Perl 6 - the future is here, just unevenly distributed

IRC log for #perl11, 2014-12-11

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:19 willthechill joined #perl11
05:30 Reini joined #perl11
07:40 rurban joined #perl11
08:25 travis-ci perl11/potion#809 (numcmp - 8437d06 : Reini Urban): The build is still failing. http://travis-ci.org/perl11/potion/builds/43689858
08:26 basiliscos joined #perl11
08:32 willthechill rurban: my build is still passing on local but failing on travis, I'm working to figure out why  :/
08:33 rurban I'm still too busy with my jit problem...
08:36 willthechill that's okay
08:55 rurban sigh, got it finally. 20min
08:57 willthechill rurban++  :)
09:04 rurban still a tiny logical error somewhere. but at least the jumps and ops are now ok
09:04 willthechill nice!
09:08 rurban Can't locate Inline.pm in @INC
09:09 rurban Inline is still pretty broken
09:10 travis-ci perl11/potion#810 (numcmp - 9541158 : Reini Urban): The build is still failing. http://travis-ci.org/perl11/potion/builds/43693068
09:10 rurban Inline::Files still accesses my 5.6
09:10 willthechill wait, Inline is broken in your potion code???
09:10 willthechill I didn't know you were using Inline
09:10 rurban no, cpan Inline::C fails
09:11 rurban I'm testing rperl now
09:12 rurban I even fixed it already, but my fixes didnt go upstream apparently
09:12 willthechill wait, you just now fixed rperl?
09:12 rurban Ingy still thinks that there cannot be only one and only perl
09:12 rurban Inline
09:13 willthechill OH you made some fix to Inline but it has not been accepted by ingy yet?
09:14 rurban He rewrote it in the meantime, jhave to check
09:15 rurban Inline is ok.
09:18 rurban Where does Inline::Files come from?
09:18 willthechill um I don't know Inline::Files
09:18 willthechill I only know Inline::Filters
09:18 willthechill does rperl build on your machine?
09:20 rurban Ok, after deleting Inline::Files, Inline::C can be installed
09:20 rurban So this requires special code in the Inline Installer, similar to Moose
09:21 willthechill wait, is any of that related to rperl?
09:21 willthechill I've never heard of Inline::Files?
09:21 rurban When it's installed rperl will not work
09:24 rurban testing rperl now
09:24 willthechill great, it should have 1457 tests pass
09:24 rurban yep, can repro the exacty same travis error
09:26 willthechill okay I have no idea why it can't find Inline in @INC
09:26 willthechill :(
09:26 willthechill do we need ingy's help?
09:26 rurban even that: Can't locate RPerl.pm in @INC
09:27 rurban no, that's your problem
09:28 willthechill I don't see "can't locate RPerl.pm" anywhere in the travis output
09:31 rurban $ pb lib/RPerl/Test/AlgorithmMath/geometric_algebra_ipvve_good_08.pl => 137 (good)
09:32 rurban running it via the test driver: t/08_interpret_run_program.t passes no @INC to it
09:32 rurban -Mblib esp.
09:33 willthechill I don't understand what is causing this...  due to the makefile and the blib directory?
09:33 rurban no, because of your open3
09:33 willthechill oh so it IS open3()!
09:34 rurban And it will not work in Windows, I guess
09:34 rurban use IPC::Cmd instead
09:35 rurban the problem is: $test_file, yiou just cannot run the test file, you need to pasds the perl and the INC to it. "$^X -Mblib $test_file"
09:36 rurban yep, works now
09:36 willthechill okay great!
09:36 willthechill do you want to push code?
09:37 rurban It's still horrible code
09:38 willthechill what is horrible code?
09:38 rurban the open3 code. unix only
09:40 rurban rperl_to_xsbinary__compile misses the compilation still, right?
09:41 rurban I don't even see a definition for it
09:41 willthechill it misses AST->C++ source still
09:41 willthechill then C++->binary is via Inline::CPP
09:41 willthechill I am debugging grammar now to finalize AST formation
09:42 willthechill then AST->C++ is pretty easy
09:42 willthechill then DONE!!!
09:42 willthechill at least, done with v1.0 finally  :P
09:42 willthechill I was considering use of Module::Compile but haven't got far enough yet
09:42 rurban pushed
09:43 willthechill may or may not need M::C
09:43 willthechill okay great!
09:43 willthechill will pull and test now
09:43 willthechill testing now
09:44 willthechill if you know how to make the code work on non-Unix environments, by all means please help!
09:44 willthechill I only know Unix
09:44 willthechill bulk88: did we ever get a build on Windows?
09:45 rurban It's just the test, you can convert to IPC::Cmd or skip it
09:46 rurban it should work manually via cmd.exe cmdline: prog >log.out 2>log.err
09:46 rurban same as on unix. do you need a timeout?
09:49 rurban1 joined #perl11
09:53 travis-ci RPerl build passed. Reini Urban says 'Test, pass our used perl and blib @INC paths to t/08_interpret_run_program.t'
09:53 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43695896 https://github.com/wbraswell/rperl/compare/d09b1aa4059b...b72acb41cee0
09:53 rurban See, fixed
09:53 willthechill yay!
09:53 willthechill thank you my friend!
09:53 willthechill rurban++
09:53 willthechill you are asking if I need a timeout for open3()?
09:54 rurban yes
09:54 willthechill I'm not sure, I had a delay of 0.3 seconds per test but when I moved the sysread() after waitpid() then I could remove the delay and it worked fine
09:54 willthechill is that kind of what you're asking?
