00:01:07  * dan336quit (Quit: Leaving.)
00:24:47  * ^vquit (Quit: http://i.imgur.com/Akc6r.gif)
00:25:04  * ^vjoined
00:27:33  * cledevjoined
00:57:10  * UniOnquit (Remote host closed the connection)
02:42:34  * cledevquit (Ping timeout: 255 seconds)
02:51:29  * hdmsquit (Quit: hdms)
03:07:32  * cledevjoined
03:12:16  * ^vquit (Ping timeout: 258 seconds)
03:32:29  * travis-cijoined
03:32:29  <travis-ci>luvit/luvit#1256 (luvi-up - 3d01e70 : Tim Caswell): The build passed.
03:32:29  <travis-ci>Change view : https://github.com/luvit/luvit/compare/7600ff04b447...3d01e70af9c4
03:32:29  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/42157975
03:32:29  * travis-cipart
03:35:35  * cledevquit (Ping timeout: 272 seconds)
04:04:58  * cledevjoined
04:16:20  * hdmsjoined
04:31:05  * cledevquit (Ping timeout: 255 seconds)
04:42:25  * cledevjoined
04:53:29  * cledevquit (Ping timeout: 264 seconds)
05:02:39  * ^vjoined
05:12:38  * ^vquit (Ping timeout: 258 seconds)
05:26:46  * cledevjoined
05:58:43  * cledevquit (Ping timeout: 265 seconds)
06:18:53  * ^vjoined
06:42:43  * ^vquit (Ping timeout: 258 seconds)
06:53:01  * torpor1quit (Read error: Connection reset by peer)
06:53:14  * torporjoined
07:04:37  * ^vjoined
07:09:08  * ^vquit (Client Quit)
08:18:42  * hdmsquit (Quit: hdms)
08:19:12  * DarkGodjoined
10:55:50  * cledevjoined
11:06:58  * typedlambdaquit (Ping timeout: 250 seconds)
11:09:23  * typedlambdajoined
12:15:30  * hdmsjoined
13:02:37  * cledevquit (Ping timeout: 240 seconds)
13:16:35  * cledevjoined
13:57:08  * jonitrythalljoined
14:10:50  * rphillipsquit (Ping timeout: 258 seconds)
14:13:03  * rphillipsjoined
14:20:53  * jonitrythallquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
14:22:09  * DarkGodquit (Ping timeout: 244 seconds)
14:27:31  * DarkGodjoined
14:30:05  * jonitrythalljoined
14:38:49  * DarkGodquit (Ping timeout: 258 seconds)
14:39:26  * DarkGodjoined
14:44:57  * DarkGodquit (Ping timeout: 258 seconds)
14:45:13  * DarkGodjoined
14:49:02  * a_lejoined
15:28:45  * a_le_joined
15:29:06  * a_lequit (Read error: Connection reset by peer)
15:29:54  * jonitrythallquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
15:30:17  * dan336joined
15:50:08  * a_le_quit (Ping timeout: 255 seconds)
15:52:00  <creationix>Cool, libuv moved to it’s own org! https://github.com/joyent/libuv/issues/1573
15:53:43  * a_lejoined
16:03:30  <rch>yeah, can't believe it took so long
16:04:16  <creationix>I guess they wanted to finish libuv 1.x first
16:05:51  * jonitrythalljoined
16:08:54  * mattlgyjoined
16:19:00  * jonitrythallquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:28:41  * jonitrythalljoined
16:48:23  * jonitrythallquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:59:13  * cledevquit (Ping timeout: 244 seconds)
17:27:25  * cledevjoined
17:37:18  * torpor1joined
17:40:08  * torporquit (Ping timeout: 256 seconds)
17:58:54  * cledevquit (Ping timeout: 264 seconds)
18:01:37  * cledevjoined
18:05:02  * DarkGodquit (Quit: Leaving)
18:06:59  * jonitrythalljoined
18:07:55  * jonitrythallpart
18:14:21  * UniOnjoined
18:25:00  <creationix>rch, what do you think about this manifest format? https://github.com/luvit/lit/blob/master/sample-lib/package.manifest
18:25:20  <creationix>it highlights well with a yaml highlighter, but isn't full yaml since that's a pain to parse
18:31:52  <rch>why not lua or json?
