the usual case
[16:19] Illourr| has joined the channel
[16:19] SubStack: but the deploy is intelligent about satisfying the service dependencies across the whole cluster
[16:20] mmalecki: SubStack: not sure how stagecoach drone looks like, but you could replace it with haibu, I think
[16:20] psudo has joined the channel
[16:20] arduix has joined the channel
[16:20] psudo: _mdp 
[16:20] psudo: hello
[16:20] geojeff: for stylus, i am getting an, expected "indent", got "outdent" error, on line 37 in https://gist.github.com/1864482
[16:20] kitt has joined the channel
[16:21] lenage has joined the channel
[16:22] wink_ has joined the channel
[16:23] Illourr|: Try resetting spacing around there?
[16:23] Illourr|: I have that problem with coffeescript
[16:25] geojeff: must be something like that, but i don't see it
[16:25] psudo: :wink:
[16:25] lenage has left the channel
[16:26] psudo has left the channel
[16:27] st_luke has joined the channel
[16:27] Octayn: geojeff: even though it's optional in stylus I find using braces is still helpful
[16:29] geojeff: Octayn: i removed them to try to solve another glitch :) but am doing this in a build program, so easy to leave them in for another try
[16:30] sreeix has joined the channel
[16:30] geojeff: same error if i leave the braces though
[16:31] Octayn: odd
[16:34] wink___ has joined the channel
[16:36] mAritz has joined the channel
[16:36] isao has joined the channel
[16:37] CarterL has joined the channel
[16:38] draginx has joined the channel
[16:38] kbhit has joined the channel
[16:43] wink_ has joined the channel
[16:43] mmalecki: is there a way to capture heap snapshot on node v0.6 without using node-inspector?
[16:48] wink___ has joined the channel
[16:53] wink_ has joined the channel
[16:53] stagas has joined the channel
[16:54] ckknight has joined the channel
[16:56] adrianFa has joined the channel
[16:56] ningu: if I've been developing with express on node 0.6, will I still be able to deploy to sites that run 0.4.x?
[16:56] ningu: I have the current version of everything. just wondering.
[16:57] josh-k has joined the channel
[16:58] wink___ has joined the channel
[16:58] margle has joined the channel
[16:58] tonist has joined the channel
[16:59] pandeiro has joined the channel
[17:00] blup has joined the channel
[17:00] Wizek has joined the channel
[17:01] mAritz: ningu: try it? nvm is awesome ;)
[17:02] ningu: what is nvm?
[17:03] mAritz: don't know
[17:03] mAritz: maybe i should ask google? hmm...
[17:03] ningu: I just tried.
[17:03] ningu: nodejs + nvm didn't produce obvious results.
[17:03] ningu: oh, wait. now it did.
[17:04] ningu: before it corrected it to npm.
[17:04] insin: https://github.com/creationix/nvm
[17:04] __mdp has joined the channel
[17:04] ningu: yeah, found it now.
[17:05] piscisaureus_ has joined the channel
[17:05] trcarden has joined the channel
[17:05] cha0s has joined the channel
[17:05] cha0s has joined the channel
[17:06] pizthewiz has joined the channel
[17:06] ningu: I'm thinking of using heroku as a way to test deployent
[17:06] ningu: deployment*
[17:06] ningu: but not sure what other options there are -- I want something that won't be too expensive, and I don't need a lot of resources or bandwidth
[17:07] igl: ....cheap testing 
[17:07] _eddyb_ has joined the channel
[17:07] ningu: ?
[17:07] bankq has joined the channel
[17:07] mmalecki: ningu: nodejitsu maybe?
[17:07] igl: heroku starts at $36 i think?
[17:07] ningu: no, if you have just one web dyno it appears to be free
[17:08] ningu: and mongodb is free for 16mb, $5 for 256mb
[17:08] bankq has left the channel
[17:08] tommyvyo has joined the channel
[17:09] dscape: ningu: are you doing nodejs?
[17:09] hij1nx has joined the channel
[17:09] ningu: most of the node hosting sites I saw liseted were in beta and seemed to be kind of uncertain (uncertain whether they'll be around etc)
[17:09] ningu: dscape: yes, why else would I be in here? :P
[17:09] dscape: you were saying heroku i thought about ruby
[17:09] dscape: :)
[17:09] WeeJeWel has joined the channel
[17:09] dscape: ningu: i work for nodejitsu and you can deploy with us for free if you like
[17:10] draginx: why not just get ur own vps? O_o
[17:10] draginx: and learn how ot sysadmin :D
[17:10] Broolucks has joined the channel
[17:11] ningu: draginx: time, plus, minimum cost is too much afaik
[17:11] draginx: 15/mo? O_o
[17:11] ningu: draginx: for what?
[17:11] draginx: and less than 5 hours of ur time? :P for a full haproxy + nginx + node + redis + mysql setup O_o
[17:11] ningu: I don't mind that much setting it up
[17:11] draginx: 256mb of ram me believes
[17:11] draginx: which is plenty if u know what ur doing O_o
[17:12] ningu: but EC2 for example scares me because it's hard to know how much it will cost
[17:12] kirbysayshi has joined the channel
[17:12] markwubben has joined the channel
[17:12] draginx: http://6sync.com/ here u go :) If u do decide to go with them i can always send a referral code if ya want :P (their support i scurrently online and mario is awesome)
[17:12] looopy has joined the channel
[17:12] Illourr| has joined the channel
[17:13] danielss89_ has joined the channel
[17:13] insin: could check if there are any no.de smartmachines going, too
[17:13] ningu: draginx: ok, I will check it out
[17:14] st_luke has joined the channel
[17:14] draginx: ningu: or #6sync :)
[17:14] mehlah has joined the channel
[17:14] igl: after comparing prices i went with rackspace over amazon
[17:14] ningu: I do prefer having the control of my own machine, but it seemed like EC2 for example wasn't realistic
[17:14] draginx: does anyone know if session-web-sockets works with socket.io 0.8.x?
[17:14] igl: 256mb *nix is 0,01€ per hour there
[17:14] draginx: igl: *looks at his stocks* thanks fo rmaking that deciision :D
[17:14] igl: lol :P
[17:15] cjm has joined the channel
[17:15] igl: invest in facebook and loose it all \o/
[17:15] insin: haven't tried mongo, but it was trivial to install redis on one
[17:15] draginx: also in socket.io would I need to do socket.leave('room') if I wanted to leave a room?
[17:15] igl: if you can setup redis, you can setup mongo and mysql too
[17:16] ningu: draginx: can you use mongo on 6sync? their site is a bit... unclear
[17:16] draginx: ya O_o
[17:16] draginx: its ur own box mate :P
[17:16] ningu: I figured
[17:16] ningu: just checking
[17:16] draginx: u can put anything u want on it a slong as its not illegal in ur country
[17:16] igl: what if the datacenter is in another country?
[17:16] igl: ^^
[17:17] ningu: and what would I need nginx for? serving static files I guess?
[17:17] hz has joined the channel
[17:17] igl: rackspace and london, texas and hong kong
[17:17] igl: you can let nginx do the static files
[17:17] igl: and you can let nginx also do the clustering
[17:17] draginx: igl: I meant the datacenter's country buuuut i do believe america has setup laws to prevent US citizens of doing crap like that
[17:18] draginx: so u cant make ur own priatebay by setting up a server in sweden for example
[17:18] draginx: u can let nginx server static files, upstream to node, etc.
[17:18] draginx: BUT if u r using websockets nginx currently does not support that
[17:18] draginx: so ull need haproxy
[17:18] draginx: to upgrade :) basically if u need sockets use apache :P
[17:18] draginx: unless u go ttime to learn and expand ur skillset (which is always a good idea, specialization is for insects)
[17:18] enmand has joined the channel
[17:18] igl: isnt nginx-dev http1.1 now too?
[17:19] igl: saw that a few weeks ago actually
[17:19] ningu: draginx: I'm happy to learn new things, but I am also in the process of writing a dissertation
[17:19] bsdguru has joined the channel
[17:19] draginx: ningu: two full time jobs, MMA training everyday, and client work suck it up ;x
[17:19] pthm_ has joined the channel
[17:19] igl: ningu do you have any linux terminal experience?
[17:20] jakehow has joined the channel
[17:20] draginx: igl: not sure i think so but websockets may not even be released in 1.1 which is sad :(
[17:20] indistylo has joined the channel
[17:20] brianseeders has joined the channel
[17:21] `3rdEden has joined the channel
[17:21] ningu: igl: sure, I'm not afraid of linux, and I've also set up apache before for fastcgi, mod_perl, that sort of thing. I know it's not that hard, it's just another thing to worry about
[17:21] igl: aye
[17:22] ningu: and I know you can do some stupid things if you have no idea how to configure apache, etc.
[17:22] sriley: cant see anything on the changes list about http 1.1 backend support in nginx dev
[17:23] draginx: Anyone know how I can get a list of rooms a certain client is in with socketio?
[17:24] ningu: draginx: anyway, you're right that 6sync's lowest option is a good deal
[17:25] draginx: :) its great for starting out
[17:25] draginx: the prices are comparable ot linode
[17:25] igl: "HTTP/1.1 proxying is currently available in the development version"
[17:25] draginx: igl: shibby
[17:25] ningu: I prefer the resources to be limited by the provider so I can't go over by accident
[17:25] draginx: hmm bandwidht might be soft capped
[17:25] draginx: but u can request hard cap
[17:25] ningu: draginx: so why go with them over linode?