09:54 rurban I believe IPC::Cmd would be easier to use. But let's no hurry
09:55 willthechill I use open3() because it is PBP
09:55 willthechill ;)
09:56 rurban Looks like OS X is now officially supported by travis: http://docs.travis-ci.com/user/osx-ci-environment/
10:00 willthechill okay great, shall I just enable it or must somebody with OSX experience first look at rperl?
10:00 willthechill didn't you do some OSX work on rperl a while back?
10:19 rurban I think so. I'll try it in on of my repos. so far it didn't trigger the osx tests
10:27 travis-ci perl11/potion#811 (numcmp - 46c1c2a : Reini Urban): The build is still failing. http://travis-ci.org/perl11/potion/builds/43700049
10:46 willthechill okay well maybe I have to change some settings in travis, let me know
10:54 willthechill oh and also yes, the code passes all tests with your fix, thanks again!
10:54 willthechill rurban+++
11:18 dalek p2/p2: aae5b0b | rurban++ | 3rd/libuv:
11:18 dalek p2/p2: 3rd: joyent => libuv/libuv change
11:18 dalek p2/p2: review: https://github.com/perl11/p2/commit/aae5b0bb8e
11:19 willthechill rurban: I am very desirous to combine together rperl and p2 as soon as possible
11:19 willthechill do we need B::C any more, or can we throw it away forever?
11:19 willthechill or do we keep B::C, and just replace the internals with p2 & rperl?
11:20 travis-ci perl11/p2#217 (p2 - aae5b0b : Reini Urban): The build was broken. http://travis-ci.org/perl11/p2/builds/43704832
11:28 rurban aeh??
11:28 willthechill what does aeh mean?
11:28 rurban B::C is for perl5. p2 needs still a long way to go. I don't even have a C or native compiler for it
11:30 rurban But I'm sceptical if 5.22 support will be doable
11:31 willthechill p2 will support both p5 and p6 natively?
11:55 rurban we will see. hopefully
11:55 rurban at least parse it
12:11 willthechill great!
12:11 willthechill how long until p2 will be usable?
12:11 rurban still cannot do heredocs in p5, but should work in p6
12:12 rurban thinking about rewriting the parser to be self-hosting, and add a Parser class and rules as methods, as in python pyrser. => parser  macros and fixed GC (eval)
12:12 rurban will be a bit slower, but more powerful. can extend the parser at run-time
12:13 rurban and the type inferencer will also come soon.
12:13 rurban similar to Bla
12:15 willthechill never heard of Bla
12:15 basiliscos joined #perl11
12:17 basiliscos1 joined #perl11
12:42 rurban1 joined #perl11
12:43 rurban see our perl11 github
12:43 rurban https://github.com/perl11/Bla
12:45 willthechill yes I see that
12:45 willthechill functional languages: blech!  haha
12:56 rurban It's very similar to potion. a simplier compiler and vm, but with gradual types
12:56 willthechill but you already have potion, why care about Bla?
12:57 rurban take the type inferencer and checker
12:58 willthechill you are going to copy the code out of bla and into potion?
12:58 rurban no, not possible. but I use it as example
12:58 rurban and Bla was never published, so I did
12:59 rurban it's from 1995 or so
13:00 willthechill oh okay cool
13:01 willthechill well it is 0700hrs, time for me to eat dinner and go to sleep for a little while!  :)
13:02 rurban I am also just cooking Tacos and Guacamole for Mirjam
13:05 willthechill mmm, sounds nice!
13:06 willthechill tell Mirjam I say hello and enjoy the guacamole!
13:06 rurban We got really good taco spices over here
13:06 rurban thanks
13:06 willthechill either chili or hamburger for me!
13:07 willthechill but nobody can beat the taco spices in Texas
13:07 rurban I believe we import them from Texas
13:13 travis-ci RPerl build passed. Will Braswell says 'Preprocessor, Interpreted Type Checking, Part 9'
13:13 travis-ci http://travis-ci.org/wbraswell/rperl/builds/43713507 https://github.com/wbraswell/rperl/compare/b72acb41cee0...33e0118919ad
13:14 willthechill good, Texas spices are a valuable import item, haha!  :)
17:10 ozzmotik|mobile joined #perl11
17:57 ozzmotik joined #perl11
18:07 basiliscos joined #perl11
18:41 bulk88 "[04:44] <@willthechill> bulk88: did we ever get a build on Windows?" yes once, long time ago
18:53 ozzmotik sometimes i tend to forget windows exists
19:23 willthechill ozzmotik: you and me both, haha!
19:23 willthechill bulk88: can you please try again now with the latest rperl code?  lots of changes!  :)
20:00 Reini joined #perl11
20:27 ozzmotik joined #perl11
20:28 willthechill joined #perl11
20:29 ozzmotik joined #perl11
21:02 travis-ci perl11/potion#812 (numcmp - a1bde2b : Reini Urban): The build is still failing. http://travis-ci.org/perl11/potion/builds/43763851
21:11 travis-ci perl11/potion#813 (numcmp - 8e1fe29 : Reini Urban): The build is still failing. http://travis-ci.org/perl11/potion/builds/43764417
21:34 rurban Fixed now my SSE assembler code for numerical comparisons
21:44 travis-ci perl11/potion#814 (numcmp - bf8ca57 : Reini Urban): The build was fixed. http://travis-ci.org/perl11/potion/builds/43768124
21:46 d4l3k_ joined #perl11
21:57 travis-ci perl11/potion#815 (master - 8960f82 : Reini Urban): The build passed. http://travis-ci.org/perl11/potion/builds/43769631
22:05 Reini joined #perl11
22:37 willthechill joined #perl11
22:51 willthechill joined #perl11
23:13 Reini joined #perl11

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary