Perl 6 - the future is here, just unevenly distributed

IRC log for #pr-challenge, 2016-02-18

| Channels | #pr-challenge index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 neilb_ joined #pr-challenge
01:38 veryrusty joined #pr-challenge
02:09 itcharlie joined #pr-challenge
02:37 vroom joined #pr-challenge
02:42 vroom joined #pr-challenge
02:48 ilbot2 joined #pr-challenge
02:48 Topic for #pr-challenge is now for more info see http://cpan-prc.org | public logs for this channel: http://irclog.perlgeek.de/pr-challenge/today
03:16 rvandam joined #pr-challenge
03:38 vroom joined #pr-challenge
03:58 vroom joined #pr-challenge
04:40 itcharlie left #pr-challenge
08:28 veryrusty joined #pr-challenge
08:29 neilb joined #pr-challenge
09:22 neilb joined #pr-challenge
10:23 veryrusty joined #pr-challenge
10:44 veryrusty joined #pr-challenge
11:53 G_SabinoMullane joined #pr-challenge
13:47 gansh left #pr-challenge
13:56 vroom joined #pr-challenge
15:55 itcharlie joined #pr-challenge
16:02 thrig joined #pr-challenge
16:10 vroom joined #pr-challenge
16:30 gryphon joined #pr-challenge
17:29 deven joined #pr-challenge
17:37 rvandam joined #pr-challenge
17:44 ether csson: App::Software::License is failing a tests now on a lot of smokers, after the Moo conversion. any ideas?
17:44 ether it looks like a problem with MooX::Options
18:06 csson ether: My first guess (undef from Config) was already taken care of.. Some of the tests doesn't seem to have any Config modules installed (http://www.cpantesters.org/cpan/report/d94b9d16-d5b0-11e5-aa20-495baef69d38), but that is not the case for all of them..
18:07 ether that shouldn't be a factor since these tests aren't reading from a config file, only from the command line, I think?
18:07 ether I added the recording of Config::* versions in case that might be a factor in another issue I was seeing
18:08 ether but the one that happens most often is dying with "Holder is missing", which means that MooX::Options isn't parsing @ARGV
18:10 csson There are a couple of tests that use a config file, and those on those tests 'holder' is set in the config file
18:11 csson Maybe I did something platform un-agnostic on those tests
18:16 ether not in the test that is failing though
18:16 ether t/options.t doesn't use a config file
18:18 csson yes: https://metacpan.org/source/ETHER/App-Software-License-0.06-TRIAL/t/options.t#L58
18:19 ether oh hey, you're right, sorry :)
18:22 ether from the error message "..just after 5", that is indeed the test that is dying.
18:22 ether (confirmed by adding a die() to that test_opts call)
18:23 ether what Config:: module is required to read the data in t/etc/software_license.conf ?
18:24 ether it looks like INI format to me
18:24 csson Config::General i guess
18:25 ether hmm I moved Config::INI aside and the test still passed. let's try Config::General
18:25 ether I don't have that installed.
18:26 csson I don't either, maybe it was INI after all..
18:26 ether oh ho ho, I moved Config::Tiny aside and now I get:
18:26 ether Can't use string ("holder = A.Holder") as a HASH ref while "strict refs" in use at lib/App/Software/License.pm line 171.
18:26 ether which is another failure I saw
18:27 ether I was suspecting that some config modules returned hashes rather than hashrefs
18:27 ether but it looks like it just failed to parse the string entirely.
18:29 ether if Config::General is installed, it erases the problem caused by a missing Config::Tiny.
18:29 ether but I still can't get "holder is missing"
18:29 ether hmm
18:30 ether so complicated! :)
18:30 vroom joined #pr-challenge
18:32 csson A test prereq on Config::General or ::Tiny would solve it?
18:35 ether no
18:35 ether removing those modules doesn't cause the error
18:36 ether I'm going to move aside my entire Config/ directory and just reinstall Config::Any on its own
18:37 ether doing that gives me:   Can't use string ("holder = A.Holder") as a HASH ref while "strict refs" in use at lib/App/Software/License.pm line 171.
18:37 ether so there is definitely a problem with Config::Any on its own, with the format used in that test file, at least
18:38 ether but not the same error as on cpantesters
18:39 ether switching the test file to using json makes the test pass again.
18:39 ether I can release that change as another -TRIAL to see if it makes things better..
18:39 ether strange that I don't see the same error message though
18:42 csson it's not the same error message as when one runs software-license without holder either
18:43 csson but hopefully the config format change works
18:44 ether huh! I just got 'holder is missing' -- while trying to do the release even :)
18:44 ether I broke something in my PERL5LIB Config/
18:46 ether now it's parsing the config file to:   '{"holder":"A.Holder"}' => undef,
18:46 ether well isn't that nice.
18:47 ether it can't parse a yaml-formatted config file either
18:47 ether Config::Any is dumb!
18:51 ether .conf is parsed via Config::Any::General
18:51 ether which tries to load Config::General and then has no good fallback
18:51 ether but, peculiarly -- I did not have Config::General installed before
18:51 ether I'm going to find the (passing) cpantesters report that came from my own machine to confirm that
18:53 ether ah, apparently I wasn't able to submit any, because libintl doesn't install anymore on my smoker. *grr*
18:53 * ether asks analysis.cpantesters.org for correlations
18:57 ether well, all the passing reports I've found so far have Config::General installed
18:58 ether so this is a reasonable working theory...
19:01 ether this is worhty of a trial
19:01 thrig Jury of connection reset by peer
19:01 ether but, one thing should be fixed before going stable... in the Moose version, config files of any format were supported. now it has to be a .conf
19:01 preaction to the arena!
19:02 ether MooseX::SimpleConfig uses Config::Any with use_ext=>1 so we should do the same.
19:03 * ether will look at that tonight. for now I must return to the salt mines
19:03 csson yes, absolutely on use_ext => 1, dont know why i didn't
19:10 ether you wouldn't be able to do a -e 'software-license.conf' if we don't know the extension
19:10 ether but there's globs for that
19:55 neilb joined #pr-challenge
20:05 neilb joined #pr-challenge
20:08 veryrusty joined #pr-challenge
20:15 itcharlie anyone here knows how to integrate https://coveralls.io/  on a Perl project in Github?
20:16 rvandam joined #pr-challenge
20:18 preaction itcharlie: http://blogs.perl.org/users/graham_knop/2014/10/travis-ci-perl-helpers.html
20:20 itcharlie thanks preaction
20:39 itcharlie ah so basically setup coverage to run via travis.yml and we are all set to go
20:41 veryrusty joined #pr-challenge
20:44 neilb_ joined #pr-challenge
21:25 veryrusty joined #pr-challenge
21:59 itcharlie left #pr-challenge
22:38 veryrusty joined #pr-challenge
22:57 rvandam joined #pr-challenge
23:25 G_SabinoMullane joined #pr-challenge

| Channels | #pr-challenge index | Today | | Search | Google Search | Plain-Text | summary