00:10:14  * felixgequit (Ping timeout: 252 seconds)
00:10:39  * felixgejoined
00:10:39  * felixgequit (Changing host)
00:10:39  * felixgejoined
00:21:00  * indexzerojoined
00:35:11  * felixgequit (Quit: felixge)
00:41:53  * brsonquit (Quit: leaving)
00:52:16  * ericktjoined
01:00:11  * indexzeroquit (Quit: indexzero)
01:29:05  * bnoordhuisquit (Ping timeout: 252 seconds)
02:20:52  * dapjoined
02:39:59  * dapquit (Quit: Leaving.)
02:53:03  * isaacsjoined
02:57:00  * isaacsquit (Client Quit)
03:03:48  * ericktquit (Quit: erickt)
03:04:59  * isaacsjoined
03:05:09  * isaacsquit (Client Quit)
03:05:59  * isaacsjoined
03:20:54  * isaacsquit (Quit: isaacs)
03:25:55  <CIA-95>node: Ryan Dahl v0.6 * r9d1bdbc / (Makefile tools/osx-pkg.pmdoc/01local.xml tools/osx-dist.sh): Fixes #2084. Makeify tools/osx-dist.sh - http://git.io/8fY6bw
04:28:04  * igorzi_joined
04:43:45  <igorzi_>ryah piscisaureus___: you there?
05:02:24  * igorzi_quit (Quit: Page closed)
05:06:58  <ryah>igorzi: yes
05:13:23  * igorzi_joined
05:13:42  <igorzi_>ryah: https://github.com/joyent/node/issues/2149
05:13:52  <igorzi_>have you seen it?
05:14:43  <igorzi_>it seems it would be good to punt it until https://github.com/joyent/node/issues/2136
05:15:14  <igorzi_>ryah: on windows, including headers with the application is very uncommon
05:15:23  <igorzi_>that's what skds are for
05:15:30  <igorzi_>*sdks
05:16:31  <igorzi_>also, on windows including just headers is not enough.. we'll also need to include import libraries for node, libuv, and v8
05:16:54  <igorzi_>maybe there should be a separate native addon sdk?
05:51:27  <ryah>igorzi_: yes
05:52:36  <ryah>we will not ship headers for a while
05:52:56  <ryah>hopefully that will drive pure js libs up :)
05:53:36  <igorzi_>ryah: ok, cool
06:05:45  <pquerna>oh, easy to say when you don't depend on any
06:22:18  * paddybyersjoined
06:55:18  * felixgejoined
06:55:18  * felixgequit (Changing host)
06:55:18  * felixgejoined
06:58:12  * igorzi_quit (Quit: Page closed)
07:11:38  * paddybyersquit (Quit: paddybyers)
07:25:30  * paddybyersjoined
08:20:13  * kuebkjoined
09:08:14  * paddybyers_joined
09:10:49  * paddybyersquit (Ping timeout: 244 seconds)
09:10:49  * paddybyers_changed nick to paddybyers
09:48:05  * bnoordhuisjoined
11:16:07  * piscisaureus_joined
11:22:48  <piscisaureus_>mornin'
11:35:00  <CIA-95>node: Maciej MaƂecki master * r3f862cf / test/simple/test-stdin-resume-pause.js :
11:35:00  <CIA-95>node: test: `stdin` isn't closed after `resume()` and `pause()`
11:35:00  <CIA-95>node: This works on `node v0.4.12`, but doesn't work on `node v0.6.2` - http://git.io/IK32iQ
11:40:36  <bnoordhuis>piscisaureus_: sup
13:04:02  <CIA-95>node: Roman Shtylman v0.6 * r4d05f56 / src/node_constants.cc :
13:04:02  <CIA-95>node: tls: expose ssl bug workaround constants
13:04:02  <CIA-95>node: The internet is filled with servers which have various bugs and other
13:04:02  <CIA-95>node: ssl handshake issues. These constants can be used with the Agent
13:04:02  <CIA-95>node: secureOptions option to get around such server limitations when performing
13:04:03  <CIA-95>node: https requests - http://git.io/QsPyCA
13:04:03  <CIA-95>node: Ben Noordhuis v0.6 * r05de01d / src/node_constants.cc : tls: expose more openssl SSL context options - http://git.io/LAu5cQ
14:40:20  * ericktjoined
14:51:29  <bnoordhuis>piscisaureus_: so what have you been up to?
