[00:04] jackyyll: what's the best way to check if a file exists with node?
[00:05] RayMorgan has joined the channel
[00:09] Cainus has joined the channel
[00:12] aryounce: jackyyll: path.exists
[00:13] aryounce: jackyyll: but you'll have to roll your own synchronous version
[00:13] aryounce: jackyyll: or take the one from the src/node.js
[00:13] aryounce: file
[00:21] jackyyll: yeah it doesn't make sense to me to have an async version of that ...
[00:21] jackyyll: dont see the point
[00:23] abadr has joined the channel
[01:07] _ry: jackyyll: fs.stat
[01:11] konobi: jackyyll: depends on how I/O bound your machine is... it may take a very long time to get a response back from the filesystem about wether something exists... time that you app could actually be doing something
[01:12] jackyyll: i suppose
[01:14] Tim_Smart has joined the channel
[01:15] jackyyll: hey tim
[01:15] Tim_Smart: yo
[01:15] jackyyll: hows it going
[01:15] Tim_Smart: Not bad. Have been in the world of Firefox extension development. The side of Javascript that give ehasaches
[01:15] Tim_Smart: *headaches
[01:16] Tim_Smart: all this stupid scope and security stuff
[01:16] jackyyll: lol
[01:20] dekz has joined the channel
[01:21] brapse has joined the channel
[01:26] alisdair has joined the channel
[01:27] eelco has joined the channel
[01:30] konobi: Tim_Smart: I once had to do a thunderbird extension... _definitely_ not fun
[01:30] Tim_Smart: konobi: Yeah, atm I'm working some on greasemonkey
[01:31] Tim_Smart: Trying to find some alternate ways of injecting scipts in page land
[01:36] aryounce: Just wrote my own .ini parsing function, and then looked in lib/ because of a namespace conflict... doh
[01:37] MattJ: Heh
[01:38] pavelz has joined the channel
[01:38] jacobat has joined the channel
[01:38] beppu has joined the channel
[01:38] cadorn has joined the channel
[01:43] aryounce: On the plus side the included version isn't parsing properly
[01:43] aryounce: so maybe it wasn't a waste of time after all
[01:43] _ry: aryounce: patch please :)
[01:44] aryounce: _ry: Will do :)
[01:44] _ry: aryounce: or at add to test/simple/test-ini.js
[01:47] bpot has joined the channel
[01:55] MattJ: _ry: Have you ever looked at Lua?
[01:55] aho has joined the channel
[01:55] MattJ: Curious, after hearing you just "need closures" :)
[01:59] _ry: MattJ: a bit
[01:59] Cainus has joined the channel
[02:00] konobi: io language looks kinda interesting too
[02:00] _ry: MattJ: javascript has additional benifits (vm arms race, large programmer base)
[02:00] MattJ: Yeah, both attractive
[02:00] _ry: *benefits
[02:00] alisdair: io is really cool, but no one uses it
[02:01] MattJ: Still tempted though, if it's not hard, to see if it would be possible to s/V8/Lua/ in Node
[02:01] _ry: MattJ: lua comes with some io libraries - which is unfortunate
[02:01] _ry: difficult to force people to do non-blocking i/o
[02:01] MattJ: Standard library?
[02:01] _ry: yeah
[02:02] _ry: i guess you could hack it to not expose those
[02:02] konobi: _ry: did you see my pull request at all?
[02:02] MattJ: Sure
[02:02] alisdair: lua's vm is orders of magnitude faster than v8
[02:02] _ry: konobi: yes
[02:02] konobi: not sure it's the right approach, but something along those lines might be good
[02:02] MattJ: alisdair: That's one reason I'm tempted
[02:03] softdrink has joined the channel
[02:03] _ry: konobi: what are you trying to do?
[02:04] _ry: (with --paths)
[02:04] konobi: _ry: be able to ask node itself where it believes it's install prefix and library paths should be
[02:05] ditesh|cassini has joined the channel
[02:06] _ry: konobi: but why? i ask because i was thinking about adding a "-I" flag (to modify require.paths) and i wodner if that'd help?
[02:07] Tim_Smart: konobi: process.installPrefix?
[02:07] _ry: i'd also like to add a "-e" flag to execute from the command line
[02:07] konobi: Tim_Smart: only available if you have a script to run
[02:08] Tim_Smart: Right, I see where you are coming from
[02:08] _ry: if that was done "./node -e 'process.stdio.write(process.installPrefix)' would work
[02:08] konobi: i'm needing the details for a makefile, for example
[02:08] konobi: it would also simplify some of the stuff in node-waf
[02:08] konobi: it could just ask node directly instead of doing a lot of legwork
[02:09] creationix has joined the channel
[02:09] creationix: _ry: how do I create a write-only tcp connection using evcom
[02:09] _ry: creationix: write-only?
[02:09] creationix: non-duplex
[02:09] creationix: or is that possible
[02:10] _ry: creationix: the other side has to shutdown
[02:10] _ry: afaik
[02:10] creationix: so then what is an evcom_writer?
[02:10] _ry: it's for pipe()
[02:10] abadr has left the channel
[02:11] voodootikigod_ has joined the channel
[02:11] _ry: creationix: use evcom_stream for tcp
[02:11] creationix: ok
[02:11] creationix: if I'm only interesting in writing to the tcp socket, do I need to wait for connect before writing
[02:12] _ry: konobi: okay 
[02:13] _ry: i'll merge it after dinner - i maybe combine it with --cflags
[02:13] _ry: like ./node --vars
[02:13] _ry: or something
[02:13] konobi: _ry: perl -V, or the like
[02:14] konobi: that's what i was sort of thinking like... but again... may not be the best approach, etc.
[02:27] peregrine81 has joined the channel
[02:28] unomi has joined the channel
[02:41] masuidrive has joined the channel
[02:41] binary42 has joined the channel
[02:45] BryanWB has joined the channel
[02:47] RayMorgan has joined the channel
[02:49] BryanWB: I don't know much about messaging systems, but I have been working w/ a rails app that is kluged together
[02:49] BryanWB: w/ apache activemq
[02:50] BryanWB: are folks in the node community thinking about tools to replace such kluges?
[02:53] alisdair: aren't node and messaging systems rather orthagonal?
[03:00] BryanWB: alisdair, umm, u could use one pool of node.js servers for handling requests and another for messages
[03:00] BryanWB: node.js doesn't only have to be used for handling http requests
[03:08] hassox has joined the channel
[03:22] BryanWB has joined the channel
[03:22] aryounce has joined the channel
[03:31] voodootikigod_ has joined the channel
[03:33] BryanWB has joined the channel
[03:39] voodootikigod_ has joined the channel
[03:40] aryounce has joined the channel
[03:41] unomi has joined the channel
[03:46] dandean has joined the channel
[03:48] konobi: _ry: about?
