Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-03-20

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

All times shown according to UTC.

Time Nick Message
00:43 ether_ joined #metacpan
01:56 omega dolmen, oalders: I don't know if NEWS gets parsed for the "oarsed" changes or not to be honest :)
02:31 oiami joined #metacpan
02:48 klapperl_ joined #metacpan
03:07 rwstauner it should, it uses the api endpoint to get the text
03:09 omega yeah, I would assume so, but I don't know for sure either way :)
03:09 omega I still haven't done anything on the API side :(
03:12 talina_ joined #metacpan
03:13 rwstauner omega: it works: https://metacpan.org/release/Unicode-LineBreak
03:13 rwstauner wait, that was premature
03:13 dipsy [ Unicode-LineBreak-2013.11 - UAX #14 Unicode Line Breaking Algorithm - metacpan.org ]
03:13 rwstauner i need to refine my filter
03:14 rwstauner the hard part will be finding a dist that uses NEWS *and* has a format that can be parsed
03:15 omega thats some whack format on the Changes file though, haha
03:18 rwstauner :-)
04:14 rwstauner omega: https://metacpan.org/release/Net-FTP-RetrHandle
04:14 dipsy [ Net-FTP-RetrHandle-0.2 - Tied or IO::Handle-compatible interface to a file retrieved by FTP - metacpan.org ]
04:15 * rwstauner &
04:24 ether_ joined #metacpan
04:25 omega rwstauner: yeah :/
05:12 nbezzala joined #metacpan
06:27 [Sno] joined #metacpan
07:09 dpetrov_ joined #metacpan
08:31 dolmen joined #metacpan
08:35 neilb joined #metacpan
09:00 DerAlex joined #metacpan
09:20 neilb joined #metacpan
09:27 sjn joined #metacpan
09:57 neilb joined #metacpan
10:56 neilb joined #metacpan
11:42 metacpan joined #metacpan
11:42 metacpan [cpan-api] monken pushed 1 new commit to master: http://git.io/3T3uLA
11:42 metacpan cpan-api/master 67b190e Moritz Onken: reduce bulk size to reduce timeouts
11:42 metacpan left #metacpan
11:42 dipsy [ reduce bulk size to reduce timeouts · 67b190e · CPAN-API/cpan-api · GitHub ]
11:45 rashi joined #metacpan
11:53 metacpan joined #metacpan
11:53 metacpan [cpan-api] monken pushed 1 new commit to master: http://git.io/-tQU7A
11:53 metacpan cpan-api/master 397bf17 Moritz Onken: Revert "Merge pull request #305 from nadoshmado/gh-272"...
11:53 metacpan left #metacpan
11:53 dipsy [ Revert "Merge pull request #305 from nadoshmado/gh-272" · 397bf17 · CPAN-API/cpan-api · GitHub ]
12:34 oiami joined #metacpan
13:47 rwstauner Is it a bootstrap thing to use one li as the nav-header for following sibling li's ? i'd much rather have a nested ul for that
13:48 mo rwstauner: check the bootstrap docs :)
13:48 rwstauner i'm looking, but haven't found any mention of the header yet
13:50 mo it is in the examples somewhere, iirc
13:50 mo also have a look at the 2.x docs
13:50 rwstauner ok
13:50 rwstauner i just went there and it said "try the latest!"
13:52 rwstauner ok, i did find a crappy example of the way we have it
13:53 rwstauner but then right below that it does say  " For nesting within a nav list, include class="nav nav-list" on any nested <ul>. "
13:53 rwstauner using css instead of semantic html seems like a step backwards
13:55 oalders upgrading to bootstrap 3 could be part of a summer project for a student
13:55 rwstauner ooh
13:55 oalders now i've got you thinking...
14:01 rwstauner on the new SCO links we're linking to AUTHOR/DIST... should we be using author-independent links like /dist/DIST like we do on mc?
14:02 rwstauner i was about to change it but thought i'd ask first
14:02 oalders ribasushi: ^^^
14:03 rwstauner for example, if on mc i browse to doy's last release of Moose, then follow the sco link that says "Latest" i end up at the last time doy released moose
14:03 rwstauner which is many versions behind
14:04 ranguard /dist/ links ++
14:04 rwstauner done.
14:04 oalders ah, i see
14:04 oalders latest should go to the proper latest
14:05 metacpan joined #metacpan
14:05 metacpan [metacpan-web] rwstauner pushed 2 new commits to master: http://git.io/F3jUgQ
14:05 metacpan metacpan-web/master 10ad5c0 Randy Stauner: Restrict xpath test so we get only the link we want...
14:05 metacpan metacpan-web/master a4e6a7f Randy Stauner: Use /dist/ links for SCO to get author-independent latest
14:05 metacpan left #metacpan
14:05 dipsy [ Comparing be416bd35647...a4e6a7f19b8b · CPAN-API/metacpan-web · GitHub ]
14:14 neilb A question related to how MetaCPAN indexes new releases:
14:15 neilb I know that MetaCPAN is mirroring CPAN from either the master or a tier 1, and that's how it sees releases so quickly. Do you use CPAN::DistnameInfo to decide whether a *release* is developer and thus not index all packages in the release, or do you wait for 02packages to catch up and base your decision on that?