14:52:22  <piscisaureus_>Up to no good
14:52:40  <piscisaureus_>In the meantime I am trying to get a buildslave going for msvc
14:52:56  <bnoordhuis>ooh, a challenge
15:04:50  <bnoordhuis>piscisaureus_: for some reason i'm getting a lot of php job offers lately, what about you?
15:05:17  <bnoordhuis>it may be a double dip but php contract work is still booming
15:26:39  <piscisaureus_>bnoordhuis: I get a lot of job offers but nothing php-related
15:26:54  <piscisaureus_>bnoordhuis: remember the dude that did the node talk @ hackersnl
15:27:01  <bnoordhuis>piscisaureus_: yes?
15:27:02  <piscisaureus_>his company sent me something lately
15:27:08  * indutnyquit (Remote host closed the connection)
15:27:15  <bnoordhuis>hah, really?
15:27:17  <bnoordhuis>what was it?
15:27:34  <piscisaureus_>work on their realtime tv stuff
15:27:38  <piscisaureus_>boooring
15:28:02  <bnoordhuis>so what did you mail back?
15:28:10  <piscisaureus_>politely
15:28:12  <bnoordhuis>"thanks, but i don't really do anything with node.js"
15:28:15  <piscisaureus_>that they can go fuck themselves
15:28:22  <piscisaureus_>node.js is a retarded platform
15:28:31  * piscisaureus_changed nick to piscisaureus
15:39:54  <bnoordhuis>piscisaureus: https://github.com/bnoordhuis/libuv/commit/d919c07 <- review?
15:42:24  <bnoordhuis>the one thing i hate about thread apis is that there's no good cross-platform way to get notified of thread death
15:42:30  <bnoordhuis>windows wins this one, i'm afraid
16:35:54  * kuebkpart
16:40:54  <piscisaureus>bnoordhuis: hey
16:41:05  <piscisaureus>bnoordhuis: fs.watch is broken on sunos
16:41:13  <piscisaureus>says my colleages
16:42:40  <piscisaureus>file changes are not detected
16:42:58  <piscisaureus>and adding a file to a directory triggers an unstoppable spree of rename events
16:43:09  <piscisaureus>ryah -^
16:48:02  <piscisaureus>bnoordhuis: please use _beginthreadex instead of CreateThread
16:48:13  <piscisaureus>it allows the thread user to call crt functions without leaking memory
16:50:51  * dapjoined
16:51:24  * indutnyjoined
16:51:40  * isaacsjoined
16:53:00  * ericktquit (Quit: erickt)
17:14:06  <piscisaureus>bnoordhuis: ping
17:14:49  * ericktjoined
17:19:48  * dapquit (Quit: Leaving.)
17:20:09  * dap1joined
17:26:55  * ericktquit (Ping timeout: 244 seconds)
17:31:38  <bnoordhuis>piscisaureus: pong
17:32:06  <piscisaureus>bnoordhuis: do you know if something's broken about fs.watch on sunos?
17:32:27  <bnoordhuis>piscisaureus: it shouldn't be, obviously
17:32:32  <piscisaureus>I know
17:32:36  <piscisaureus>I cannot debug that
17:32:36  <bnoordhuis>do you have test cases i can try?
17:32:50  <piscisaureus>but read up, this is what Mike observed?
