[00:01] caolanm_ has joined the channel
[00:02] dguttman_ has joined the channel
[00:02] brettgou_ has joined the channel
[00:03] creationix has left the channel
[00:03] springmeyer has joined the channel
[00:03] niftylettuce: Substack: this doesnt seem right though... it has the browserify JS twice... http://pastie.org/private/nfgfb1bvbe0rpiw8izdw
[00:04] niftylettuce: Substack: when the files r called, JS is redundant
[00:07] a|i_ has joined the channel
[00:07] a|i_ has joined the channel
[00:11] josephboyle has joined the channel
[00:12] brettgoulder has joined the channel
[00:13] spadin has joined the channel
[00:13] azeroth__ has joined the channel
[00:13] dnolen has joined the channel
[00:13] AAA_awright_ has joined the channel
[00:16] kriszyp has joined the channel
[00:17] tylerstalder has joined the channel
[00:19] ryanj has joined the channel
[00:19] a|i: how does node-xmpp work? do I need to run an xmpp server, or is it standalone?
[00:21] k1ttty has joined the channel
[00:21] JianMeng has joined the channel
[00:22] yukihr has joined the channel
[00:24] tonymilne has joined the channel
[00:25] thalll has joined the channel
[00:25] littlemagician has joined the channel
[00:27] dcelix has joined the channel
[00:29] mscdex: a|i: seems like it's a server itself
[00:29] a|i: mscdex: xmpp is a protocol, not a server, I think.
[00:29] mscdex: xmpp server
[00:29] romainhuet has joined the channel
[00:30] mscdex: a|i: https://github.com/astro/node-xmpp/blob/master/lib/xmpp/server.js
[00:30] mscdex: i've never used it, so i can't tell you 100% for sure
[00:31] rytas has joined the channel
[00:31] a|i: seems you're right.
[00:31] mscdex: client and server it seems
[00:34] puffpio: hi guys..i was wondering what the latest hotness is in terms of running a node daemon
[00:34] puffpio: i have just been using upstart..but i was wondering what else is out there since i haven't really been looking at that area
[00:38] tobias has joined the channel
[00:38] mikeric has joined the channel
[00:38] tobias: tmpvar, you there?
[00:41] davidcoallier has joined the channel
[00:41] davidcoallier: tswicegood: ping
[00:41] gtramont1na has joined the channel
[00:43] neoesque has joined the channel
[00:43] puffpio has left the channel
[00:44] cloudhea1 has joined the channel
[00:44] cloudhea1_ has joined the channel
[00:44] EyePulp has joined the channel
[00:48] xsyn has joined the channel
[00:48] pifantastic has joined the channel
[00:49] jkridner has joined the channel
[00:51] draginx has joined the channel
[00:51] draginx: Anyone know how to run coffee script inside of nodejs?
[00:51] draginx: #coffeescript = dead
[00:51] slloyd has joined the channel
[00:52] zcopley has joined the channel
[00:56] ezl has joined the channel
[00:57] Determinist has joined the channel
[00:58] jesusabdullah: draginx: coffee proggy.coffee
[00:58] draginx: ty
[00:58] justinTNT: puffpio : forever
[00:58] jesusabdullah: npm should give you this executable when you install cS
[00:58] brownies has joined the channel
[00:59] theblob has joined the channel
[01:00] LowValueTarget has joined the channel
[01:01] Guest78954: wasnt there a lib for that
[01:01] thalll has joined the channel
[01:01] Marak has joined the channel
[01:01] jesusabdullah: a lib for what, Guest74518 ?
[01:01] Marak has left the channel
[01:01] deedubs: tjholowaychuk: is there a way to dump routes for an express app? trying to debug a express-resource issue
[01:01] Tobsn: https://github.com/mauricemach/zappa
[01:02] Tobsn: jesusabdullah coffeescript
[01:02] tjholowaychuk: deedubs haha :p see arent high level abstractions fun!
[01:02] Tobsn: lol
[01:02] jesusabdullah: ah
[01:02] tjholowaychuk: regular routing ftw
[01:02] jesusabdullah: well yeah npm install coffeescript
[01:02] tjholowaychuk: deedubs but yeah https://github.com/visionmedia/express-trace
[01:02] Tobsn: yeah
[01:02] jesusabdullah: or is it coffee-script ? EITHER WAY
[01:02] Determinist: ACTION lighst a smoke and screams inside his head
[01:02] Tobsn: i dont see the benefits either way ;)
[01:03] tjholowaychuk: deedubs that might help, it's not even remotely finished and the output only looks good when you have a single request via curl etc but looks nasty with assets etc
[01:03] Tobsn: of coffeescript
[01:04] deedubs: tjholowaychuk: ahh so that'll show me how a request traverses express. Problem is I don't know if the nested route is being set properly and don't know the url to hit :D
[01:04] tjholowaychuk: hm
[01:04] tjholowaychuk: well there is no nesting really
[01:04] tjholowaychuk: express-resource stuff just expands to regular app.get() etc calls
[01:04] hij1nx has joined the channel
[01:07] samsonjs has joined the channel
[01:08] alek_br_ has joined the channel
[01:08] deedubs: tjholowaychuk: Yeah just looking at the forums example with forums.add
[01:09] zcopley has joined the channel
[01:09] flippyhead has joined the channel
[01:10] abraxas has joined the channel
[01:10] yhahn has joined the channel
[01:15] devrim has joined the channel
[01:19] devrim has joined the channel
[01:21] devrim has joined the channel
[01:23] Tobsn: is anyone working with cluster?
[01:23] Tobsn: with socket.io
[01:23] Tobsn: cluster( server ).use( bridge( socket ) ).listen( 3000 ); - cause that throws an error that it cant use one of it as function
[01:24] tonymilne: Anyone know any way i can listen for Mongoose model save events? E.g. for model Example, every time an example instance is saved, my event listener runs.
[01:24] mikeal has joined the channel
[01:25] RyanD has joined the channel
[01:25] tonymilne: (The idea being, that I can socket new examples to the user ... or is this something that I should do myself at a higher up level in my code?)
[01:26] F1LT3R has joined the channel
[01:26] sebbie has joined the channel
[01:26] Yuffster_work has joined the channel
[01:28] Tobsn: tjholowaychuk, still there?
[01:29] mikeal: isaacs: how do i link globally?
[01:29] niftylettuce: he is watering his garden of goodness
[01:29] isaacs: npm link
[01:29] Tobsn: -g
[01:29] Tobsn: oh
[01:29] isaacs: mikeal: all links are global
[01:29] Tobsn: https://gist.github.com/d7e3f5329e3c2c9583bb can someone tell me if that is proper
[01:29] Tobsn: cause its not working...
