[00:00] herbySk has joined the channel [00:00] flukes1: yes, with cat it doesn't work [00:00] mscdex: jtsnow: https://gist.github.com/797791 [00:01] Ond has joined the channel [00:02] micheil: ryah: what's the difference between WriteStream.end and WriteStream.destroy? [00:03] bbttxu has left the channel [00:04] ryah: micheil: one happens on drain, the other happens now [00:04] micheil: okay [00:04] flukes1: is there a way to make console.log output in tests appear regardless of whether the test succeeds [00:04] flukes1: -v doesn't appear to do it [00:05] micheil: flukes1: you'd need to say what test runner [00:05] micheil: if it's nodes, then no. [00:05] micheil: unless you manually run the test [00:05] flukes1: python tools/test.py -v [00:06] flukes1: sorry if this is a stupid question - started looking at node about 10 mins ago ;p [00:06] sechrist has joined the channel [00:06] ryah: flukes1: not that i know of [00:10] sechrist has joined the channel [00:11] micheil: ryah: is there any reason for utterly everything being queued? [00:11] bingomanatee: Okay TJ you got your sensible sort order [00:11] tjholowaychuk: WAHOO [00:12] bingomanatee: but it was REALLY HARD becuase the f**king christkov-native sort order is NOT consistent with the mongodb docs and it took WAY too long to figure that out! and I'm gonna go cry a little now [00:12] Matt___: Quick application-side question: Has anyone had any success combining backbone.js with express? [00:12] bingomanatee: I can see why you wanted it - you rank way up there! [00:12] ryah: micheil: memory [00:12] unomi has joined the channel [00:13] tjholowaychuk: bingomanatee: just couldnt tell what was what at all haha [00:13] flukes1: ryah: in test-tls-server-verify, shouldn't all of the certs be rejected because they don't have client authentication specified in extended key usage [00:13] tjholowaychuk: whats the link? [00:13] micheil: ryah: okay. [00:13] heavysixer has joined the channel [00:13] flukes1: or do you not care / check for that [00:13] bingomanatee: Yeah it wasn't meaningful data [00:13] sirganya has joined the channel [00:13] bingomanatee: I mean it was accurate but without a freq sort its kind of bull**t. [00:14] bingomanatee: but you're not more popular than 3sus :D [00:14] MikhX has joined the channel [00:15] sechrist has joined the channel [00:15] bingomanatee: Really though - I gotta get better stopword filtering. [00:15] ryah: flukes1: i don't check for that. i don't even know about that. can you tell me more? [00:16] bingomanatee: and nick filters too. [00:17] sirganya: hi, I'm having a problem with fs.write. I have a thing that watches a couchdb database and writes to a file each time it gets a new entry - problem is it writes it six times instead of once. Is there something I'm missing? Any ideas would be gratefully welcomed, I've been scratching googles head for hours.. [00:17] flukes1: ryah: a certificate shouldn't be usable as a client cert unless it either has a) OBJ_client_auth in the "extended key usage" extension, or b) digitalSignature in the [not extended] "key usage" extension [00:17] ryah: sirganya: there is no bug like that. [00:17] ryah: sirganya: i suspect it's your code [00:18] flukes1: rejectUnauthorized = true should cause this to be checked, I think [00:18] pyrotechnick has joined the channel [00:18] sirganya: I have a log in the write callback that shows whats being written, it only appears once and the entry is correct.. :( [00:18] sechrist has joined the channel [00:19] ryah: flukes1: it's only doing SSL_get_verify_result() [00:19] sirganya: it's definitely only getting to the open and write callbacks once [00:20] ryah: sirganya: could it be you're modifying the buffer that you fs.write() with before the callback it made? [00:20] dguttman_ has joined the channel [00:20] flukes1: ryah: sorry, ignore me - your agent certs have key usage "any" [00:21] flukes1: which of course includes digitalSignature [00:21] sechrist has joined the channel [00:22] ddollar has joined the channel [00:22] MikhX has joined the channel [00:22] sbellity has joined the channel [00:23] sirganya: ryah: if you have a moment.. http://pastebin.com/QmJCjvUd [00:24] sirganya: ryah: line 29 [00:25] sechrist has joined the channel [00:26] sechrist has joined the channel [00:27] shiawuen has joined the channel [00:27] ryah: sirganya: fs.WriteStream('/home/virtual/vibrantrioja.com/var/www/html/grandtasting/signups.csv').write(csvEntry + "\n").end(); [00:30] sirganya: ryah: Thanks, I'll try that.. [00:30] ryah: sirganya: or something like that... [00:30] jacobolus has joined the channel [00:31] Solsys has joined the channel [00:31] unomi has joined the channel [00:32] admc1 has joined the channel [00:34] shiawuen has joined the channel [00:34] flukes1: ryah: i've added "ext_key_usage" to the object returned by getPeerCertificate, which is just an array of extended key usage OIDs on the certificate. it works fine and I've written a test which also works. is it worth submitting a pull request? [00:36] Me1000 has joined the channel [00:36] jimt has joined the channel [00:36] micheil: ryah: I'm still missing the point of having fs.open in the _queue [00:39] jakehow has joined the channel [00:39] stepheneb has joined the channel [00:39] m0rganic1 has left the