[17:25] draginx: ningu: I like to try out new things :) I have had absolutely zero problems with linode in the past
[17:25] draginx: but $15 < $20 :p
[17:26] ningu: hehe
[17:26] draginx: And Im hoping 6sync will be a much better future bet
[17:26] draginx: Kind of like how Apple used ot suck compared to Mcsft, etc.
[17:26] ningu: let's just hope it isn't a 10 year period of darkness like apple
[17:27] felixge has joined the channel
[17:27] felixge has joined the channel
[17:27] draginx: haha i dont think so :P if u got ays to improve their cpanel pygi on #6sync wil listen
[17:28] ningu: rackspace's lowest option is cheaper
[17:28] ningu: redhat? ugh
[17:28] ningu: maybe not :P
[17:29] jmaya has joined the channel
[17:29] phidah has joined the channel
[17:29] draginx: Anyone here good with socket.io? Basically, I'm trying to make a system where i dont reconnect when opening a new pop up window (basically I want someone to be able to join 5 poker tables)
[17:29] draginx: actually nvm thats impossible meh x-x
[17:29] draginx: ningu: who offers redhat?
[17:29] ningu: draginx: rackspace... but I'm confused, they have a box "number of red hat servers" which seems to be different from "linux server"
[17:29] draginx: O-o lol
[17:29] draginx: prob hae a deal with RH
[17:30] ph^ has joined the channel
[17:30] bradleymeck has joined the channel
[17:30] ningu: anyway, $15/mo for something easy to set up is better than $11/mo for something crappy
[17:30] ajjajaja has joined the channel
[17:30] ningu: but I have no idea what any of these services are really like
[17:31] ningu: or how important their control panel thingy really is
[17:31] draginx: O_o dont know rackspace's control panel
[17:31] ningu: I mean, if it takes care of dns routing that's good
[17:31] draginx: but 6sync's is very easy t use
[17:31] draginx: yup theres a dns manager
[17:31] ningu: but otherwise I don't need it for all that much
[17:31] ningu: or maybe I do and don't realize :P
[17:32] quijote has joined the channel
[17:32] mandric has joined the channel
[17:32] blueadept has joined the channel
[17:32] spion1 has joined the channel
[17:33] eukreign has joined the channel
[17:34] eukreign: i would like to use node.js to test some javascript that will run in the browser
[17:35] eukreign: i'm trying to figure out how require(); works and can't figure out how to make it work more like an include instead of a module system
[17:35] wink_ has joined the channel
[17:35] SubStack: mmalecki: https://gist.github.com/1864760
[17:35] draginx: u coming from a php background? O_o
[17:36] eukreign: no, python
[17:36] djcoin has joined the channel
[17:36] zemanel has joined the channel
[17:36] eukreign: not sure how that's relevant though
[17:36] ajjajaja has joined the channel
[17:36] mmalecki: SubStack: hrm... what shell is that?
[17:36] mmalecki: SubStack: also, is '~/prefix/bin' yours?
[17:36] SubStack: GNU bash, version 4.2.8(1)-release (x86_64-pc-linux-gnu)
[17:37] SubStack: yes I install most things with --prefix=$PREFIX
[17:37] tonist_ has joined the channel
[17:37] Ajdare has joined the channel
[17:37] Illourr| has joined the channel
[17:38] Ajdare: Bet and Win with us : http://bet-tips.cogia.net/?ref=285
[17:39] mmalecki: SubStack: looking, this totally should spawn a new subshell with a different path :/
[17:39] eukreign: when you do <script> in HTML does that not do a simple include?
[17:40] ningu: eukreign: if it has a link to a file rather than the script in its content
[17:40] mmalecki: SubStack: you've installed it from npm?
[17:41] disappearedng has joined the channel
[17:42] eukreign: so, how can i replicate this kind of include in nodejs?
[17:42] mmalecki: eukreign: you shouldn't
[17:42] eukreign: what is a better way?
[17:42] ningu: dscape: so is nodejitsu still in beta?
[17:43] mmalecki: eukreign: require
[17:43] dscape: ningu: yes, private beta
[17:43] ningu: eukreign: it really depends what you are trying to do, but in node you have to explicitly declare your exports -- so when you require('foo') it returns an object with those exports only
[17:43] ningu: how to make use of that is up to you, there are various patterns
[17:43] ningu: not so many really
[17:44] eukreign: but i also use that included file in the browser
[17:44] eddyb: og
[17:44] eddyb: *oh
[17:44] eddyb: then you may want a require-like system for the browser
[17:44] zomgbie has joined the channel
[17:44] eukreign: so if i put everything in exports, how would that work when running in a browser environment
[17:44] eddyb: or you could check for a browser environment
[17:45] SubStack: mmalecki: yes I did `npm install -g give`
[17:45] ningu: yeah, I never thought about it really but eddyb's suggestions are decent
[17:45] SubStack: it does spawn a subshell
[17:45] SubStack: it's just that $PATH isn't any difference
[17:45] eddyb: wrap the whole file in (function(exports){...})(exports||SomeNamepaceNameForTheBrowser)
[17:46] eddyb: (require would wrap it anyway in node, so it's a bit overkill)
[17:47] eukreign: hmm, i was hoping there was a way to do this without having to deal with checking if i'm running in browser or in node.js
[17:47] eukreign: cus i'm only using node.js for unit testing
[17:48] ningu: dscape: ok, and would I be able to redirect my dns to your servers and also host static files?
[17:48] ningu: eukreign: browser-safe js is rather different from what you might prefer to write in node, though...
[17:48] ningu: but I take your point.
[17:48] xaq_ has joined the channel
[17:48] jiboumans_ has joined the channel
[17:49] wilgroin has joined the channel
[17:49] ningu: I wish js had a better way of checking if a variable existed other than (if typeof foo != "undefined")... seems rather ugly :)
[17:49] TheJH has joined the channel
[17:49] Guest24248 has joined the channel
[17:49] ningu: I've only ever used that kind of code client-side though
[17:49] eukreign: i would have even been okay with passing the names of the files on the command line to 'js'
[17:49] eukreign: but it appears js only accept one .js file at a time
[17:50] ningu: eukreign: huh?
[17:50] ningu: not sure what the question is, but a required module can easily require other modules
[17:50] eukreign: i guess i could have my unit test runner just concatenate the files to emulate what the browser does
[17:50] ningu: and re-export if necessary
[17:50] ningu: hmm
[17:50] ningu: I suppose that could work
[17:50] eukreign: for example, i have core.js and then core-test.js
[17:51] ningu: but you won't have the browser DOM
[17:51] eukreign: core.js is included by the browser
[17:51] eukreign: so is pretty straight forward
[17:51] eukreign: core-test.js is the node.js specific code that tests core.js
[17:52] eukreign: when i run it i guess i could 'cat core.js > core-test-concat.js; cat core-test.js >> cores-test-concat.js' or something?
[17:52] simenbrekken has joined the channel
[17:52] eukreign: i don't need access to DOM
[17:53] draginx: ningu: dpeending on ur variable there is a better way than typeof O_o
[17:53] draginx: typeof is actually slow
[17:54] ArchWeb has joined the channel
[17:55] ningu: draginx: that's for things where I have a particular javascript file I load in the browser that I use on a number of pages, but there are hooks for things that only some pages do
[17:55] ningu: so I'll do something conceptually like if(foo) foo();
[17:55] ningu: but to check whether foo has been defined I have to do that
[17:55] erikzaadi has joined the channel
[17:56] ningu: actually it sort of feels like a hack, I can probably do better
[17:56] TheJH: ningu, `if (window.foo) foo()` :D
[17:56] ningu: ah
[17:56] ningu: ok
[17:56] eukreign: well, i found this: https://github.com/sstephenson/stitch/
[17:56] draginx: or window.hasOwnProperty("foo") :P
[17:56] eukreign: i guess that does exactly what i just suggested
[17:57] eukreign: seems kind of annoying to have to jump through these hoops though
[17:57] cha0s has joined the channel
[17:57] cha0s has joined the channel
[17:57] Slashbunny has joined the channel
[17:57] ningu: these are usually little functions that I put in an inline script tag since they are one or two lines
[17:58] ningu: starting autocompleters and the like
[17:58] warz has joined the channel
[17:58] warz has joined the channel
[17:58] bradleyg has joined the channel
[17:58] ningu: actually, I'd just do that directly with $(document).ready(...), these are for things that fit into a control flow of some sort. anyway the if (window.foo) thing works well :)
[17:58] joshsmith has joined the channel
[17:59] dscape: ningu: yes
[18:00] isao has joined the channel
[18:00] ningu: dscape: and mongodb presumably... well, can you send me some info about getting started, or should I just fill out the form on the site?
[18:00] zemanel has joined the channel
[18:00] ningu: I might move to a vps eventually but it would be great to have something to get my app working on first
[18:00] orlandovftw has joined the channel
[18:00] ningu: and it's quite low resource
[18:01] kbhit has joined the channel
[18:01] draginx: Anyone know how to get redis store to work in socket.io with a specific host/port/auth?