[03:57] masuidrive has joined the channel
[04:03] aryounce has joined the channel
[04:17] isaacs has joined the channel
[04:17] rektide: BryanWB: ryah is working on an AMQP client, which is about as pure MQ as it gets.  its just a library written in node.js, there's other MQ libraries.
[04:26] BryanWB: rektide, neat
[04:32] rektide: "good luck" with ActiveMQ
[04:35] masuidrive: Yes! I almost coded node-mysql! I'll build real application for testing.
[04:35] cedricv has joined the channel
[04:37] masuidrive: BTW I found http://github.com/Sannis/node-mysql project name is conflict. maybe users will be confuse.
[04:38] masuidrive: I'll change my project name to node-mysql-js or node-mysql-async . Do you have good suggestion?
[04:41] PyroPeter has joined the channel
[04:45] tlrobinson_ has joined the channel
[04:51] steadicat has joined the channel
[04:57] kriskowal has joined the channel
[05:00] Tim_Smart: _ry: Got around to the registerExtension patch http://github.com/Tim-Smart/node/commit/f9545cc9b9f0d867f22e5b6a593cdbc499614ee2
[05:00] Tim_Smart: Comments / improvements / etc welcome by anyone
[05:01] _ry: Tim_Smart: okay - i'll have a look at it tomorrow
[05:06] isaacs: hey, this is odd: http://github.com/Tim-Smart/node/commit/f9545cc9b9f0d867f22e5b6a593cdbc499614ee2#L0R524
[05:06] isaacs: i thought it was .node, not .addon
[05:07] Tim_Smart: yeah, it was previously in the source :/
[05:07] Tim_Smart: so I left it
[05:09] micheil has joined the channel
[05:20] gwoo has joined the channel
[05:20] creationix has joined the channel
[05:21] creationix: _ry: I'm almost done with this silly homework
[05:21] creationix: I've got one question
[05:21] gwoo has joined the channel
[05:21] creationix: how do I pass an arguments to the callback of an ev_timeout
[05:22] creationix: I tried enabling nesting functions in gcc, but the closures don't work the same as javascript's
[05:23] BryanWB has joined the channel
[05:25] Tim_Smart: creationix: I remember using ev_timeout's a little while ago, I'll see if I can find my code on it
[05:25] creationix: ok, thanks
[05:29] Tim_Smart: creationix: I think I tossed the code, but this was extremely useful http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#code_ev_timer_code_relative_and_opti
[05:29] Tim_Smart: but no doubt you looked there
[05:29] creationix: yep
[05:30] creationix: I'm so close, but this project is due in 30 minutes
[05:30] creationix: all I need is simple closures
[05:30] Tim_Smart: from memory, I called a callback from the callback
[05:30] Tim_Smart: :p
[05:30] creationix: hmm, manual closure type stuff
[05:33] Tim_Smart: creationix: Ah, this might help http://github.com/ry/node_postgres/blob/master/binding.cc#L523
[05:33] Tim_Smart: not exactly a timeout callback, but they are structured in a similar fashion
[05:34] creationix: I'm using plain C, I'm not if I can use that
[05:35] Tim_Smart: Ah right, I'm pretty novice to C and C++, so hopefully someone else can answer in 30 minutes :p
[05:37] creationix: here's the relevent snippet
[05:37] creationix: http://pastie.org/859089
[05:37] masuidrive has joined the channel
[05:42] masuidri_ has joined the channel
[05:43] dekz has joined the channel
[05:43] _ry: creationix: pack stuff into a struct and set watcher.data
[05:43] creationix: watcher.data, awesome
[05:44] creationix: I just need an int
[05:44] creationix: do you have a quick example
[05:45] _ry: mm..
[05:47] _ry: http://github.com/ry/evcom/blob/612fc6227e729d50a7686ff38071e301ad6df770/evcom.c#L996 ?
[05:50] _ry: konobi: pon
[05:50] _ry: g
[05:55] BryanWB has joined the channel
[05:56] konobi: _ry: nm... looks like there's no way to doing something along the lines of lwatchFile()
[05:58] creationix: _ry: That worked, thanks!
[05:59] creationix: git clone http://git.creationix.com/ricart
[05:59] creationix: if you want to see my trials and tribulations
[05:59] creationix: I haven't slept in 36 hours
[05:59] _ry: :
[06:00] creationix: the good news is I know a lot more about evcom and libev now
[06:00] creationix: so next time I give a talk on node I won't be completely pulling it out of thin air
[06:02] _ry: is it working
[06:02] _ry: ?
[06:03] creationix: it is on my mac, I'm testing on the solaris machine now
[06:03] creationix: it takes a while to compile everything
[06:05] _ry: creationix: were you supposed to use event loops?
[06:05] creationix: it was our option
[06:05] creationix: I chose event loops over threads for fun
[06:05] Tim_Smart: creationix: What was the project?
[06:06] creationix: just a school project to simulate Ricart Agrawala algorithm
[06:09] kennethkalmer has joined the channel
[06:10] creationix: crap, does sparc have different endieness than osx
[06:11] creationix: I'm getting bug errors on every process
[06:11] creationix: s/bug/bus/
[06:15] creationix: simulation.c:45: warning: cast to pointer from integer of different size
[06:15] creationix: I'll bet that's it, it's using the wrong end of the pointer 
[06:15] creationix: oh well, off to bed
[06:17] _ry: yes
[06:20] halorgium: 
ACTION wonders why people write C without considering endianz

[06:24] blastbyte has joined the channel
[06:25] creationix: fixed the pointer cast issue, now the tcp connections are being refused
[06:25] Tim_Smart: modula is a nice alternative to C as well
[06:25] Tim_Smart: You can use C libraries with it also
[06:25] creationix: I hear go is pretty nice
[06:25] creationix: I wonder if it works with v8 yet
[06:26] steadicat has joined the channel
[06:31] tlrobinson_ has joined the channel
[06:33] BryanWB has joined the channel
[06:34] brapse has joined the channel
[06:41] bronson has joined the channel
[06:52] aguynamedben has joined the channel
[06:53] abadr has joined the channel
[06:53] abadr: anyone know when net2 is going to be merged?