channel [00:40] mikeal: only MongoDB would add a feature for better durability and the keep it off by default http://www.mongodb.org/display/DOCS/Journaling [00:40] mikeal: wtf is wrong with these guys [00:41] saschagehlich has joined the channel [00:41] ryah: flukes1: please [00:41] saschagehlich: hey folks, I've got a (pretty simple) problem. Shouldn't the output be "a, b" here instead of "b, b"? I remember that it worked in older versions of v8... https://gist.github.com/aa2241cb250ab4a4892c [00:41] konobi: mikeal: the hardcoded default is fuck_data_royally=1 all the time though, right? [00:41] mikeal: well, more accurately it's "corrupt on shutdown" [00:41] mikeal: but yes, that's the default [00:42] mikeal: just like the default write API call has no response [00:42] mikeal: you just hope it's there by the time you wanna read it [00:42] konobi: saschagehlich: no [00:42] skm has joined the channel [00:42] saschagehlich: so how can I make sure that _val is the same one as in the for-loop? [00:43] dguttman has joined the channel [00:43] dguttman has joined the channel [00:43] konobi: saschagehlich: do the _var assignment within the function [00:43] Aikar: saschagehlich: for(var i;;) { (function(){ var val = blah[i]; // do stuff })(); } [00:44] Aikar: gotta wrap a function block around it [00:44] kaelan has joined the channel [00:44] sechrist has joined the channel [00:44] kaelan-w: i'm following the windows build instructions for node off ry's github wiki, and the sconscript for v8 isn't working [00:44] SwiftLayer has joined the channel [00:44] micheil: ryah: I've pushed fs.open to be inline, no need to queue it really. [00:45] kaelan-w: by default, KeyError: 'os' when it's checking the settings from the environment. if i patch the sconscript to set the os, then i get an error that 'os must be specified' but it doesn't say where to specify it [00:45] saschagehlich: Aikar: well, but is "i" the right one then? [00:45] Aikar: saschagehlich: your having an issue where when your callback is ran that i is the last value in the array instead of what it was when you registered the callback correct? [00:45] benburkert has joined the channel [00:46] Aikar: paste your code if you can [00:46] sirganya has joined the channel [00:46] saschagehlich: Aikar: that's correct, one second.. [00:46] nooder has joined the channel [00:47] Aikar: yeah its cause every callback registered inside the for() are all pointing to same var, which last value was the last index, so you got to copy the index/value to its own scope (that wont change) [00:47] admc has joined the channel [00:48] Aikar: saschagehlich: https://github.com/Aikar/node-nova/blob/master/lib/utils.js or copy the forEach function out of my file here [00:48] Aikar: then just do blah.forEach(yourarray, function(value, key) { }); [00:48] sirganya has joined the channel [00:48] skm has joined the channel [00:49] saschagehlich: Aikar: https://gist.github.com/19d40b96e558e5eb53f2 it's something like that. this.store.servers.getRandom() is a function that is doing some redis stuff and calls the callback afterwards [00:49] sechrist has joined the channel [00:50] micheil: ryah: how were you measuring the number of actual sys calls that that test was making? [00:50] sechrist has joined the channel [00:50] Aikar: saschagehlich: https://gist.github.com/a6f430462a18c6abb9e8 [00:51] ryah: micheil: iwasn't [00:51] micheil: oh, right. [00:51] pyrotechnick: FUCK I AM AWESOME [00:51] pyrotechnick: sorry [00:51] sechrist has joined the channel [00:51] maushu has joined the channel [00:51] pyrotechnick: sometimes i just cant handle it and need to tell people [00:51] ryah: micheil: but you can do "strace node x.js" [00:51] micheil: time node test-slow-writestream-write? [00:51] micheil: okay [00:51] micheil: I think I can cheat on the solution a little. [00:52] Aikar: saschagehlich: thats a common 'gotcha' due to js having function level scope instead of block level scope. so you have to wrap it in a function (thats executed immediately) to give it its own local scope so that the url value has not changed when the callback runs [00:52] kaelan-w: nvm, the tag i checked out was just busted [00:52] kaelan-w has left the channel [00:52] sechrist has joined the channel [00:53] saschagehlich: Aikar: hm okay, I'll try that. thanks so far! :) [00:54] micheil: ryah: where'd you get the strace command? [00:54] LoRe has joined the channel [00:54] devrim has joined the channel [00:54] Aikar: saschagehlich: btw you can use the .bind propery of a function in node instead of that wrapper __bind you got. like so: https://gist.github.com/a6f430462a18c6abb9e8 [00:54] saschagehlich: Aikar: but that means that I can't do "break" inside the loop... right? [00:55] sechrist has joined the channel [00:55] tanepiper: pyrotechnick: nodejs brings out the awesomeness [00:56] Aikar: v8: a=[1,2,3]; for(var i = 0; i (/Users/micheil/Documents/git/miksago.github.com/node/test/simple/test-file-write-stream.js:56:12) [03:25] micheil: at EventEmitter.emit (events.js:59:20) [03:25] micheil: Command: build/default/node /Users/micheil/Documents/git/miksago.github.com/node/test/simple/test-file-write-stream.js [03:25] micheil: is the only error I get. [03:25] micheil: fuck. hit bad args again.. [03:27] jdalton has joined the channel [03:27] sechrist has joined the channel [03:28] Bosmon2 has joined the channel [03:32] sechrist has joined the channel [03:32] chapel: pyrotechnick: have you really used couch that much? [03:33] keks: hey, working on fuse, how can i pass the json object {getattr: function(..){...