[18:01] spolu has joined the channel
[18:01] simenbrekken has joined the channel
[18:02] ningu: btw why do people use redis for sessions rather than mongodb? faster?
[18:03] draginx: imo yes
[18:03] ningu: I just used mongo because it was easier, already had it running and didn't have redis installed
[18:04] wink___ has joined the channel
[18:05] insin: becuase I was already using redis and it was just a case of using the redis session impl
[18:05] looopy has joined the channel
[18:05] sven_oostenbrink: How I need to do a regex match on a variable, so like String.match(/user.name + ':'/)  is this possible? I tried this, didnt really work..
[18:05] themiddleman_itv has joined the channel
[18:06] ningu: insin: ok, but is there a performance or other reason for using redis over mongo for sessions?
[18:06] ningu: sven_oostenbrink: you need to do something like foo.match(RegExp(user.name + ':'))
[18:06] insin: dunno, I've never used mongo, nor been in a traffic situation where performance was an issue
[18:06] EhevuTov has joined the channel
[18:06] ningu: sven_oostenbrink: but be careful about escaping metachars, XRegExp provides a facility for that
[18:07] ningu: there might be another existing way but I don't know of one yet
[18:07] devaholic: hello
[18:07] JJasonClark has joined the channel
[18:07] ningu: insin: ok, just curious
[18:07] ningu: I doubt it will matter in my case
[18:07] sven_oostenbrink: thanks a lot!
[18:08] ningu: for example a dot in a username should be escaped
[18:08] devaholic: hmm someone pinged me but my scrollback cleared it
[18:08] xaq has joined the channel
[18:10] wink_ has joined the channel
[18:10] isao has joined the channel
[18:10] sven_oostenbrink: ningu: how would I escape?
[18:11] draginx: Is there a wya to sned client information to socket.io upon disconnect?
[18:12] lohkey has joined the channel
[18:12] ph^ has joined the channel
[18:12] st_luke has joined the channel
[18:13] ullmark has joined the channel
[18:14] AviMarcus has joined the channel
[18:15] RobWC has joined the channel
[18:15] RobWC has left the channel
[18:17] simenbrekken has joined the channel
[18:20] nicki9knuckles has left the channel
[18:20] simenbrekken has joined the channel
[18:20] izepstan has joined the channel
[18:21] wink_ has joined the channel
[18:22] chadskidmore has joined the channel
[18:22] criswell has joined the channel
[18:23] hipsterslapfight has joined the channel
[18:24] tonist has joined the channel
[18:24] xaq has joined the channel
[18:25] margle has joined the channel
[18:25] draginx: Why wont a socket.emit() function be called on socket.on('disconnect') client side?
[18:25] draginx: but a console.log('whatever') shows up in my console really quickl;y
[18:26] maxogden: disconnect gets called once the user has been.... disconnected
[18:26] ArchWeb: hi!
[18:26] maxogden: so they are disconnected
[18:26] simenbrekken has joined the channel
[18:27] ArchWeb: what do I need to build or test something like jekyll ?!
[18:27] _baton_ has joined the channel
[18:27] maxogden: ArchWeb: http://blog.nodejitsu.com/introducing-blacksmith is kinda like that i think (havent used it )
[18:28] simenbrekken_ has joined the channel
[18:28] dscape: ningu: just fill up the form in the site
[18:28] dscape: then gimmesomebeta.jit.su
[18:28] dscape: and u should be happily deploying :)
[18:28] ningu: cool :)
[18:28] Felixfox has joined the channel
[18:28] ArchWeb: @mazogden ill try
[18:29] zeade has joined the channel
[18:29] maletor has joined the channel
[18:30] tkahn6_ has joined the channel
[18:32] lohkey has left the channel
[18:32] lohkey has joined the channel
[18:33] EuroNerd has joined the channel
[18:33] EuroNerd has joined the channel
[18:35] simenbrekken has joined the channel
[18:36] Sahua has joined the channel
[18:38] ArchWeb: how to fix the permission denied when exec npm?
[18:38] simenbrekken_ has joined the channel
[18:39] Qzen: ArchWeb: have you tried doing "sudo npm {commands}"
[18:39] shanem_ has joined the channel
[18:39] ArchWeb: but is it secure, I read something to recommend not to use sudo for npm
[18:39] _baton_ has joined the channel
[18:40] theCole has joined the channel
[18:40] Qzen: well you are only giving it permissions
[18:40] Qzen: sudo stands for "SuperUser DO"
[18:40] Qzen: so if you trust the application you can do it
[18:40] Qzen: and as far as i know npm is safe
[18:40] simenbrekken has joined the channel
[18:40] sriley: that was a while ago it recommended not running as sudo, now its safe to do so
[18:41] redir has joined the channel
[18:41] mmalecki: SubStack: sorry, been on a call. do you set your PATH explicitely anywhere?
[18:41] mmalecki: SubStack: that was my problem - bash reads rc files
[18:41] Qzen: when i am developing i always use a VirtualBox machine with ubuntu or something installed
[18:41] ArchWeb: ok ill try
[18:41] Qzen: just so i dont mess something up
[18:42] TheJH: ArchWeb, the "no sudo" warning was in ancient versions of npm
[18:42] TheJH: ArchWeb, isaacs totally changed his opinion in the meantime
[18:42] Qzen: so i would recommend installing VirtualBox or other similar program where you can have a virtual machine
[18:42] ArchWeb: ok thnx
[18:42] Qzen: that way you cant break something in the system
[18:43] ArchWeb: Im doing with amazon images :P
[18:43] Qzen: if it breaks the system you can just reinstall the OS on the virtual machine and then have another go
[18:43] ArchWeb: I broke like 5 vm on amazon
[18:43] wingie: there was a way to do: require('foo/bar') but i can't recall how to specify this in package.json
[18:44] Qzen: Well, "Oracle" has their "VM VirtualBox" available for Win/Unix and i think Mac
[18:44] Qzen: thats simple and fast to use
[18:45] Qzen: sorry, Linux*
[18:45] Qzen: people dont like to call Ubuntu unix
[18:45] ArchWeb: yeah ill try that for offline test, lol
[18:46] draginx has joined the channel
[18:47] ArchWeb: and which alternatives do you know to build something like jekyll, blacksmith looks 0k 
[18:48] Qzen: none
[18:48] Qzen: i only work backbone
[18:48] cleric has joined the channel
[18:48] Qzen: so i cant help there
[18:49] ArchWeb: k
[18:49] igl1 has joined the channel
[18:49] cleric: do i still need require('jsdom'); even though node.io has jsdom built in?
[18:49] Qzen: if it has it built in you dont need it
[18:50] cleric: but i cant use the jsdom functions
[18:50] Qzen: you just need to hook to the node.io jsdom
[18:50] TheJH: cleric, yes, I'm pretty suure you have to
[18:50] phidah has joined the channel
[18:50] jakehow has joined the channel
[18:50] cleric: ok thanks
[18:51] wink___ has joined the channel
[18:51] Oi-san has joined the channel
[18:52] draginx: Why wont a socket.emit() function be called on socket.on('disconnect') client side? but a console.log('whatever') shows up in my console really quickl;y
[18:54] Qzen: you might have something wrong in the code
[18:54] ceej has joined the channel
[18:54] Qzen: if you post it i might be able to help
[18:54] samBiotic:  /join #cyanogenmod
[18:54] samBiotic: oops
[18:55] churp has joined the channel
[18:55] Qzen has left the channel
[18:55] Qzen has joined the channel
[18:55] frewsxcv has joined the channel
[18:56] frewsxcv: anyone recommend an xml parser that includes xpath?
[18:56] wink_ has joined the channel
[18:57] Qzen: well, i dont use xml but you should check out https://github.com/joyent/node/wiki/modules#wiki-parsers-xml
[18:57] frewsxcv: yes, i saw that, but i can't decide on one
[18:57] rauchg has joined the channel
[18:57] draginx: Qzen: was that comment to me?
[18:57] Qzen: try some of them
[18:57] Qzen: draginx no
[18:58] draginx: oh
[18:58] jtsnow has joined the channel
[18:58] Dotan has joined the channel
[18:58] joshthecoder has joined the channel
[18:58] tonist has joined the channel
[18:59] Qzen: frewsxcv, just because one of those xml parsers fits my needs it doesnt need to fit yours...
[18:59] Qzen: so just try them and see which one you like
[18:59] frewsxcv: yes, but i value recommendations
[19:00] hackband has joined the channel
[19:00] Qzen: yeah me to, but sometimes its nice to have a opinion of your own ;)
[19:00] churp has joined the channel
[19:01] Qzen: thats why i often try them all untill i know what i want
[19:01] Qzen: but i have all the time in the world since i work from home :P
[19:02] joshkehn has joined the channel
[19:02] lohkey has left the channel
[19:02] joshkehn has left the channel
[19:03] Qzen: 
ACTION is now away

[19:03] Onkeltem has joined the channel
[19:03] criswell has joined the channel
[19:04] dnjaramba has joined the channel
[19:05] snowinferno has joined the channel
[19:05] frewsxcv has left the channel
[19:06] eukreign has joined the channel
[19:07] Sly__ has joined the channel
[19:09] wink___ has joined the channel
[19:09] isaacs_mobile has joined the channel
[19:09] manuj has joined the channel
[19:09] snowinferno: Hi all, I created a module to add a custom object to a project of mine, but when I require the module, the object I get from require is empty. How can I go about debugging this kind of thing?