17:32:58  <piscisaureus>it was easy
17:33:06  <piscisaureus>fs.watch a directory, then create a file in it
17:33:22  * bnoordhuislogs in on a sunos machine
17:34:38  <bnoordhuis>piscisaureus: yes, reproduced
17:38:56  <ryah>hello
17:39:06  <ryah>piscisaureus: do we have an issue for that?
17:40:23  <ryah>what's thread depth?
17:41:57  <bnoordhuis>ryah: you mean thread death?
17:43:03  <ryah>oh sorry i read that wrong
17:43:15  <ryah>im landing npm in node now
17:43:25  <isaacs>yikes!
17:43:56  <ryah>its going to increase our tarball size by 1mb -_-
17:45:03  <isaacs>yeah.
17:45:50  <isaacs>it'd be a bit smaller if i didn't include the node-tar tests.
17:50:34  * indexzerojoined
17:52:28  <ryah>isaacs: for the .pkg - where should the npm source go?
17:52:37  <ryah> /usr/local/lib/node/npm ?
17:52:42  <isaacs>no..
17:52:49  <isaacs>wait, where does the pkg put node?
17:52:58  <ryah> /usr/local/bin/node
17:53:07  <isaacs>not /usr/local/bin/nodejs/node?
17:53:13  <ryah>no
17:53:15  <isaacs>oh, right, it symlinks
17:53:36  <isaacs>right?
17:53:42  <ryah>no
17:53:49  <ryah>it just puts it at /usr/local/bin/node
17:54:02  <ryah>(it change)
17:54:04  <ryah>d)
17:55:06  <isaacs>anyway... drop the tarball contents into /usr/local/lib/node_modules/npm, and symlink /usr/local/bin/npm to ./lib/node_modules/npm/bin/npm-cli.js
17:56:36  <ryah>ok..
17:57:16  <isaacs>may need to unpack with --strip-components=1
18:00:54  <ryah>bnoordhuis: your patch looks okay to me
18:01:26  <ryah>.msi docs could be better...
18:02:01  <bnoordhuis>ryah: the thread functions? thanks. looking at piscisaureus's suggestion now
18:02:53  <piscisaureus>bnoordhuis: you also have to CloseHandle the thread handle at some point.
18:03:01  <piscisaureus>or the thread will remain as a "zombie" thread
18:03:28  <bnoordhuis>piscisaureus: uv_thread_join() is a good place for that, right?
18:03:56  <piscisaureus>bnoordhuis: if that implies that the thread is dead and unreferenceable, yes
18:04:12  <piscisaureus>bnoordhuis: the thread handle can be used to retrieve the thread's exit code
18:04:20  <piscisaureus>that's why you may want to keep it around
18:04:31  <bnoordhuis>piscisaureus: i don't do that anyway
18:04:38  <piscisaureus>ah, ok
18:04:41  <piscisaureus>yes, then it's finm
18:04:43  <piscisaureus>*fine
18:07:08  * mikealjoined
18:07:41  <ryah>god it's amazing how fucking complex the MSI stuff is
18:25:33  * ericktjoined
18:35:29  <igorzi>ryah: yes, the msi stuff is pretty painful.. do you need help with figuring something out in particular?
18:36:53  * AvianFlujoined
18:38:30  <ryah>igorzi: should i be using heat.exe to install a whole directory?
18:40:29  <igorzi>ryah: i haven't used it.. can you just add Directory element(s) to product.wxs? (http://wix.sourceforge.net/manual-wix2/wix_xsd_directory.htm)
18:51:21  <CIA-95>node: Ryan Dahl v0.6 * rb488be1 / (449 files in 75 dirs):
18:51:21  <CIA-95>node: Include NPM, update .pkg to install it.
