Perl 6 - the future is here, just unevenly distributed

IRC log for #native, 2017-01-16

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

All times shown according to UTC.

Time Nick Message
00:03 plicease latest Pegex no longer builds on 5.8 breaking travis :/
00:03 plicease AB travis uses it indrectly via Inline::C
00:07 plicease I worked around this by installing an older version of Pegex
00:07 plicease https://github.com/Perl5-Alien/Alien-Base/commit/d0a0941be281a67b682092d4ed579b4da68b3d2d
00:07 plicease but in the near future I plan on altering the travis config to skip the Inline tests on 5.8.
00:29 willthechill plicease: RPerl has always been based on Inline, and thus has never worked in Perl v5.8
00:29 willthechill it _is_ on my to-do list, but so are several hundred other line items as well :-P
12:53 plicease Inline has worked for years on 5.8 this is new
12:54 dcmertens Besides Inline::C, what is Pegex used for?
12:54 * dcmertens grumbles
12:55 * dcmertens admits he could look up the reverse dependencies himself
12:56 plicease nothing that I use myself
12:56 plicease https://metacpan.org/requires/module/Pegex?sort=[[2,1]]&size=500
12:58 * dcmertens checks
12:58 plicease metacpan used to provide a count of the dists listed
12:59 plicease nevermind it has it on the bottom
21:10 mohawk joined #native
21:10 mohawk evening
21:10 plicease mohawk: welcome!
21:11 mohawk thanks!
21:12 mohawk i'm going to look at plicease's PRs against NativeCall, the world-conquering albeit barely-functional thing that is mentioned in the /topic!
21:12 mohawk if my mailer ever starts responding to inputs
21:12 plicease ha that topic was set before the existence of perl5 NativeCall
21:13 plicease i only mentioned it way back when with the thought that some might recognize the term coming from other languages
21:13 mohawk doh
21:13 * mohawk pretends he knew that
21:13 plicease lol
21:14 mohawk anyway, it's hardly a coincidence - someone linked me to the p6 one and i immediately saw how to (sort of) have it in p5 too
21:14 plicease i like the simplicity of it.  I never really liked FFI::Platypus::Declare that much which was my attempt at a simple FFI interface.
21:15 mohawk if i have seen far, it is because i have stolen ideas from giants
21:17 mohawk plicease, looking at PR#4 first
21:17 mohawk why do we need no warnings 'redefine' ?
21:22 plicease i'm overriding the attach method
21:23 plicease cleaner way would be to use Monkey::Patch or something like that but I it didn't seem worth the extra prereq
21:24 plicease Probably should have scoped it for just that sub definition
22:24 mohawk sorry, was afk
22:34 mohawk plicease, i don't understand why override attach there?
22:35 plicease I just want to test that it calls attach in the expected way.
22:35 plicease rather than have to depend on a specific symbol
22:36 mohawk yeah, just getting there myself
22:36 mohawk fair enough
22:36 plicease there are lots of tests in FFI-Platypus with regard to symbol look up
22:36 plicease didn't seem necessary to replicate that here
22:36 plicease i'm off for a bit may be on later.
22:36 mohawk ok
22:39 mohawk plicease, not sure why we need 5.14 for PR#1
23:24 mohawk that's moot since i believe plicease that 5.16 is needed because of attribute.pm bug
23:25 mohawk merged all but the closures one which needs conflicts resolved, but also a little doc
23:41 mohawk and i've released 0.005

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