[01:29] isaacs: mikeal: when you do `npm link express`, it installs express globally, then links it locally
[01:30] jtsnow has joined the channel
[01:30] isaacs: mikeal: when you do `npm link ../foo` it links it globally, and then links it locally
[01:30] mikeal: i'm in the directory
[01:30] isaacs: yep
[01:30] mikeal: i only want to link globally :)
[01:30] isaacs: do `npm link` and it'll be "installed" globally
[01:30] isaacs: in /usr/local/lib/node_modules
[01:30] jmoyers: Tobsn: looks proper
[01:30] jacter has joined the channel
[01:31] mikeal: % npm link ../../git/morestreams ~/Documents/slides/node-intro
[01:31] mikeal: /usr/local/lib/node_modules/morestreams -> /Users/mikeal/Documents/git/morestreams
[01:31] mikeal: npm ERR! Error: ENOENT, No such file or directory 'morestreams@0.0.1'
[01:31] mikeal: wtf?
[01:31] dyer has joined the channel
[01:32] avalanche123 has joined the channel
[01:32] tonymilne: Tobsn: do you require connect?
[01:32] davidcoallier: mikeal: I assume you aren't running on Windows? :)
[01:32] Tobsn: not really
[01:33] Tobsn: well it looks like its working
[01:33] Tobsn: but if i telnet on 8000 and send a single string the client gets disconnected
[01:33] Tobsn: and no console.log() appears
[01:33] jmoyers: uh
[01:33] tonymilne: socket.io works by connecting from the client JS to the server.
[01:33] jmoyers: im sure there is a handshake
[01:33] tonymilne: you kinda should be using their protocol.
[01:34] Tobsn: oooooh
[01:34] jmoyers: telnet isn't going to do the trick for connecting via ws
[01:34] Tobsn: i see
[01:34] mike5w3c_ has joined the channel
[01:34] maqr has joined the channel
[01:34] aaronblohowiak has joined the channel
[01:34] jmoyers: also, socket.io introduces heartbeat and some other business
[01:35] aaronblohowiak: SubStack: THANK YOU for making base work in package.json
[01:35] ByteCrunch has joined the channel
[01:35] jmoyers: so unless you have a client that does that stuff, it'll disconnect after 10s~
[01:35] Tobsn: yeah i see
[01:35] Tobsn: havent thought about that
[01:35] tonymilne: Whats your big picture Tobsn?
[01:36] stellar678 has joined the channel
[01:36] Tobsn: oh just messing around
[01:36] Tobsn: for now
[01:36] draginx has left the channel
[01:36] stellar678: hello everybody
[01:37] stellar678: !
[01:37] tonymilne: hi stellar :)
[01:37] stellar678: how goes?
[01:38] tonymilne: Good. Whatsup?
[01:40] stellar678: newish at playing around with nodejs, and figured i might come ask the pros some questions instead of guessing and endlessly googling!
[01:40] stellar678: which is of course the standard
[01:42] eliziario has joined the channel
[01:43] stellar678: i've got a nodejs app that is pretty simple, loads i think 8 libraries, and it amazes me that it shoots to the top of the list for virtual mem usage right after i start it
[01:44] Tobsn: nice
[01:44] Tobsn: works now
[01:44] Tobsn: thanks tonymilne
[01:44] fr0stbyte has joined the channel
[01:44] Tobsn: brb, driving home ;)
[01:44] Tobsn: btw. what was setPath used for?
[01:45] Tobsn: btw. does socket.io just disconnects after 10 secondsß
[01:45] Tobsn: ah vermind
[01:45] Tobsn: nevermind
[01:45] Tobsn: i killed the node process :P
[01:47] tonymilne: stellar678: It's kind of hard for us to tell you what's chewing your mem without knowing what libs you're using them, and what code you've got in your app.
[01:50] draginx has joined the channel
[01:51] draginx: Whats the best way to ensure people can only grab broadcast data from nowjs within a certain room and not be able to get information until they officially leave one room and enter in the next?
[01:51] draginx: Store things in redis?
[01:51] MikeMakesIt has joined the channel
[01:52] jesusabdullah: no idea :S I use dnode for my rpc needs
[01:54] draginx: interesting
[01:58] niftylettuce: jesusabdullah: u try browserify?
[01:59] SubStack: those are different niches
[02:00] tk has joined the channel
[02:01] SubStack: aaronblohowiak: yeah the codebase is getting really hideous but it passes all its tests so I don't care enough to clean it up yet
[02:01] niftylettuce: SubStack: what is solution to not loading browserify's JS twice.. :/
[02:01] niftylettuce: SubStack: should I modify browserify to support an array of mounts?
[02:02] pandeiro has joined the channel
[02:02] SubStack: niftylettuce: I'm not sure why you'd want to do that
[02:02] jesusabdullah: niftylettuce: I haven't had a chance to try browserify yet.
[02:02] niftylettuce: SubStack: in HTML5 boilerplate, you include modernizr and respond at the top, and then the rest of the JS like jquery etc. at the bottom below
[02:02] SubStack: niftylettuce: just use http redirects
[02:03] niftylettuce: SubStack: what do u mean?
[02:03] SubStack: niftylettuce: not sure what that is
[02:03] heavysixer has joined the channel
[02:03] SubStack: browserify generates one file. That's not going to change any time soon
[02:03] SubStack: too hard to change
[02:04] aaronblohowiak: SubStack: i call wrap manually =)
[02:04] eyesUnclouded has joined the channel
[02:04] SubStack: aaronblohowiak: oh that's probably a bad idea since I don't have any tests that call wrap directly
[02:05] aaronblohowiak: SubStack: i dont have any problems yet! :-D it lets me debug easier
[02:05] SubStack: yeah debug mode might be useful
[02:05] SubStack: no idea how to make that work though
[02:05] gavin_huang has joined the channel
[02:06] aaronblohowiak: SubStack: ideally you could just pass "skip preamble" to bundle
[02:07] bad_at_math has joined the channel
[02:07] aaronblohowiak: SubStack: this way you could use the publicly supported bundle interface multiple times. additionally, you'd probably want to have a "preample" function that would return the string of just the shim and require definition stuff.
[02:07] isaacs has joined the channel
[02:07] aaronblohowiak: this wouldn't avoid duplication in the strings that bundle() calls produce but it would let you have some finer-grained control of the output
[02:07] draginx has left the channel
[02:08] pandeiro: is 'npm search ...' working?
[02:08] aaronblohowiak: (without requiring a complicated interface for having browserify produce mutliple files)
[02:08] jacobolus has joined the channel
[02:08] gavin_huang has joined the channel
[02:08] littlemagician has left the channel
[02:08] Spion has joined the channel
[02:08] SubStack: could work
[02:08] dgathright_ has joined the channel
[02:08] gavin_huang has joined the channel
[02:08] kriszyp has joined the channel
[02:09] niftylettuce: so basically a check if var is true or false and if its true then dont include preamble
[02:09] aaronblohowiak: SubStack: right now, using browserify to share *application code* is really tricky.
[02:10] aaronblohowiak: SubStack: it would basically be putting this in an if(){} block: https://github.com/substack/node-browserify/blob/master/index.js#L57-60
[02:10] SubStack: aaronblohowiak: what is tricky about it?