18:51:21  <CIA-95>node: .msi update coming soon. - http://git.io/XHELFQ
18:51:48  <ryah>isaacs: you'll get props for every kb you remove from npm
18:51:50  * brsonjoined
18:51:56  <isaacs>ryah: hahah, i'm sure :)
18:52:32  <ryah>[email protected]:~/projects/node-v0.6% du -sh deps/http_parser
18:52:32  <ryah>3.0M deps/http_parser
18:52:32  <ryah>[email protected]:~/projects/node-v0.6% du -sh deps/zlib
18:52:32  <ryah>668K deps/zlib
18:52:32  <ryah>[email protected]:~/projects/node-v0.6% du -sh deps/npm
18:52:34  <ryah>4.0M deps/npm
18:53:44  <ryah>actually the source code isn't so bad - somehow it grows the tarballs greatly
18:53:58  <ryah>(and the .pkg)
18:54:12  <ryah>% ls -lh out/node-v0.6.3.pkg
18:54:13  <ryah>-rw-r--r-- 1 ryan staff 3.4M Nov 21 10:49 out/node-v0.6.3.pkg
18:54:42  <ryah>previous .pkg was 2369883
18:56:08  <isaacs>ryah: it could be because it has a bunch of tarballs in the tar test fixtures.
18:56:30  <isaacs>but even most of that should be gzipped away
18:56:42  <ryah>isaacs: oh, yeah maybe - i wonder if i can selectively exclude those with PackageMaker... hm
18:57:05  <isaacs>yeah, you really don't need any node_modules/*/test* folders in there
18:57:19  <isaacs>might cut down a bit on it.
19:00:35  <bnoordhuis>call?
19:00:53  <ryah>yes
19:01:38  <piscisaureus>igorzi
19:02:33  <igorzi>yep
19:03:29  <piscisaureus>I am going to initiate the call
19:04:16  <ryah>hm - everyone died?
19:04:37  <piscisaureus>ryah: pick up!!
19:05:51  <ryah>you're not calling me
19:06:03  <ryah>or am i supposed to start it?
19:06:51  <piscisaureus>ryah: we are calling
19:13:33  <isaacs>skype?
19:14:48  <piscisaureus>isaacs: I am trying to add you
19:14:55  <isaacs>i'm isaac.schlueter
19:15:04  <isaacs>piscisaureus: i don't see you online
19:15:59  <piscisaureus>isaacs: you're isaacschlueter
19:17:23  <isaacs>not getting added
19:17:26  <isaacs>try adding isaac.schlueter
19:17:28  <isaacs>with teh dot
19:19:20  <piscisaureus>isaacs: when I try to add isaac.schlueter it says that that skype name does not exist
19:23:10  <piscisaureus>shit
19:23:13  <piscisaureus>Now I dropped
19:23:14  <bnoordhuis>okay, now i can't hear you any more :)
19:32:48  <ryah>/usr/include/sys/dirent.h:#define __DARWIN_MAXPATHLEN1024
19:38:38  <bnoordhuis>on linux it's mostly 1K or 4K
19:47:31  <CIA-95>libuv: Ben Noordhuis master * r5728bd4 / test/test-fs-event.c : test: touching a file should generate only one fs event - http://git.io/dauprg
19:47:43  <isaacs>found an awesome bug in node-tar
19:47:49  <isaacs>> "10 uid=501\n".length
19:47:49  <isaacs>11
19:47:55  <isaacs>that 10 should be 11
19:54:30  <ryah>so
19:54:41  <ryah>in wix you need to list every file you install :)
19:54:45  <ryah>great.
19:55:20  <isaacs>ryah: explicit lists are explicitly explicit.
19:55:29  <ryah>yeah..
19:55:40  <isaacs>you wouldn't want things to not be explicit would you?
19:57:09  <ryah>i appriciate the thought - but it works against us here
20:00:36  <isaacs>hehe, i'm being a little facetious
20:03:12  <isaacs>found the crash that mmalecki was hitting
20:03:16  <isaacs>probably thejh, as well
20:08:00  <ryah>http://blog.chromium.org/2011/11/game-changer-for-interactive.html
20:08:43  * mikealquit (Quit: Leaving.)