[06:54] r11t has joined the channel
[06:56] kennethkalmer has joined the channel
[06:56] BryanWB has joined the channel
[06:56] kennethkalmer has joined the channel
[07:05] tlrobinson_ has joined the channel
[07:21] r11t_ has joined the channel
[07:27] kriskowal: _ry i'd really like to get the main() decoupled from most of the other stuff in node.cc
[07:29] kriskowal: i'm presently doing a manual patch to my copy of it for the narwhal main()
[07:31] kriskowal: i'm more than happy to do the work; just need a hint as to what you'd consider merge-worthy.
[07:32] michaelk^ has joined the channel
[07:32] kriskowal: perhaps break it into node.{h,cc} and node_common.{h,cc}?
[07:47] mikeal has joined the channel
[07:49] markwubben has joined the channel
[07:52] syd_ has joined the channel
[07:54] unomi has joined the channel
[07:54] unomi has joined the channel
[07:54] syd_: http://pastie.org/859187
[07:54] syd_: any idea what might be causing that?
[08:06] konobi: why does http server lowercase header names?
[08:07] felixge has joined the channel
[08:07] felixge has joined the channel
[08:10] kjeldahl has joined the channel
[08:12] Tim_Smart: syd_: Have you tried re-cloning the latest HEAD?
[08:13] syd_: I only pulled for the first time about half an hour ago
[08:13] Tim_Smart: hmm ok. What OS you running?
[08:13] r11t has joined the channel
[08:13] syd_: ubuntu jaunty
[08:15] Tim_Smart: konobi: HTTP doesn't explicitly say what casing needs to be used, so it is lower cased for 
[08:15] Tim_Smart: *so you don't run into surprises I guess
[08:16] Tim_Smart: syd_: Hmm ok, I have run node on jaunty before, all I needed was build-essential and node's source
[08:16] syd_: strange, I have build-essential in there
[08:16] Tim_Smart: so I assume maybe something is gone weird with waf or gcc
[08:17] syd_: maybe I'll git revert back a bit and see if it worked at a later point
[08:18] Tim_Smart: OK. Give it a shot, otherwise start a thread in the mailing list
[08:25] syd_: yeah, no luck.
[08:25] syd_: I'll post on the mailing list.
[08:27] michaelk^ has joined the channel
[08:47] erikvold has joined the channel
[08:58] markwubben_ has joined the channel
[09:10] pdelgallego has joined the channel
[09:11] piranha has joined the channel
[09:12] davidjrice has left the channel
[09:24] christkv has joined the channel
[09:31] rtomayko has joined the channel
[10:10] Mandar has joined the channel
[10:12] ithinkihaveacat has joined the channel
[10:14] felixge has joined the channel
[10:14] felixge has joined the channel
[10:29] Tim_Smart has left the channel
[10:38] spoob has joined the channel
[10:53] jed_ has joined the channel
[11:28] micheil_mbp has joined the channel
[11:32] JonGretar has joined the channel
[11:48] rolfb has joined the channel
[12:58] alex-desktop has joined the channel
[13:03] peregrine81 has joined the channel
[13:11] binary42 has joined the channel
[13:17] softdrink has joined the channel
[13:31] kriszyp has joined the channel
[13:33] felixge has joined the channel
[13:33] felixge has joined the channel
[13:41] cedricv has joined the channel
[13:42] peregrine81 has joined the channel
[14:01] sztanpet has joined the channel
[14:11] n8o has joined the channel
[14:12] gwillickers has joined the channel
[14:14] davidsklar has joined the channel
[14:15] gwillickers has left the channel
[14:18] pdelgallego has joined the channel
[14:30] steadicat has joined the channel
[14:39] cedricv has joined the channel
[14:42] gf3 has joined the channel
[14:48] mattly has joined the channel
[14:48] Booster has joined the channel
[14:59] mattly has joined the channel
[15:00] MattJ has joined the channel
[15:01] alex-desktop has joined the channel
[15:26] bronson has joined the channel
[15:29] hassox has joined the channel
[15:35] dnolen has joined the channel
[15:36] binary42 has joined the channel
[15:37] binary42 has joined the channel
[15:38] dnolen has joined the channel
[15:45] steadicat has joined the channel
[15:52] aryounce has joined the channel
[15:55] cloudhead has joined the channel
[15:55] piranha has joined the channel
[15:58] around has joined the channel
[16:02] cozby has joined the channel
[16:06] cozby has joined the channel
[16:13] brapse has joined the channel
[16:14] alexiskander has joined the channel
[16:34] siculars has joined the channel
[16:49] mattly has joined the channel
[16:51] joshbuddy has joined the channel
[16:56] bpot has joined the channel
[16:58] kennethkalmer has joined the channel
[17:07] RayMorgan has joined the channel
[17:07] dnolen has joined the channel
[17:12] felixge has joined the channel
[17:14] CIA-77: node: 03Ryan Dahl 07master * r209b219 10/ src/node.cc : Add environmental varibles to help text - http://bit.ly/ctQYXS
[17:14] CIA-77: node: 03Felix Geisendörfer 07master * rc473915 10/ doc/api.txt : Fix typo in fs.writeStream docs - http://bit.ly/dvuJfX
[17:14] CIA-77: node: 03Felix Geisendörfer 07master * r9a9f08b 10/ (4 files in 3 dirs): 
[17:14] CIA-77: node: Add callbacks to stream methods
[17:14] CIA-77: node: Allows for more fine graining, especially finding out about an individual
[17:14] CIA-77: node: chunk of data being flushed in a write stream rather than the whole queue.
[17:14] CIA-77: node: This commit also fixes a bug causing forceClose to fail on a readStream that
[17:14] CIA-77: node: did not finish opening yet. - http://bit.ly/apNaHf
[17:14] CIA-77: node: 03Ryan Dahl 07master * r40e42e8 10/ src/node.cc : Replace --cflags with --vars
[17:15] mikeal has joined the channel
[17:23] aguynamedben has joined the channel
[17:29] christkv: felixge: there's a pretty bad bug in the latest v8 you guys merged into master
[17:29] christkv: http://code.google.com/p/v8/issues/detail?id=636
[17:30] felixge: christkv: I'd say "pretty weird", not sure what real implication this would have
[17:30] christkv: it breaks the mongodb bson serialization
[17:30] felixge: christkv: did you experience it in any node code?
[17:30] felixge: christkv: fair enough :)
[17:31] felixge: Well, usually v8 fixes these things quickly, but _ry could also revert the v8 upgrade
[17:31] christkv: felixge: a better explenation http://github.com/christkv/node-mongodb-native/issues#issue/3
[17:31] christkv: felixge: seems to be some sort of bug in the jitter is my thought since adding a "nop" call seems to work around it
[17:32] _ry: there's been a few releases 
[17:32] _ry: i can update 
[17:32] felixge: _ry: not sure if this would help, seems like the bug has not been specifically adressed
[17:32] felixge: does the changelog say anything ?
