Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2015-01-07

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

All times shown according to UTC.

Time Nick Message
00:35 vroom joined #metacpan
00:37 punter joined #metacpan
00:42 vroom joined #metacpan
00:59 mrhorner joined #metacpan
01:40 vroom joined #metacpan
02:06 vroom joined #metacpan
02:13 vroom joined #metacpan
02:42 klapperl_ joined #metacpan
03:48 vroom1 joined #metacpan
03:57 vroom joined #metacpan
08:17 dp_ joined #metacpan
09:34 neilb joined #metacpan
09:43 castaway joined #metacpan
10:17 castaway joined #metacpan
11:39 alnewkirk joined #metacpan
12:03 dp_ joined #metacpan
12:26 dp_ joined #metacpan
12:36 vroom joined #metacpan
13:00 andreeap joined #metacpan
13:52 metacpan joined #metacpan
13:52 metacpan [metacpan-web] oalders pushed 2 new commits to master: http://git.io/5nvX1w
13:52 metacpan metacpan-web/master 1e8902f Andreea Pirvulescu: Update CONTRIBUTING.md
13:52 metacpan metacpan-web/master a4c8101 Olaf Alders: Merge pull request #1434 from andreeap/master...
13:52 metacpan left #metacpan
13:52 dipsy [ Comparing 72eb0ab4f8da...a4c810117e4c · CPAN-API/metacpan-web · GitHub ]
15:06 rwstauner riveter: you've got that backwards. the model in the lib sets the default to 'cpan' (b/c there's a separate job on the server that sets the 'latest' flag)
15:07 rwstauner however, the test model defaults to 'latest' because for most of the fake test dists,  the release will be marked as latest
15:07 rwstauner unless there is actually more than one version, and you are testing the older one
17:51 andreeap yesterday, I got this error  https://gist.github.com/andreeap/4b16c457b97d5e8c64f5 while indexing a release the file that gives the error it's a symlink and this is why it doesn't work in the first place. If I skip it, works like a charm. What should I do?
17:51 dipsy [ gist:4b16c457b97d5e8c64f5 ]
17:52 oalders rwstauner: any reason we should care about symlinks?
17:54 rwstauner i wonder what PAUSE does with them
17:54 neilb_ joined #metacpan
17:55 oalders no idea
17:57 oalders andreeap: you could ask that question in #toolchain
17:57 oalders "what does PAUSE do with symlinks in an upload?"
18:00 rwstauner the problem isn't that it's a symlink... it's that it's a broken symlink
18:00 rwstauner the call to stat returns undef
18:00 rwstauner we should check that and skip to the next file
18:00 oalders works for me
18:00 oalders andreeap ^^
18:02 rafl or perhaps you wanna use lstat rather than stat?
18:15 andreeap if it's undef, then I will skip it
18:19 trs for indexing, using lstat might make more sense.
18:24 trs my $s = $child->stat || $child->lstat; preserves the existing behaviour of looking at a symlink's target, but falls back to the symlink itself if the target is missing.
18:24 trs not sure if that's more useful than skipping, though it likely only matters in edge cases.
18:28 andreeap that will give the same output as just having my $s = $child->stat, or am I missing something else here?
19:13 oalders andreeap: right, but it will cover more cases
20:24 ether hi, could the MooseX-App-Cmd dist please be reindexed? I've just reindexed it on PAUSE after getting some perm issues fixed.
20:24 ether thank you!
20:35 oalders ether: done
20:35 ether oalders++
20:57 mrhorner joined #metacpan
22:40 vroom joined #metacpan
22:43 vroom joined #metacpan
23:24 vroom joined #metacpan
23:57 vroom joined #metacpan

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