Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-08-08

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

All times shown according to UTC.

Time Nick Message
00:16 jwang joined #metacpan
01:32 klapperl_ joined #metacpan
01:34 FROGGS__ joined #metacpan
01:53 ether rwstauner++ # issue archaeology
02:34 talina_ joined #metacpan
03:32 jnbek joined #metacpan
03:37 oiami joined #metacpan
07:01 rashi joined #metacpan
07:04 neilb joined #metacpan
07:17 [Sno] joined #metacpan
07:54 neilb joined #metacpan
08:07 talina_ joined #metacpan
08:27 neilb joined #metacpan
10:28 n0body joined #metacpan
10:45 n0body joined #metacpan
11:18 n0body joined #metacpan
11:28 oiami1 joined #metacpan
13:07 oiami joined #metacpan
13:21 n0body joined #metacpan
14:51 [Sno] joined #metacpan
16:51 neilb joined #metacpan
17:20 dolmen left #metacpan
18:48 neilb_ joined #metacpan
19:01 Mithaldu how can i get dev releases i made reindexed after i got COMAINT for them?
19:03 haarg dev release?  indexed?
19:06 mst I've just beat on the PPI perms
19:07 mst Mithaldu's devrel is showing UNAUTHORIZED on metacpan
19:07 mst since PAUSE doesn't index devrels, this must be a metacpan thing
19:07 mst so the question is: how do we get it to retry?
19:07 Mithaldu or more simply put, can i push a button to make the red go away here? https://metacpan.org/release/MITHALDU/PPI-1.217_01
19:07 dipsy [ PPI-1.217_01 - Parse, Analyze and Manipulate Perl (without perl) - metacpan.org ]
19:08 Mithaldu or do i simply need to rerelease?
19:08 Mithaldu or wait?
19:20 trs first pause needs to rewrite 06perms
19:21 trs then someone with server access can kick the indexer to retry
19:21 trs either pause hasn't rewritten 06perms yet or mst missed PPI::Util
19:22 mst GODDAMNIT
19:22 mst also, I thought we were going to have the indexer auto-retry when people got new perms?
19:23 mst Added MITHALDU to co-maintainers of PPI::Util.
19:24 * trs waits for pause
19:24 Mithaldu https://metacpan.org/release/MITHALDU/PPI-1.217_01 looks plenty blue now :)
19:24 dipsy [ PPI-1.217_01 - Parse, Analyze and Manipulate Perl (without perl) - metacpan.org ]
19:27 trs mst: I thought we'd only talked about auto reindexing when we noticed a pause reindex in 02packages.
19:27 trs Mithaldu: weird... I ran the indexer, but it said "release PPI-1.217_01 contains unauthorized modules: PPI::Util" which I assumed meant the dist remained unauthorized.
19:28 Mithaldu trs: maybe it is
19:28 trs because this is still missing you: curl -s cpan.metacpan.org/modules/06perms.txt.gz | zgrep PPI::Util
19:28 Mithaldu the provided modules don't seem to get marked with authorization
19:28 Mithaldu and the module name is still red
19:29 trs module name in red is because it's a developer release.
19:29 Mithaldu ok
19:29 mst trs: yeah, I just added it explicitly
19:29 trs "authorized" : false, from the API though
19:30 mst trs: as in, when I pasted that line 6 minutes ago was when I added that one
19:30 mst I did miss it the first time around
19:30 trs mst: right, I know, but pause hasn't rewritten 06perms yet.
19:31 trs s/because // might make my statement more sensical.
19:33 trs wtf indexer: http://api.metacpan.org/file/MITHALDU/PPI-1.217_01/lib/PPI/Util.pm?fields=authorized
19:33 dipsy urgh. long url. Try http://tinyurl.com/k68jtz6
19:34 Mithaldu joy :D
19:34 trs pause just rewrite, I reran the indexer for good measure ("consistency!")
19:35 trs I have no idea why the indexer reported "unauthorized modules: PPI::Util", but then left it in an authorized state.
19:35 trs (before 06perms was updated)
19:35 Mithaldu well according to mst's outburst it was not authorized before
19:35 Mithaldu the rest is all race conditions
19:36 trs it wasn't authorized, I reran the indexer the first time, it reported unauthorized, but then left the file as authorized in ES
19:36 trs I don't see the race, but *waves hands*
19:37 Mithaldu i'm not really following well either, this allergy stuff is hitting kind of hard
19:42 trs basically, the file was considered authorized, but the package inside the file wasn't.
19:43 trs which it turns out is because the file is only marked unauthorized if there is a =head1 NAME tying it to a specific package within the file and that package is unauthorized.
19:43 Mithaldu haha
19:43 trs I'm sad I looked.
19:43 trs and now I'm going back to work.
19:43 Mithaldu so it's not actually a bug that undocumented files are not marked
19:43 trs I guess
19:44 trs the file's authorization status is tied to the "primary" package (module in metacpan's terms) of the file.
19:44 trs unlike a release, the authorization of which is tied to there being no unauthorized packages (modules)
19:45 trs I'm not sure why exactly a file can be authorized but contain unauthorized packages.
20:45 ether is the unauthorized package not considered for indexing?
20:46 ether IMO, metacpan shouldn't bother marking 'unauthorized' on dev releases anyway
20:46 ether the same thing happened here - https://metacpan.org/release/ETHER/Dist-Zilla-Plugin-ReadmeAnyFromPod-0.142180-TRIAL
20:46 dipsy [ Dist-Zilla-Plugin-ReadmeAnyFromPod-0.142180-TRIAL - Automatically convert POD to a README in any format for Dist::Zilla - metacpan.org ]
20:47 ether this isn't my dist; I created a release for testing purposes. I marked it -TRIAL because I wanted to signal that I never intended it to be considered for indexing
20:47 ether (knowing it would fail anyway)

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