18:32:28  <rphillips>i vote lua or json
18:32:54  <creationix>package.lua?
18:33:13  <rphillips>+1
18:33:17  <creationix>I don
18:33:33  <creationix>I don’t like JSON because it’s not very human friendly (no comments, super strict syntax, etc)
18:33:35  <rch>i mean one of the sad things about lua vs json is that you can run arbitrary code in a lua file
18:33:46  <rch>yeah i agree, json has some awkward limitations
18:33:48  <creationix>but lua has it’s own problems, it’s not machine friendly. If you eval it, it’s turing complete
18:34:10  <creationix>and editing existing lua code automatically is tricky
18:34:14  <rch>right
18:34:17  <creationix>like bumping the version through a script
18:34:25  <rch>at least at the present time, the tooling is not as convenient in lua as in js
18:35:01  <rch>there's the package manifest file format and then there's what it expresses (deps, versions, contents, other metadata)
18:35:48  <creationix>yaml is too big
18:35:51  <rch>for format we like easy, which means easy to parse, easy to write, easy to use/edit as a person, also maybe availability of existing tools/parsers
18:35:58  <rch>yeah yaml has things like references
18:36:06  <creationix>a superset of JSON would work, but then it’s not json anymore
18:36:17  <creationix>(add comments and allow js style keys without quotes)
18:36:21  <rch>right
18:36:39  <rch>too bad there isn't an object literal syntax / file format for lua
18:36:49  <creationix>lon
18:36:55  <creationix>I wrote a parser once, it’s not hard
18:37:03  <creationix>then you don’t have the eval issue
18:37:10  <rch>which is what json was fixing
18:37:22  <creationix>and since it’s a custom subset of lua I can decide to include comments
18:37:26  <rch>heh
18:38:14  <creationix>with my jack language I had JON format
18:38:25  <creationix>it looked like nicer JSON essentially
18:38:35  <creationix>no trailing commas, lua-style comments allowed
18:39:29  <rch>i'm not worried about whatever happens right now in lit
18:39:34  <rch>can always revisit it
18:40:00  <rch>having the tools would be great
18:40:30  <creationix>I think I’ll use LON, it’s a compramize
18:40:37  <creationix>*compromise
18:40:42  <rch>i feel like in a perfect world we would write pure lua lua ast parsing/serializing/linting tools and be able to assert about a package.lua that it has no side effects but that's a long road
18:40:49  <creationix>and for MVP I can just eval
18:41:06  <creationix>writing a LON parser is pretty easy
18:41:06  <rch>cool
18:41:34  <creationix>actually, lua isn’t terribly hard to sandbox
18:41:44  <creationix>replacing _G with an empty table goes a long ways
18:41:51  <rch>well it would be nice to do it statically rather than at runtime
18:42:05  <rch>i.e. this ast contains only assignment statements
18:42:06  <creationix>sure, but that’s harder
18:42:10  <rch>right
18:42:24  <creationix>yeah, I’ll just eval with a neutered _G for now
18:42:31  <rch>cool
18:43:04  <creationix>it’s stricter syntax than I would like, but still better than JSON
18:44:24  <creationix>should the ignore and include paths be glob syntax or lua patterns?
18:44:31  <creationix>glob is much easier to write and optimized for paths
18:44:51  <rch>patterns are much harder
18:45:08  <rch>globs are so much better
18:45:27  <creationix>* and ** translate directly to patterns too
18:45:31  <creationix>{foo,bar} is harder
18:46:08  <creationix>I’ll just leave out {foo,bar} glob support for now
18:46:30  <rch>right, you need lexicographic permutations like in the case of {a,b}c{d,e}
18:50:04  <creationix>ok, compare https://github.com/luvit/lit/blob/32ccb23969fd278064922c5a81915f48cab00ca7/sample-lib/package.lua with https://github.com/luvit/lit/blob/4f94252b646e33c8197b34008c7d428ba1a8ae76/sample-lib/package.manifest
18:50:59  <rch>looks great to me
18:51:01  <creationix>oh, deps versions all assume semver ranges. The idea is to encourage authors to follow semver
18:51:17  <rch>i was hoping. semver operators?