20:10:53  <ryah>igorzi: i need help
20:12:12  <ryah>igorzi: you can't have a <Directory> child of <Component>
20:12:39  <ryah>it seems the only way to install a directory of stuff is to apply some hacks..
20:14:51  <igorzi>ryah: what hacks did you find?
20:14:55  <piscisaureus>Actually FF gc seems to be better than chrome's
20:15:19  <piscisaureus>FF8 stable is much smoother and runs at more fps for me
20:16:49  <ryah>igorzi: one is http://wix.sourceforge.net/manual-wix3/heat.htm which will generate XML for all the files in the npm folder. This http://blog.genom-e.com/PermaLink,guid,8044083d-8153-4046-8601-ec3556f7adaf.aspx also
20:17:04  <ryah>the second seems to be some custom WiX addon
20:17:31  <ryah>the problem with heat is that we're going to have to carefully monitor when files are changed in deps/npm
20:17:48  <ryah>unless we include it in the build process - but that seems to be discouraged
20:19:05  <igorzi>ryah: let me find out what's the best practice.. i have a friend that knows msi stuff better than me
20:20:31  <ryah>igorzi: thanks
20:23:07  <isaacs>ryah: yeah, this could be a huge pita if changes to npm cause build failures in node.
20:27:03  <bnoordhuis>100000 threads created in 3.78 seconds (26427/s) <- on my puny 2009 laptop
20:35:29  <bnoordhuis>100000 threads created in 1258.28 seconds (79/s) <- windows xp...
20:36:57  <bnoordhuis>piscisaureus: https://github.com/bnoordhuis/libuv/compare/5728bd4...master
20:38:33  <bnoordhuis>well, i'm just going to land it
20:39:00  <CIA-95>libuv: Ben Noordhuis master * r8e4ed88 / (9 files in 7 dirs): Wrap platform thread APIs. - http://git.io/muDcfg
20:39:00  <CIA-95>libuv: Ben Noordhuis master * r45180fe / (test/benchmark-list.h uv.gyp test/benchmark-thread.c): bench: add thread creation benchmark - http://git.io/dniwbA
20:40:21  * piscisaureusquit (Read error: Operation timed out)
20:57:44  * piscisaureusjoined
20:58:05  * piscisaureuschanged nick to piscisaureus_
21:16:45  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
21:17:04  * isaacsquit (Quit: isaacs)
21:18:47  * piscisaureus_joined
21:21:46  <ryah>piscisaureus_: what time should we meet for the release
21:22:07  <ryah>piscisaureus_: friday 00:00 PST ?
21:25:36  <piscisaureus_>ryah: let me compute that in cest :-)
21:27:26  <piscisaureus_>ryah: ah 9am
21:27:30  <piscisaureus_>seems okay
21:27:49  <piscisaureus_>8am is also fine I think
21:28:03  <piscisaureus_>If I can anticipate
21:30:11  <ryah>let's aim for 9am/midnight
21:30:18  <ryah>i'll try not to be extremely drunk
21:31:05  <piscisaureus_>:-)
21:31:10  <piscisaureus_>Fine with me
21:31:38  <piscisaureus_>I'll try to not have a terrible hangover
21:42:31  <bnoordhuis>piscisaureus_: keep next wednesday's late afternoon, early evening free
21:47:48  <felixge>is make test supposed to test on OSX 10.7 in master?
21:47:50  <piscisaureus_>bnoordhuis: for what?
21:47:51  <felixge>getting a bunch of fails
21:49:04  <bnoordhuis>piscisaureus_: beer!
21:49:36  <bnoordhuis>felixge: yes, not everything passes - i had 3 or 4 fails last week
21:50:12  <felixge>bnoordhuis: thanks.