[17:33] _ry: no
[17:33] felixge: christkv: do you know which v8 version introduced the bug?
[17:34] _ry: but, yeah. i think they'll fix this very quickly if it doesn't match other js engines
[17:34] christkv: can't tell you unfortunately. I got it from one of the users.
[17:35] christkv: but after 0.1.31 that's for sure
[17:35] felixge: christkv: it says you found a workaround right?
[17:35] christkv: one I'm not applying lol
[17:35] felixge: :)
[17:35] christkv: work around is a do nothing operation in the for loop
[17:36] felixge: christkv / everybody: We should all "star" the issue so it gets attention from the v8 guys
[17:37] christkv: star == add a comment right
[17:37] felixge: christkv: no, there is a little star icon on the top left
[17:38] christkv: my bad found it
[17:38] _ry: felixge: http://github.com/ry/node/issues#issue/75
[17:38] felixge: christkv: you may still want to consider the workaround for the time being
[17:39] felixge: _ry: huh, I think he's right. I wonder why this worked locally for me so
[17:39] spot__ has joined the channel
[17:39] felixge: _ry: fixing it
[17:39] jviereck has joined the channel
[17:39] felixge: _ry: ah, the test suite exports 'fs' as a global, right?
[17:39] felixge: damn, global variables FTL
[17:39] jviereck has left the channel
[17:40] _ry: felixge: yeah - i suppose that should be fixed
[17:40] _ry: i do that all the time
[17:40] _ry: fucking javascript
[17:40] felixge: _ry: :)
[17:41] felixge: _ry: let me fix the test suite, then the bug
[17:41] orlandov: amen
[17:41] CIA-77: node: 03Ryan Dahl 07master * rc98b079 10/ (46 files in 9 dirs): Upgrade V8 to 2.1.2.6 - http://bit.ly/aSSNkF
[17:43] jviereck has joined the channel
[17:43] voodootikigod has joined the channel
[17:46] around: dont mess around
[17:46] around: know when 2 pull out
[17:50] _ry: felixge: thanks
[17:53] dandean has joined the channel
[17:57] qFox has joined the channel
[17:57] technoweenie has joined the channel
[18:00] felixge: christkv: I think the v8 bug got noticed :) http://code.google.com/p/v8/issues/detail?id=636&colspec=ID%20Type%20Status%20Priority%20Owner%20Summary%20HW%20OS%20Area%20Stars&start=100#makechanges
[18:01] unomi has joined the channel
[18:02] dnolen has joined the channel
[18:07] felixge: _ry: http://github.com/felixge/node/commits/file2
[18:07] piranha has joined the channel
[18:08] felixge: _ry: I actually found another semi-bug while doing this
[18:08] felixge: http://github.com/felixge/node/commit/10fc429eb63fb84f0a8f118ac5a9d529fddfb20a#L12L2
[18:12] kriszyp has joined the channel
[18:15] bronson has joined the channel
[18:15] RayMorgan_ has joined the channel
[18:18] RayMorgan__ has joined the channel
[18:21] paulca has joined the channel
[18:22] CIA-77: node: 03Ryan Dahl 07master * re898c1f 10/ (38 files in 5 dirs): Upgrade WAF to 1.5.14 - http://bit.ly/abPlDi
[18:23] CIA-77: node: 03Felix Geisendörfer 07master * r44dde5f 10/ (19 files in 2 dirs): 
[18:23] CIA-77: node: Remove fs/path globals from test suite
[18:23] CIA-77: node: It is way too easy to not notice bugs due to those modules being
[18:23] CIA-77: node: exported as globals while running the test suite.
[18:23] CIA-77: node: See: http://github.com/ry/node/issues#issue/75 - http://bit.ly/bSKXTW
[18:23] CIA-77: node: 03Felix Geisendörfer 07master * r3947270 10/ lib/fs.js : 
[18:23] CIA-77: node: fs.createWriteStream throws fs undefined error
[18:23] CIA-77: node: The improved test suite now catches this bug, so hopefully thing like
[18:23] CIA-77: node: this will not happen again.
[18:23] CIA-77: node: See: http://github.com/ry/node/issues#issue/75 - http://bit.ly/bX6VZN
[18:23] _ry: felixge: thanks--^
[18:23] alisdair has joined the channel
[18:24] felixge: _ry: np. Makes the tests a little more verbose, but keeps idiots like me from breaking the build unnoticed :)
[18:24] _ry: including waf, scons, v8, libev, libeio in node has worked out really well. 
[18:24] _ry: so nice to know explicitly what code is being run
[18:24] felixge: _ry: yeah the build system is very nice : )
[18:24] felixge: also the new test suite is sweet
[18:27] JAAulde has joined the channel
[18:44] bajeczka has joined the channel
[18:54] alisdair has joined the channel
[19:08] tlrobinson_ has joined the channel
[19:08] tilgovi has joined the channel
[19:08] jviereck has joined the channel
[19:09] r11t has joined the channel
[19:10] javajunky has joined the channel
[19:10] javajunky: lo ;)
[19:13] isaacs has joined the channel
[19:13] CIA-77: node: 03Ryan Dahl 07master * rc9c13e0 10/ test/simple/test-http-tls.js : require('fs') in test-http-tls - http://bit.ly/bSGAFs
[19:13] javajunky: Before I embarrass myself by trying to fix it, is process.mixin currently working fully ? ..i.e. I expected the outcome of sys.puts(sys.inspect(process.mixin(true,  {b: {foo:"bar"}}, {a: 'x', b: {c:"d"}}) ));  to be   { b: { foo:"bar", c:"d"}, a:'x'}   but I actually get { b: {  c:"d"}, a:'x'}
[19:13] mikeal has joined the channel
[19:14] r11t has joined the channel
[19:15] isaacs: javajunky: it doesn't do a deep copy
[19:15] isaacs: javajunky: so, the "b" member from the second is overwriting the "b" member on the first
[19:16] javajunky: so the first argument 'deep' isn't actually listened to ?
[19:17] felixge: _ry: sorry, the tls test doesn't run on my system
[19:17] felixge: javajunky: I think this is a bug
[19:18] javajunky: I think I can prolly fix it and submit a patch.
[19:18] r11t has joined the channel
[19:19] felixge: javajunky: sweet : )! Ping me and I'll review
[19:20] mikeal has joined the channel
[19:27] maritz has joined the channel
[19:28] maushu has joined the channel
[19:29] sveisvei has joined the channel
[19:31] voodootikigod: i'd pay for some action...