[19:10] maxogden: snowinferno: nodejs.org/docs/latest/api/modules.html
[19:10] Leemp has joined the channel
[19:10] Dotan has joined the channel
[19:10] manuj has left the channel
[19:11] bartt1 has joined the channel
[19:11] ashishg has joined the channel
[19:12] clarkfischer has joined the channel
[19:12] snowinferno: maxogden: I've been reading that page, but haven't found anything there that helps me
[19:12] khrome has joined the channel
[19:14] patcito has joined the channel
[19:14] dnjaramba has joined the channel
[19:15] ashishg has joined the channel
[19:17] maxogden: snowinferno: based on the information on that page you should have: a node_modules folder at the root of your project, within that a folder containing your module (complete with package.json and and index.js that has a proper module.exports). then in your project you can require('packagename') (where 'packagename' is defined in your custom modules package.json)
[19:17] maxogden: snowinferno: so if you did all that and its still not working you can try using require.resolve('packagename') to see where npm is loading the package from
[19:18] snowinferno: yeah, I have all that but the object returned from require is empty, when i run it in node debug and watch the var i assign the require to, i get {} and require.resolve shows it is accessing ./node_modules/my_package
[19:18] isaacs has joined the channel
[19:19] maxogden: im guessing its a problem with the syntax in my_package then
[19:19] maxogden: assuming you can put any other node module into ./node_modules and require() it okay
[19:20] snowinferno: hmm I'll try to require the my_package.js file directly instead of as a module, see if that helps
[19:20] bhunt has joined the channel
[19:21] theBrettman has joined the channel
[19:22] lohkey has joined the channel
[19:22] superlou has joined the channel
[19:22] Lingus: snowinferno: And you're adding stuff into exports?
[19:22] snowinferno: yup
[19:23] Lingus: Can you gist your code?
[19:25] snowinferno: Lingus: for the module I presume?
[19:25] Lingus: Yeah
[19:25] snowinferno: it's in an open github project: https://github.com/snowinferno/accum/blob/master/accum.js
[19:25] rauchg has joined the channel
[19:26] frewsxcv has joined the channel
[19:26] frewsxcv: does node support for...of loops?
[19:27] Lingus: > require('./accum')
[19:27] Lingus: { Accum: { [Function: Accum] super_: [Function: EventEmitter] } }
[19:27] Lingus: v0.6.8
[19:27] ^robertj: anyone used alfred?
[19:29] snowinferno: hmmm that's troubling. Thanks for checking that Lingus, I clearly have something deeper going on than I initially thought
[19:31] Skyjoker has joined the channel
[19:31] lennyx has joined the channel
[19:33] stonebranch has joined the channel
[19:34] chadskidmore has joined the channel
[19:34] bfredrikb has joined the channel
[19:34] wink_ has joined the channel
[19:35] Dotan_ has joined the channel
[19:35] TooTallNate has joined the channel
[19:36] izepstan has joined the channel
[19:36] eddyb: snowinferno: I once wrote something like that
[19:37] cjm has joined the channel
[19:39] innociv has joined the channel
[19:39] margle has joined the channel
[19:42] EhevuTov has joined the channel
[19:43] ph^ has joined the channel
[19:43] jetienne has joined the channel
[19:44] isaacs: frewsxcv: i don't thikn for..of is a javascript thing
[19:44] simoon has joined the channel
[19:45] maxogden: you got an enumerable question i feel bad for you son, i got 99 elements but underscore.js isnt incapable of enumerating one
[19:45] Cromulent has joined the channel
[19:45] jetienne: maxogden: add one more ?
[19:47] simoon: lol maxogden 
[19:47] simoon: o wow
[19:47] gregpascale has joined the channel
[19:47] level09 has joined the channel
[19:48] maxogden: 
ACTION mic drop

[19:48] GabiPurcaru has joined the channel
[19:48] simoon: 
ACTION all my internets

[19:49] bfredrikb has left the channel
[19:49] fermion has joined the channel
[19:50] slaskis has joined the channel
[19:50] imsky has joined the channel
[19:51] jetienne has left the channel
[19:51] jetienne has joined the channel
[19:52] ohtogo has joined the channel
[19:52] mandric has joined the channel
[19:54] ryan1 has joined the channel
[19:54] GabiPurcaru: 
ACTION has quit [Ping timeout: 255 seconds]

[19:54] johnhamelink has joined the channel
[19:54] kbhit has joined the channel
[19:55] elnn has joined the channel
[19:55] avalanche123 has joined the channel
[19:55] Sephi-Chan has joined the channel
[19:55] Sephi-Chan: Hi
[19:56] Sephi-Chan: Do someone use Mustache as template engine? I wonder how to use numeric conditions. My use case: add the class "urgent" to the element if "some_attribute" is < 5 ?
[19:57] zomg: Sephi-Chan: did you try just if x < 5?
[19:58] zomg: Actually I don't even know if mustache has conditional blocks to begin with
[19:58] Sephi-Chan: In a Mustache template??
[19:58] Sephi-Chan: Mustache doesn't have if statements
[19:59] Sephi-Chan: As some people using Node use it for templating, I hoped that I could find help about that here ;)
[19:59] zomg: You probably would need to do the checks before rendering and then assign bools to the template
[19:59] st_luke has joined the channel
[20:00] zomg: I think it can selectively render things based on true/false but not sure if it has anything besides that
[20:00] zomg: Personally I just use ejs templates
[20:01] ohtogo has left the channel
[20:01] st_luke has joined the channel
[20:02] tricon_ has joined the channel
[20:04] sven_oostenbrink has joined the channel
[20:04] johnhamelink has joined the channel
[20:05] dnjaramba has joined the channel
[20:05] cjm has joined the channel
[20:06] Sephi-Chan: Ok ;)
[20:06] Sephi-Chan: Thanks for the tip
[20:06] frewsxcv: isaacs: https://developer.mozilla.org/en/JavaScript/Reference/Statements/for...of
[20:06] isaacs: frewsxcv: ECMAScript 6th Edition proposal 
[20:07] isaacs: frewsxcv: it's not ES (yet)
[20:08] clarkfischer has joined the channel
[20:08] marvin_ has joined the channel
[20:09] frewsxcv: isaacs: yes, but i wasn't sure if, like mozilla, they already implemented
[20:09] PhilK has joined the channel
[20:10] isaacs: frewsxcv: nope.  v8 usually follows the spec, rather than implement everything ever.
[20:10] Lingus: No, v8 seems to wait for the standard.
[20:10] YayOo has joined the channel
[20:10] Broolucks: which is wise, though that's a particularly useful construct
[20:11] dnjaramba has joined the channel
[20:11] frewsxcv: i don't see the harm in implementing it ahead of time
[20:11] draginx: is there a way to set a cookie session id before socket.on('connection') gets called?
[20:12] Lingus: AFAIK there are flags to enable some of the not-spec-yet-stuff
[20:12] Lingus: Check the v8 docs
[20:12] YayOo: hey ;) i don't know if i'm right here... i have a problem in installing node-proxy on my Mac
[20:12] nhunzaker has joined the channel
[20:13] ohtogo has joined the channel
[20:13] kainosnoema has joined the channel
[20:13] sven_oostenbrink has joined the channel
[20:14] sandfox: draginx: sort of yes
[20:14] sandfox: depends on what your doing...
[20:14] draginx: sandfox: how would I go about doing this?
[20:14] pandark_: Sephi-Chan: I think you are supposed to use an helper for that https://github.com/janl/mustache.js/blob/helpers/examples/helpers.js
[20:14] draginx: well i wantot check a cookie in io.set('authorization')
[20:14] draginx: want to*
[20:14] Sephi-Chan: pandark_: i think so :s
[20:15] sandfox: and you want a way to set it first?
[20:15] wingie: it's stupid that s3 doesn't support JSON
[20:15] simoon has joined the channel
[20:15] frewsxcv has left the channel
[20:15] wingie: who wanna work with xml
[20:15] ohtogo_ has joined the channel
[20:16] sandfox: wingie: plenty of libraries out there...
[20:16] draginx: sandfox: yup
[20:17] dnjaramba has joined the channel
[20:17] sandfox: draginx: i'll see if i've got some example somewhere (they are probably rubbish but they might help..)
[20:17] russell_h has joined the channel
[20:17] draginx: sandfox: ty
[20:18] esundahl has joined the channel
[20:18] ohtogo has joined the channel
[20:18] sandfox: dragnix: in short the cookie gets set before the web -socket connection is attempted, you can either do that on page load, or via some ajax calls after the page had loaded but before you start your web-sockets
[20:19] wingie: sandfox: most of them seem outdated
[20:19] wingie: which one is the most popular for s3/cloudfront?
[20:20] diogosnows has joined the channel
[20:20] draginx: sandfox: not sure what u  mean
[20:20] draginx: sockets work on all pages for my project
[20:21] wink___ has joined the channel
[20:22] sandfox: draginx: here ia gist https://gist.github.com/1865578
[20:22] mh415 has joined the channel
[20:22] sandfox: it uses connect etc but you ca ignore them for the moment as they are background magic
[20:22] lz has joined the channel
[20:22] draginx: sandfox: does it have ot be in the createServer block or can I do it in the app.use block?
