Camelia, the Perl 6 bug

IRC log for #moe, 2013-02-14

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

All times shown according to UTC.

Time Nick Message
00:32 am0c joined #moe
02:24 pingup3rl joined #moe
03:08 moe [moe] stevan pushed 8 new commits to builtins: http://git.io/Y1VWGQ
03:08 moe moe/builtins fcad9c1 Stevan Little: merging in master
03:08 moe moe/builtins 868885c Stevan Little: note to self
03:08 moe moe/builtins b3ea67f Stevan Little: don't need this anymore
06:18 clintongormley joined #moe
06:23 clintongormley joined #moe
06:57 hiratara joined #moe
11:31 clintongormley joined #moe
12:16 hiratara joined #moe
12:45 stevan joined #moe
13:12 bphillips joined #moe
13:47 bphillips left #moe
13:47 bphillips joined #moe
13:48 bphillips is src/main/scala/org/moe/runtime/MoeRuntime.scala where many of the `perldoc perlfunc` stuff should be implemented?
13:49 gizmomathboy joined #moe
13:54 bphillips or, perhaps many of perl5's builtin functions belong as methods on the builtin types (i.e. substr, chomp, shift, push, pop, etc)
13:55 ilbot2 joined #moe
13:55 Topic for #moe is now https://github.com/MoeOrganization/moe | http://moeorganization.github.com/moe-docs/ | logs at http://irclog.perlgeek.de/moe/today
13:56 jasonmay bphillips: hmm I wonder if it makes sense to make them methods, prefixed as unary operators
13:57 jasonmay bphillips: stevan expressed interest in bootstrapping them as subs to corePackage, so that's probably a good place for now
13:58 jasonmay that might have been for just say and warn though
13:58 jasonmay I think this is the difference between hard and soft perl5 keywrods
13:58 bphillips right, I was wondering the same (if those are special-cased)
13:59 jasonmay I vote for making them methods :)
14:01 * jasonmay has a three day weekend, hopes to get a lot of hacking done
14:01 jnap joined #moe
14:09 bphillips having no experience with this level of programming (compilers and language implementations) I have a hard time imagining how the parser is going to know the sub-version of "substr" should really be translated into the method version on the target object
14:10 jasonmay we special-cased '+' in the parser
14:10 jasonmay I think that is up for discussion for how that will be implemented inthe long run
14:10 jasonmay how does perl 6 do it :)
14:11 bphillips I think they have function versions of builtins that delegate to the method in the runtime, but I'm not certain
14:14 moritz that's correct
14:14 moritz though we also have function versions of builtins that directly do stuff, which is more performant when we know the exact type
14:14 moritz functions are much easier to statically analyze (and even inline) than methods
14:16 bluescreen joined #moe
14:21 masak that's their main semantic difference, in my view. functions can be statically known (and often are unless you pass them around in variables), methods tend to be late-bound (and often are unless the compiler does some pretty deep tricks).
14:30 jasonmay in theory could you have both available, and bind the method to the statically known function directly (instead of adding a layer of delegation) during the bootstrap process?
14:33 jasonmay ... that's kind of indirect object syntax, isn't it. oops!
14:34 bphillips jasonmay: I had the same train of thought :-)
14:36 jasonmay but I don't see the harm in special-casing it for a few things, as long as it's not a feature
14:36 jasonmay and it would be going the other way :)
14:37 jasonmay though, how badly do we *need* to make these functions available from CORE
15:36 stevan joined #moe
15:55 jnap left #moe
15:56 jnap joined #moe
16:29 clintongormley joined #moe
17:26 bphillips joined #moe
18:41 clintongormley joined #moe
18:59 moe [moe] prakashk pushed 3 new commits to runtime-improvements: http://git.io/148adw
18:59 moe moe/runtime-improvements 2ecec1d Prakash Kailasa: implemented magical string increment function. Used in auto-increment, and enumerating string ranges as well.
18:59 moe moe/runtime-improvements 785210b Prakash Kailasa: preserve length/format of the incremented string ("01" -> "02")
18:59 moe moe/runtime-improvements 56509b2 Prakash Kailasa: implemented string range enumeration using magicalStringIncrement function and added more tests
19:01 jnap joined #moe
19:35 jnap joined #moe
19:41 sahadev jasonmay: please take a look at my comment on https://github.com/MoeOrganization/moe/pull/54 when you get a chance.
19:42 doy sahadev: you should make your changes on a new branch off of master
19:42 * sahadev == prakashk@github, btw
19:42 jasonmay what doy said :)
19:42 sahadev ok.
19:43 jasonmay heh, my PR reply had a pretty high hyphen-to-word ratio
19:43 sahadev i guess it was a bad idea to have started with the runtime-improvements branch.
19:43 jasonmay it's our fault for not removing (or updating) it
19:44 sahadev is it being used at all by anyone?
19:45 jasonmay nope, it was merged into master a while ago
19:46 sahadev so, since I merged master into this branch and resolved conflicts, shouldn't it have only my changes (compared to master)?
19:48 sahadev git diff tells me that's not the case. never mind.
20:41 bphillips1 joined #moe
21:01 moe [moe] prakashk created prakashk/string-ranges (+2 new commits): http://git.io/dHVP6A
21:01 moe moe/prakashk/string-ranges 2898757 Prakash Kailasa: merged with master
21:01 moe moe/prakashk/string-ranges e8bdae5 Prakash Kailasa: Merged changes made to branch 'runtime-improvements' (same as pull request #54)
21:08 jnap1 joined #moe
21:09 jnap joined #moe
21:09 * jasonmay just noticed doy's DI scala project
21:09 sahadev I created a new branch prakashk/string-ranges off of master, and applied my changes to that branch. Should I create a new pull request? Or, is there a way the existing one can be reused?
21:10 bphillips1 anyone know where this came from? http://perl8.org/
21:12 jasonmay ah I thought it was another implementation of Bread::Board ;)
21:13 jasonmay (his repo, not perl8.org)
21:13 jasonmay bphillips: I was wondering that myself
21:13 doy jasonmay: yeah, i'm just fiddling around with scala a bit
21:13 doy that repo is based a decent bit on bread::board
21:13 doy but really only conceptually
21:14 jnap bphillips:  wow, a page with frames
21:16 doy it might be something to consider for the structure of the compiler though
21:16 jnap https://gist.github.com/jjn1056/4956477
21:16 doy supposedly it's similar to how the scala compiler is structured
21:41 jnap joined #moe
23:43 gizmomathboy joined #moe
23:43 sivoais joined #moe

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