[19:31] felixge: voodootikigod: what kind of action? :)
[19:32] voodootikigod: something better than "ncsa-mosica is on github"
[19:33] maushu: Besides require what are the other ways to insert external code?
[19:33] isaacs: javajunky: ah, didn't catch that you passed "true" as the first arg.  yes, that looks bugish
[19:33] felixge: maushu: fs.cat() + process.compile()
[19:33] maushu: felixge, thanks.
[19:33] felixge: * fs.readFile()
[19:34] isaacs: javajunky: hey, btw, i have a feeling that process.mixin could be a lot faster if it used Object.create rather than copying in a for/in loop
[19:34] felixge: maushu: check src/node.js for inspiration on rolling your own module system
[19:34] malte_ has joined the channel
[19:34] felixge: process.mixin = the devil
[19:34] felixge: could really need some refactoring badly
[19:34] felixge: :)
[19:34] javajunky: currently my head is melting ;) 
[19:34] maushu: Not my own module system, I want to create a (web) manager that allows me to load, unload node programs.
[19:34] isaacs: javajunky: something like this: http://gist.github.com/306377
[19:35] isaacs: javajunky: the issue then is that it won't be an "own property" of the target object
[19:35] felixge: isaacs: that looks lovely, you should make a patch for node
[19:35] felixge: :)
[19:35] felixge: and replace mixin
[19:35] isaacs: _ry: thoughts? ^
[19:36] javajunky: ;) suits me
[19:37] malte_: felixge: How is your work on dirty/.30 coming along?
[19:37] felixge: malte_: flush() is a little more difficult to get right than I thought
[19:38] malte_: Do you have a branch on github?
[19:38] felixge: malte_: but it should be easier now that I got file streams done
[19:38] felixge: malte_: no, I could push one so
[19:38] malte_: would be cool
[19:38] felixge: fuck, open source distracts me from transloadit too much :)
[19:38] felixge: anyway, let me spend a dirty hour
[19:38] felixge: and see what I can do
[19:38] felixge: :)
[19:38] felixge: 
ACTION could need a club mate

[19:39] felixge: 
ACTION got addicted recently

[19:39] gwoo:  NO
[19:39] gwoo: felixge: don't do it
[19:39] gwoo: you remember what happend to the last time
[19:39] felixge: gwoo: no, was it good?
[19:40] gwoo: if you consider 48 coding sessions and 48 sleeping sessions to be a good idea
[19:40] gwoo: 48hr
[19:40] sxs has joined the channel
[19:40] sxs: hello
[19:40] sveisvei: felixge:  zip it, and work it - i need _dirty_ for dirty app here :P
[19:41] felixge: sveisvei: ok, headed out to get a club mate! Let's do some hacking
[19:42] sxs: i compiled nodejs on two macs but on the third machine the compile hangs up. no error. all three machines have osx 10.6. how can i debug this?
[19:42] gwoo: and you thought felixge was bad after a few beers
[19:42] gwoo: wait until the club mate kicks in
[19:42] gwoo: :)
[19:45] mau has joined the channel
[19:49] malte_ has joined the channel
[19:49] kjeldahl_ has joined the channel
[19:50] charlenopires has joined the channel
[19:51] pedrobelo has joined the channel
[19:52] joshbuddy has joined the channel
[19:52] joshbuddy has joined the channel
[19:53] felixge: ok, all fueled up, time to get dirty
[19:54] gwoo: oh no
[19:54] gwoo: just when the voodootikigod was looking to pay for something
[19:55] gwoo: felixge: would there be any benefit to writing dirty in c++ and including it in node as an embedded db?
[19:56] JimBastard has joined the channel
[19:56] spot__: FYI... http://twitter.com/spotnyk/status/10185260832
[19:56] gwoo: oh
[19:56] drostie has joined the channel
[19:56] felixge: gwoo: yes, it could be made faster and more memory efficient. But for now that's my lowest priority since it's already "fairly" good in both categories
[19:57] gwoo: and i guess it is so easily included that being embedded in node would not be all that advantageous
[19:57] felixge: V8 might be getting slower due to rolling some Ecma5 stuff *wild guess*
[19:57] voodootikigod: what about me
[19:58] felixge: but I'm mean it's only the laws of physics in between google's resources spend on V8 and optimal performance - so I'm not worried in the long run
[20:01] isaacs: yikes, there's a continue statement inside a forEach in process.mixin
[20:01] isaacs: that's no good
[20:02] isaacs: js> f = {foo:"bar"}; g = {f:f}; process.mixin(true, f, g)
[20:02] gbot2: isaacs: Error: ReferenceError: process is not defined 
[20:02] isaacs: (well, it'd throw if you had access to process)
[20:02] Guest30751: FIX IIIIIIT
[20:02] orlandov: does anyone know if the HTML4 webdatabase query method supports multiple sqlstatements per single call?
[20:02] orlandov: s/HTML4/HTML5/
[20:03] orlandov: its not immdediately obvious from the w3 site
[20:03] maushu: orlandov, it shouldn't.
[20:03] orlandov: kk... i know that sqlite can support that via prepare and step
[20:04] orlandov: and tail
[20:04] maushu: Technically sqlite has nothing to do with web databases.
[20:04] maushu: Thats how some people decided to implement in the back.
[20:05] maushu: Chrome also uses sqlite for localStorage. :3
[20:05] orlandov: yeah, that's why i figured it'd be good to ask :)
[20:05] malte_: maushu: Its specced as: The same behavior as sqlite that we might eventually document so that there could be other implementations
[20:06] maushu: It doesn't say to use sqlite. It just says that it's the same behavior. :p
[20:08] javajunky: hey felixge, how'd that gist look to you ?
[20:10] felixge: javajunky: could you paste it with a '.diff' filename so github does syntax highlight?
[20:12] bpot has joined the channel
[20:13] javajunky: is it any better now ?
[20:13] mikeal: why not use gist.github ?
[20:13] mikeal: there is a drop down
[20:14] javajunky: I did ;)
[20:15] paulca has joined the channel
[20:18] brandon_beacher has joined the channel
[20:19] javajunky: i've a patch, hot to trot if thats ok with you felixge (the *limited* existing tests continue to pass)
[20:20] felixge: javajunky: the patch looks good to me. But yes, the limited tests on this function are worrying :)
[20:20] javajunky: coolI'll submit the patch to the ML :)
[20:23] _ry: we have to get rid of process.mixin
[20:23] gwoo: yes
[20:23] gwoo: :)
[20:24] gwoo: felixge: how do you feel about process.mixin?