[02:11] aaronblohowiak: SubStack: I'd have to create package.json files for my application code
[02:11] SubStack: wha?
[02:12] mikeal has joined the channel
[02:12] SubStack: you can just browserify.bundle(__dirname + '/code'), no package.jsons required
[02:12] SubStack: or likewise with browserify() if you want a middleware
[02:12] aaronblohowiak: SubStack: that only works if i *just* want the code in one directory
[02:13] galaxywatcher has joined the channel
[02:13] SubStack: well it's a recursive traversal
[02:13] aaronblohowiak: SubStack: no multiple root directories
[02:13] aaronblohowiak: SubStack: also, sometimes i want a dependency to vary between server and client contexts.. like data_store.js =)
[02:13] SubStack: ah well you can use base : { foo : 'somedir', bar : 'someotherdir' } now
[02:13] aaronblohowiak: SubStack: i know this might be out of scope of browserify
[02:14] aaronblohowiak: SubStack: yes, which is why i thanked you so hard =)
[02:14] SubStack: it also supports node_modules
[02:14] colinsullivan has joined the channel
[02:14] SubStack: so you can npm install somemodule@6.6.6 in your base'd somedir
[02:14] ryanallenbobcat has joined the channel
[02:14] tilgovi has joined the channel
[02:14] tilgovi has joined the channel
[02:14] colinsullivan has left the channel
[02:15] SubStack: and the browser-side resolver will be able to pick it up
[02:15] gmanika has joined the channel
[02:15] sonnym has joined the channel
[02:16] SubStack: just don't pass that dep to the require field and you're good so long as it's in node_modules
[02:16] aaronblohowiak: SubStack: cool. that isnt something i'm currently concerned with..
[02:19] Swimming_bird has joined the channel
[02:20] aaronblohowiak: SubStack: https://gist.github.com/1be0cb795fa435ae6add that is how i am using browserify today
[02:20] aaronblohowiak: https://gist.github.com/raw/1be0cb795fa435ae6add/fd0fd2c42e53865805650383922eb3d47a15b8cd/gistfile1.txt - with word wrap
[02:22] pifantastic has joined the channel
[02:26] mischievious has joined the channel
[02:30] mischief has joined the channel
[02:33] niftylettuce: aaronblohowiak: what is wrap()?
[02:33] neoesque has joined the channel
[02:33] aaronblohowiak: niftylettuce: black magic voodoo. not supported =)
[02:33] niftylettuce: ...
[02:33] aaronblohowiak: https://github.com/substack/node-browserify/blob/master/index.js#L155
[02:34] niftylettuce: example use
[02:34] niftylettuce: ?
[02:34] aaronblohowiak: niftylettuce: it is an internal browserify function
[02:34] niftylettuce: is there doc on usingi t?
[02:34] niftylettuce: what does it do
[02:34] aaronblohowiak: niftylettuce: no, it iss not supported
[02:34] stepheneb has joined the channel
[02:34] niftylettuce: oh
[02:35] aaronblohowiak: niftylettuce: it is the internal implementation of taking a library name and returning the string of that library and its dependencies
[02:35] niftylettuce: ah
[02:35] niftylettuce: aaronblohowiak: did u have any advice on how I can remain consistent with html5 boilerplate? basically you have modernizr and respond javascript included in the
and then before you include javascript like on ready states or jquery functions
[02:36] Emmanuel has joined the channel
[02:36] aaronblohowiak: niftylettuce: no, sorry
[02:36] niftylettuce: aaronblohowiak: I guess I can just include them as minified JS files
[02:37] Emmanuel has joined the channel
[02:38] gtramont1na has joined the channel
[02:40] amaudy has joined the channel
[02:41] lakin has joined the channel
[02:42] brownies has joined the channel
[02:43] unomi has joined the channel
[02:43] Transformer has joined the channel
[02:46] Tobsn has joined the channel
[02:49] mynyml has joined the channel
[02:51] [[zz]] has joined the channel
[02:53] yukihr has joined the channel
[02:56] mikeal has joined the channel
[02:58] seivan has joined the channel
[03:01] dipser_ has joined the channel
[03:03] Corren has joined the channel
[03:03] e6nian has joined the channel
[03:05] vikstrous: is it just me or does vows -w not seem to work the way it should?
[03:06] brimster has joined the channel
[03:06] bbttxu has joined the channel
[03:06] vikstrous: it seems like every time i change a test or the app it always uses the new test but it never uses the new app..
[03:07] mikeric has joined the channel
[03:07] k1ttty has joined the channel
[03:07] unomi has joined the channel
[03:08] EyePulp has joined the channel
[03:09] yukihr has joined the channel
[03:09] dget_ has joined the channel
[03:11] iFire has joined the channel
[03:11] zeade has joined the channel
[03:12] defstream_ has joined the channel
[03:12] caseywstark has joined the channel
[03:13] harth has joined the channel
[03:16] isaacs has joined the channel
[03:16] Dinosaurus has joined the channel
[03:17] jacobolus has joined the channel
[03:18] dnolen has joined the channel
[03:18] xsyn has joined the channel
[03:21] defstream has joined the channel
[03:21] Aria has joined the channel
[03:22] defstream: anybody familiar with everyauth?
[03:22] defstream: i keep getting this Error: Step rememberTokenSecret of `twitter` is promising: requestTokenSecret ; however, the step returns nothing. Fix the step by returning the expected values OR by returning a Promise that promises said values.
[03:26] RyanD has joined the channel
[03:30] Emmanuel: anybody knows a convenient way to load javascript in the fly with Node ? Say you have a running node server (that you don't want to stop), according to a request, the server should load some external JS
[03:30] Emmanuel: (so, without restarting the server, we can change it's behaviour)
[03:30] Aria: Read and eval? use the Script object?
[03:31] Emmanuel: I'll take a look at that
[03:31] Emmanuel: (the Script object)
[03:31] Emmanuel: hm. What do you mean by script object actually :)
[03:31] Dinosaurus: is it possible to use sessions in a mongoid scope
[03:31] Dinosaurus: like scope :mine, where(user_id: session[:user][:_id])
[03:32] chjj: Emmanuel, what are you trying to do exactly?
[03:32] Emmanuel: we have a product with basic features
[03:32] superjudge has joined the channel
[03:32] Emmanuel: and we want to extend it "on the fly", without restarting the server
[03:33] Emmanuel: that means typically adding some new JS script and ask the server to load them depending on the request
[03:33] Emmanuel: say, if client requires page X, we load the associated script for page X, that tells the server how to process it
[03:33] Dinosaurus: write everything in assembly and maintain your stack by hand
[03:33] chjj: Emmanuel: it sounds like you might be going about things the wrong way
[03:33] Emmanuel: maybe read and eval is good enough actually
[03:34] vikstrous: Emmanuel: what's the problem with just using requre?
