Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-10-10

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

All times shown according to UTC.

Time Nick Message
00:00 vroom joined #metacpan
00:12 vroom joined #metacpan
00:39 vroom joined #metacpan
01:27 vroom joined #metacpan
01:30 klapperl_ joined #metacpan
01:40 bvierra joined #metacpan
01:42 vroom joined #metacpan
02:39 andreea joined #metacpan
02:49 vroom joined #metacpan
04:26 zostay joined #metacpan
04:28 zostay joined #metacpan
04:33 metacpan joined #metacpan
04:33 metacpan [metacpan-developer] oalders pushed 1 new commit to master: http://git.io/19inYA
04:33 metacpan metacpan-developer/master c042c67 Olaf Alders: Create metacpan_server_local.conf if it doesn't already exist.
04:33 metacpan left #metacpan
04:37 dipsy [ Create metacpan_server_local.conf if it doesn't already exist. · c042c67 · CPAN-API/metacpan-developer · GitHub ]
04:43 metacpan joined #metacpan
04:43 metacpan [metacpan-developer] oalders pushed 1 new commit to master: http://git.io/P1DU4w
04:43 metacpan metacpan-developer/master 2264eea Olaf Alders: Add note about setting the path to CPAN in API config.
04:43 metacpan left #metacpan
04:46 metacpan joined #metacpan
04:46 metacpan [cpan-api] oalders created oalders/cpan (+3 new commits): http://git.io/rgjwuw
04:46 metacpan cpan-api/oalders/cpan f71e840 Olaf Alders: Add Data::Printer to cpanfile for easier debugging.
04:46 metacpan cpan-api/oalders/cpan 0d93df7 Olaf Alders: Minor lib cleanup.
04:46 metacpan cpan-api/oalders/cpan aec0f74 Olaf Alders: Remove default path to fakepan for CPAN....
04:46 metacpan left #metacpan
04:46 dipsy [ Add note about setting the path to CPAN in API config. · 2264eea · CPAN-API/metacpan-developer · GitHub ]
04:49 metacpan joined #metacpan
04:49 metacpan [cpan-api] oalders force-pushed oalders/cpan from aec0f74 to 5c429de: http://git.io/hi_MVA
04:49 metacpan cpan-api/oalders/cpan 5c429de Olaf Alders: Remove default path to fakepan for CPAN....
04:49 metacpan left #metacpan
04:49 dipsy [ Comparing f71e840e7019^...aec0f74dce08 · CPAN-API/cpan-api · GitHub ]
04:51 dipsy [ Commits · CPAN-API/cpan-api · GitHub ]
06:53 dp_ joined #metacpan
07:01 kentnl Is the "Provides" listing on the top level "release" authoritative?  ie: Can one assume all things there are indexed? Or do I have to find some other way ?
07:05 * kentnl was hoping there was some easier way to resolve module -> package deps for ranges than the horrible monstrosity that cpanm and similar have to use
07:05 kentnl I've done it that way before, but every time I re-visit it, I go "huh, wtf is this even doing"
07:40 haarg kentnl: haven't you done things extending CPAN::Changes?
07:40 haarg i'm currently looking at rebuilding it from scratch, but want to keep existing uses in mind
07:57 kentnl haarg: yeah, there's a few changes I supplied and had patches merged for for allowing subclassing of parts.
07:58 haarg what were you doing with the subclasses
07:58 haarg ?
07:58 kentnl Though I'm not sure I'm completely in favour of doing things entirely my way, my way optimised for being lazy, but that proves to be counter-productive if you want to have segements generated from data and then cached
07:59 kentnl https://metacpan.org/pod/CPAN::Changes::Group::Dependencies::Details
08:00 kentnl =~ slowly been teasing out all the guts of my Changes generator
08:00 dipsy [ CPAN::Changes::Group::Dependencies::Details - Full details of dependency changes. - metacpan.org ]
08:00 neilb joined #metacpan
08:01 haarg my general plan is to deprecate ::Group and instead have ::Release->entries which gives an arrayref of ::Entry objects.  and those have ->entries which gives an arrayref on ::Entry objects, etc.