[20:24] _ry: i just copied it in from jquery as a quick hack - now everyone is hacking it
[20:24] gwoo: _ry: yeah it's a slippery slope
[20:25] _ry: which means it needs to be user-land
[20:25] alisdair has joined the channel
[20:25] _ry: unfortunately it's a bit difficult to get rid of. 
[20:25] _ry: the tests use it very much
[20:26] felixge: _ry: I think we should provide a good deep copy 
[20:26] felixge: JS sorely lacks an implementation
[20:26] felixge: and everybody kind of wants the same thing, just bug free
[20:26] mrd` has joined the channel
[20:26] malte_: JSON.parse(JSON.stringify({ deep: { deeper: [1,2,3]}}))
[20:27] javajunky: *ironically* thats where I started out 'fixing' my problem this evening...
[20:28] _ry: doesn't ES5 have a deep copy?
[20:28] _ry: Object.clone or something?
[20:28] gwoo: i think so
[20:30] _ry: felixge: feel free to help me write http://wiki.github.com/ry/node/gsoc-application
[20:30] _ry: felixge:  :)
[20:30] mrd`: _ry: Can I suggest updating to waf-1.5.14a? It fixes a compatibility bug with older Python versions (e.g., OS X 10.5's).
[20:30] _ry: i'll have a look at it tonight
[20:30] _ry: mrd`: i just ran into that issue
[20:30] felixge: _ry: I was just waiting for their form to become available, wasn't this morning for me
[20:31] _ry: (and so did the build bots: http://buildbot.nodejs.org:8010/waterfall
[20:31] gwoo: felixge: usually it's around the 13th
[20:37] gwoo: _ry: i'll help with the gsoc application if you want
[20:38] rictic has joined the channel
[20:43] _gcr has joined the channel
[20:43] _gcr: Hey guys! I have the worst idea ever!
[20:43] _gcr: How's THIS for an anti-templating system? http://dpaste.com/169596/
[20:43] _gcr: Now, somebody convince me not to make that.
[20:49] isaacs: _ry: so, i'm looking through this a bit more... deep copy is pretty broken.
[20:49] isaacs: there's three things i've found so far that completely hose it.
[20:49] Booster has joined the channel
[20:49] isaacs: also, the "prevent infinite loop" thing very much does not prevent the infinite loops
[20:49] isaacs: and there's a continue inside a closure... ugh.
[20:49] isaacs: do the tests really use *deep* copy? or just copying?
[20:50] _ry: isaacs: they just copy
[20:50] isaacs: _ry: how would you feel about removing the deep functionality?  let that live in a module somewhere outside node core
[20:50] isaacs: it could simplify the logic quite a lot
[20:51] _ry: isaacs: does es5 have any copyin?
[20:51] javajunky: arggghh this is turning out badly for me ;) … 
[20:51] isaacs: _ry: it has Object.create, and with maybe just a few lines, you can do Object.deepCreate
[20:51] isaacs: _ry: i think perhaps a good solution is to have a process.deepCopy that makes a deep copy of an object, and process.mixin that just copies properties from source to target shallowly
[20:52] _ry: but what if i want to copy require('sys') into global?
[20:52] _ry: (my main use-cse for process.mixin)
[20:52] isaacs: _ry: well, in that case, you'd do process.mixin(require("sys")) just like before
[20:52] isaacs: and it'd do a shallow copy
[20:52] isaacs: if you want a deep copy that doesn't share refs, you could do process.mixin(process.deepCopy(require("sys")))
[20:52] _ry: i think i want to remove process.mixin
[20:53] isaacs: _ry: what about putting it into the tests/mjsunit/common.js
[20:53] maushu: Anyone knows a realtime javascript syntax highlighter in javascript?
[20:53] isaacs: node itself doesn't really use mixin anywhere, that i'm aware of.
[20:53] _ry: isaacs: there might be a few places
[20:53] _ry: yeah that'd work
[20:53] isaacs: _ry: i'd volunteer to hunt it down and eradicate.
[20:54] _ry: just have common.js copy sys manually into global
[20:54] isaacs: _ry: the better solution is to point people to the facilities already available in ES5 and supported by v8
[20:54] isaacs: "process.mixin was removed.  please see this gist for how to do it yourself: ..."
[20:54] _ry: yeah i think Object.clone can be used in many cases where process.mixin is now
[20:54] javajunky: *sob* this is going down as my least successful bug-find&patch cycle ever :( 
[20:55] isaacs: _ry: sure, or setting __proto__ manually, etc.
[20:55] _ry: we're going to have to keep process.mixin for a while and just add a message
[20:55] isaacs: javajunky: fwiw, this would've happened eventually, i think.
[20:55] _ry: i think many people use process.mixin for options
[20:55] _ry: this.options = Object.clone(options);
[20:55] isaacs: _ry: (it's Object.create)_
[20:55] _ry: er
[20:55] isaacs: also, that's faster anyhow
[20:56] _ry: this.options = Object.clone(defaults, options);
[20:56] _ry: ^-- is that right?
[20:56] _ry: oh
[20:56] _ry: create
[20:56] isaacs: since it's not iterating over the object, and uses v8's shadow objects
[20:56] _ry: :P
[20:56] isaacs: _ry: so, the diff there is that process.mixin creates "own" properties, since it's copying over.
[20:56] isaacs: _ry: so if you then do Object.getOwnPropertyNames, it'll show the new ones.  the __proto__ tree is flattened.
[20:57] kjeldahl_: Anybody having any luck with js-haml partials working, for instance the sample_app that comes with the picard framework (load "/partial"). It bombs at my end.
[20:57] isaacs: that's convenient for some things, but also slower.
[20:57] javajunky: isaacs, that example from before, will that do a deep-clone effectiely anyway ? (I'm fixing someone else' code as it happens… and that job just got work ..
[20:57] jspiros has joined the channel
[20:57] isaacs: javajunky: according to the docs, yes, what you expected was valid.  however, the code wasn't right.
[20:57] isaacs: javajunky: it looked like you actually want a deep copy-and-merge
[20:59] javajunky: 'deep-merge' it would seem ;)   I can just write one I just uess his code was depending on it previously (fwiw its the 'express' framework) 
[21:00] Tim_Smart has joined the channel
[21:00] isaacs: javajunky: sure, that's probably best.
[21:00] javajunky: cheers.
[21:03] Tim_Smart: _ry: You get around to looking at that registerExtension thing?