[03:34] Dinosaurus has left the channel
[03:34] Emmanuel: require should do it too
[03:34] chjj: why not have everything loaded in a module to begin with?
[03:34] Emmanuel: (that's a new idea, I'm thinking it out, asking for suggestion)
[03:35] vikstrous: Emmanuel: chjj is right... there's rarely any point of loading only a part of the application
[03:35] vikstrous: maybe if you had different servers that had different purposes and you wanted them to only load what's necessary for their job... but then you might as well have different apps
[03:36] Emmanuel: the app is a proxy server
[03:36] Emmanuel: a user get the page through the server, the server make some change (or provide a brand newly generated page)
[03:36] Emmanuel: and we want to be able to add more features in the feature
[03:36] Emmanuel: future *
[03:37] Emmanuel: like, we have that specific case that require the page to be process in a new way, we didn't expected before
[03:37] Emmanuel: so we had the associated script for that
[03:37] Emmanuel: but maybe just require'ing the associated script according to the request should do it
[03:38] vikstrous: hmm... so you have some kind of a node app hosting platform and don't want to restart the server when you make modifications to the apps hosted?
[03:40] dguttman has joined the channel
[03:40] justinTNT: Emmanuel : I've got some ideas about how you should do that.
[03:41] Emmanuel: vikstrous: yes
[03:41] justinTNT: I reckon for the kind of functionality you're describing, the way to go ahead is to setup a handler on a config port.
[03:41] Emmanuel: because we /know/ we will have more features, or more exceptions, or more hacky stuff to add in the future
[03:41] justinTNT: listen on that config port for connections from a utility, that sends various runtime config commands and settings to the main app.
[03:41] Emmanuel: we just can't think of everything ahead
[03:41] justinTNT: including one to load new modules from a preconfigured path.
[03:42] rchavik has joined the channel
[03:42] justinTNT: this is the method I'm using to add new sites to my webserver without restarting.
[03:43] Silly_Wabbit has joined the channel
[03:43] rchavik has joined the channel
[03:43] Emmanuel: that sounds like a good idea
[03:43] Dinosaurus has joined the channel
[03:44] wayne530 has joined the channel
[03:44] timmywil has joined the channel
[03:45] niftylettuce has joined the channel
[03:47] Emmanuel: thanks people !
[03:48] febits has joined the channel
[03:48] sub_pop has joined the channel
[03:51] Bradleymeck has joined the channel
[03:51] htoothrot has joined the channel
[03:54] jakehow has joined the channel
[03:54] superjudge has joined the channel
[03:58] eyesUnclouded has joined the channel
[03:59] tilgovi has joined the channel
[03:59] tilgovi has joined the channel
[03:59] ryah has joined the channel
[04:02] temp01 has joined the channel
[04:03] criswell has joined the channel
[04:03] matyr has joined the channel
[04:06] shanez has joined the channel
[04:06] philtor has joined the channel
[04:07] ryah has joined the channel
[04:07] dhasenan: Any suggestions for a UI/RIA framework?
[04:08] themiddleman has joined the channel
[04:08] wayne530 has left the channel
[04:10] devrim has joined the channel
[04:12] azeroth__ has joined the channel
[04:12] jmoyers has joined the channel
[04:12] temp01 has joined the channel
[04:13] schulz has joined the channel
[04:14] schulz: hey quick question: I'm looking at this page: https://github.com/joyent/node/wiki/modules#web-frameworks-routers which router should I use?
[04:14] JackeyChan has joined the channel
[04:15] jmoyers: schulz: whats the project entail?
[04:15] schulz: me screwing around with node.js
[04:15] jmoyers: heh
[04:15] jmoyers: express
[04:16] jmoyers: https://github.com/visionmedia/express - simple sinatra like syntax, plugins etc etc
[04:16] Me1000 has joined the channel
[04:16] schulz: sweet, thanks a lot
[04:17] Aria: Yeah, for sure. Express is pretty delightfully light.
[04:18] blkcat: express is a thing of beauty
[04:19] zhangkaizhao has joined the channel
[04:21] brownies: building a node.js app *without* express just seems masochistic
[04:21] danfo has joined the channel
[04:21] jmoyers: i wouldn't go that far ;-)
[04:22] brownies: jmoyers: what're some other "good" approaches?
[04:22] mikeric has joined the channel
[04:22] brownies: ACTION is a node noob
[04:22] danfo: hi - what's the 'best' method to get mongodb data into a dedicated full-text search service (solr/sphinx/elasticsearch/whatever)?
[04:22] jmoyers: it depends on what you're doing
[04:22] jmoyers: node is good for a class of application
[04:22] jmoyers: not just straight rest-based web apps
[04:22] jmoyers: its good for data heavy real-time stuff
[04:23] brownies: like web chat?
[04:23] jmoyers: you can see interesting projects that have nothing to do with http servers, like mikeal's spider or things that are entirely client side like sammy.js
[04:23] jmoyers: not ENTIRELY client side
[04:24] jmoyers: i suppose sammy doesn't even have anything to do with node per se :: lines of javascript are blurring mind aaaaah ::
[04:24] slloyd_ has joined the channel
[04:24] brownies: heh
[04:24] jmoyers: you can also build straight udp/tcp servers
[04:25] dhasenan: The sammy.js website runs on sinatra. Dogfood much?
[04:26] jmoyers: wha?
[04:26] jmoyers: its sinatra inspired syntax
[04:26] jmoyers: not sinatra
[04:26] dhasenan: http://sammyjs.org/ -- look at the last line.
[04:27] dhasenan: "This site is powered by Sinatra and is 100% Open Source"
[04:27] jmoyers: oh, the website
[04:27] yukihr has left the channel
[04:27] yukihr has joined the channel
[04:28] jmoyers: as far as i know, wrt sammy
[04:28] jmoyers: its more about client side routing
[04:28] jmoyers: and local storage and such
[04:28] jmoyers: reason it popped into my brain is because i just saw a demo of it last weekend
[04:29] sechrist has joined the channel
[04:29] nilcolor has joined the channel
[04:30] dhasenan: Ah, that's actually kind of strange. Looked a bit like a server-side framework at first.
[04:31] jmoyers: exactly
[04:31] jmoyers: a little mind blowing
[04:31] jmoyers: strange class of application
[04:32] dguttman has joined the channel
[04:32] k1ttty has joined the channel
[04:34] svnlto has joined the channel
[04:37] dhasenan: I'm not sure how it's helpful.
[04:37] dhasenan: I guess it gives you a global command namespace...
[04:38] davidbanham: An EC2 micro instance takes FOREVER to build node.
[04:38] mertonium has joined the channel
[04:38] Emmanuel: davidbanham: so true !
[04:38] nilcolor has joined the channel
[04:39] mscdex: davidbanham: if you're on ubuntu, there's a PPA
[04:39] davidbanham: mscdex: I'm just using the default micro instance image so that it's covered under the free tier. I've used the PPA previously though and it's been great.
[04:40] alcuadrado has joined the channel
[04:41] zhangkaizhao has joined the channel
[04:41] alcuadrado: Hi there! How can I test that an http client is localhost? ipv6's localhost "::1" is not the unique way of representing that
[04:42] ryanallenbobcat has joined the channel
[04:42] reid has joined the channel
[04:42] ryanallenbobcat has left the channel
[04:42] JoshC has joined the channel
[04:45] kaichenxyz has joined the channel
[04:45] stagas has joined the channel
[04:46] mscdex: alcuadrado: hmm, well one way would be to use dns.reverse(), but it seems it only returns localhost/loopback for ipv6 addresses... for 127.0.0.1 it just returns an empty array for the hostnames
[04:46] mscdex: kind of strange
[04:46] mscdex: :S
[04:47] olympum has joined the channel
[04:47] alcuadrado: request.connection.remoteAddress has the necessary info, but I don't know the correct way for checking it's content
[04:48] alcuadrado: as it could have, for example, ::1 or :0:1 for localhost
[04:48] alcuadrado: afaik
[04:48] mscdex: right
[04:48] mscdex: well, you can get the list of network adapters on the machine from node and their addresses and if they are internal or not
[04:48] mscdex: maybe you can just parse that list for internal addresses and then just compare them on connection
[04:49] mscdex: oh wait, nevermind that's only available in the master branch
[04:49] mscdex: sorry ><
[04:49] alcuadrado: I think making a generic ipv6 compare algorithm would be more useful
[04:49] alcuadrado: oh :( haha
[04:49] alcuadrado: I'm with 4.7
[04:50] mscdex: it'll be in 0.6 though! ;)
[04:50] alcuadrado: v0.5.0-pre sorry
[04:50] mscdex: oh
[04:50] alcuadrado: strange, I thought I was in 4.7 :s
[04:50] mscdex: if you're on v0.5.0-pre then you may have that capability
[04:52] mscdex: alcuadrado: see here: https://github.com/joyent/node/blob/master/doc/api/os.markdown
[04:52] alcuadrado: but 0.4.7 on the server :p
[04:52] mscdex: ah ok
[04:52] alcuadrado: I'd have to learn about the alternative ways to write ipv6 ips and make a generic algorithm
[04:52] alcuadrado: that would be cool :)
[04:53] alcuadrado: it's a shame my deadline is.. tomorrow haha
[04:53] mscdex: i don't think you can really write an algorithm
[04:54] mscdex: because you can multiple ipv6 loopback addresses
[04:54] mscdex: s/can/can have/
[04:55] mscdex: i think so anyway...
[04:55] alcuadrado: well, maybe I asked for the wrong problem. I'm making an API which has a whitelist of authorized ips (currently only localhost)
[04:55] jaket has joined the channel
[04:55] alcuadrado: so what I would need in fact is a more generic ipv6 comparison
[04:56] davidwalsh has joined the channel
[04:56] gf3: howdy
[04:56] Swimming_bird has joined the channel
[04:56] fakewaffle has joined the channel
[04:57] fakewaffle: i read somewhere not to use for (var i in data) {....} and instead use Object.keys(data).forEach(function(key) {...}) ? What is the difference?
[04:58] mscdex: alcuadrado: here's what you can check for ipv6: http://www.freebsd.org/doc/handbook/network-ipv6.html#AEN42407
[04:58] SubStack: Object.keys() only does enumerable properties
[04:58] alcuadrado: thanks mscdex !!!
[04:58] temp01 has joined the channel
[04:59] SubStack: oh hmmm, for in does that too
[04:59] SubStack: maybe for..in looks at the prototype chain?
[04:59] gf3: fakewaffle: it's faster
[04:59] chjj: for...in gets all properties in the prototype chain, but it doesnt get non-enumerables
[04:59] SubStack: yeah that's it
[04:59] fakewaffle: ok, thanks guys
[05:00] gf3: also, the faster part
[05:00] chjj: if you really want to see ALL the properties, you would do for...in + getOwnProperties
[05:00] indutny has joined the channel
[05:00] mscdex: 0wnedProperties
[05:01] brownies has joined the channel
[05:01] jmoyers: i thought thats what enumerable meant in this context
[05:01] jmoyers: cant create an iterator to walk it
[05:01] jmoyers: i mean, you can manually
[05:01] chjj: i actually think its called getOwnPropertyNames
[05:01] chjj: but whatever
[05:01] jmoyers: but not by the default language constructs like for/in
[05:02] reid has joined the channel
[05:05] temp01 has joined the channel
[05:06] sechrist_ has joined the channel
[05:06] vikstrous: guys i'm writing a module and in this module there's a function which is asynch and it can have optional parameters... so i need to decide what order i should put the parameters in. should i put the callback first? i've always seen the callback being last but i need space for my optional parameters... maybe i should pass all parameters as a single object?
[05:06] amaudy has joined the channel
[05:06] mischief has joined the channel
[05:07] jmoyers: options hash, callback last imo
[05:07] SubStack: vikstrous: var args = [].slice.call(arguments)
[05:07] SubStack: var cb = args.slice(-1)[0]
[05:07] SubStack: or even:
[05:08] SubStack: var cb = args.filter(function (arg) { return typeof arg === 'function' })[0]
[05:08] SubStack: any order
[05:08] vikstrous: so having both optional arguments and callback last is good?
[05:08] flippyhead has joined the channel
[05:08] jmoyers: Sub's method allows you to have any number of optional parameters.
[05:09] vikstrous: yeah i just wasn't sure if that's a good way of doing it or something that people would expect
[05:09] jmoyers: lots of modules have variable method signatures
[05:09] jmoyers: i like the options hash because you can use _.defaults to populate it
[05:09] SubStack: vikstrous: you can get pretty fancy https://github.com/substack/dnode-protocol/blob/master/index.js#L221-L264
[05:09] jmoyers: rather than doing a bunch of arguments object slicing
[05:10] jmoyers: also SubStack, do you know if accessing arguments is 'slow'? I overheard something of the sort
[05:10] temp01 has joined the channel
[05:11] vikstrous: i think i'll go with the object with parameters in any case because it's easier to read what's what... better than having like 3 parameters +3 optional ones
[05:11] SubStack: jmoyers: I don't listen to those people.
[05:11] SubStack: optimize the pieces that are measurably slow and only after you've written it the nice way first
[05:12] dguttman has joined the channel
[05:12] SubStack: reasoning about performance is doubly hard because the v8 engine is doing all sorts of crazy unpredictable optimizations itself
[05:13] mikeric has joined the channel
[05:13] jmoyers: nod, i agree -- its similar to arguments about scope resolution imo
[05:13] jmoyers: we're talking about 1 extra lookup
[05:13] cloudhead has joined the channel
[05:14] vikstrous: wait, how did performance come up?