}, read: function(..){...},...} to the cpp code? [03:33] springmeyer has joined the channel [03:35] derferman has joined the channel [03:35] sechrist has joined the channel [03:35] sechrist has joined the channel [03:38] micheil: ryah: some how I manage to get an object into the queue, but I have no idea how [03:38] sechrist has joined the channel [03:39] whyme has joined the channel [03:39] sechrist has joined the channel [03:39] sechrist has joined the channel [03:41] zomgbie has joined the channel [03:42] sechrist has joined the channel [03:42] tim_smart: keks: Using the v8 API [03:43] keks: okay, ill look for their docs, didn't find anything on nodejs. but that actually makes more sense... [03:43] keks: thanks [03:44] tim_smart: keks: Take a look at some existing node addons. [03:44] keks: I did before for general understanding, but nothing came to mind concerning that object. [03:44] mscdex: or core code [03:45] c4milo1 has joined the channel [03:45] sechrist has joined the channel [03:47] Me1000 has joined the channel [03:47] sechrist has joined the channel [03:48] mscdex: keks: what in particular do you need help with? i mean, you pass it to the c++ function just like you would a js function [03:49] keks: mscdex: basically how to handle it: what type is it, how do i address the object's contents, etc. [03:49] mscdex: it's just of type Object [03:49] dguttman_ has joined the channel [03:49] sechrist has joined the channel [03:49] keks: hm, okay, i'll look into it [03:50] micheil: ryah: okay, this definitely works on some level. [03:50] clarkfischer has joined the channel [03:50] mscdex: usually someone has uploaded a version that's been run through doxygen... let me see if i can find one [03:51] sechrist has joined the channel [03:52] fallsemo has joined the channel [03:52] chrischris has joined the channel [03:53] mscdex: keks: i've got an older copy of the generated docs, but it should still be usable: http://mscdex.net/v8/ [03:53] mscdex: keks: look under the data structures tab [03:53] sechrist has joined the channel [03:53] keks: mscdex: okay, thanks [03:55] micheil: ryah: node time build/default/node ~/tmp/test-slow-write.js master* [03:55] micheil: done [03:55] micheil: build/default/node ~/tmp/test-slow-write.js 1.43s user 0.18s system 91% cpu 1.762 total [03:55] micheil: number of write()s? 25. [03:56] keks: mscdex: child_process accepts an object when called, but doesn't pass it as an object to v8. are you sure this is the right way? [03:56] jakehow has joined the channel [03:57] mscdex: keks: where at? [03:57] keks: nodejs git, /lib/child_process.js:50 [03:58] sechrist has joined the channel [03:58] micheil: ryah: http://cl.ly/06042Z1z1U2t2P0v3f2v [04:00] liar has joined the channel [04:01] mscdex: keks: i'm not seeing it. InternalChildProcess is the C++ object though, and no call to any of its functions includes plain object arguments that i can see [04:01] ryah: micheil: it's cause they're all in different processes [04:02] noahcampbell has joined the channel [04:03] keks: mscdex: in the for loop at ~50 it iterates the object and appends it to optionArg, which are arguments given to the c++ functor [04:03] siculars has joined the channel [04:04] keks: mscdex: but your right, it's probably better to just use an object [04:04] sechrist has joined the channel [04:04] beawesomeinstead has joined the channel [04:05] mscdex: keks: that block of code is just overwriting the default options, it's not actually passing 'options' or 'optionsArg' to a C++ function [04:05] micheil: ryah: although, do you agree that I took the slightly write approach in allowing the data to be buffered ever so slightly? [04:06] micheil: s/write/right [04:08] perlmonkey2 has joined the channel [04:08] keks: mscdex: hum, okay. anyway, /src/node_child_process.cc is reading arg[0..5] and no objects. instead what is passed in the object is args[3..5]. that's why i assumed that for loop did that. [04:09] keks: mscdex: that's line 100-150 [04:09] sechrist has joined the channel [04:10] sechrist has joined the channel [04:10] ryah: micheil: s/process/threads [04:11] micheil: okay, I'm confused slightly. [04:11] ryah: micheil: i haven't read it yet. i'll check it out tomorrow [04:11] micheil: okay. I'll have an updated version tomorrow. [04:11] micheil: basically I do the same writeQueueing as what the network module does [04:12] admc has joined the channel [04:12] brianmario has joined the channel [04:14] sechrist has joined the channel [04:16] sechrist has joined the channel [04:16] JianMeng has joined the channel [04:17] blueadept has joined the channel [04:19] richcollins has joined the channel [04:21] sechrist has joined the channel [04:22] Aria has joined the channel [04:25] richcollins has joined the channel [04:27] sechrist has joined the channel [04:27] daniellindsley has joined the channel [04:28] sechrist has joined the channel [04:29] ddollar has joined the channel [04:29] sechrist has joined the channel [04:29] tilgovi has joined the channel [04:31] sechrist has joined the channel [04:31] hij1nx has joined the channel [04:31] jmckind has joined the channel [04:33] sechrist has joined the channel [04:34] brianleroux has joined the channel [04:36] mscdex: so is there a simple http/https proxy example out there that uses the new tls api? [04:36] sechrist has joined the channel [04:39] warz: i'm pretty new to nodejs, and am wondering if there's a way to clean up this typical nested callback block of code: http://jsfiddle.net/Lrjgz/1/ [04:39] warz: is this something that async.js, and those types of libs, would help me with? [04:40] warz: i know that's not ugly on its own, but when i had a lot of that kind of stuff in my code it adds up [04:40] sechrist has joined the channel [04:45] warz: hrm. looks like it. think ill try async.js. [04:48] pengo has joined the channel [04:48] keks: warz: you also could return promises yourself [04:50] mscdex: warz: there are modules like that out there to help with that. you could roll your own and do something like this: https://gist.github.com/798083 [04:50] mscdex: warz: here's some existing flow control modules: https://github.com/ry/node/wiki/modules#async-flow [04:51] liar has joined the channel [04:51] keks: yeah that promise idea was stupid... [04:51] mscdex: warz: node-seq, Step, and chainsaw might be ones to look at [04:51] warz: alright, checking them out now. thanks. [04:52] pHcF has joined the channel [04:52] richcollins has joined the channel [04:52] slickplaid: What is the default 404 handling redirect? 301 redirect -> 404.html or whatever the page is? or just render the 404 page without a redirect? [04:52] slickplaid: Hopefully that makes sense :D [04:52] hunterloftis has joined the channel [04:53] Aria: slickplaid: in what system? [04:53] slickplaid: Like if they go to site.com/notreal.html, should it redirect 301 to the 404.html page? Or just render the 404 without redirection? [04:54] Aria: In what system? [04:54] slickplaid: I'm using express [04:54] Aria: Ah, express. [04:54] Aria: Just for the record, node.js isn't express -- express is just a library. [04:54] Aria: (I don't actually know the answer in that case.) [04:54] chrisdickinson: slickplaid: you should 404. [04:54] slickplaid: throw new NotFound; ? [04:54] mscdex: i would say render the 404 as well [04:54] Aria: And yeah, you should 404 and not redirect. It's bloody annoying when people redirect. A simple typo becomes a whole retype. [04:55] slickplaid: :) [04:55] slickplaid: Yeah, agreed with the no redirect [04:56] slickplaid: excellent... works like a charm (throw new NotFound;) [04:56] sechrist has joined the channel [04:58] brianc: can anyone point me in the direction of how to declare development only dependencies in npm? [04:58] brianc: i.e. my unit tests depend on 'should.js' (which is hotness) but not the actual library [04:59] sechrist has joined the channel [05:01] sechrist has joined the channel [05:02] sechrist has joined the channel [05:03] stepheneb has joined the channel [05:03] sechrist has joined the channel [05:03] sechrist has joined the channel [05:04] Yuffster_work has joined the channel [05:05] bingomanateeIpho has joined the channel [05:09] Fuld has joined the channel [05:09] asimismo has joined the channel [05:11] sechrist has joined the channel [05:12] sechrist has joined the channel [05:12] Lorentz: Anyone got a good tutorial on using mongoose? [05:12] Lorentz: I tried http://mongoosejs.com, but looks like page isn't up. [05:12] muk_mb: did one of you build http://dominion.isotropic.org ? seems like something node.js would be good for [05:13] losing has joined the channel [05:17] mikeal has joined the channel [05:17] sechrist has joined the channel [05:19] fugoki has joined the channel [05:20] sechrist has joined the channel [05:21] sechrist has joined the channel [05:23] dstufft: is there a "prefered" irc lib for node.js? [05:25] sechrist has joined the channel [05:26] sechrist has joined the channel [05:27] aaronblohowiak has joined the channel [05:27] aaronblohowiak: hellooo [05:28] dnyy has joined the channel [05:28] ossareh has joined the channel [05:29] brianc: I ran into a good deal of async "mind-benders" when writing tests for node-postgres and what resulted was a sort of hybrid from the qunit style of "expect" testing. https://github.com/brianc/drain [05:30] aaronblohowiak: ahoy! [05:30] aaronblohowiak: mind-benders? [05:31] isaacs has joined the channel [05:31] brianc: aaronblohowiak: indeed maytee. trying to test connection pooling of a database driver and shutting down each client after a specific number of queries have executed qualifies as a mind-beinder in my feeble-minded universe [05:32] aaronblohowiak: brianc: ooh, fun [05:32] sechrist has joined the channel [05:33] brianc: aaronblohowiak: would you know how to specifiy 'dev' dependencies for an npm package [05:33] brianc: like you can with rubygems [05:33] isaacs: brianc: {"devDependencies": {...