21:50:26  <felixge>removing the pipe from two sources logic right now
21:55:48  * ericktquit (Ping timeout: 258 seconds)
21:58:02  <piscisaureus_>bnoordhuis: are you coming to amsterdam?
21:58:06  <piscisaureus_>:-)
21:59:15  <bnoordhuis>piscisaureus_: yep, meeting up with an ex co-worker
22:01:20  <piscisaureus_>ah, ok
22:01:35  <piscisaureus_>bnoordhuis: you are not going to have beer with him?
22:01:54  <piscisaureus_>bnoordhuis: or do you mean you're coming to amsterdam to convey the message that I am fired?
22:02:41  <bnoordhuis>piscisaureus_: no, i would've just texted that
22:02:54  <bnoordhuis>piscisaureus_: i'm going to have a beer with him but you're invited too
22:03:11  <bnoordhuis>he's got a date with several women later on so yeah
22:03:52  <piscisaureus_>oh this was that company that you used to work for
22:04:09  <piscisaureus_>that paid in hooker hours, not money
22:04:31  * ericktjoined
22:04:44  <felixge>ripped out the multi source thing from pipe(): https://github.com/joyent/node/pull/2161
22:07:18  <felixge>so npm is in core now?
22:08:00  <piscisaureus_>I think it's not going to be developed in core
22:08:09  <piscisaureus_>will just be part of the core package
22:08:19  <piscisaureus_>so we can easily build installers etc for the whole thing
22:08:41  <piscisaureus_>with NPM that is actually tested with that particular version of node
22:09:57  <felixge>ok
22:10:05  <felixge>if I install node on nix, will it install npm?
22:10:06  * isaacsjoined
22:12:08  <piscisaureus_>If you install it on mac with the pkg file, it will
22:12:34  <piscisaureus_>felixge: git checkout will still be the same, and I think `make install` will not change either.
22:12:57  <felixge>ah ok
22:13:02  <felixge>that seems reasonable
22:19:38  * felixgequit (Quit: felixge)
22:20:39  * felixgejoined
22:22:17  <piscisaureus_>ryah and isaacs can tell you all the details
22:23:27  * felixgequit (Client Quit)
22:26:28  <piscisaureus_>I am still not enthousiastic about message passing between dok=mains
22:29:24  <piscisaureus_>Message passing is usually used for a request-response kind of thing... When a domain dies you either ignore requests (so the response never comes) or reject the request ("socket not writable").
22:29:32  <piscisaureus_>Not Good
22:32:22  * indexzeroquit (Quit: indexzero)
22:32:44  <bnoordhuis>piscisaureus_: what's so bad about rejecting requests? at least it gives a clear signal that the receiver is no longer interested
22:32:59  <piscisaureus_>bnoordhuis: well, that's the issue
22:33:04  <piscisaureus_>bnoordhuis: what would you do?
22:33:05  <piscisaureus_>THROW
22:33:17  <piscisaureus_>So the failure cascades
22:33:52  <piscisaureus_>suddenly you get an exception in your face and the error is no longer isolated to one domain, another dies as well
22:34:08  <piscisaureus_>(unless you want to wrap all interdomain communication in try-catch blocks)
22:34:19  <bnoordhuis>only if the sending domain doesn't handle it, right?
22:34:20  <bnoordhuis>right
22:34:43  <bnoordhuis>try/catch? are sends sync?