14:17 rwstauner I believe CPAN::DistnameInfo contributes
14:17 rwstauner the version of a release is a factor
14:17 neilb Specific example: rjbs has discovered that if PAUSE sees a non-developer (ie stable) release, but one of the packages in it has a developer release number (eg 1.01_01), then that package won't make it into 02packages, but the rest will.
14:17 rwstauner 02packages will tell us when a dist should be "latest"
14:17 rwstauner interesting
14:18 neilb So I then wondered which side of things MetaCPAN falls on
14:18 rwstauner if PAUSE doesn't put it in 02packages, metacpan won't see it as "latest"
14:18 rwstauner whether or not it will be marked as "Dev" i'm not sure
14:18 neilb So if you searched on that package, you'd see the previous release, you reckon?
14:20 rwstauner yes, it would not be considered for search
14:35 metacpan joined #metacpan
14:35 metacpan [metacpan-web] rwstauner pushed 2 new commits to master: http://git.io/nH_nKQ
14:35 metacpan metacpan-web/master 0adbaa4 Randy Stauner: Make "This version" test more specific for /pod, too...
14:35 metacpan metacpan-web/master 6bf5456 Randy Stauner: Test the location header on the uc-author redirect
14:35 metacpan left #metacpan
14:35 dipsy [ Comparing a4e6a7f19b8b...6bf54562bc2a · CPAN-API/metacpan-web · GitHub ]
14:39 rwstauner aw, just realized the release page doesn't show changes summary on some of my -TRIAL dists :-/
14:42 metacpan joined #metacpan
14:42 metacpan [metacpan-web] rwstauner pushed 1 new commit to master: http://git.io/n1zIZg
14:42 metacpan metacpan-web/master e174e9e Randy Stauner: Move recent change log parsing to release controller...
14:42 metacpan left #metacpan
14:42 dipsy [ Move recent change log parsing to release controller · e174e9e · CPAN-API/metacpan-web · GitHub ]
14:44 * rwstauner tries git grep contributor, gets nothing, then finds that the file is named contributor... GIT GREP Y U NO DO WHAT I MEAN?
14:53 oiami joined #metacpan
14:56 oalders rwstauner++ # fixing broken build
15:05 oiami left #metacpan
15:06 oiami joined #metacpan
15:15 oiami left #metacpan
15:39 oalders http://explorer.metacpan.org/
15:39 oalders mo++
15:48 ranguard mo++
15:51 priodev joined #metacpan
17:05 trs rwstauner: metacpan doesn't actually use 02packages.
17:05 trs but it should.
17:06 rwstauner i understand that to be the definition of "latest"
17:09 rwstauner trs: https://github.com/CPAN-API/cpan-api/blob/master/lib/MetaCPAN/Script/Latest.pm#L34
17:09 dipsy [ cpan-api/lib/MetaCPAN/Script/Latest.pm at master · CPAN-API/cpan-api · GitHub ]
17:09 trs rwstauner: yes, was just grepping.
17:09 trs obviously my state of the world was wrong. :)
17:09 rwstauner the "latest" script is run, like hourly or so
17:09 rwstauner separate from the indexer
17:09 trs I recalled previous greps for 02packages turning up nil, but I don't know what I was smoking.
17:09 rwstauner :-)
17:10 trs anyway, \o/
17:10 punter joined #metacpan
17:24 neilb joined #metacpan
17:46 neilb rwstauner: following up on our chat earlier, I'm guessing that 02packages is loaded up less often than you look for new releases?
17:47 neilb If you look in 02packages for Acme::Playpen right now, you'll see two lines:
17:47 rwstauner right, releases are indexed within minutes
17:47 neilb Acme::Playpen                      0.14  N/NE/NEILB/Acme-Playpen-0.14.tar.gz
17:47 neilb Acme::Playpen::Utilities           0.12  N/NE/NEILB/Acme-Playpen-0.12.tar.gz
17:47 rwstauner i think latest happens every hour or so
17:47 rwstauner hmm
17:47 rwstauner ok
17:47 neilb But if you look at Acme::Playpen::Utilities in MetaCPAN, you'll get the dev version of it which isn't listed in 02packages
17:48 neilb The last non dev version of the module was in 0.12, but MC is showing the version that was in 0.14. So will MetaCPAN "catch up" when the next load of 02packages happens?
17:56 oalders http://explorer.metacpan.org/#/1e0ef0b311d3e711fd2a
17:56 trs neilb: metacpan only cares about dev/nondev releases, not modules, afaik
18:04 neilb yeah, I suspected that was the case, hence my question earlier...
18:42 [Sno] joined #metacpan
23:05 ether the 'issues' link on this new release says there are 3 issues, but there are no github issues at all.. https://metacpan.org/pod/release/SUKRIA/File-Scan-0.001/lib/File/Scan.pm
23:05 dipsy [ File::Scan - file parser intended for big files that doesn't fit into main memory. - metacpan.org ]
23:47 alnewkirk joined #metacpan
23:55 nbezzala joined #metacpan
23:57 trs ether: it's pulling from rt.cpan.org - http://explorer.metacpan.org/#/5a6232016a25e9f2ae01
23:58 trs and showing github info on hover is likely bad
23:58 trs s/bad/bug/
23:59 trs weird.

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