}} [05:33] aaronblohowiak: brianc: as isaacs he just arrived =) [05:33] aaronblohowiak: *ask [05:33] isaacs: brianc: then `npm install blah --dev` [05:33] brianc: isaacs: you're my daddy [05:33] pyrotechnick: what's a reasonable price for a full-time node dev? someone with 6 months experience in high-traffic production [05:33] isaacs: brianc: npm link always is --dev'ed [05:33] isaacs: since link packages are primarily for development [05:33] isaacs: put your test harness dependencies there [05:34] brianc: isaacs: yah was depending something on 'should' for testing [05:34] brianc: and then will depend node-postgres on this other library [05:34] Aria: pyrotechnick: In what country? In what city? [05:34] isaacs: yeah, ideally you shouldn't depend on a test framework unless you ARE a test framework [05:34] brianc: so I assume node-postgres -> other_thing -> should will pull it all down [05:35] brianc: i could drop 'should' requirement, but it's just so much move lovely (imho) to write the .should style assertions [05:35] pyrotechnick: Aria: anywhere. telecommuting [05:35] isaacs: brianc: meh. if it's a dev thing, then whatever. [05:35] isaacs: depend on whatever helps you develop it [05:35] Aria: Heh. I'd think it'd depend highly on the cost of living. [05:35] brianc: cool. cool. I'll fiddle with it tomorrow and see how it goes [05:36] pyrotechnick: Aria: is that what it's like in the states? [05:36] pyrotechnick: here it really doesn't matter where you live. i think i like it like that [05:36] pyrotechnick: is there really that much of a difference between the tech hubs? [05:36] pyrotechnick: im a fan of paying for what people are worth. not how much it seems to cost them to live [05:36] sechrist has joined the channel [05:36] brianc: isaacs: as always thank you so so much for your help. =D [05:37] Aria: Oh man. San Francisco costs a fortune to live in. Middle US? Not nearly so much. [05:37] Aria: New York is even worse. [05:37] brianc: Aria: but in the middle of the US you've got to factor in the cost of your crystal meth habbit. [05:37] aaronblohowiak: isaacs: is there a good way to say "never ever publish this, i just want to use package.json for tracking my project deps?" [05:38] brianc: it's awesome when you misspell 2 syllable words [05:38] aaronblohowiak: pyrotechnick: does anyone have 6 months experience with high-traffic node ? [05:39] pyrotechnick: aaronblohowiak: yes [05:39] pyrotechnick: aaronblohowiak: i'm the employee in this instance [05:40] Aria: Heh, I was wondering that, too. It's a rarity. [05:40] pyrotechnick: we worked on adservers, 3.5M clicks a day [05:40] dgathright_ has joined the channel [05:40] sechrist has joined the channel [05:40] Aria: 3.5? Heh. [05:40] pyrotechnick: when we left [05:40] pyrotechnick: it's probably more now without changing the code if you know what i mean [05:40] pyrotechnick: it was quite stable [05:40] aaronblohowiak: pyrotechnick: well, charge the most you can get. ^.^ rails devs with 5 years experience are getting 100-150 for in-office in sf [05:40] bgilbert has joined the channel [05:41] pyrotechnick: wow really? [05:41] pyrotechnick: thank fuck we rolled with node [05:41] pyrotechnick: we're on more than double that [05:41] pyrotechnick: and we're 21 [05:41] pyrotechnick: i was just checking we wernt getting screwed [05:42] pyrotechnick: it's hard to find the work. we were very lucky TBH [05:42] pyrotechnick: all i will say is it pays to help out with FOSS [05:42] pyrotechnick: that's how we landed these jobs [05:42] Aria: Oh heck yes. [05:42] pyrotechnick: im not sure why i did it [05:43] pyrotechnick: but i basically ran myself into the ground for 6 months learning and hacking on node for free [05:43] pyrotechnick: ruined my credit etc [05:43] pyrotechnick: teaching the comrades etc [05:43] pyrotechnick: you cant just find noders you have to breed them [05:43] sechrist has joined the channel [05:44] pyrotechnick: so i've bred half a dozen and now we're getting more than we could have dreamed of to hack on node everyday [05:44] warz: breed me? [05:44] pyrotechnick: atm we have enough noobs, im happy to help anyone getting started though [05:44] pyrotechnick: right now we're looking for a coffee/dnode wizard [05:44] sechrist has joined the channel [05:44] pyrotechnick: if anyone knows one [05:45] aaronblohowiak: pyrotechnick: more than double 100k/year ? [05:45] pyrotechnick: we'll see [05:45] pyrotechnick: that would be one of our full wages ATM [05:45] pyrotechnick: for some of us [05:45] pyrotechnick: (some are getting less) [05:45] pyrotechnick: (some are part time) [05:45] pyrotechnick: but yes comparatively as good of a deal [05:46] isaacs: pyrotechnick: what do you do? [05:46] pyrotechnick: ahh we're making someones dream come true [05:46] pyrotechnick: MMO/WebGL, not our feisty projects but similar [05:46] Lorentz: Why would you breed programmers [05:47] pyrotechnick: because i live in brisbane, australia [05:47] pyrotechnick: and wanted friends who know node/coffee [05:47] aaronblohowiak: isaacs: any "never publish this package" magicsauce that will still let me use package.josn for deps but be free from worrying about accidently publishing proprietary code? [05:47] aaronblohowiak: pyrotechnick: that sounds like a dream come true [05:48] pyrotechnick: hey it was difficult for a long time [05:48] pyrotechnick: like i said [05:48] sechrist has joined the channel [05:48] aaronblohowiak: ^.^ [05:48] pyrotechnick: very much drove ourselves into the ground, dropped out of uni to do it [05:48] pyrotechnick: left THQ [05:48] pyrotechnick: doing what i thought i loved doing [05:48] warz: thought? [05:49] Lorentz: I'd join your team if you were in sydney [05:49] chapel: :) [05:50] warz: id let him join my team if he were in TX [05:50] warz: and he could breed me [05:50] sechrist has joined the channel [05:50] Lorentz: Why would anyone breed with you [05:50] warz: so he could have another node friend [05:50] slickplaid: I know there's a node app for development that will reload it when it's been updated... anyone know offhand what the name of it is? [05:50] pyrotechnick: restartr [05:50] slickplaid: ty [05:50] pyrotechnick: theres a few [05:50] pyrotechnick: but we dont mind restartr [05:51] pyrotechnick: im not sure its node [05:51] chapel: Ive used node-dev [05:51] pyrotechnick: might be [05:51] pyrotechnick: yeah? [05:51] pyrotechnick: any good? [05:51] chapel: I like it [05:51] chapel: simple [05:51] chapel: doesn't loop when hitting an erro [05:51] chapel: error [05:51] pyrotechnick: chapel: in 3 months we'll be able to consult nodehub for the answer to which is better ;) [05:51] chapel: waits for another file update [05:51] chapel: :) [05:52] sechrist has joined the channel [05:53] pyrotechnick: are there any node job boards? [05:53] chapel: dont think so [05:53] pyrotechnick: chapel: wave that [05:53] Lorentz: Make one [05:53] Lorentz: In node [05:54] pyrotechnick: for git? [05:54] sechrist has joined the channel [05:54] pyrotechnick: Lorentz: ? [05:54] pyrotechnick: like a github clone? [05:55] Lorentz: Sooorta, I guess, there are a bunch of tech job boards. [05:55] pyrotechnick: yeah but nothing node specific [05:55] sechrist has joined the channel [05:55] Lorentz: But github is probably the best bet right now, probably because the way most node projects are hosted there. [05:55] pyrotechnick: i think node is new and different enough to warrant its own [05:56] Gruni has joined the channel [05:57] sechrist has joined the channel [05:59] aaronblohowiak: pyrotechnick: restartr is pure node i guarantee it =) [05:59] aaronblohowiak: chapel: would you rater it waited for another file change instead of auto-restart? [05:59] aaronblohowiak: *rather [06:00] aaronblohowiak: ACTION is the restartr maintainer [06:01] chapel: aaronblohowiak: well thats what it does, if it errors, it waits for you to change the file again before running [06:01] pyrotechnick: aaronblohowiak: you wrote it? [06:02] aaronblohowiak: pyrotechnick: https://github.com/aaronblohowiak/restartr [06:02] aaronblohowiak: pyrotechnick: https://github.com/aaronblohowiak/restartr/blob/master/restartr [06:02] slickplaid: nice, checking that out now :D [06:03] sechrist has joined the channel [06:03] aaronblohowiak: i could make it wait for another file change instead of auto-restart on exit by adding an option flag check around https://github.com/aaronblohowiak/restartr/blob/master/restartr#L79-81 [06:03] pyrotechnick: aaronblohowiak: sick man. we use it everyday [06:03] pyrotechnick: you are now ON THE LIST [06:03] caligula_ has joined the channel [06:03] aaronblohowiak: pyrotechnick: sweet. i quit my job last week to spend more time contributing to open-source [06:04] pyrotechnick: you're a brave man aaronblohowiak [06:04] pyrotechnick: it's people like you who will forge the early success of node [06:04] slickplaid: paired with dropbox and the auto update it entails... pretty sweet combo [06:04] warz has joined the channel [06:04] aaronblohowiak: pyrotechnick: i am going to contract part-time to pay the billz [06:04] pyrotechnick: hows ur dnode? [06:04] pyrotechnick: slash coffee [06:04] sechrist has joined the channel [06:04] chapel: aaronblohowiak: I would propose the default behavior wait for file change on errors, just because its more elegant, and if someone wants to bypass that, have a flag [06:05] aaronblohowiak: pyrotechnick: what do you think of chapel's suggestion? [06:06] pyrotechnick: i care when i'm on battery [06:06] pyrotechnick: but when im plugged in it doesnt really bother me [06:06] pyrotechnick: maybe if u just throttled it a bit more [06:06] chapel: just saying its more elegant [06:06] pyrotechnick: that would be a nice trade-off [06:07] aaronblohowiak: chapel: yes, i agree. the bug isnt going to fix itself magically [06:07] chapel: having to ctrl+c isn't hard [06:07] pyrotechnick: well our apps never throw errors our coffee is always perfect [06:07] pyrotechnick: so it's a non-issue for us [06:07] pyrotechnick: ;) [06:07] aaronblohowiak: i could just delete those lines [06:07] aaronblohowiak: pyrotechnick: lol [06:07] chapel: but the reason you use something like restartr is to not have to use ctrl+c [06:07] chapel: :) [06:07] aaronblohowiak: chapel: yes, those lines are about to be destroyed [06:07] chapel: kk [06:07] chapel: Ill bbl, off to work [06:08] pyrotechnick: ill leave you to it lads [06:08] aaronblohowiak: l8r, thanks for the advice [06:08] pyrotechnick: i too must leave [06:08] aaronblohowiak: pyrotechnick: aaron.blohowiak@gmail.com [06:08] pyrotechnick: sick [06:08] pyrotechnick: i got a proposal for you [06:08] pyrotechnick: help you pay the bills maybe [06:08] aaronblohowiak: sweet! [06:08] pyrotechnick: or at least maybe put you into contact [06:08] pyrotechnick: cant promise anything so pipe down [06:08] pyrotechnick: but you've helped us out [06:08] pyrotechnick: so we'll try our best [06:08] aaronblohowiak: coo, appreciate it [06:09] aaronblohowiak: later [06:09] pyrotechnick: bbl [06:13] sechrist has joined the channel [06:14] creationix has joined the channel [06:14] sechrist has joined the channel [06:14] sechrist has joined the channel [06:15] warz: hrm i think pyro may have convinced me to try out coffeescript. [06:22] sechrist has joined the channel [06:28] ddollar has joined the channel [06:28] ddollar has joined the channel [06:31] sechrist has joined the channel [06:32] sechrist has joined the channel [06:33] bronson_ has joined the channel [06:33] sechrist has joined the channel [06:34] sechrist has joined the channel [06:34] fangel has joined the channel [06:35] sechrist has joined the channel [06:35] mikeal has joined the channel [06:38] nilcolor has joined the channel [06:38] Ond has joined the channel [06:38] void_ has joined the channel [06:38] orospakr: Hi! Mocking library/patterns recommendations? The one I'm the most to is rspec's. I see a few things recommended on the node.js wiki: gently, sinon, espionage, etc. [06:39] admc: hey all, is there a node equivalent of pythons sys.platform? [06:40] whyme: admc: try process.platform [06:41] admc: hah awesome, should have tried that [06:41] admc: thanks [06:45] hunterloftis has joined the channel [06:46] sechrist has joined the channel [06:46] sechrist has joined the channel [06:47] Metapony has joined the channel [06:47] mfernest has joined the channel [06:51] k04n has joined the channel [06:51] sechrist has joined the channel [06:57] sechrist has joined the channel [06:58] m64253 has joined the channel [06:58] techwraith has joined the channel [06:58] sechrist has joined the channel [07:02] Metapony: Is there any general consensus between using mongodb vs couchdb with node.js? I was leaning towards mongo for my app, but now think I might use couchdb and just port the db layer to mongo if I get fed up with couchdb... [07:03] sechrist has joined the channel [07:04] whyme has left the channel [07:06] sechrist has joined the channel [07:09] Matt__ has joined the channel [07:09] sechrist has joined the channel [07:09] mynyml has joined the channel [07:11] mikedeboer has joined the channel [07:14] creationix: ACTION needs a way to register event source callbacks [07:14] sechrist has joined the channel [07:15] sechrist has joined the channel [07:16] hzin has joined the channel [07:17] eresair has joined the channel [07:19] Fuld: If one could have an infinite number of threads, would blocking be better because it simplifies programming? [07:19] muhqu has joined the channel [07:20] tim_smart: creationix: Event source? [07:21] sechrist has joined the channel [07:21] sechrist has joined the channel [07:22] teddy_ has joined the channel [07:22] creationix: yeah, anything that creates a new stack is an event source [07:22] creationix: like a setTimeout's timer expiring [07:22] creationix: or a data packet coming in on a stream [07:22] Fuld: The the benefit of async seems to be that the same thread can handle another client instead of doing nothing while waiting on some I/O operation to complete [07:22] creationix: or a new http request [07:23] tim_smart: And you want to keep track of them? [07:23] creationix: not keep track of them, but be able to insert a shim when they get registered [07:24] creationix: a timer event will never fire unless there is some code that tells node to set a timeout [07:24] creationix: so that timeout belongs to another stack [07:25] mscdex: admc: there is also other information in the os module in recent versions of 0.3.x [07:25] creationix: tim_smart: here is how I can hook into setTimeout and setInterval https://gist.github.com/798197 [07:25] creationix: but other events aren't so 1:1 [07:26] creationix: especially with event emitters [07:27] creationix: there can be events emitted that nobody registered for directly [07:27] creationix: but someone somewhere started the process that eventually resulted in that event [07:29] Bonuspunkt has joined the channel [07:29] sechrist has joined the channel [07:30] sechrist has joined the channel [07:30] tim_smart: Sounds like a lot of small hacks are needed. [07:31] mike5w3c_ has joined the channel [07:32] tim_smart: You can take advantage of the whole callback is the last argument paradigm though. [07:32] tim_smart: And modify the EventEmitter prototype [07:32] tim_smart: And you got most things covered. [07:34] creationix: possibly [07:34] creationix: well, better get off to bed [07:37] saschagehlich has joined the channel [07:39] peol has joined the channel [07:40] jimt_ has joined the channel [07:40] sechrist has joined the channel [07:41] liar has joined the channel [07:42] sechrist has joined the channel [07:43] SamuraiJack has joined the channel [07:43] yx has joined the channel [07:43] sechrist has joined the channel [07:44] aku has joined the channel [07:44] sechrist has joined the channel [07:47] kolbyjack has joined the channel [07:50] springify has joined the channel [07:51] Metapony_ has joined the channel [07:51] GasbaKid has joined the channel [07:51] sechrist has joined the channel [07:53] k3yvn has joined the channel [07:54] sechrist has joined the channel [07:55] qFox has joined the channel [07:55] sechrist has joined the channel [07:57] dekz has