[05:14] superjudge has joined the channel
[05:14] SubStack: arguments
[05:14] vikstrous: oh
[05:16] mitko has joined the channel
[05:16] AdmGre has joined the channel
[05:16] Adman65 has joined the channel
[05:17] AdmGre has joined the channel
[05:17] ryah has joined the channel
[05:18] mykul has joined the channel
[05:19] seivan has joined the channel
[05:20] aaronblohowiak: SubStack: args.pop()
[05:22] Zarathust: I don't get why I have a second executable "node.io"
[05:23] jmoyers: node.io as in the scraping thing?
[05:23] Zarathust: yes
[05:23] Zarathust: i can run node modules correctly and node.io related code correctly
[05:23] jmoyers: because it wraps up its functionality in a command line utility
[05:24] Zarathust: but when I combine both, nothing works
[05:24] jmoyers: im a little confused by what you mean
[05:24] jmoyers: node.io can be require()'d in a general node script
[05:24] jmoyers: there is also node.io as a cli tool which can be used to do scraping from the shell
[05:25] jmoyers: shouldn't need to combine the two
[05:25] jacobolus has joined the channel
[05:25] Zarathust: yes, I think the problem is when I require('node.io') in a script ran from node
[05:25] Zarathust: require('node.io') only works when invoked with node.io
[05:25] jmoyers: error?
[05:25] jmoyers: what happens
[05:26] Zarathust: I'm trying to scrape stuff and store it in a mysql datable
[05:27] mitko has joined the channel
[05:29] Zarathust: hum... I think I found the problem, it was rather unrelated
[05:29] tikva has joined the channel
[05:29] boehm has joined the channel
[05:33] SamuraiJack has joined the channel
[05:34] ryah has joined the channel
[05:34] fbzhong has joined the channel
[05:34] context_ has joined the channel
[05:35] wonginator1221: Howdy. I'm trying to "glue" a java program that spits numbers out with a node server. Is there a recommended way to do this? I was originally going to write the server in java before I realized how much of a mess that would be having never written a server from scratch before. Should I try to send 'messages' from the java program to node, or should I try to use child_processes? Thanks.
[05:36] context has joined the channel
[05:36] wonginator1221: P.S. This is my first time trying to put two different programming languages together.\
[05:36] temp01 has joined the channel
[05:37] dguttman has joined the channel
[05:37] admc has joined the channel
[05:38] mertonium has joined the channel
[05:39] fbzhong has joined the channel
[05:39] mscdex: wonginator1221: why not do it all in node?
[05:40] devrim has joined the channel
[05:40] fbzhong: Hi, could anyone help me on this issue of node.js? https://github.com/joyent/node/issues/1105
[05:41] asabil has joined the channel
[05:43] mscdex: fbzhong: google isn't defined....
[05:43] mscdex: in your sandbox you only have 'window' defined
[05:44] aaronblohowiak: fbzhong: might be a jsdom issue
[05:44] indutny: fbzhong: you run code in different contexts
[05:44] context: it looks like all he did was merge the two runInNewContext() lines together
[05:44] wonginator1221: mscdex: well I'm somewhat dependent on a jar for the java side of it.
[05:44] fbzhong: actually it run in the same sandbox
[05:45] wonginator1221: Additionally, how is windows support for node?
[05:45] Bonuspunkt: fbzhong: try window.google.test = "test"
[05:45] fbzhong: Yes, window.google.test = "test" is ok.
[05:45] aaronblohowiak: fbzhong: because "
[05:45] Skola has joined the channel
[05:45] aaronblohowiak: window is not the global scope in the new context
[05:46] fbzhong: how to set window as global scope in the new context?
[05:46] fbzhong: I think the sandbox.__proto__ = window, the sandbox is global scope, the window is included as well.
[05:48] Zarathust: when executing with node instead of node.io, this line is not executed : exports.job = new nodeio.Job({timeout:10}, methods);
[05:49] Bonuspunkt: fbzhong if i remember this right, just remove the window - sandbox should be the global scope
[05:49] Zarathust: The line is probably executed, but my job doesn't start
[05:51] context: hmm
[05:51] context: fbzhong: is this possibly.... (if v8 does this) moving var definitions to the top of the script, and possibly doing it in alphabetical order or something ?
[05:51] fbzhong has joined the channel
[05:51] context: fbzhong: is this possibly.... (if v8 does this) moving var definitions to the top of the script, and possibly doing it in alphabetical order or something ?
[05:51] context: im fairly new to js and node but :x
[05:51] brownies has joined the channel
[05:51] sivy has joined the channel
[05:53] Zarathust: How can I start my nodeio job?
[05:53] __tosh has joined the channel
[05:55] fangel has joined the channel
[05:55] fbzhong: Bonuspunkt:
[05:56] fbzhong: Bonuspunkt: if rmove the window, the test1 will throw error: 'window is not defined'.
[05:57] dguttman has joined the channel
[05:57] fbzhong: Actually, I just want the Node.JS can simulate as the standard browser.
[05:58] fbzhong: indutny: yes, it runs in different context. The strange thing is that test 1 works well, test 2 does not...
[05:58] pen has joined the channel
[05:59] indutny: indutny: oh, i see now
[05:59] indutny: indutny: probably you should try to do it w/o __proto__
[06:00] fbzhong: indutny: without __proto__, in the test 1, the window will be not defined. That's why I write the code "window.window = window".
[06:02] indutny: fbzhong: why are you using Context bindings?
[06:02] indutny: fbzhong: why not just a regular object
[06:03] context: heh
[06:03] fbzhong: indutny: do you means the new Context() ?
[06:03] indutny: fbzhong: and please try this with require('vm')
[06:03] indutny: fbzhong: yes
[06:03] fbzhong: indutny: yes, I tried the vm and regular object, same error.
[06:04] context: he is saying stop using bindings()
[06:04] context: and this works very find without the Context object
[06:04] fbzhong: Let me submit the code using vm.
[06:04] fbzhong: context: really?
[06:04] context: runInNewContext(blah, window) works fine
[06:04] zhangkaizhao has joined the channel
[06:05] indutny: fbzhong: require('vm').runInNewContext('window.google = {};google.test =1', new P);
[06:05] sonnym1 has joined the channel
[06:05] indutny: fbzhong: where P is : function P() {this.window = this};
[06:05] indutny: fbzhong: if you want it to be a object with a prototype :)
[06:06] context: oddly i cant find any docs of any kind for said Context object
[06:06] fbzhong: indutny: let me have a try~
[06:06] w3dot0 has joined the channel
[06:06] indutny: fbzhong: you can substitute P with var window = {};window.window = window;
[06:06] brianseeders has joined the channel
[06:07] skm has joined the channel
[06:08] fbzhong: indutny: that's good, let me have a try in jsdom. Thanks for your help :)
[06:08] indutny: np
[06:08] ditesh|cassini has joined the channel
[06:09] context: what exactly is Context and where are there any docs for it :x
[06:09] indutny: context: it's an old API
[06:09] context: ahh
[06:10] langworthy has joined the channel
[06:10] indutny: context: no docs and it'll be deleted someday
[06:10] context: is there a lot without docs that should ? or do the docs do a pretty good job at covering everything that exists right now. i know dgram doesnt have docs still :x
[06:11] context: or is dgram gonna get pulled into 'net' eventually :x
[06:11] mscdex: huh?
