01:52:42  * piscisaureusquit (Quit: ~ Trillian Astra - www.trillian.im ~)
02:37:54  * bradleymeckjoined
03:01:58  * alejandromgjoined
03:35:24  * bnoordhuisquit (Ping timeout: 244 seconds)
04:04:42  * alejandr1mgjoined
04:05:07  * alejandromgquit (Ping timeout: 248 seconds)
04:05:11  * alejandr1mgquit (Client Quit)
04:05:27  * alejandromgjoined
04:41:52  * dvvjoined
05:15:59  * tilgovijoined
06:24:07  * Raynosquit (*.net *.split)
06:34:10  * alejandromgquit (Quit: leaving)
07:58:06  * Raynosjoined
08:01:31  * mmalecki[zzz]changed nick to mmalecki
08:39:43  * dvvquit (Ping timeout: 244 seconds)
08:57:30  * dvvjoined
09:08:46  <indutny>yeeeeeeeeeah
09:08:50  <indutny>creationix: yt?
09:08:57  <indutny>I've just learned how to insert Phi functions
09:09:10  <indutny>well, they ain't optimal, but it seems to be working
09:53:32  * tilgoviquit (Remote host closed the connection)
10:06:28  * bnoordhuisjoined
10:09:43  <indutny>now it's time to calculate variable liveness
10:55:02  * mmaleckichanged nick to mmalecki[plane]
11:56:54  <creationix>indutny, cool
12:00:54  <indutny>creationix: have you seen this: https://github.com/creationix/candor.io/wiki/Standard-API ?
12:00:56  <indutny>bradleymeck: ++
12:00:56  <kohai>bradleymeck has 36 beers
12:09:41  <creationix>indutny, will that be candor core of up to implementations like candor.io?
12:10:04  <indutny>creationix: that's up to implementations
12:10:14  <indutny>creationix: though as I told before, string stuff may be moved to VM
12:10:31  <indutny>creationix: slicing may be faster, if I'll introduce SliceStrings
12:10:37  <indutny>i.e. lazy-slicing
12:43:42  <creationix>right
12:45:44  * bradleymeckquit (Quit: bradleymeck)
12:46:19  * bradleymeckjoined
13:25:36  <bradleymeck>yea the standard api is just stuff I had to do while killing time, I am unsure on DateTime stuff for now since that gets crazy complex API wise quickly
13:26:08  * piscisaureus_joined
13:28:26  <bradleymeck>an array slice syntax per dimension via [start to end] would be nice still for both arrays and strings, not sure about CData or something akin to Node's buffers
13:28:56  <indutny>yeah, a[1..2] sounds good to me
13:28:58  <indutny>or a[1..]
13:29:03  <indutny>or a[..2]
13:32:11  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
13:33:30  * piscisaureus_joined
13:34:06  <bradleymeck>indutny would advice against .. , it makes for odd reading and due to having the double meaning of '.' for property access and '..' for range, you can read more about using '..' from digital mar's D language
13:34:28  <indutny>hm...
13:34:47  <indutny>what do you think may serve this purpose better?
13:35:16  <bradleymeck>x=.1 and a[0 .. .2] for example give me nightmares from D
13:36:08  <bradleymeck>indutny: straight up give a keyword when describing ranges, make it something that is not used when combining anything '1 to 4' , limit the expression to integer types (not always possible at run time :/)
13:38:31  <bradleymeck>indutny: but syntax is easy to change once you get it coded since its generally agreed it will be arr[x ?? y]
13:38:56  <indutny>I don't like idea of adding new keyword
13:39:14  <indutny>but I need to think about it, sorry head is blown up with HIR/LIR/SSA stuff right now
13:39:53  <bradleymeck>no different than a new operator really, no worries, just don't use punctuation that can be between 2 alphanumeric tokens and it should be ok
13:40:20  <indutny>WIP: https://github.com/indutny/candor/compare/feature-ssa
13:41:24  <indutny>if only someone could review that :D
13:42:24  <bradleymeck>unused binary ascii punctuation we have is ~ , ! , @, $, ?
13:42:42  <indutny>$ can be in variable names
13:42:51  <indutny>!true === false
13:43:00  <indutny>other yes
13:43:09  <indutny>~ may be used for binary not
13:43:18  <bradleymeck>$ i see, x ! y is not valid syntax right now
13:43:18  <indutny>so @ and ? :D
13:43:27  <indutny>yes, but it'll confuse people
13:43:34  <bradleymeck>sure
13:43:39  <indutny>it's often used in actor-model
13:43:42  <indutny>in other languages
13:43:47  <bradleymeck>do we have a comma operator?
13:44:28  <indutny>well, comma is used in arguments for both declarations and calls
13:44:48  <indutny>btw, what do you think about a[0, 2] ?
13:44:56  <bradleymeck>yea but that is mutually exclusive from being in same place as an array index, since both are wrapped in () or []
13:46:02  <bradleymeck>as long as we never allow crazy a[0, 2, 3] etc that was in Python for a bit its fine since we don't have a comma operator and do not have multiple returns
13:46:20  <indutny>heh
13:46:23  <indutny>right
13:47:03  <bradleymeck>fairly similar to think about lookup via [] being similar to invocation via () which is a plus when using a comma
13:47:40  <bradleymeck>just one is read based rather than computation
13:49:01  <bradleymeck>also i need to read your Jit's assembler sometime haha
13:49:53  <bradleymeck>just had a crazy thought but would be horrifiyingly slow since ranges could be treated like invocations (by eye) arr[ .. range ] where range = [ 0, -1 ]
13:50:44  <bradleymeck>though i havent even taken a look at how to invoke rest params / variadics
13:51:11  <bradleymeck>nm i havent really slept in like 38 too many of these would cause runtime checks
13:52:26  <indutny>haha
13:52:42  <indutny>don't look at JIT
13:52:46  <indutny>I'm rewriting it
13:53:16  <indutny>because I'm not really happy with how it is implemented (no register allocation, no cfg)
15:00:17  * alejandromgjoined
16:11:48  * alejandromgchanged nick to alejandro|out
16:52:08  * bradleymeckpart
17:44:01  <indutny>https://gist.github.com/3ccd6421219aaf3beb37
18:06:53  * alejandro|outquit (Quit: leaving)
18:52:32  * mmalecki[plane]changed nick to mmalecki
20:12:23  * alejandromgjoined
20:16:18  <dvv>hi! wonder if {} can be dropped in favor of significant indentation?
20:22:57  <indutny>dvv: I think no
20:23:27  <indutny>dvv: candor is ES-like, not coffeescript (python) -like
20:23:52  <indutny>and significant indentation is changing semantics dramatically
20:26:39  <indutny>going to sleep, ttyl
20:26:59  <indutny>if someone will have free time - please take a look at https://github.com/indutny/candor/compare/feature-ssa
20:27:14  <indutny>any C++ style fixes or anything would be really welcome!
20:27:15  <indutny>thank you!
20:33:00  * alejandromgquit (Quit: leaving)
20:36:29  * alejandromgjoined
20:53:57  * dvvquit (Ping timeout: 250 seconds)
22:28:31  * tilgovijoined
22:41:38  * piscisaureus_quit (Ping timeout: 240 seconds)
22:47:29  * piscisaureus_joined