joined the channel [07:58] ChrisPartridge has joined the channel [07:59] MikhX has joined the channel [08:00] ROBOTARMY has joined the channel [08:00] abraham has joined the channel [08:05] unomi has joined the channel [08:07] sriley has joined the channel [08:08] mraleph1 has joined the channel [08:09] whyme has joined the channel [08:09] peol has joined the channel [08:09] peol has joined the channel [08:10] mAritz has joined the channel [08:13] sechrist has joined the channel [08:14] sechrist has joined the channel [08:17] kubrow has joined the channel [08:18] ossareh has joined the channel [08:19] skot has joined the channel [08:19] keks: does anyone have a v8 doxygen of the version node is using? I get weird errors and don't know if I got the api right [08:19] mscdex: what's the error? [08:20] SamuraiJack has joined the channel [08:20] sechrist has joined the channel [08:21] syntheze has joined the channel [08:22] jbpros has joined the channel [08:22] romainhuet has joined the channel [08:22] sechrist has joined the channel [08:23] romainhuet has joined the channel [08:24] tfcoding has joined the channel [08:24] stephank has joined the channel [08:24] Moominpapa has left the channel [08:25] ErikCorry has joined the channel [08:25] pedrobelo has joined the channel [08:25] romeo_ordos has joined the channel [08:25] romeo_ordos has left the channel [08:26] sechrist has joined the channel [08:26] mikedeboer has joined the channel [08:26] ROBOTARMY has joined the channel [08:27] keks: mscdex: hang on [08:27] ivanfi has joined the channel [08:28] keks: mscdex: http://dpaste.com/358337/ [08:29] keks: apostrophes are broken for me.. [08:31] keks: already tried without casting the string of cause, that was an act of desparation [08:31] sudoer has joined the channel [08:33] mAritz has joined the channel [08:33] sechrist has joined the channel [08:34] Fullmoon has joined the channel [08:34] mscdex: what is the data type for the argument to mp()? [08:35] mscdex: is it a c string ? [08:35] mscdex: er wait nvm [08:36] keks: hmk [08:36] mscdex: you should be using Local and Local [08:36] keks: okay ill try that [08:36] mscdex: for the first two vars [08:37] [AD]Turbo has joined the channel [08:37] pyrotechnick has joined the channel [08:37] sechrist has joined the channel [08:37] adambeynon has joined the channel [08:37] mscdex: and the third var should instead be set to the value of mp->Utf8Value() [08:37] [AD]Turbo: yo all [08:38] mscdex: [AD]Turbo: yo [08:38] [AD]Turbo: ciao mscdex [08:39] peol has joined the channel [08:39] peol has joined the channel [08:40] JianMeng has joined the channel [08:40] luke` has joined the channel [08:40] keks: mscdex: still not really as expected.. http://dpaste.com/358370/ [08:40] keks: :( [08:41] mscdex: oh, you changed the data type for the last var :) [08:41] mscdex: 1 sec [08:41] keks: i'll need it that way for fuse [08:42] mscdex: i thought so [08:42] svenlito has joined the channel [08:42] keks: interestingly, that's the only type conversion he's not complaining about: the one that are not entirely his own types... [08:42] sechrist has joined the channel [08:43] Druid_ has joined the channel [08:45] muhqu has joined the channel [08:46] sechrist has joined the channel [08:46] keks: check the print function in this example: http://mscdex.net/v8/shell.cc-example.html they don't use Local<> either [08:46] GasbaKid has joined the channel [08:46] Fullmoon has joined the channel [08:47] keks: actually, that line works... [08:47] mscdex: keks: they don't need to use Local in that function [08:47] mraleph1 has joined the channel [08:47] mscdex: keks: the ToCString function in there is handy to have [08:48] keks: hm okay [08:48] mscdex: keks: you could probably do something like: char* mountpoint = ToCString(args[0]->ToString()->Utf8Value()); [08:48] sbellity has joined the channel [08:48] keks: hm okay [08:48] keks: I'll try that [08:49] mscdex: then for your array you can do: Local fuseArgs_v8 = Local::Cast(args[2]); [08:50] aklt has joined the channel [08:50] fly-away has joined the channel [08:50] keks: uhm [08:50] keks: error: ‘ToCString’ was not declared in this scope [08:50] mscdex: yeah, you have to copy and paste it in there [08:51] mscdex: :) [08:51] mscdex: it's in the v8 shell example [08:51] keks: uh, damn im blind sometimes [08:52] jimt has joined the channel [08:52] keks: but I don't see why they put that in an extra function [08:53] keks: they just call it differently... [08:54] mscdex: it's just a convenience function, makes things a little more readable too [08:54] keks: hm okay [08:55] virtuo has joined the channel [08:55] sechrist has joined the channel [08:56] gJ|Alex has joined the channel [08:57] clarkfischer has joined the channel [08:57] devrim has joined the channel [08:57] sechrist has joined the channel [08:57] sechrist has joined the channel [08:59] q_no has joined the channel [08:59] keks: okay now there still is an array and an object that was commented out before. [08:59] keks: but there are no ToArray() or ToObject methods [09:01] jimt_ has joined the channel [09:02] keks: mscdex: http://dpaste.com/358422/ [09:03] mscdex: keks: line 34 has to be: Local fuseArgs_v8 = Local::Cast(args[2]); [09:03] keks: ahh [09:04] mscdex: keks: line 33 has to be: Local