08:04 kentnl oh, so destructure it so it can have arbitrary nesting of groups instead of simple single-level groups?
08:04 haarg correct
08:06 kentnl So I think for cross-compat the thing to look for there would be the reasonably recent ->attach() method, In *theory* you could make groups emulate entries in some way, but I haven't thought about it too deeply
08:06 kentnl Though keep me appraised and I'll of course endeavour to change any of my things as need be to make it work, etc.
08:07 haarg for the most part, groups are going to be dispatching to the appropriate entry object
08:09 haarg https://github.com/haarg/CPAN-Changelog this has the rough structure of how i want it to work, although it doesn't have the mutating methods yet.
08:09 dipsy [ haarg/CPAN-Changelog · GitHub ]
08:09 haarg but no compatibility yet
08:13 * kentnl doesn't quite understand how you're serializing it at this time
08:13 kentnl But I just wanted to say I would grealty like to side step 2 problems presently imposed by the Text::Wrap usage
08:13 kentnl 1. Changing the wrap length is not easy
08:14 kentnl 2. Text::Wrap has bad defaults and breaks on chr(0xA0), non-breaking-space, and also snaps long tokens like URLs into multiple lines
08:14 kentnl https://metacpan.org/source/KENTNL/Dist-Zilla-PluginBundle-Author-KENTNL-2.021000/utils/dep_changes.pl#L314 # Hence why I have these 3 lines.
08:15 haarg the serializing is a bit half baked.  i'm wrapping manually right now.
08:15 dipsy [ utils/dep_changes.pl - metacpan.org ]
08:15 haarg it doesn't cut long tokens, but i don't remember how i'm handling whitespace
08:15 kentnl You appear to have lots of \s . \s matches 0xA0 :/
08:15 haarg easy to change though
08:30 neilb joined #metacpan
09:09 daxim joined #metacpan
10:17 rashi joined #metacpan
11:02 rashi joined #metacpan
11:37 vroom joined #metacpan
11:45 vroom joined #metacpan
11:51 vroom joined #metacpan
11:56 vroom joined #metacpan
12:12 rashi joined #metacpan
12:35 vroom joined #metacpan
13:22 rashi joined #metacpan
13:49 vroom joined #metacpan
13:53 tokuhirom joined #metacpan
14:26 haarg kentnl: could you test out any of your extensions with https://github.com/haarg/CPAN-Changelog
14:26 dipsy [ haarg/CPAN-Changelog · GitHub ]
14:27 haarg (repo name is wrong-ish, modules are still CPAN::Changes)
14:31 haarg the thing in CPAN-Changelog is a rewrite from scratch.  then i added a backcompat layer that works well enough for almost all of the CPAN::Changes tests to pass.
14:32 haarg i adjusted the code in the main CPAN::Changes slightly to match the serialization from my module, which i think is more sensible
14:32 kentnl haarg: will do =). Thanks.
14:33 haarg and after that the only failures were from two features i plan on removing with no replacement
14:35 rwstauner ++ # code removal
14:35 dipsy Thanks!
14:37 haarg given that the new concept is unlimited nesting, with "entries" rather than "groups" and "changes", a method called "delete_empty_groups" doesn't make any sense
14:40 haarg rwstauner: not so much "removal" as "not re-implementing".  the back-compat code is already longer than the code for the new api.
14:40 rwstauner :-)
14:59 vroom left #metacpan
15:00 vroom joined #metacpan
16:42 neilb joined #metacpan
16:46 neilb joined #metacpan
16:53 dp_ joined #metacpan
19:04 bvierra joined #metacpan
19:07 bvierra_ joined #metacpan
19:27 bvierra joined #metacpan
19:38 bvierra joined #metacpan
20:00 vroom joined #metacpan
20:22 FROGGS joined #metacpan

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