Camelia, the Perl 6 bug

IRC log for #moe, 2013-03-11

| Channels | #moe index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:37 sawyer_ joined #moe
00:47 am0c joined #moe
01:58 bphillips joined #moe
04:19 stevan joined #moe
05:41 hiratara joined #moe
06:14 stevan joined #moe
12:22 bphillips joined #moe
12:47 gizmomathboy joined #moe
12:47 gizmomat` joined #moe
12:51 gizmomathboy joined #moe
12:53 gizmomat` joined #moe
12:57 gizmomathboy joined #moe
13:28 bluescreen joined #moe
13:38 stevan joined #moe
13:59 jnap joined #moe
14:05 jnap joined #moe
14:29 stevan joined #moe
15:28 moe [moe] prakashk created prakashk/strclass-methods (+1 new commit): http://git.io/LuuJVQ
15:28 moe moe/prakashk/strclass-methods 5207a2b Prakash Kailasa: concat (".") and repeat ("x") operators for strings
15:36 moe [moe] prakashk opened pull request #67: string concat (".") and repeat ("x") operators (master...prakashk/strclass-methods)  http://git.io/Y3xegA
15:37 * jasonmay looks up what PackratParser is
15:51 jasonmay so, anon subs... will they be similar to C perl5 but with signatures? I don't see any in the proposed spec
15:51 stevan jasonmay: I would love pointy subs
15:51 stevan -> ($x) { ... }
15:51 stevan ala Perl 6
15:51 stevan but I am in a meeting so I can't really talk
15:52 jasonmay mmm stabby lambdas
15:53 perigrin obviously you should just core lambda.pm
15:56 doy there's no reason we couldn't start with just anon subs like perl 5 has now
15:57 doy well, i guess it wouldn't matter much either way
15:57 doy it'd really just be s/sub/->/
15:58 doy anyway, yeah, anon subs should work identically to named subs, except without the name
15:59 jasonmay it looks like pointy blocks just lets you omit parens around the signature
16:04 doy well, the point of pointy blocks in perl 6 is that they look sort of like real syntax
16:04 doy so for @things -> $x { ... }
16:05 doy is just FOR LIST ANONSUB
16:05 jasonmay ahh
16:05 doy basically
16:06 doy i'm not sure if they are necessarily worth it if we aren't going to have syntax like that
16:06 doy since i think sub ($x) { ... } is more readable on its own
16:07 doy pretty similar to how ruby has { |x| ... }
16:08 doy or scala's { x => ... }
16:08 doy just something that looks nicer at the end of a statement/expression
16:08 jasonmay I agree
16:08 doy but perl 5's syntax as it stands doesn't really allow for things like that in many places
16:09 jasonmay I get a little sad when I see ->foo(sub {...})->bar(sub {...})
16:09 doy right
16:09 stevan yeah, we should probably start with sub { ... }
16:09 stevan but I would like to see -> { ... } at some point
16:10 doy i think we should support both, since i think "foo(bar => 1, cb => -> $x { ... })" is basically unreadable
16:11 sartak didn't gfx have a module for nice rubyish sub syntax in perl5?
16:11 sartak or was it a core patch or something..
16:11 doy PerlX::MethodAsBlock or something
16:11 doy it was something under PerlX::
16:11 * phaylon will be using -> and => for single-expr lambdas in his toy thingy
16:11 sartak yeah, that
16:12 sartak oops, it was gugod
16:12 sartak https://metacpan.org/module​/PerlX::MethodCallWithBlock
16:12 doy https://metacpan.org/module​/PerlX::MethodCallWithBlock
16:12 doy yeah
16:12 jnap "This version is released as a proof that it can be done. However, the internally parsing code for translating codes are very fragile at this moment."
16:12 doy devel::declare + ppi
16:12 doy oh god
16:12 jasonmay eeeek
16:13 sartak this is why moe needs to exist :P
16:13 doy (:
16:14 jasonmay speaking of parser hooks
16:14 jasonmay awwaiid: was it ClassLoader you were messing around with?
16:20 awwaiid you mean for hitting java code?
16:21 jasonmay yeah
16:22 sahadev joined #moe
16:26 awwaiid I don't think I got far enough to say I was actually making progress. was looking at random snippets such as http://stackoverflow.com/questions/206039​5/is-there-any-scala-feature-that-allows-​you-to-call-a-method-whose-name-is-stored
16:27 doy i tried looking into it last time you mentioned it
16:28 doy but reflection is crazy complicated it seems
16:28 awwaiid I think there is some confusion in the docs -- reflection from java->scala is crazy complicated, or scala->scala even. but I get this feeling that scala->java isn't any harder than java->java
16:28 awwaiid but I haven't proven it :)
16:29 doy well, java reflection is still pretty complicated(:
16:29 awwaiid there is that :)
16:29 jasonmay oh yeah, I didn't think to the point beyond loading the class.. yeah reflection is important
16:29 doy yeah, loading classes is trivial
16:29 doy it's doing anything with them once you load them that gets hard
16:30 doy especially for things like java.io, since scala doesn't have any concept of static methods, so things don't really map particularly cleanly
16:30 jasonmay TopCoder's client has a working implementation of this, heh
16:30 doy also, the scala reflection stuff was just rewritten for 2.10, and is basically completely undocumented
16:30 awwaiid doy -- I think that the 2.10 stuff is for ->scala reflection though
16:30 awwaiid all that Mirror stuff
16:31 doy awwaiid: well, it'd be nice to not have to write the reflection code twice
16:31 awwaiid oh
16:31 awwaiid point
16:31 doy and i think the scala reflection api should be able to handle both java and scala, in theory at least
16:42 am0c joined #moe
16:43 am0c joined #moe
17:13 am0c joined #moe
17:27 bluescreen_ joined #moe
17:41 stevan joined #moe
17:48 jnap joined #moe
18:14 perigrin completely undocument reflection APIs? Scala *is* just like Perl!
18:15 doy haha
18:16 gizmomathboy joined #moe
18:16 doy jasonmay: oh, by the way, http://doy.github.com/scala-test-more/#com.iinteractive.test.package is possibly worth looking at
18:17 stevan yeah I want to convert Moe tests to that
18:18 stevan eventually
18:19 jasonmay nice
18:25 moe [moe] stevan pushed 1 new commit to master: http://git.io/p6zoUA
18:25 moe moe/master 4f003a0 Stevan Little: Merge pull request #67 from MoeOrganization/prakashk/strclass-methods...
18:28 moe [moe] prakashk deleted prakashk/strclass-methods at 5207a2b: http://git.io/8qR64g
19:14 bluescreen joined #moe
20:07 gizmomathboy joined #moe
20:28 jnap joined #moe
21:35 stevan joined #moe
22:59 bphillips1 joined #moe

| Channels | #moe index | Today | | Search | Google Search | Plain-Text | summary