18:51:42  <rch>also why explicitly include files in the package rather than just packaging everything by default?
18:51:54  <rch>basically in my mind i compare everything to npm
18:51:55  <creationix>so 1.2.5 will match >= 1.2.5 and < 2.0.0
18:52:10  <rch>cool
18:52:32  <creationix>Also in the registry, anything under 1.0.0 won’t be publicly visible
18:52:49  <rch>interesting
18:53:09  <creationix>according to semver you should be at least 1.0.0 if others are using your code
18:53:21  <creationix>and the semantics are undefined for anything under 1.0.0
18:53:30  <creationix>I’ll probably do like npm’s ~ for < 1.0.0
18:53:40  <creationix>0.2.3 will match >= 0.2.3 and < 0.3.0
18:54:17  <creationix>so it’s like npm’s ^ for >= 1.0.0 and ~ for < 1.0.0
18:55:19  <creationix>isaac made his ^ act like = for < 1.0.0 versions as a way to encourage people to bump their versions to 1.0.0
18:55:40  <creationix>but I think simply hiding pre-release versions is a better way
18:58:04  <creationix>rch: the reason I make people explicitly include files and not include everything by default is to keep the package sizes smaller
18:58:16  <creationix>too many pakage maintainers are lazy and include way more than they need
18:58:35  <creationix>as long as I support glob, it won’t be too much burdon to add new files to a package
18:58:43  <creationix>*burden
18:59:40  <rch>interesting
19:00:11  <creationix>I mean they could just add “**” and be done with it
19:00:33  <creationix>but that line is there in their manifest reminding them to add negative rules when they want to not include something
19:00:57  <creationix>I don’t think I’ve ever used .npmignore or even know if that’s the right thing to use
19:01:10  <rch>yep it is and i mostly don't forget to use it…
19:01:45  <creationix>well that’s good
19:02:14  <creationix>I’m not totally against adding a “**” line in auto-generated package.lua files
19:03:10  <creationix>though I might suggest a little bit of conventions with things like skip the “tests” and “docs” folders
19:04:04  <creationix>ideally, tests and docs integrate with the repository in a more integrated way somehow so you can browse docs offline for all packages in your local cache, but not have the docs in your modules folder
19:04:59  <rch>right
19:13:17  <mattlgy>I like the package.lua. Much more than json. Comments and not exploding on trailing commas is nice +1
19:13:57  <mattlgy>Also, just sort of stumbled across luvit recently and have been super impressed with it.
19:14:21  <creationix>and with full lua support (inside the sandbox) you can declare variables and use them in multiple places
19:14:28  <creationix>not sure why, but it’s there
19:14:44  <creationix>mattlgy: glad you enjoy it. I’ll rest at night when the new version is out
19:21:17  <creationix>RFC https://github.com/luvit/luv/issues/94 rphillips rje
19:43:17  * DarkGodjoined
20:21:14  <mattlgy>I've been looking through some of the code and looking at the mailing list. Do you think you are keeping with the Node-y callbacks for async stuff?
20:23:14  <rch>luvit will keep the node api, there was a big decision about that recently
20:24:08  <rch>https://groups.google.com/forum/#!topic/luvit/_TI-TjAFtwU
20:24:57  <rch>you'll be able to do basically any async style you prefer with luv though
20:25:26  * torpor1quit (Quit: Leaving.)
20:26:07  <rch>heh google translate offers to translate wouter's comment at the bottom "npm!" into "Above sea level!"
20:26:31  * torporjoined
20:26:41  * torporquit (Client Quit)
20:26:44  <mattlgy>Hah, just saw that too.
20:27:14  <rch>pretty sure that's not what he meant
20:27:59  <mattlgy>> Above sea level!
20:28:58  <creationix>I’m going to start naming projects after positive polish sayings
20:50:46  * DarkGodquit (Remote host closed the connection)
22:35:03  * DarkGodjoined
23:09:51  * mattlgyquit (Remote host closed the connection)
23:56:08  * dan336quit (Quit: Leaving.)