[06:11] mscdex: dgram has docs
[06:11] context: oh wait nm. there are dgram docks
[06:11] context: kk im just a tard
[06:12] mscdex: the only thing that isn't documented is the sync http and tcp functions
[06:12] mscdex: :-D
[06:12] context: oh. thats a good thing.
[06:13] context: mscdex: remove the methods, dont add docs :p
[06:13] mscdex: that was the joke, there are no sync network functions in node ;)
[06:13] context: oh. good. haha
[06:14] a2800276 has joined the channel
[06:14] chjj: the javascript tag on stackoverflow should just be renamed to
[06:14] chjj: jquery
[06:14] chjj: more accurate
[06:19] coreb has joined the channel
[06:22] stonebranch has joined the channel
[06:23] nivoc has joined the channel
[06:26] fbzhong: indutny: thank you, I resolve this issue :)
[06:26] indutny: fbzhong: great,np
[06:28] mikeal has joined the channel
[06:28] fbzhong: indutny: however, I still wonder to know why the test 2 is not correct. In the test 1, those code execute in two context, with the same sandbox is works well. The test 2 does not.
[06:28] indutny: fbzhong: probably that's a prototype problem
[06:29] riven has joined the channel
[06:29] er1c_ has joined the channel
[06:29] fbzhong: indutny: so you think there may be a issue in node.js, don't you?
[06:30] indutny: fbzhong: nope, there can be issue within old API, but that doesn't matter
[06:30] fbzhong: indutny: no, it throw error in new vm api either.
[06:30] indutny: fbzhong: and probably __proto__ feature works not how you expect it work. Anyway it's depricated
[06:30] indutny: __proto__ is depricated
[06:31] amerine has joined the channel
[06:31] fbzhong: indutny: Orz, what's the recommended style to insead of __proto__ in NodeJS?
[06:31] ryah has joined the channel
[06:33] indutny: fbzhong: window.constructor.prototype
[06:34] jmoyers: erm
[06:35] fson: Object.getPrototypeOf(object)
[06:35] mikeric has joined the channel
[06:35] CIA-90: node: 03Brian White 07v0.4 * r9b34726 10/ doc/api/crypto.markdown :
[06:35] CIA-90: node: Crypto documentation fixes
[06:35] CIA-90: node: Fixes #1104. - http://bit.ly/kPh3yM
[06:36] mikeric has joined the channel
[06:36] qFox has joined the channel
[06:36] fbzhong: indutny: thank you, very appreciate~
[06:36] indutny: fbzhong: it works?
[06:37] indutny: fson: Object.getPrototypeOf returns same result
[06:37] jmoyers: constructor.prototype doesn't let you walk the actual prototype chain if i remember correctly
[06:37] fbzhong: indutny: sorry I don't test it.
[06:37] jmoyers: (no deep inheritance?)
[06:37] mraleph has joined the channel
[06:37] zentoooo has joined the channel
[06:37] indutny: jmoyers: yep, but you can do something like that:
[06:37] `3rdEden has joined the channel
[06:38] mcluskydodallas has joined the channel
[06:39] fson: here is how you can also walk the prototype chain: http://ejohn.org/blog/objectgetprototypeof/
[06:39] indutny: v8: function a() {};function b() {};function c(){};a.prototype = new ;b.prototype = new c;var A = new a, _p = A.constructor.prototype, newp = new d();A.constructor.prototype = newp;newp.constructor.prototype = _p;console.log(A);
[06:39] v8bot: indutny: SyntaxError: Unexpected token ;
[06:39] mertonium has joined the channel
[06:39] indutny: oh shit
[06:39] indutny: k, no time to fix it
[06:39] indutny: later
[06:41] herbySk has joined the channel
[06:43] temp01 has joined the channel
[06:45] k1ttty has joined the channel
[06:49] pigmej has joined the channel
[06:49] gavin_huang has joined the channel
[06:51] mhausenblas has joined the channel
[06:51] mhausenblas_ has joined the channel
[06:52] Charuru has joined the channel
[06:52] liar has joined the channel
[06:54] groom has joined the channel
[06:55] dargo_simpson has joined the channel
[06:57] makuchaku has joined the channel
[06:58] jaket_ has joined the channel
[06:59] eee_mk has joined the channel
[07:00] fbzhong has joined the channel
[07:01] tdegrunt has joined the channel
[07:01] sreeix has joined the channel
[07:01] flottis has joined the channel
[07:02] flottis: hi everyone, i need to know why a child process dies, any ideas how to debug that?
[07:03] marshall has joined the channel
[07:04] caseywstark_ has joined the channel
[07:04] ivanfi has joined the channel
[07:05] ExsysTech has joined the channel
[07:06] djcoin has joined the channel
[07:06] flippyhead has joined the channel
[07:06] zentoooo has joined the channel
[07:10] bene1 has joined the channel
[07:10] yozgrahame has joined the channel
[07:10] fangel has joined the channel
[07:10] bene1 has left the channel
[07:10] darshanshankar has joined the channel
[07:11] beawesomeinstead has joined the channel
[07:11] beawesomeinstead has joined the channel
[07:11] jbpros has joined the channel
[07:13] sriramk has joined the channel
[07:15] steffan has joined the channel
[07:15] mAritz has joined the channel
[07:16] ditesh|cassini has joined the channel
[07:16] niftylettuce: night owl catches the snake...
[07:17] zhangkaizhao_ has joined the channel
[07:18] xsyn has joined the channel
[07:18] level09 has joined the channel
[07:19] rictic has joined the channel
[07:20] matjas has joined the channel
[07:25] Zeffer has joined the channel
[07:26] sriramk has joined the channel
[07:28] thalll has joined the channel
[07:31] JianMeng has joined the channel
[07:31] gausby has joined the channel
[07:32] Druide_ has joined the channel
[07:33] mike5w3c_ has joined the channel
[07:34] temp01 has joined the channel
[07:34] [AD]Turbo has joined the channel
[07:34] msucan has joined the channel
[07:38] AAA_awright_ has joined the channel
[07:38] fbzhong: indutny: there is someone reply this issue on github. He said the __proto__ property does not copy proper in runInNewContext code. https://github.com/joyent/node/issues/1105#issuecomment-1233739
[07:38] indutny: fbzhong: great
[07:38] indutny: fbzhong: as I said
[07:38] indutny: fbzhong: so does my workaround with constructor.prototype works?
[07:39] fbzhong: indutny: and he said that the getter/setter property does not copied either. Later I will write some code to test it.