[21:05] RayMorgan has joined the channel
[21:05] jviereck_ has joined the channel
[21:07] javajunky: arggg process.mixin is *everywhere* :( 
[21:08] unomi has joined the channel
[21:09] _ry: Tim_Smart: looking at it now
[21:10] isaacs: javajunky: yeah, replacing one line of code with 2 all over the place is lame.  but process.mixin(true, ...) is really quite rarte.
[21:10] isaacs: and that's what's so broken
[21:12] felixge: _ry: not perfect yet, but it's a start: http://wiki.github.com/ry/node/gsoc-application
[21:14] _ry: felixge: thanks
[21:14] _ry: Tim_Smart: this seems kind of complicated
[21:15] felixge: _ry: I think the fact that you ported V8 to solaris looks very nice in the project description :)
[21:15] _ry: maybe i don't get it
[21:15] javajunky: sorry to be so dumb, I get 'deep' is borked, but what are you suggesting I replace http://github.com/visionmedia/express/blob/master/lib/express/core.js#L6 with as an example ?
[21:17] _ry: Tim_Smart: i was thinking just a hook directly before process.compile
[21:17] _ry: Tim_Smart: seems like it would be 10 line change at most
[21:18] _ry: Tim_Smart: also, did you have an evalcx patch?
[21:19] Cainus has joined the channel
[21:23] Tim_Smart: Tim_Smart: Yeah I did.
[21:23] unomi has joined the channel
[21:24] indiefan has joined the channel
[21:24] Tim_Smart: lolwut, I mentioned myself? _ry: Yeah I did
[21:27] Tim_Smart: _ry: The reason I didn't check it before process.compile, is so I could change the way it does things depending if it was require.async or just a normal require
[21:29] jfherdman has joined the channel
[21:33] maushu: Anyone here used John Resig's microtemplating?
[21:35] CIA-77: node: 03Ryan Dahl 07master * r5809483 10/ tools/wafadmin/Tools/config_c.py : Upgrade waf to 1.5.14a - http://bit.ly/buEhRV
[21:35] felixge: _ry: do you have any pointers for how big chunks one should send to fs.write() for optimal performance?
[21:36] felixge: optimal performance = writing stuff to disk as fast as possible
[21:36] felixge: maushu: I have, it's nice
[21:36] felixge: maushu: have not used it within node yet if that's what you mean so
[21:36] maushu: It destroys the tabs and newlines and I don't know what for.
[21:37] maushu: I'm trying to understand all these splits and stuff.
[21:37] maushu: str.replace(/[\r\t\n]/g, " ") <- what is this for? What is the problem of \r\t\n?
[21:38] felixge: maushu: no idea - the code is quite hackish :)
[21:39] maushu: I find it silly.
[21:39] maushu: I'm trying to recreate it.
[21:40] felixge: Tim_Smart: it seems like you overwrote my node application stuff in the wiki?
[21:40] Tim_Smart: >.<
[21:40] maushu: Should we get the pitchforks and torches?
[21:41] felixge: looks like github does not let me restore it
[21:41] felixge: Tim_Smart: can you make sure the old version gets restored?
[21:42] Tim_Smart: yup
[21:46] marcus__ has joined the channel
[21:47] maushu: felixge, I just need to fix the single quote problem and my version should supposedly work without all that anti-space stuff.
[21:47] maushu: I still don't see _why_ all that stuff is needed. It's like he found a weird bug or something that I haven't found yet.
[21:49] indiefan: maushu: and it isn't unit tested?
[21:49] abadr has joined the channel
[21:49] abadr: what's the right way to manage db connections?
[21:50] maushu: indiefan, I'm unit testing my version. Till now I haven't found anything but I haven't really tried with something really hard.
[21:50] Tim_Smart: felixge: That better? :p
[21:50] felixge: Tim_Smart: yes! Thanks
[21:51] Tim_Smart: The wiki system needs rebase >.>
[21:51] Tim_Smart: and conflict warnings
[21:52] technoweenie: i dont think the github wiki is stored in git
[21:53] jfherdman: i once started to write a wiki with a git backend. it proved kind of awkward
[21:53] technoweenie: http://github.com/al3x/git-wiki
[21:53] jfherdman: touche ;)
[21:53] maushu: Can I replace specific parts in a regex replace instead of the whole regex match?
[21:54] indiefan: ([subpattern]), no?
[21:54] maushu: indiefan, wat?
[21:54] maushu: I'm trying to replace all single quotes between those two "%>" "<%".
[21:55] maushu: Something like str.replace(/%>(')*<%/, "\'");
[21:56] indiefan: catch subpatterns in parens then refer to them with backreferences
[21:56] maushu: Huh.
[21:57] Tim_Smart: `git add --patch` is freaken awesome
[21:58] Tim_Smart: You can commit little chunks of changes within a file, and leave the reset unstaged
[21:58] Tim_Smart: *rest
[21:58] maushu: indiefan, I don't understand, how do I do that with replace?
[22:03] indiefan: Not sure what the best way to do it would be, but i'd use a callback. pass it the subpattern and then do an additional replace on that
[22:04] indiefan: maushu: here's an example of passing subpattern matches into a callback on replace: http://ejohn.org/blog/search-and-dont-replace/
[22:04] indiefan: of course, you'd return the replacement in your case
[22:05] maritz: why not do a global match, go through all matches and add a ' before/after the match?
[22:06] maushu: ...
[22:07] maritz: ?
[22:09] tilgovi has joined the channel
[22:12] erikvold has joined the channel
[22:13] indiefan: maritz: he wants to add a \ before all single quotes between %> and <%. Is there a global match that would do that?
[22:16] maritz: na, sorry. though match did something it doesn't do. stupid me
[22:16] maritz: :D
[22:16] maritz: *thought
[22:16] indiefan: cool. maushu: that callback strategy make sense?
[22:17] maushu: It's weird.
[22:18] maritz: hm, i can't think of any way this would work. *brainfart*
[22:18] maushu: Regex failed me.
[22:19] maushu: The humanity.
[22:19] maushu: I mean, *cough*, Oh The Humanity!
[22:20] maritz: you could take out the entire content in %>content<%, pass it to the callback, and then do the replace in the callback. would that work? :D
[22:20] maushu: ...isn't the copy in the callback a copy by value?
[22:21] MattJ: Use a closure!
[22:21] Tim_Smart: _ry: Braces :p Done, while I was there I did the whole lot, which you might like / not like, not sure
[22:21] MattJ: (I've no idea what your problem is, but closures make everything better)
[22:21] indiefan: maushu: something like this? http://pastium.org/view/dbeb1b9a0112d8867d8928829b63210f
[22:21] maushu: MattJ, it's still a copy by value, a substring of a string if you will. Not the string.
