Perl 6 - the future is here, just unevenly distributed

IRC log for #inline, 2014-12-24

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

All times shown according to UTC.

Time Nick Message
00:44 davido__ joined #inline
00:52 sivoais it's up!!! <https://metacpan.org/release/ZMUGHAL/Statistics-NiceR-0.02>, <https://metacpan.org/release/ZMUGHAL/Data-Frame-0.001>. Time to write a blog post!
01:01 ingy go sivoais!
01:59 sivoais blogged <http://enetdown.org/dot-plan/posts/2014/12/24/a_fast_and_natural_interface_to_R_from_Perl/>
03:30 ingy sivoais: *click*
03:33 ingy sivoais: I'll link to it from TheFinalReport™!
03:33 ingy great work
03:35 sivoais now I'm trying to get it working on Windows :-)
04:53 GitHub104 [inline-c-pm] jelty comment on issue #21: Hi,... http://git.io/tI9FiA
06:08 mohawk ingy, i'm planning to skip_all on I::C's 26fork.t and release
06:08 mohawk thoughts?
06:08 mohawk woops
06:09 mohawk skip_all if on WinXP
06:29 GitHub194 [inline-c-pm] mohawk2 pushed 1 new commit to master: http://git.io/QHCZXA
06:29 GitHub194 inline-c-pm/master 256e470 Ed J: skip_all if WinXP as test hangs there
06:30 GitHub130 [inline-c-pm] mohawk2 comment on issue #21: `skip_all` if WinXP in 256e4706885d7663d22193549bbbb647fca1fe4b. Waiting for clearance from @ingydotnet to release. http://git.io/nuMS5w
06:31 mohawk ingy, as you can see above - waiting to release if you're happy
07:54 ingy mohawk: did you get 27 working?
07:54 * ingy tests
07:55 mohawk works for me
07:55 mohawk while i'm there, may as well fix that fucker too
07:56 ingy ?
07:58 mohawk 27* works for me since someone made it skip_all
07:59 mohawk trying without
08:01 mohawk ok, getting a fail
08:02 ingy mohawk: I'd like this know when it started failing and why
08:03 ingy also didn't bulk88's PR/32 fix the 26 hang?
08:04 ingy well go ahead and release when you want to. we should at least make it installable again.
08:04 mohawk his PR/32 might fix it
08:04 mohawk but someone said they wanted to wait for a fix to 27
08:04 mohawk https://github.com/ingydotnet/inline-c-pm/pull/32#issuecomment-67603479
08:05 mohawk and no-one merged it
08:06 ingy your call. just be sure rebase or cherry-pick it if you use it.
08:06 mohawk yeah\
08:06 ingy ingy out &
08:07 mohawk glad you're keen on git purity
08:07 mohawk less glad on leaving PRs just sitting there "waiting for (x)"
08:18 GitHub54 [inline-c-pm] mohawk2 comment on issue #32: @bulk88, Win32::Mutex is a CPAN module and I'd like to avoid dependencies. Can you please redo this using eg `flock()`? http://git.io/xazf0g
08:20 mohawk ingy ^
08:22 ingy mohawk: go ahead with your skip fix for now
08:22 mohawk ok
08:22 ingy I don't want the dep
08:22 mohawk nor me
08:23 mohawk i'm having a quick go at t/27*
08:23 mohawk looks like rcopy (from File::Copy::Recursive) that's failing
08:23 mohawk that's disappointing
08:23 ingy although we could skip the tests unless Win32::Mutex installed
08:23 mohawk that's a bodge
08:24 ingy ?
08:24 mohawk because Win32::Mutex is used in the actual module
08:24 ingy which module?
08:24 mohawk I::C
08:24 mohawk so we need a dynamic config thing, which i don't think Z::D can handle
08:24 mohawk (ie a dep only on win32)
08:25 ingy I'm confused. anyway, too late for me
08:25 mohawk deps are normally fixed
08:26 mohawk there's a facility to have dynamic config, where a module writes out MYMETA.json with its deps "now"
08:26 mohawk that might differ from its CPAN-distrib-ed META.json
08:26 mohawk that's a dynamic config
08:26 mohawk it allows platform-specific deps, like here
08:30 GitHub170 [inline-module-pm] ingydotnet pushed 1 new commit to master: http://git.io/hYrq-Q
08:30 GitHub170 inline-module-pm/master 9fa27a7 Ingy döt Net: Add 'makestub => 1' to config. `perl Makefile.PL`
08:32 mohawk oh dear, t/pegex-parser.t is trying to use "rm"
08:32 mohawk shockingly that doesn't work on windows
08:36 mohawk a big problem with t/27* is it's using glob, which can't handle spaces
08:36 mohawk nice work, p5p
08:38 mohawk wait, yes it can
08:38 mohawk danging
08:39 mohawk no it can't
08:40 mohawk if the pattern is 'space name/*'
08:41 mohawk not on windows, anyway
08:43 mohawk doh, it's documented to act that way
08:55 mohawk ok, works now - replaced with opendir/readdir
08:55 GitHub129 [inline-c-pm] mohawk2 pushed 1 new commit to master: http://git.io/hyIXgw
08:55 GitHub129 inline-c-pm/master e247419 Ed J: Replace glob as fails if space in dir name
09:11 mohawk passes all tests on windows
09:12 mohawk albeit with some TODO, especially from test/TestInlineC.pm's use of diff - not impressed
09:17 GitHub91 [inline-c-pm] mohawk2 pushed 1 new commit to master: http://git.io/R5JBLQ
09:17 GitHub91 inline-c-pm/master 392d378 Ed J: CPAN Release 0.68...
09:17 GitHub96 [inline-c-pm] mohawk2 tagged 0.68 at master: http://git.io/u9-LJg
09:20 mohawk w00t
09:20 mohawk that should be slightly less shit now
09:21 mohawk ingy, i'm assuming you're ok with me using my judgement on fixes for t/26 and releases based on that
09:27 GitHub153 [inline-c-pm] mohawk2 comment on issue #20: The skip_all was a wrong action. The whole point of making perl stuff including Inline::C work in places-with-spaces is to enable the poor Windows user whose temp dir is in such a place, who'd like to use Perl and install CPAN modules. Surprised you guys hate Windows users so much.... http://git.io/r1spTw
09:31 mohawk ingy, i'm thinking of a new Inline keyword like "add_module" to enable eg Inline::Struct (or Filters) without hard-coded support in Inline[::C] - are you in favour?
10:08 GitHub199 [inline-c-pm] mohawk2 comment on issue #32: From IRC #win32:... http://git.io/fKQotw
10:11 GitHub66 [inline-c-pm] bulk88 comment on issue #32: 26t became stable because I used Win32's native Kernel provided synchronization APIS. Not a hacked up API of using flock foor syncronixation,. Usunt Natuve Win32 IPC is going to be better than try hacked up aattempts that dont even reach the claims they show off. http://git.io/N6NB1Q
11:16 rurban1 joined #inline
11:19 rurban1 joined #inline
14:50 priodev joined #inline
16:52 bulk88_3 joined #inline
17:07 bulk88_2 joined #inline
18:37 GitHub151 [acme-math-xs-pm] ingydotnet pushed 1 new commit to eumm: http://git.io/M3dz_Q
18:37 GitHub151 acme-math-xs-pm/eumm 6f987ef Ingy döt Net: Add `makestub` param.
20:38 mohawk ingy, i'd like you to give urgent attention to https://github.com/ingydotnet/module-compile-pm/pull/4
20:41 mohawk i've EVEN updated Meta and Changes
20:41 mohawk all you should have to do is merge, then zild make release
20:52 ingy mohawk: ok. I'll probably get to it sometime today
20:54 ingy why is Module::Compile urgent?
20:55 mohawk it breaks PDL installation
20:55 ingy ok
20:55 mohawk i'm hoping this will be a genuine no-brainer
21:00 ingy it's the wrong solution
21:00 ingy but maybe ok for interim
21:00 mohawk that's all we need
21:01 mohawk it uses a module, so we declare that
21:01 mohawk if there's another change that's better in the long-term, great
21:01 mohawk but two separate issues, shurely
21:01 ingy I'll probably just do it right the first time
21:03 mohawk what if you did the interim literally now
21:03 mohawk THEN got started on the right one?
21:05 mohawk the way i see it, you put this thing on the CPAN
21:05 mohawk that's a promise it will actually install and maybe even work
21:05 mohawk right now we're breaking that promise
21:06 mohawk so please, i'm asking nicely, add the dep
21:06 mohawk and why not benefit from me fixing up the zild update at the same time
21:06 mohawk this is a real win for you
22:56 priodev joined #inline

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