22:34:46  <piscisaureus_>before you know it the whole thing crumbles like an apple pie
22:35:02  <piscisaureus_>well.. I don't expect people to write that many error handlers
22:35:06  <piscisaureus_>that's the bottom line
22:35:16  <piscisaureus_>sends could be sync
22:35:31  <piscisaureus_>it would probably something similar to process.nextTick
22:35:50  <piscisaureus_>since you want to be on a new stack but it's not going to be multithreaded
22:36:23  <bnoordhuis>okay, then i don't see the issue - let people handle it (or not) in their error listeners
22:40:48  <CIA-95>node: Ben Noordhuis v0.6 * r7c180d7 / src/node_file.cc :
22:40:48  <CIA-95>node: fs: fix uninitialized fs req struct use
22:40:48  <CIA-95>node: uv_fs_t struct is not initialized on fs error so don't try to clean it up. - http://git.io/FvZ2ng
22:46:21  <piscisaureus_>shaven: http://nodesummit.com/wp-content/uploads/ryan_dahl_joyent_headshot_black-185x185.jpg
22:47:14  <bnoordhuis>fuck, libuv mallocs - i'm going to revert that commit
22:50:21  * ericktquit (Quit: erickt)
22:59:28  * ericktjoined
23:09:11  <CIA-95>libuv: Ben Noordhuis master * r92c9e95 / src/unix/fs.c :
23:09:11  <CIA-95>libuv: unix: fix stray pointer free() when HAVE_FUTIMES=0
23:09:11  <CIA-95>libuv: uv_fs_futime() failed to initialize req.path when HAVE_FUTIMES=0.
23:09:11  <CIA-95>libuv: uv_fs_req_cleanup() later on tried to free the bogus pointer. - http://git.io/jsloAg
23:09:11  <CIA-95>libuv: Ben Noordhuis master * r0834e73 / (config-unix.mk src/unix/fs.c src/unix/stream.c uv.gyp): Merge branch 'v0.6' - http://git.io/V8asTw
23:09:17  <CIA-95>libuv: Ben Noordhuis v0.6 * r92c9e95 / src/unix/fs.c :
23:09:17  <CIA-95>libuv: unix: fix stray pointer free() when HAVE_FUTIMES=0
23:09:17  <CIA-95>libuv: uv_fs_futime() failed to initialize req.path when HAVE_FUTIMES=0.
23:09:17  <CIA-95>libuv: uv_fs_req_cleanup() later on tried to free the bogus pointer. - http://git.io/jsloAg
23:22:18  <igorzi>ryah: if we want to have something automated - heat.exe is the way to go
23:23:26  <igorzi>ryah: some people (who are very anal about what they ship in the msi) just maintain the list by hand.. we probably don't want that
23:24:28  <piscisaureus_>Hmm. When did we stop being anal about random stuff?
23:27:42  <bnoordhuis>piscisaureus_: about that fs.watch bug on sunos
23:27:54  <bnoordhuis>i can reproduce it but i can't quite fix it :/
23:28:12  <bnoordhuis>the kernel appears to simply keep sending events
23:28:13  <igorzi>piscisaureus_: we just want to include whatever is in npm, right.. will you ever want to exclude something from node msi that's in npm?
23:31:34  * paddybyersquit (Read error: Connection reset by peer)
23:31:55  <piscisaureus_>igorzi: *shrug*
23:31:57  * paddybyersjoined
23:32:07  <piscisaureus_>igorzi: I don't see any reason to exclude stuff
23:32:09  <piscisaureus_>tests, maybe?
23:32:48  <piscisaureus_>anarchist propaganda that isaacs accidentally landed in npm
23:32:58  <piscisaureus_>{ :-p }
23:33:47  <isaacs>piscisaureus_: excluding npm's tests will be very nice.
23:33:59  <piscisaureus_>igorzi: ^
23:34:05  <isaacs>i'm probably going to make a change to npm to not include tests of bundled dependencies, but i can't figure out a way to do so reliably
23:34:13  <isaacs>the tar tests are actually worse than the npm tests.
23:34:41  <isaacs>igorzi: if there's a way to exclude node_modules/*/test* from npm, that'd be worthwhile
23:36:59  <piscisaureus_>sounds like `dir /ad/s/b node_modules\*\test*`
23:38:32  <igorzi>piscisaureus_ isaacs: ok
23:44:20  <piscisaureus_>pls review https://github.com/joyent/libuv/commit/39d31a100f542d3e4d50cf22ee3147563eb1b5c7