[22:22] maushu: Huh, that might work.
[22:22] maritz: indiefan: wouldn't you then have %>%><%<%? but else it's exactly what i meant and was just about to gist too :D
[22:22] maushu: Actually, if it works like I think it *should* work. Thanks indiefan!
[22:23] maritz: ARG, disregard me, i suck cocks
[22:24] jfherdman has left the channel
[22:24] maushu: Wait, what just happened.
[22:24] indiefan: may need to tell the pattern not to be greedy?
[22:25] kriskowal has joined the channel
[22:25] maushu: At least now I'm on the right path.
[22:26] mikeal: isaacs: did you get the hosting worked out for js-registry?
[22:27] rolfb has joined the channel
[22:30] maritz: maushu indiefan: it needs to be (.*) in the first pattern. :)
[22:30] maushu: maritz, yeah, I just found out.
[22:30] maushu: Silly bug.
[22:30] maritz: not a bug, correct behavious
[22:30] maritz: *r
[22:34] bpot has joined the channel
[22:42] dekz has joined the channel
[22:49] maushu: Huh, why is the option "s" not working on regex?
[22:53] maushu: ...javascript doesn't have s.
[22:53] christkv has joined the channel
[22:55] maushu: It seems [\s\S]* works.
[23:02] javajunky: quick q re ordering, do 'sequential' calls to fs.write (note not fs.writeSync) ..occur in the order they're called or are there no guarantees? .. currently I have an asyc callback feeding me some data, but was unsure whether I had to call fs.writeSync inside the callback to avoid corrupting the file ?
[23:09] dekz has joined the channel
[23:10] felixge: club mate session was successful, got the basics of dirty working again on node 0.1.30+ : http://github.com/felixge/node-dirty/tree/v0.2
[23:23] derferman has joined the channel
[23:25] dekz has joined the channel
[23:26] derferman: Anyone have experience with creating a streaming api with node?
[23:27] technoweenie: it shouldnt be that hard should it? just keep the response socket open 
[23:28] technoweenie: the twitter api periodically sends \n's if there are no tweets
[23:28] isaacs: mikeal: re: hosting for js-registry: nope, not yet
[23:29] mikeal: ok
[23:29] mikeal: I'm going to make some changes
[23:29] isaacs: mikeal: i'm playing with it on localhost atm, and i'm gonna probably stick it on registry.izs.me or something soonish
[23:29] mikeal: not to what actually goes in to couchdb
[23:29] isaacs: mikeal: awesome
[23:29] mikeal: but it's going to use the new couchapp stuff that i wrote in node
[23:29] isaacs: so the data is still gonna work?
[23:29] mikeal: it'll be like 1 file that is 50 lines instead of directories with small files all annoying like it is now
[23:29] isaacs: ie, not schema change
[23:30] mikeal: there is no schema :)
[23:30] isaacs: well, sure
[23:30] siculars: so im trying to grok fu.js from http://github.com/ry/node_chat/blob/master/fu.js and it seams that once a static file is loaded via fu.staticHandler() it will not reload until the node process is restarted. i wonder if updating fu.staticHandler to keep an mtime check somewhere could fix that and reload if mtime changes. any thoughts? anyone know where i could thief some code?
[23:30] mikeal: the design doc will be identical
[23:30] mikeal: just the sync script and the definition that is checked in to source will change
[23:30] mikeal: in fact that sync script will go away
[23:30] mikeal: instead you'll do
[23:30] isaacs: mikeal: well, there IS a schema, like there is such a thing as 7, it's just that the database doesn't know what it is, necessarily
[23:30] mikeal: npm install couchapp
[23:30] mikeal: couchapp sync app.js
[23:30] isaacs: zomgawesome
[23:30] isaacs: yay
[23:30] mikeal: actually
[23:30] mikeal: couchapp —design app.js --sync
[23:31] mikeal: and you can do —test to run the tests
[23:31] isaacs: mdash? really?
[23:31] isaacs: :P
[23:31] isaacs: -- ≠ —
[23:31] mikeal: adium globs that
[23:31] Tim_Smart: technoweenie: Do you think a patch for dynamically adding extra tracking channels for twitter-node is ideal?
[23:31] isaacs: colloquy ftw!
[23:31] mikeal: that's weird
[23:31] mikeal: colloquey is too good at crashing
[23:31] technoweenie: extra tracking channels
[23:31] Tim_Smart: colloquy :D
[23:31] technoweenie: oh so you could have multiple streams open?
[23:32] technoweenie: well twitter wont let you have multiple connections, you'd need to use multiple twitter accounts i guess
[23:32] Tim_Smart: technoweenie: Well, so you could on-the-fly add an extra user to listen for tweets
[23:32] technoweenie: oh so it'd drop and recreate the connenction
[23:32] technoweenie: i think that would be cool
[23:32] technoweenie: i bet you could just create the new one and i'td automatically drop the first one
[23:32] Tim_Smart: technoweenie: Yeah, but it will do it a non-destructive way
[23:34] Tim_Smart: technoweenie: I'll look into patching it soon. I will need it later on for a app I'm working on
[23:34] maushu: It worked! O_O
[23:34] Cainus has joined the channel
[23:36] cloudhead has joined the channel
[23:37] pedrobel_ has joined the channel
[23:38] maushu: Yay! My templater works! ...and it doesn't desorganizes the source code.
[23:41] gf3 has joined the channel
[23:41] voodootikigod_ has joined the channel
[23:41] javajunky has joined the channel
[23:44] technoweenie: Tim_Smart: cool, i didnt actually do the app i wrote twitter-node for :)
[23:44] indiefan: technoweenie: what was it going to be?
[23:44] derferman: so, I have noticed that on the modules, there are three different  web sockets implementations on top ob node js. Any idea which one is the best?
[23:44] technoweenie: cant tell you
[23:44] indiefan: hahaa, fair enough
[23:44] technoweenie: ah we had a few ideas, nothing major :)
[23:47] mjijackson has joined the channel
[23:48] isaacs: mikeal: lmk when you push some changes for js-registry.  i gotta run for a bit
[23:48] mikeal: will do
[23:56] Tim_Smart: derferman: http://github.com/ncr/node.ws.js looks good
[23:57] derferman: thanks tim
[23:58] Tim_Smart: as for client side, there is a cross browser implementation that falls back to flash somewhere
[23:59] derferman: yeah, I have actually used that as well
[23:59] derferman: it is just tough deciding on what ws server side module to use, as three are linked