[20:23] draginx: err app.configure
[20:23] lz: anyone here use mocha?
[20:23] cjm: NodeUp is live http://www.ustream.tv/channel/nodeup
[20:23] lz: can you not define tests inside callbacks? mines not working with require :(
[20:23] cjm: and we're in #nodeup
[20:24] dnjaramba has joined the channel
[20:24] sandfox: draginx: not sure, thetas just way i used it
[20:24] draginx: hmm right
[20:24] draginx: ok thanks will try to figure this out :S
[20:26] nhunzaker: hmmm
[20:26] nhunzaker: anyone ever have difficulty configuring iriscouch with Flatiron's Resourceful?
[20:27] sandfox: draginx: my workflow for the page is: index page gets loaded, javascript etc gets loaded, user gets login page, that page sends an ajax request to /auth which sets a value on their session, this value is then looked for when the sockets start. It's an overly simple example but it might help. The session ID itself is created when the user first loads the index page, and a cookie sent to them with it in.
[20:27] nhunzaker: huh
[20:27] nhunzaker: that's silly
[20:28] pangratz has joined the channel
[20:28] ningu: nhunzaker: say that to someone who doesn't do web dev and they'll think you're insane
[20:28] nhunzaker: haha
[20:29] dodo has joined the channel
[20:29] sven_oostenbrink has joined the channel
[20:29] rauchg has joined the channel
[20:30] githogori has joined the channel
[20:30] maxogden: do websockets have access to http-only cookies?
[20:30] bsdguru has joined the channel
[20:31] mraleph has joined the channel
[20:31] ningu: why is everyone using websockets? to pull data from other sites in an app?
[20:31] maxogden: websockets have a 2 byte overhead whereas http can be up to 8 kilobytes
[20:31] sandfox: maxogden: web sockets start with a a normal http request with an 'upgrade' header inside it
[20:32] maxogden: sandfox: oh right, nice
[20:32] draginx: http://pastie.org/private/jovuwtdfzgae8wsensyg why wouldnt this work?
[20:32] sambao21 has joined the channel
[20:32] draginx: (req is undefined)
[20:33] maxogden: app.use(startCookies);
[20:34] pangratz has left the channel
[20:34] sandfox: maxogden: +1 
[20:35] ritch- has left the channel
[20:35] localhost has joined the channel
[20:35] Revernd has joined the channel
[20:36] sandfox: whats the best way to debug node's memory usage anyone?
[20:36] sandfox: (and not too long winded, just something quick sand dirty)
[20:37] hipsterslapfight has joined the channel
[20:37] ningu: a sandfox should know sand dirty
[20:37] sandfox: dammit… :-)
[20:38] koxa has joined the channel
[20:38] Renegade001 has joined the channel
[20:38] ningu: I guess node-inspector just has debugging facilities
[20:38] ningu: I'm curious too what's out there, I'm just getting started so I haven't worried about memory usage and profiling yet
[20:38] koxa has joined the channel
[20:39] maxogden: sandfox: https://github.com/joyent/node-panic has some useful tools
[20:39] Broolucks: there is no space in quicksand
[20:39] Broolucks: :)
[20:40] sandfox: thanks maxogden, i'll give it a look, ningu: am on first big production app and memory has just become an issue..
[20:40] koxa has joined the channel
[20:41] maxogden: sandfox: protip: use more Buffers
[20:41] draginx: so even whe nusing app.configure(app.use(startCookies)) that *still* doesn't get sent before socket.on('authorization
[20:41] ningu: buffers instead of what?
[20:41] mansoor has joined the channel
[20:41] ningu: buffers for file read/write you mean?
[20:42] maxogden: Buffers are allocated outside of node's heap which is very limited in size
[20:42] hotchkiss has joined the channel
[20:42] koxa has joined the channel
[20:42] sandfox: draginx: it's the client (browser) that starts the web socket connection,
[20:42] pvankouteren has left the channel
[20:42] draginx: sandfox: right but express *should* set those cookies before any request is finished
[20:42] maxogden: for instance: var upload = ""; req.on('data', function(data) { upload += data }) allocates nodes memory
[20:43] koxa has joined the channel
[20:43] maxogden: but var buffers = []; req.on('data', function(data) { buffers.push(data) }) keeps them in Buffers and they get to use system memory
[20:43] ningu: maxogden: hmm, I see... but that wouldn't work if you loaded a bunch of records from mongodb, right? unless you rewrote the driver
[20:44] sandfox: draginx: hmmmm, maybe… what if you stop the web socket connection and inspect the web page, has it's cookie been set?
[20:44] maxogden: ningu: i dont know about mongo or its libs but if its written well it will give you Buffer streams
[20:44] logicalparadox has joined the channel
[20:44] draginx: sandfox: it actually hasnt been set
[20:44] draginx: but why? :(
[20:44] koxa has joined the channel
[20:44] draginx: even when i refresh the page
[20:45] diogogmt has joined the channel
[20:45] sandfox: maxogden: I have a bigger problem… I'm out of memory full stop on the server, can buffers hep me much in this case?
[20:45] dnjaramba_ has joined the channel
[20:45] st_luke has joined the channel
[20:46] maxogden: sandfox: well thats a different problem. what is the main thing eating up your memory (if you know yet)
[20:46] sandfox: draginx: not sure… if you strip out the startCookies bit and just do a normal request does anything get set at all? if not then your first problem is a total lack of cookies
[20:46] draginx: well connect.sid works
[20:47] sandfox: maxogden: it's a few (20) million objects on a giant KDTree
[20:47] draginx: and according to app.get('/', .. req.session.user is fine ("1")
[20:47] maxogden: draginx: if you inspect the HTTP that express is returning the auth ajax there should be a Set-Cookie header
[20:47] draginx: but chrome debugger doesnt show anything except connect.sid
[20:47] draginx: theres no ajax O_o its websockets..
[20:47] joshkehn has joined the channel
[20:48] maxogden: sandfox: a custom in memory kdtree in node? or some other db?
[20:48] joshkehn has left the channel
[20:48] sandfox: dragninx: thats all it should show...
[20:48] maxogden: draginx: oh sorry i was referring to sandfox's description (thought it was your setup)
[20:48] draginx: hmm
[20:48] Chel has joined the channel
[20:49] aaronmcadam has joined the channel
[20:49] sandfox: maxogden: in memory (has to be for speed), each object has x/y values but a ton of other data that never gets used except to throw down the wire as an http response
[20:50] tdegrunt has joined the channel
[20:50] sandfox: draginx: the only thing stashed on the client side will be the session id
[20:50] logicalparadox has joined the channel
[20:51] joshkehn has joined the channel
[20:51] joshkehn has left the channel
[20:52] draginx: sandfox: interesting… i wonder if epress has a way of retrieving cookies based on connect.sid
[20:52] sandfox: draginx: do you really care what's in the cookie? apart from the ID?
[20:53] draginx: sandfox: yes the "token"
[20:53] sandfox: draginx: surely that token should only be stashed on the server in the session?
[20:54] adrianF has joined the channel
[20:54] izepstan has joined the channel
[20:54] draginx: sandfox: its a token generated upon login stores in the DB and in Redis
[20:54] draginx: that token is passed each time i want to send the socket a command to authenticate
[20:55] draginx: this all happens on SSL so middle attacks are not a concern
[20:55] joeytwiddle has joined the channel
[20:55] dnjaramba has joined the channel
[20:55] sandfox: draginx:you only need the session id from the cookie to retrieve the auth token?
[20:55] maxogden: sandfox: if you can reproduce it on OSX you can use Instruments.app a la http://journal.paul.querna.org/articles/2011/04/05/openssl-memory-use/
[20:55] draginx: sandfox: how so?
[20:55] maxogden: sandfox: also the kdtree sounds cool, is it on github?
[20:57] draginx: sandfox: i think i got it thanks
[20:57] sandfox: maxogden: not yet as I haven't finished hacking it up, but i can point you in the direction of a friend of mines who made version 1 (only handles 2 dimensions and probably lots of bugs, but works well enough) : https://github.com/dbotha/Javascript-k-d-tree
[20:58] maxogden: sandfox: cool. i hack on datacouch.com in my spare time and im rewriting it in node and have been looking for useful in-memory data crunching tools to hook up to it
[20:59] chjj: https://github.com/seedifferently/the-great-web-framework-shootout - i think theyre forgetting one
[20:59] draginx: actually no i dont the only example i have is using memorystore but i want to use redisstore *sighs*
[20:59] mraleph has joined the channel
[20:59] sandfox: draginx: the session store shouldn't make any difference (in theory anyway)
[21:00] henrw has joined the channel
[21:00] draginx: how would iretrieve info based on session key then?
[21:00] draginx: sessionStore.get(data.sessionID, function (err, session) { works but only if ur using MemoryStore...
[21:01] sandfox: maxognden: sounds interesting, there is a lot of optimisation to be done on the KDTree (and it really needs to be done in c++)
[21:01] wink_ has joined the channel
[21:01] henrw: hi, does anyone have any experience with using node.io in connection with node.js?