[07:39] indutny: k
[07:39] fbzhong: indutny: sorry that I didn't test it yet... :(
[07:39] Bj_o_rn has joined the channel
[07:40] indutny: fbzhong: probably constructor property won't copy too in this case
[07:40] thron7 has joined the channel
[07:40] fbzhong: indutny: let's write a code to verify it. Any progress, I will let you know :)
[07:42] thron7 has left the channel
[07:42] MrTopf has joined the channel
[07:43] indutny: fbzhong: no neeed, wait a sec I'll answer why
[07:43] indutny: fbzhong: https://github.com/joyent/node/blob/master/src/node_script.cc#L336-345
[07:44] igl has joined the channel
[07:44] indutny: fbzhong: constructor and __proto__ ain't in properties list
[07:45] fbzhong: indutny: so you means the sandbox->GetPropertyNames will return all keys without the __proto__ and setter/getter propery, right?
[07:45] indutny: yep
[07:45] jhurliman: does anyone else have issues with the line numbers in node-inspector drifting (becoming incorrect) while stepping through many methods?
[07:46] indutny: V8EXPORT Local v8::Object::GetPropertyNames( ) Returns an array containing the names of the enumerable properties of this object, including properties from prototype objects. The array returned by this method contains the same values as would be enumerated by a for-in statement over this object.
[07:46] uchuff has joined the channel
[07:46] indutny: b/c they are not enumerable
[07:46] fbzhong: ok, I got it...
[07:47] superjudge has joined the channel
[07:47] ntk1000 has joined the channel
[07:47] fbzhong: indutny: why not use the sandbox object as the global object in the new context? It's less code, maybe less bugs as well.
[07:47] [AD]Turbo: hi there
[07:48] azeroth__ has joined the channel
[07:49] lackac has joined the channel
[07:49] indutny: fbzhong: probably thats because Context->Global() can't be set to a object
[07:49] fbzhong: indutny: refer to this http://bespin.cz/~ondras/html/classv8_1_1Context.html#dd11dd18ee8e7f03d383afa79e87d1e6
[07:50] fbzhong: indutny: static Persistent v8::Context::New ( ExtensionConfiguration * extensions = NULL,
[07:50] fbzhong: Handle< ObjectTemplate > global_template = Handle< ObjectTemplate >(),
[07:50] fbzhong: Handle< Value > global_object = Handle< Value >()
[07:50] indutny: yep I'm looking at it too
[07:50] MikeMakesIt has joined the channel
[07:50] fbzhong: indutny: the third parameter of Context::New is the global object.
[07:51] indutny: I see, I see
[07:51] fbzhong: indutny: however, this morning, I am trying to pass sandbox as the new Context global object, the script->run throws exception: segment fault...
[07:51] beseku has joined the channel
[07:51] indutny: hm....
[07:51] gozala has joined the channel
[07:52] troessner has joined the channel
[07:52] fbzhong: indutny: maybe my code is wrong...
[07:52] beseku has left the channel
[07:53] indutny: try to pass it as second argument
[07:53] temp01 has joined the channel
[07:54] indutny: fbzhong: https://github.com/joyent/node/blob/master/deps/v8/samples/shell.cc#L67
[07:54] fbzhong: indutny: let me have a try.
[07:54] indutny: k
[07:55] test123432432 has joined the channel
[07:55] fbzhong: indutny: wait me for a moment :)
[07:56] sechrist has joined the channel
[07:56] indutny: fbzhong: np, I'm doing my regular job :) Planning to be here for sometime
[07:56] fbzhong: indutny: me too :)
[07:58] FireFly|n900 has joined the channel
[08:04] micheil has joined the channel
[08:05] stonebranch has joined the channel
[08:06] mehlah has joined the channel
[08:08] davidbanham has joined the channel
[08:08] indutny: fbzhong: so, any results?
[08:11] jeremyselier has joined the channel
[08:12] fbzhong: indutny: sorry, I am working other stuff...
[08:12] mc_greeny has joined the channel
[08:12] indutny: k, np
[08:12] fbzhong: indutny: any progress, I will let you know :)
[08:15] Xano has joined the channel
[08:16] adambeynon has joined the channel
[08:16] ewdafa has joined the channel
[08:18] adambeynon has joined the channel
[08:20] ExsysTech has joined the channel
[08:22] indutny: fbzhong: ok, that won't work
[08:22] indutny: fbzhong: second argument needs to be ObjectTemplate
[08:23] fbzhong: indutny: I use the Handle< ObjectTemplate >() as the second argument, it failed...
[08:24] beppu has joined the channel
[08:24] fbzhong: indutny: I think it may be the global_template of sandbox object does not the same with the Handle< ObjectTemplate >().
[08:24] indutny: fbzhong: of course that will fail
[08:25] fbzhong: indutny: global_object An optional global object to be reused for the newly created context. This global object must have been created by a previous call to Context::New with the same global template. The state of the global object will be completely reset and only object identify will remain.
[08:27] a2800276 has joined the channel
[08:28] secoif has joined the channel
[08:28] burningdog has joined the channel
[08:28] adrianmg has joined the channel
[08:29] pomodoro has joined the channel
[08:33] jonathantaylor has joined the channel
[08:35] Xano_ has joined the channel
[08:36] josephboyle has joined the channel
[08:36] sechrist has joined the channel
[08:36] hellp has joined the channel
[08:37] jonathantaylor has joined the channel
[08:39] thalll has joined the channel
[08:46] TomY has joined the channel
[08:49] prettyrobots has joined the channel
[08:49] markwubben has joined the channel
[08:50] test123432432_ has joined the channel
[08:50] fbzhong_ has joined the channel
[08:55] bzinger has joined the channel
[08:58] fbzhong has joined the channel
[09:00] romainhuet has joined the channel
[09:00] stonebranch has joined the channel
[09:01] gozala has joined the channel
[09:01] dies_el has joined the channel
[09:03] littke has joined the channel
[09:03] bzinger_ has joined the channel
[09:04] rictic has joined the channel
[09:06] stonebranch has joined the channel
[09:07] thron7 has joined the channel
[09:09] ExsysHost has joined the channel
[09:09] zcopley has joined the channel
[09:09] sechrist_ has joined the channel
[09:12] JackeyChan has joined the channel
[09:14] wittemann has joined the channel
[09:15] Schmallon has joined the channel
[09:16] jtsnow has joined the channel
[09:16] unomi has joined the channel
[09:19] fbzhong: indutny: I did use sandbox as the new Context global object, however, it throw segment fault error.
[09:19] wittemann has left the channel
[09:19] dgathright has joined the channel
[09:19] fbzhong: indutny: And I think we could refer the code from Chromium...
[09:20] indutny: fbzhong: you should use persistent handles
[09:20] indutny: fbzhong: for storing every object related to that Context
[09:20] slaskis has joined the channel
[09:20] fbzhong: indutny: what's that? do you have more detail about it?
[09:22] indutny: fbzhong: it's a class, like Local or Handle
[09:22] indutny: fbzhong: actually morelike Local
[09:22] indutny: fbzhong: Persistent