Perl 6 - the future is here, just unevenly distributed

IRC log for #perl11, 2017-01-13

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

All times shown according to UTC.

Time Nick Message
00:07 mako93r PASSED: https://ci.appveyor.com/project/wbraswell/rperl
00:08 mako93r What was it?
00:09 willthechill it was because you had typed `set path=BLAHBLAH;` instead of `set path=BLAHBLAH;%path%`
00:09 willthechill so the original path was lost, and AStyle.exe could not be found
00:09 willthechill it was never a problem with astyle, or RPerl
00:09 willthechill it was not even a problem with AppVeyor
00:10 mako93r hm? I just copied that stuff
00:10 willthechill it was only a problem with the appveyor.yml script deleting the path
00:10 willthechill well whatever you copied was not exactly correct, so I debugged it and fixed it
00:10 willthechill either way, let us celebrate, for the time of AppVeyor sadness is over
00:11 mako93r so including '%path%' was needed?
00:11 willthechill yes
00:11 willthechill that fixed it
00:11 travis-ci joined #perl11
00:11 travis-ci wbraswell/rperl#1146 (master - 671de3d : Will Braswell): The build passed.
00:11 travis-ci Change view : https://github.com/wbraswell/rperl/compare/33711f07a6e0...671de3dcda7b
00:11 travis-ci Build details : https://travis-ci.org/wbraswell/rperl/builds/191481678
00:11 travis-ci left #perl11
00:11 mako93r so simple?
00:11 willthechill "simple"
00:11 willthechill so simple it took weeks to debug and let to finding bugs in other unrelated software
00:12 mako93r yeah "simple" what does "simple" really mean
00:12 willthechill but yes, most bugs are "simple" after you know the answer
00:12 mako93r so getting the env from point to the other was issue?
00:14 willthechill the issue was to not delete the path
00:14 willthechill that's it
00:15 mako93r not so fast, please!!
00:15 mako93r let's do the hard way!!!
00:17 willthechill I don't understand what you're asking
00:17 willthechill the path got deleted, so AStyle.exe could not be found, that's it
00:17 willthechill there is no "hard way"
00:18 willthechill once the path got un-deleted, then AStyle.exe could be found, and the diff check tests run fine
00:18 willthechill it is a simple fix, as you say
00:19 willthechill so, moving on from AppVeyor...  next step is to finish the 4 remaining arithmetic operators in CPPOPS_CPPTYPES mode: abs, exp, log, sqrt    http://rperl.org/learning_rperl.html#Section_2.1.11%3A_Arithmetic_Operators
00:20 mako93r "hard way" means please check my current pull request
00:21 willthechill sorry, too slow
00:21 willthechill yes, I go fast
00:21 willthechill like RPerl
00:22 mako93r should give you the same results. Hey, I removed my hackery. Nobody likes hacks. That's what I'm really talking about.
00:23 willthechill I told you that I would delete the unneeded code.  I deleted it.  You made a pull request anyway, even though I said I would delete it.  Your pull request was too late.
00:24 mako93r Oh, sorry.
00:24 mako93r Over eager
00:25 willthechill that's fine, just trying not to waste time
00:25 willthechill so, again, moving on from AppVeyor...  abs, exp, log, sqrt
00:26 mako93r hope I haven't interferred with your stuff
00:26 mako93r and yes, let's move on
00:26 willthechill nope, deleting the pull request and re-explaining myself only took about 3 mins
00:26 willthechill my current RPerl system task is to fix bugs and get RPerl to pass on as many CPAN testers systems as possible
00:27 willthechill this is a huge task with dozens of bugs, mostly in other people's software
00:27 willthechill so I will leave the work on RPerl operators to you
00:27 willthechill talk to me on IRC as you work on each operator, and once you finish with these 4 arithmetic ops then I will help choose the next ops to implement, and so forth
00:28 mako93r ok, so we consider AppVeyor stuff finished?
00:28 mako93r somehow?
00:29 willthechill yes, as I have said, the time of AppVeyor is now over
00:29 willthechill so you may celebrate accordingly
00:30 mako93r thx, I hope I was of any use for you
00:30 mako93r feeling like a monkey that didn't achieve anything
00:30 willthechill oh yes, you did a very good job getting it started
00:31 willthechill and I finished it, so we worked as somewhat of a team
00:31 willthechill or at least, you did the first half, and I did the second half
00:31 willthechill so that is fair
00:31 willthechill we should both be very happy that AppVeyor is working
00:32 willthechill and now you know to always check your environmental variables for accidental deletion, trailing semicolons (Windows), trailing colons (*nix), etc.
00:32 willthechill live, learn, move on to next task
00:33 mako93r sounds good but not satisfying, I'm feeling like I failed
00:33 mako93r even though everything is fine now
00:34 willthechill nope, nobody failed
00:44 mako93r ok, no problem, but again, for the last idiots, including of '%path%' was the issue, when setting env var 'PATH'? Am I getting this right?
00:45 mako93r that chocolatey stuff mentioned by bulk88 some time ago?
00:46 willthechill yes, we were just missing %path% when setting the path env var, that is it
00:46 mako93r if so, it was never RPerl Team's fault, wasn't it?
00:48 willthechill yes it was our fault for not having the %path% correctly in appveyor.yml
00:48 willthechill but it is fixed now, so let us move on
00:54 mako93r EXCELLENT! I'll leave it that way. I hope I can concentrate on my OpenBSD stuff now and what ever else is remaining (like Operators). :-)
00:54 mako93r collaboration pays off
00:56 willthechill the newest version of RPerl on CPAN should have better BSD support as well
00:57 mako93r cool, glad to hear that
00:57 mako93r will test on OBSD, ASAP
00:57 willthechill thanks to the Alien team
00:57 willthechill helping me with Alien::astyle
00:58 mako93r Alien::astyle is relying on Alien::Base isn't it?
00:59 willthechill yes all Alien modules do as far as I know
00:59 mako93r on windows I haven't even noticed it
01:00 mako93r caused me some trouble under OpenBSD, but could resolve the issue
01:00 mako93r I'd like to come up with CI system for BSD systems...
01:01 mako93r ..., but nothing really useful at moment
01:01 willthechill yes well I am currently relying on CPAN testers for BSD testing, and that takes sometimes a few days for tests to run
01:02 mako93r And that's hardly reliable. But I think you shouldn't bother.
01:03 mako93r Concentrate on XUbuntu Linux!
01:03 mako93r That's far easier.
01:03 willthechill I already said that my current task is to get RPerl to pass on as many CPAN tester platforms as possible
01:03 willthechill not just Linux
01:04 willthechill I am not currently implementing any new RPerl system software, I am only fixing bugs so that RPerl can support all different platforms
01:04 mako93r and you did good
01:05 mako93r best option: I test RPerl on OpenBSD for you and give you as detailed report as I can.
01:06 willthechill that will help, yes, but it does not replace CPAN testers
01:07 mako93r what's the difference?
01:07 willthechill please familiarize yourself with CPAN testers
01:07 willthechill http://www.cpantesters.org/
01:07 willthechill http://matrix.cpantesters.org/?dist=RPerl+2.403000
01:08 willthechill it _would_ help if you created your own CPAN tester account and tested every version of RPerl that comes out, using the official CPAN testers system
01:08 willthechill I am not very much interested in receiving tons of different build reports from different people and places; it is enough for me to check 3 different sources already, travis & appveyor & CPAN testers
01:13 mako93r ok, accepeted
01:14 willthechill first step: make sure it passes Travis (easy);  second step: AppVeyor (difficult);  third step: CPAN testers (very difficult)
01:15 mako93r will study the stuff!!!
01:16 mako93r glad to contribute for the progress
01:17 willthechill yes you will be MUCH more valuable for testing purposes once you are an official CPAN tester
01:19 mako93r cool, will study that stuff carefully
01:19 willthechill great!
01:20 mako93r but one thing remains: Please remind me to be careful when it comes up to making promises!
01:20 willthechill okay?
01:22 mako93r I promised you to help you with AppVeyor. And as far as it looks like it worked, somehow. But next time deeper thinking is necessary.
01:23 mako93r Yes, I'm somewhat over-keen when it comes to this.
01:24 mako93r Best would be I'm cooling down.
01:26 willthechill well you did help with appveyor, so I don't see a problem
01:31 mako93r cool, now I can really go back to op dev
01:32 travis-ci perl11/cperl#2144 (smoke/master - 382c987 : Aaron Crane): The build passed. https://travis-ci.org/perl11/cperl/builds/191488913
01:34 willthechill yup
01:44 travis-ci joined #perl11
01:44 travis-ci wbraswell/rperl#1150 (master - 31554c1 : Will Braswell): The build passed.
01:44 travis-ci Change view : https://github.com/wbraswell/rperl/compare/feb26e04d472...31554c155038
01:44 travis-ci Build details : https://travis-ci.org/wbraswell/rperl/builds/191499459
01:44 travis-ci left #perl11
07:21 anton_p joined #perl11
07:55 fluca1978 joined #perl11
09:48 travis-ci perl11/cperl#2145 (smoke/master - 59ddb19 : Aaron Crane): The build passed. https://travis-ci.org/perl11/cperl/builds/191569160
10:34 travis-ci joined #perl11
10:34 travis-ci wbraswell/rperl#1153 (master - 3faf1b3 : Will Braswell): The build passed.
10:34 travis-ci Change view : https://github.com/wbraswell/rperl/compare/18fc2684359d...3faf1b3d5fea
10:34 travis-ci Build details : https://travis-ci.org/wbraswell/rperl/builds/191576653
10:34 travis-ci left #perl11
10:52 willthechill yay
11:05 anton joined #perl11
11:57 travis-ci Alien::astyle build passed. Will Braswell says 'CPAN Release, v0.007; Codename Bond, James Bond'
11:57 travis-ci https://travis-ci.org/wbraswell/alien-astyle/builds/191607295 https://github.com/wbraswell/alien-astyle/compare/ecfaa1ab070c...4957142b17f2
11:57 willthechill yay
11:57 willthechill SHAKEN, NOT STIRRED haha  :-D
12:05 travis-ci perl11/cperl#2146 (smoke/master - d400fee : Reini Urban): The build passed. https://travis-ci.org/perl11/cperl/builds/191597186
12:31 travis-ci joined #perl11
12:31 travis-ci wbraswell/rperl#1154 (master - fe3a61c : Will Braswell): The build has errored.
12:31 travis-ci Change view : https://github.com/wbraswell/rperl/compare/3faf1b3d5fea...fe3a61cd1c2a
12:31 travis-ci Build details : https://travis-ci.org/wbraswell/rperl/builds/191612011
12:31 travis-ci left #perl11
12:43 willthechill gotta wait for Alien::astyle v0.007 CODENAME Bond, James Bond  :-D
14:27 travis-ci joined #perl11
14:27 travis-ci wbraswell/rperl#1154 (master - fe3a61c : Will Braswell): The build passed.
14:27 travis-ci Change view : https://github.com/wbraswell/rperl/compare/3faf1b3d5fea...fe3a61cd1c2a
14:27 travis-ci Build details : https://travis-ci.org/wbraswell/rperl/builds/191612011
14:27 travis-ci left #perl11
15:40 travis-ci perl11/cperl#2147 (smoke/master - a316258 : Karl Williamson): The build passed. https://travis-ci.org/perl11/cperl/builds/191650616
18:00 travis-ci perl11/cperl#2148 (smoke/master - 1ec02f0 : Karl Williamson): The build passed. https://travis-ci.org/perl11/cperl/builds/191693522
18:48 anton joined #perl11
20:53 mako93r joined #perl11
20:54 willthechill joined #perl11
20:57 mako93r when it comes to 't/06': /home/mako/src/rperldev/rperl93r/_Inline/lib/auto/eval_659_ea4f/eval_659_ea4f.so: undefined symbol '_ZNKSt8__detail20_Prime_rehash_policy11_M_next_btkEm'
20:57 mako93r lazy binding failed!
20:58 mako93r what does that mean?
20:58 willthechill does it actually say "lazy binding failed", or did you make up that part yourself?
20:59 mako93r no, 'make test' gave me that
21:00 willthechill I've never seen that error before!
21:01 willthechill please clean the whole system as best as possible, then rebuild from scratch, and if the problem still persists, then we need to open a new GitHub issue with all the pertinent details
21:04 mako93r ok, will do, but a second question, 't/07' said: "GMP.so: undefined symbol '__gmpz_init_set_ui'" (left out the paths)
21:04 mako93r lazy binding failed!
21:04 mako93r as a make error msg
21:05 willthechill yeah you must have some kind of problem on your system
21:05 willthechill please clean it as well as possible, then be as careful as possible while building
21:05 willthechill and please make sure you are using RPerl v2.410 just released onto CPAN a few hours ago
21:05 willthechill AFK
21:08 mako93r ok, my sys cfg is: OpenBSD 6.0 (64bit) (g++ 4.9.3) (libgmp 6.1.2) (perl 5.20.3)
21:08 mako93r would be I set up a new machine from scratch
21:09 mako93r *would be best
21:10 mako93r BTW: what is the minimal version of 'libgmp' to work with when it comes to RPerl?
21:58 willthechill that is a very good question, I do not currently know what is the libgmp minimum version required
22:14 mako93r joined #perl11
22:16 mako93r knowing about min version of libgmp could solve at least the 't/07' problem under OBSD
22:17 mako93r because the error msg looks like this
22:19 mako93r a version problem, you know
22:19 mako93r what version of 'libgmp' are you using under linux?
22:22 willthechill just a moment, I'll check
22:25 willthechill libgmp-dev v6.1.0
22:26 willthechill in Xubuntu v16.04.1
22:27 willthechill however it should also work with libgmp-dev v5.1.3, because RPerl passes all tests in Xubuntu v14.04 as well
22:27 willthechill so we can say libgmp minimum version require v5.1.3
22:27 willthechill *required
22:28 mako93r what about l
22:28 mako93r ...
22:29 mako93r libgmp 5.0.2p3?
22:30 willthechill I have given you all the information I possess
22:30 willthechill sorry but I can't answer any more questions, because that's all I know
22:30 willthechill obviously, you can test it out yourself and give a report
22:30 willthechill and that's the only option at this time
22:31 mako93r ok, no problem, it looks like I have to dig deeper
22:31 willthechill okay
22:32 mako93r OBSD is an exotic and unorthodox platform
22:33 mako93r so you can expect unnormal behaviour
22:33 mako93r but most tests were passing
22:34 willthechill well that's good
22:34 mako93r even 't/13' with 'Alien::astyle' ran good
22:34 willthechill sweet
22:35 mako93r had tp copy 'astyle' to a dir listed in $PATH manually, but not a big issue
22:35 willthechill well that's not right
22:36 willthechill in fact, that is specifically wrong
22:36 willthechill you should never have to manually copy anything
22:37 mako93r anyway, the build process was just fine, but yes u are right
22:37 mako93r I shouldn't have to do copying
22:38 mako93r but, the build process did not require any copying
22:41 willthechill well now I'm really confused, if the build process did not require any copying, then when and why did you have to manually copy astyle?
22:41 willthechill what error did you receive which made you think you had to manually copy astyle?
22:43 mako93r to make 'astyle' available/callable (in $PATH) after installing 'Alien::astyle'
22:43 mako93r binary available after the build process of 'Alian::astyle'
22:44 mako93r had to copy the binary of 'astyle' to a path listed in env var 'PATH'
22:46 mako93r it seems the build process doesn't need that
22:47 mako93r an '_astyle'-directory is created if you call 'perl Build.PL' under 'Alien-astyle' unpacked directory
22:48 willthechill okay yes that is correct behavior, Alien::astyle does NOT actually install an astyle binary into your path
22:48 willthechill so whatever you did was not part of the normal Alien::astyle usage and falls completely outside of this discussion
22:49 willthechill in other words, your copying of the astyle binary is not related to RPerl
22:49 mako93r no, it isn't
22:49 willthechill okay well I'm glad to get to the bottom of that
22:50 willthechill you must actually review the Alien::astyle usage docs if you want to use the Alien::astyle binary from inside a Perl program
22:50 willthechill https://metacpan.org/pod/Alien::astyle
22:50 willthechill use Alien::astyle;
22:50 willthechill use Env qw(@PATH);
22:50 willthechill unshift @PATH, Alien::astyle->bin_dir();  # `astyle` command is now in your path
22:52 mako93r an 'astyle' binary was created and I could copy that binary to a location where it could be called via '$PATH'
22:52 mako93r don't bother it worked fine
22:53 willthechill yes, again, what you did is fine but not related to RPerl
22:54 willthechill so not only is is "not a big issue", it is "not an issue at all"
22:54 willthechill which makes me happy, less bugs for me to fix
23:01 mako93r absolutly right, let's stay with that, everything is pretty cool right now
23:02 mako93r but you don't the minimal version of 'libgmp', do ya?
23:03 mako93r *don't know the ...
23:03 willthechill I have already answered that question
23:05 mako93r what the heck? oh, sorry, about that.
23:05 willthechill "<willthechill> so we can say libgmp minimum version require v5.1.3"
23:06 mako93r I see some of the beers I had are taking over my mind :-D
23:06 mako93r Yes, you mentioned it.
23:07 willthechill okay well perhaps you should not try to work on IRC if you have been drinking alcohol
23:07 mako93r you're right
23:08 mako93r but I thought as long as I am able to write in english I am sober enough
23:08 willthechill as long as you are able to make sense and remember when I have already given you an answer to a question
23:09 willthechill otherwise, no
23:11 mako93r I am, but anyway it would be wise to stop. In my timezone it's time for most people to go to sleep. And I'm feeling sleepy.
23:12 mako93r that's the bigger issue
23:20 mako93r cu

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