[21:01] nathanieldavid has joined the channel
[21:02] sandfox: dragninx: does sessionStore present the same interface when backing it with redis
[21:03] sandfox: henrw: node.io ??
[21:03] chilts: node.io is writte for nodejs as far as I can tell
[21:03] chilts: *written
[21:03] henrw: scraping framework for node.js
[21:04] yayOo has joined the channel
[21:04] chilts: yeah, though I haven't used it - sorry :)
[21:05] dnjaramba has joined the channel
[21:05] henrw: k, no problem..
[21:05] cjm: NodeUp is live, talking what's next for node, http://www.ustream.tv/channel/nodeup (and were in #nodeup)
[21:06] nhunzaker: @mmalecki if you have a minute, do you know of a good example of iterating over an array using Plates?
[21:06] wink_ has joined the channel
[21:07] ramitos has joined the channel
[21:08] emilk2001 has joined the channel
[21:09] Emile has joined the channel
[21:09] jesusabdullah: nhunzaker: Collections should work now
[21:09] jesusabdullah: or at least half-work
[21:10] nhunzaker: hmm
[21:10] ningu: cjm: is that an audio feed or what? what's it actually a feed of?
[21:10] jesusabdullah: https://github.com/flatiron/plates/issues/24
[21:10] nhunzaker: hot
[21:10] WeeJeWel has joined the channel
[21:10] nhunzaker: that's exactly what I need
[21:10] ningu: I guess NodeUp is a conference
[21:10] jesusabdullah: it's just a podcast
[21:11] cjm: ningu: yea only audio, no video
[21:11] draginx: sandfox: i got it working! thanks :)
[21:11] michaelhartau has joined the channel
[21:11] sandfox: draginx: what was up?
[21:11] draginx: a few things tbh >.> 
[21:11] wink___ has joined the channel
[21:11] draginx: after parseCookie() i just did data['cookiename']
[21:12] draginx: when it shouldve been data.cookie['cookieame']
[21:12] draginx: calling wrong parameter in the actual socket.on(connection etc.
[21:12] ningu: who's the german guy talking?
[21:12] ningu: he sounds very german
[21:12] zemanel_ has joined the channel
[21:13] sandfox: draginx: hehe, it's now working at any rate
[21:13] cjm: ningu: that's felixge
[21:13] draginx: sandfox: yup :D working really well too thanks
[21:13] felixge: ningu: :)
[21:14] draginx: now i just need to hmac cookies, get some encryption  mechanism going for socket talking, etc :P
[21:14] snearch has joined the channel
[21:15] sandfox: draginx: hmac is kids play, import crypto and 3 lines later, tada. for socket encryption, just throw it all over SSL
[21:15] ningu: felixge: you?
[21:15] ningu: ah, ok
[21:15] felixge: ningu: yes
[21:15] ningu: actually your english sounded almost perfect, but I've lived in germany enough to hear the accent
[21:15] nhunzaker: @jesusabdullah hmmm, haha. So it's working, it's just outputing the raw object data. But I'll keep messing with it before I gripe more
[21:16] draginx: sandfox: yeah thats what i figured :) thanks
[21:16] zemanel_ has joined the channel
[21:17] salva has joined the channel
[21:18] dnjaramba has joined the channel
[21:18] QaDeS_ has joined the channel
[21:18] danielss89: Hmm. In my main file, which i runs with node, on top i require mysql and afterwords i require some modules which then uses the mysql.. But my modules can't find mysql? ReferenceError: mysql is not defined
[21:19] pilot19 has joined the channel
[21:20] superlou has joined the channel
[21:20] dunder-mifflin has joined the channel
[21:20] danielss89: how can i define mysql one time, and use it in all submodules
[21:20] danielss89: so i don't have to include it and connect for each module
[21:21] AkhiAbdullah has joined the channel
[21:22] EhevuTov: is the audience chat for the ustream here in irc or on the ustream site? (ustream chat isn't working well for me)
[21:22] felixge: danielss89: pass the mysql instance between your objects/functions?
[21:22] dunder-mifflin: danielss89: it's a better design to have a single module that's responsible for datastore (in your case, mysql)... it indicates low coupling http://en.wikipedia.org/wiki/Coupling_%28computer_programming%29 ... having to pass something to too many modules indicate a design that needs to be re-thought
[21:23] danielss89: dunder-mifflin: yes exactly, but if i have a module called db.js and i load it as the very first thing in my main app file.. how can the other modules use it?
[21:23] danielss89: do they need to require it to?
[21:24] danielss89: too*
[21:24] CarterL has joined the channel
[21:25] pilot19 has left the channel
[21:26] ningu: who are the main people on the podcast?
[21:27] ajackbot has joined the channel
[21:28] innociv has joined the channel
[21:29] dannyamey has joined the channel
[21:29] LordTrev has joined the channel
[21:29] wink_ has joined the channel
[21:29] dnjaramba has joined the channel
[21:31] alex_r has joined the channel
[21:31] LordTrev has joined the channel
[21:33] LordTrev has joined the channel
[21:34] wink___ has joined the channel
[21:35] LordTrev has joined the channel
[21:37] dnjaramba_ has joined the channel
[21:38] daniel: I want to define a number of common variables at the top of all my modules; is there any way to do that?
[21:39] ohtogo has joined the channel
[21:39] Axsuul has joined the channel
[21:39] bhunt has joined the channel
[21:39] daniel: other than copying and pasting a large `var foo = require("globals").foo; var bar = require("globals").bar; ...` block?
[21:39] rkrv has joined the channel
[21:40] felixhummel has joined the channel
[21:40] daniel: I tried `module.foo = "abc"; console.log(foo)` but that didn’t work
[21:40] daniel: is the module scope purely lexical?
[21:41] dve has joined the channel
[21:41] Broolucks: on the flip side, that c/p makes the files easier to understand
[21:42] Broolucks: it's easier to see where the variables come from
[21:43] adambeynon has joined the channel
[21:43] dnjaramba has joined the channel
[21:44] richardr has joined the channel
[21:44] nhunzaker: @jesusabdullah notice anything out of the ordinary here? https://gist.github.com/1865970
[21:44] illourr has joined the channel
[21:45] illourr: afternoon dudes!
[21:45] AndreasMadsen has joined the channel
[21:47] wink_ has joined the channel
[21:48] jesusabdullah: nhunzaker: Nothing stands out, no
[21:48] sorensen__ has joined the channel
[21:48] nhunzaker: @jesusabdullah okay. So I think what's happening is that Plates is iterating over the array. It's just that the array happens to be an object.
[21:48] jesusabdullah: well, except list.html and index.html aren't the same
[21:48] nhunzaker: haha. pardon
[21:48] jesusabdullah: what output do you get?
[21:48] nhunzaker: I should c/p better
[21:49] nhunzaker: I get the json directly output
[21:49] nhunzaker: it's doing a great job of iterating over it
[21:49] jesusabdullah: can you copypaste the result as well?
[21:49] nhunzaker: but it's replacing the HTML with a stringified JSON object,
[21:49] nhunzaker: sure
[21:49] jesusabdullah: word
[21:50] DrPheltRight has joined the channel
[21:50] nhunzaker: alright
[21:51] daniel: Broolucks: I agree; but some are so ubiquitous that declaring them at the top of every single file just gets tedious
[21:52] dberlinger has joined the channel
[21:52] daniel: language-extension-like things
[21:52] nhunzaker: @jesusabdullah so test-09 does a great job of verifying that arrays work, but the test seems to fail on actual collections
[21:52] zomgbie has joined the channel
[21:53] nhunzaker: @jesusabdullah well looking at it again, I don't want that to mislead you. Could it be that it's not checking nested elements?
[21:53] boltR has joined the channel
[21:54] jhurliman has joined the channel
[21:54] dberlinger has joined the channel
[21:54] koxa has joined the channel
[21:54] dnjaramba has joined the channel
[21:54] ajackbot has joined the channel
[21:55] dberlinger has joined the channel
[21:55] jhurliman: is there a popular node.js library floating around providing mutex-like functionality? i want to acquire a mutually exclusive resource, do three or four asynchronous operations (various database reads and writes), then release the resource
[21:55] ramitos_ has joined the channel
[21:56] larsschenk has joined the channel
[21:56] larsschenk has left the channel
[21:56] Broolucks: daniel: that might very well be overkill, but I figure you could do something like give your files a separate extension and register the extension with require() so that it prepends these lines to what it loads
[21:57] daniel: Broolucks: ah!
[21:57] ramitos__ has joined the channel
[21:59] TheJH: jhurliman, I think something by Sage does that
[21:59] jhurliman: TheJH, username Sage on github?
[21:59] jhurliman: i just found https://gist.github.com/713082 that seems to be enough for my needs
[22:00] TheJH: jhurliman, yes
[22:00] jhurliman: ty
[22:00] TheJH: jhurliman, I think it's related to streamlinejs
[22:02] jesusabdullah: nhunzaker: I have heard rumors that it wasn't actually resolved, which is why I wanted to see what you got
[22:02] wink_ has joined the channel
[22:02] jhurliman: i'll take a look thanks
[22:02] jhurliman: separate question.. is anyone writing in icedcoffeescript these days?
[22:02] nhunzaker: @jesusabdullah I'm creating a test case right now. I'll probably make some noise in a few.
[22:02] artusrocha has joined the channel
[22:02] dnjaramba has joined the channel
[22:03] dve has joined the channel
[22:03] igl1 has joined the channel
[22:05] lotsofcode has joined the channel
[22:05] netminder has joined the channel
[22:05] felixge has joined the channel
[22:05] felixge has joined the channel
[22:06] netminder has left the channel
[22:07] wink___ has joined the channel
[22:07] skm has joined the channel
[22:07] rauchg has joined the channel
[22:07] dnjaramba has joined the channel
[22:07] stefpb has joined the channel
[22:10] igl has joined the channel
[22:11] sandfox: jhurliman: I'd never come across iced before, seem quite nice
[22:11] simenbrekken has joined the channel
[22:11] kickingvegas has joined the channel
[22:12] wink_ has joined the channel
[22:12] zixiz has joined the channel
[22:13] innoying has joined the channel
[22:13] michaelhartau has joined the channel
[22:14] djbell has joined the channel
[22:14] igl1 has joined the channel
[22:16] zixiz has joined the channel
[22:17] igl2 has joined the channel
[22:17] bhunt has joined the channel
[22:17] jaime has joined the channel
[22:17] jtsnow has joined the channel
[22:17] hkjels has joined the channel
[22:19] dnjaramba has joined the channel
[22:19] YoY has joined the channel
[22:19] ajackbot_ has joined the channel
[22:20] ryan1 has joined the channel
[22:22] dscape: jhurliman: why do you need a mutex
[22:22] dscape: node isnt multithreaded
[22:22] jhurliman: dscape, it's asynchronous though
[22:22] willwhite has joined the channel
[22:23] dscape: jhurliman: kk
[22:23] hkjels: What is the simplest way to make a mock request-object for express?
[22:23] jhurliman: consider: database.read('field1', function(res) { if (!res) database.write('field1', 42); });
[22:24] dscape: hkjels: github.com/flatiron/nock
[22:24] maxogden: jhurliman: https://github.com/iriscouch/txn does it for couchdb well, not generic though
[22:24] jhurliman: maxogden, thanks
[22:24] dscape: jhurliman: so how do u implement a mutex then?
[22:25] jhurliman: dscape: https://gist.github.com/713082
[22:25] dscape: in a single process?
[22:25] dscape: ok, nevermind me
[22:25] dscape: :)
[22:25] dscape: im just trying to say i think you need something else :)
[22:25] hkjels: dscape: Looks good. And this is express-compatible for sure?
[22:25] jhurliman: nope, i need exactly what is in that snippet
[22:25] maxogden: dscape: you are letting your preconceived notions of what a mutex is distort the generic definition
[22:26] dscape: maxogden: yeah i know mutexes like we use them in databases
[22:26] dscape: hkjels: it works with express
[22:26] dscape: it actually overrides node http request from code
[22:26] dscape: core*
[22:26] jhurliman: i can enter a lock, do lots of asynchronous work, release the lock, and only one chain of events will be inside the critical section at once
[22:27] hkjels: ahh. This is great. Thank you!
[22:27] dscape: hkjels: you are welcome
[22:27] miceiken has joined the channel
[22:28] dscape: hkjels: check the recording functionality
[22:28] dscape: if your tests already work it will spit out working code for you
[22:28] dscape: hkjels: nock.recorder.rec()
[22:28] teslan has joined the channel
[22:28] hkjels: Will do
[22:28] teslan: declerative or programmatic - what do you choose when you have a choice and why?
[22:29] zomg: teslan: wouldn't that depend on what you're doing? =)
[22:29] Danielss89 has joined the channel
[22:30] xbayrockx has joined the channel
[22:30] dscape: teslan: whats programmatic?
[22:31] zomg: teslan: I'm just going to assume you mean dojo since you're on #dojo and were talking about that there.. Declarative is probably faster to work with, but programmatic has better performance (it doesn't need to scan the DOM for dijits)
[22:31] Renegade001 has joined the channel
[22:31] dnjaramba_ has joined the channel
[22:32] zomg: The reason declarative is faster to work with is in my opinion because you don't need to go around writing IDs on your elements and then writing snippets of JS to create widgets on those elements, instead you just toss in the attribute and done.
[22:32] tec_ has joined the channel
[22:34] blup has joined the channel
[22:34] gr4yscale has joined the channel
[22:34] felixhummel has joined the channel
[22:34] teslan: declerative is also starting to feel more and more like "clik and point" programming, without understanding much of what's going on underneath?
[22:34] zomg: heh
[22:34] ramitos has joined the channel
[22:35] zomg: it is easier to get started without really knowing how dijits are instanciated programmatically
[22:35] gr4yscale has joined the channel
[22:37] rauchg has joined the channel
[22:37] daniel: is `require.extensions` documented anywhere?
[22:38] dnjaramba has joined the channel
[22:38] rope has joined the channel
[22:39] wink___ has joined the channel
[22:40] mape: Does node have issues with fs.watch on debian? Keep getting Caught exception: Error: watch ENOSYS
[22:42] bradleyg has joined the channel
[22:45] wink_ has joined the channel
[22:46] johnnychimpo has joined the channel
[22:47] CIA-102: node: 03koichik 07master * r7ae0d47 10/ test/simple/test-tls-over-http-tunnel.js : test: fix test-tls-over-http-tunnel with v0.7 - http://git.io/RSHx4g
[22:47] sven_oostenbrink: ningu: Hey there, a while ago you gave me a good tip about how to do foo.match(RegExp(user.name  + ':'))  Question: Where does regexp come from in Node?
[22:48] dnjaramba has joined the channel
[22:49] maxogden: sven_oostenbrink: RegExp is part of javascript, not node specific at all
[22:49] Lingus: daniel: No, because you shouldn't be touching it
[22:49] wink___ has joined the channel
[22:50] sven_oostenbrink: maxogden: ah, okay! thanks!
[22:51] hkjels: dscape: I'm not seeing how I can use this as request mock-object. The result is closer to a response I guess
[22:51] simoon has joined the channel
[22:51] hkjels: it's also missing all of the methods applied by express unfortunatly
[22:52] dscape: hkjels: you dont modify your code
[22:52] dscape: just do the requests
[22:52] dscape: and this will catch it
[22:52] wookiehangover: maxogden: BUT I REUSE ALL THE CODEZ
[22:53] ningu: sven_oostenbrink: ah, I see they already answered it :)
[22:53] boltR has joined the channel
[22:53] hkjels: dscape: I am after the request-object. So that I can test wether the request is json and has certain headers set
[22:53] igl has joined the channel
[22:53] ningu: XRegExp is not part of js, though, it's in npm
[22:53] hkjels: I'll mock it manually I guess
[22:53] rauchg: dscape 
[22:54] rauchg: that looks pretty cool 
[22:54] dscape: rauchg: what?
[22:54] rauchg: /nock
[22:54] dscape: yeah its a life saver
[22:54] dscape: like imagine creating machines on rackspace
[22:54] dscape: and waiting for it to finish
[22:54] dscape: for each test...
[22:54] rauchg: `Nock works by overriding Node's http.request function.`
[22:54] rauchg: is that accurate 
[22:54] dscape: yeah
[22:54] rauchg: or does it override the constructor prototype
[22:55] wink_ has joined the channel
[22:55] confoocious has joined the channel
[22:55] confoocious has joined the channel
[22:55] rauchg: for http.Client or whatever it's caleld 
[22:55] dscape: rauchg: https://github.com/flatiron/nock/blob/master/lib/intercept.js#L72
[22:55] sven_oostenbrink: ningu: yeah.. Just one other question, you mentioned that RegExp has methods to escape, but checking, I only found .compile(), .test(), and .exec().. Which is the method to escape strings?
[22:55] dscape: http.ClientRequest = OverridenClientRequest;
[22:56] tkahn6_ has joined the channel
[22:56] dscape: hkjels: what are you trying to do?
[22:56] rauchg: i guess that should be clarified 
[22:56] dscape: rauchg: portuguese people cant write proper english i guess :P
[22:56] antono has joined the channel
[22:56] danielss89_ has joined the channel
[22:57] dscape: its built by pedro from nodetuts
[22:57] dscape: you met him at nodesummit i think
[22:57] rauchg: yup
[22:57] rauchg: briefly 
[22:57] isaacs has joined the channel
[22:57] rauchg: anyhow, i'll give you more feedback when i try it out 
[22:58] rauchg: could be really useful for testing obscure stuff without having to add needless public events and such
[22:58] hkjels: dscape: I want to make sure that all requests to my api is valid json and has X-API-Key header set
[22:58] rauchg: like transports :P
[22:58] dscape: rauchg: awesome. try recorder.rec() to start
[22:58] dscape: writing the code is super boring
[22:58] dscape: i wish someone built a tool on top of nock that would actually do the request periodically and update mocks automatically
[22:59] dscape: right now i just use process.env.NOCK to decide if i nock or not
[22:59] jnbek has joined the channel
[22:59] jesusabdullah: NOCK it off brah!
[22:59] superlou has joined the channel
[23:00] dscape: hkjels: are you building a module?
[23:00] maxogden: a cool property of mikeals request is that when you pipe() a http request to a request() the request() instance inherits all of the properties of the http.request()
[23:01] hkjels: dscape: A website
[23:01] maxogden: it would be fun to write a module that automatically creates mock servers based on that property
[23:01] dscape: maxogden: interesting idea
[23:01] maxogden: then you wouldnt have to monkeypatch
[23:02] maxogden: cause you could just inspect requests and on-the-fly spin up little servers that will catch them
[23:02] dscape: intelligent nc for request in node?
[23:03] maxogden: nc = north carolina?
[23:03] dscape: hkjels: if you arent doing an http client then nock is useless, sorry
[23:03] dscape: netcat
[23:03] zzak has joined the channel
[23:03] maxogden: oh
[23:03] maxogden: hehe right
[23:03] dscape: better than north carolina
[23:03] dscape: :D
[23:03] dscape: cause it has cat
[23:03] maxogden: hahaha
[23:04] maxogden: string.match(/cat/ig) ? return "better" : return "worse"
[23:04] dscape: :)
[23:04] nhunzaker: don't be putting down NC
[23:04] hkjels: I kind of figured. It's not that much to mock, so I'll be fine. But thanks
[23:04] nhunzaker: haha
[23:04] izepstan has joined the channel
[23:04] innoying has joined the channel
[23:05] zzak has joined the channel
[23:06] insin has joined the channel
[23:06] dscape: nhunzaker: putting it in maxogden words "string.match(/cat/ig) ? return "better" : return "worse""
[23:06] nhunzaker: haha
[23:07] maxogden: .. "north carolina".match(/cat/ig) ? return "better" : return "worse"
[23:07] catb0t:  Exception: SyntaxError: Unexpected token return
[23:07] maxogden: .. "north carolina".match(/cat/ig) ? "better" : "worse"
[23:07] catb0t:  "worse"
[23:07] maxogden: hehe
[23:07] gf3 has joined the channel
[23:07] nhunzaker: Well alright
[23:07] nhunzaker: Max your over in Cal right?
[23:07] dnjaramba has joined the channel
[23:08] maxogden: oakland, yep
[23:08] nhunzaker: Yeah,
[23:08] nhunzaker: we're losing some of our braintrust over here in NC
[23:08] nhunzaker: Cole just flew to Berlin
[23:09] maxogden: a buddy of mine lives in asheville
[23:09] nhunzaker: Yeah I"d kill to live in Asheville again.
[23:09] johnhamelink has joined the channel
[23:09] mraxilus has joined the channel
[23:11] dnjaramba has joined the channel
[23:12] ixti has joined the channel
[23:15] dnjaramba has joined the channel
[23:16] tazrover has joined the channel
[23:16] mape: Error: watch ENOSYS on 6.11, debian, https://gist.github.com/d441c0c338300d6acab4 is this a known issue?
[23:18] wink___ has joined the channel
[23:18] nicholasf has joined the channel
[23:19] SvenDowideit has joined the channel
[23:19] diva has joined the channel
[23:19] CrypticSwarm has joined the channel
[23:20] tkahn6_ has joined the channel
[23:20] blup has joined the channel
[23:22] isaacs__ has joined the channel
[23:26] dnjaramba has joined the channel
[23:27] sandfox: anyone have any insight on this… each time i run this code snippet it produces different values for memory consumption… https://gist.github.com/1866451  (It's late for me and I might be over-tired and therefore being daft). How can this be non-deterministic?
[23:29] kirbysayshi has joined the channel
[23:29] overthemike has joined the channel
[23:30] sebicas has joined the channel
[23:30] Topcat: I only have nodeup and node bits bookmarked what else can I browse? anything phigets related would be cool
[23:31] maxogden: whats a phiget
[23:32] dscape: Topcat: nodetuts.com
[23:32] mandric has joined the channel
[23:32] dscape: howtonode is better than node bits btw
[23:32] dscape: imho
[23:33] SubStack: http://reddit/r/node
[23:33] dscape: planetnodejs
[23:34] Topcat: maxogden: a phidget is like er, a programmable usb hardware interface board. http://www.phidgets.com/ -- I saw someone run node on it off reddit i think it was.
[23:34] daniel: is there any way to detect if an object is an arguments object?
[23:34] Topcat: thanks for the sites guys
[23:34] rauchg has joined the channel
[23:35] daniel: oh, you’re not even allowed to use the argument object outside of a single funtion...
[23:36] daniel: except you do that all the time: e.g., foo.apply(null, arguments)
[23:36] jldbasa has joined the channel
[23:37] dscape: daniel if working with arguments dont use for in loops
[23:37] dscape: lol
[23:37] dnjaramba has joined the channel
[23:37] daniel: basically what I want is an `if (foo instanceof Array)` that also detects arguments objects
[23:38] tkaemming has joined the channel
[23:38] dscape: arguments is not an array though daniel 
[23:38] dscape: {'0': 'something'}
[23:38] daniel: dscape: exactly
[23:38] daniel: dscape: which is why I was asking for another way to detect it
[23:39] dscape: daniel: what is it that you are trying to do?
[23:39] Sly__: var args = []; for (var key in arguments) this.args.push(argument[key]);
[23:39] dscape: if you want you can convert it to an array
[23:39] Sly__: Er.
[23:39] Sly__: *args.push
[23:39] Sly__: Not this.args
[23:39] daniel: dscape: detect whether I was passed an array and also have it work for argument objects
[23:39] dscape: [].slice.call(arguments,0)
[23:39] Sly__: /afk
[23:39] dscape: ?
[23:40] dscape: is that what ur looking for?
[23:40] daniel: I want people to be able to call my function with an arguments object and have it work like an array
[23:40] ts__ has joined the channel
[23:41] daniel: I want my function to do X if passed an array or an arguments object, or Y otherwise
[23:42] russfrank: daniel: check if it has a length property. if so, treat it like an array
[23:42] SubStack: Array.isArray(obj)
[23:42] russfrank: (or you could check if it has 'length' and '0') 
[23:42] SubStack: why do you want to pass around an arguments object?
[23:42] daniel: russfrank: yeah, that’s probably best
[23:42] Qzen has joined the channel
[23:43] SubStack: you can just construct an array of the arguments dynamically
[23:43] daniel: russfrank: can’t check "0"; it could be empty
[23:43] russfrank: daniel: well then you do nothing anyway
[23:43] SubStack: .. (function () { return Object.prototype.toString.call(arguments) })()
[23:43] catb0t:  "[object Arguments]"
[23:43] russfrank: SubStack: probably designing some api and he wants it to be versatile 
[23:43] SubStack: just var args = [].slice.call(arguments)
[23:43] russfrank: i'm sure he's not thinking man i'd love to pass around some arguments objects today
[23:43] daniel: russfrank: not in this case :-)
[23:43] SubStack: and you get a regular array
[23:44] SvenDowideit has joined the channel
[23:44] rauchg has joined the channel
[23:44] SubStack: .. (function () { return [].slice.call(arguments) })('a', 'b', 'c')
[23:44] catb0t:  [ "a", "b", "c" ]
[23:44] jazzycha_ has joined the channel
[23:44] daniel: SubStack: but I don’t *want* to have to do that all over the place
[23:44] daniel: SubStack: just like I don’t bother converting before calling `foo.apply(null, arguments)`
[23:45] sven_oostenbrink has joined the channel
[23:45] dscape: daniel: you dont need to
[23:45] dscape: https://github.com/dscape/p/blob/master/pattern.js#L36
[23:45] daniel: dscape: ?
[23:46] dscape: f.apply(this, [].slice.call(arguments,0)) works
[23:46] patcito has joined the channel
[23:46] daniel: `f.apply(this, arguments)` works too
[23:46] daniel: and is shorter
[23:47] dscape: im just saying u dont need to convert back :)
[23:47] saikat has joined the channel
[23:47] pekim_ has joined the channel
[23:47] daniel: dscape: sorry man, but you just don’t seem to understand what I’m talking about :-)
[23:47] mikeric has joined the channel
[23:48] daniel: whereas russfrank got it immediately
[23:48] SvenDowideit has joined the channel
[23:48] copongcopong has joined the channel
[23:49] dnjaramba has joined the channel
[23:49] ramitos has joined the channel
[23:49] daniel: dscape: please don’t DM :-)
[23:51] samstefan has joined the channel
[23:51] tazrover has joined the channel
[23:51] daniel: ah, got it! :D `{}.toString(foo) === "[object Arguments]"` (cc russfrank)
[23:51] daniel: sorry, `{}.toString.call(foo) === "[object Arguments]"`
[23:52] daniel: awesome
[23:53] Destos has joined the channel
[23:54] sorensen__ has joined the channel
[23:54] devaholic: ... {}.toString.call(foo) === "[object Arguments]
[23:54] devaholic: .. {}.toString.call(foo) === "[object Arguments]"
[23:54] catb0t:  Exception: SyntaxError: Unexpected token .
[23:54] wink_ has joined the channel
[23:55] daniel: .. ({}).toString.call(foo)
[23:55] catb0t:  Exception: ReferenceError: foo is not defined
[23:55] daniel: .. ({}).toString.call((function () { return arguments })())
[23:55] catb0t:  "[object Arguments]"
[23:57] dnjaramba has joined the channel