Camelia, the Perl 6 bug

IRC log for #metacpan, 2011-10-03

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

All times shown according to UTC.

Time Nick Message
00:05 castaway joined #metacpan
01:10 alh joined #metacpan
01:25 klapperl joined #metacpan
01:28 Khisanth joined #metacpan
01:50 oalders autarch: ah, ok. now i understand :)
01:50 autarch oalders: it seemed like we were kind of talking past each other
01:51 autarch I think the indexer and/or the display code needs to be able to recognize pod in strings (and ignore it)
01:51 oalders autarch: it's mostly me not being detail oriented :)
01:52 oalders yeah, there are some display issues here and there. i suppose that's likely to be happening with other modules as well
01:52 autarch I think it's likely
01:53 autarch but I suspect the indexer may _also_ be picking it up, which is a separate problem
01:53 autarch because the indexer thinks that parse_olson in DT-TZ 1.38 is associated with the DT::TZ::Catalog package
04:14 autarch left #metacpan
06:18 gabimuc|work joined #metacpan
06:50 clintongormley joined #metacpan
06:59 woldrich woah. I can no longer click on the author's name here https://metacpan.org/source​/BESSARABV/Mock-Person-0.01 , because 'BESSARABV / Mock-Person-0.01 ' is one link only. Am I allowed to fix it?
06:59 dipsy [ Browse BESSARABV/Mock-Person-0.01/ - metacpan.org ]
07:01 doy should probably make https://metacpan.org/source/BESSARABV/ go somewhere intelligent first
07:01 doy and then have each path part be a link to go up to that particular level in the directory
07:01 doy (i don't think that the author's name should link to the author page from there)
07:19 hanekomu joined #metacpan
07:59 woldrich why not? that's what it did when I was younger
07:59 doy no it didn't
08:00 doy until a day or two ago, that page was just a raw Plack::App::Directory
08:03 mo woldrich: there is a link "to author page" on the left
08:27 preflex_ joined #metacpan
08:44 daxim joined #metacpan
08:53 clintongormley Ranguard: how do you handle log rotation with plack/starman - there doesn't seem to be any automated way of doing it
08:53 clintongormley and do you use Server::Starter with starman?
08:57 mo clintongormley: have you had a look at /etc/init.d/metacpan-web?
08:57 mo it doesn't solve the logrotate though :)
08:57 avar where is /etc/init.d/metacpan-web publicly available?
08:57 * mo looks
08:57 Ranguard clintongormley: I'll ask my sysadmin :)
08:57 clintongormley ta
08:59 mo avar: https://gist.github.com/1258738
08:59 dipsy [ monken's gist: 1258738 ? Gist ]
09:00 mo meh something's missing
09:07 mo avar: http://pastie.org/2631056
09:07 dipsy [ #2631056 - Pastie ]
09:08 mo it's an opensuse start script
09:18 metacpan joined #metacpan
09:18 metacpan [metacpan-web] monken pushed 1 new commit to master: http://git.io/_aQ2Pw
09:18 metacpan [metacpan-web/master] fix table sorter for columns without sort attribute - Moritz Onken
09:18 metacpan left #metacpan
09:18 dipsy [ Commit e11e2b936510aab39c73768091434f38969dc54d to CPAN-API/metacpan-web - GitHub ]
09:22 jsut joined #metacpan
09:31 avar mo: people here are trying to solve the logging problem with starman, I think they're considering just using syslog for it.
09:32 mo metacpan uses the logrotate daemon
09:35 Ranguard clintongormley: yea, we use logrotate daemon as well: https://gist.github.com/5fd3c323a6bf2262945d
09:35 dipsy [ ranguard's gist: 5fd3c323a6bf2262945d ? Gist ]
09:36 clintongormley oh, then you just restart the app
09:36 clintongormley ok - that seems overkill
09:36 clintongormley i'd rather just send a signal to the app to make it reopen the log files
09:36 mo is that actually required?
09:37 Ranguard yea, we have 2 servers behind our proxy, and they rotate at slightly different times
09:37 clintongormley ok
09:37 Ranguard https://github.com/miyagawa/Starman/issues/34
09:37 dipsy [ #34: HUP doesn't restore the arguments properly because of the use of $0 - Issues - miyagawa/Starman - GitHub ]
09:37 Ranguard that's why we restart instead of reload/HUP
09:40 clintongormley ah ta
09:45 clintongormley Ranguard: although it'd be easy to pass the logfile name to the AccessLog middleware, and use some signal handler to just reopen the file handle in all the children
09:52 * Ranguard likes using wheels that have already been invented :)
09:53 Mithaldu joined #metacpan
10:00 Mithaldu joined #metacpan
11:53 virtualsue joined #metacpan
12:52 avar mo: Why use Starman instead of $some_webserver + fastcgi ?
12:52 avar (just curious)
12:54 virtualsue joined #metacpan
12:55 Ranguard avar: not done the comparisons, but Starman is VERY fast for my uses at work, we thought we'd try it and then if there was a problem look at alternatives (this was after Apache+MP1->Apache+MP2 migration slowed our servers to a crawl)
12:56 rafl some like simple http proxying for easier debugging
12:56 Ranguard it's just worked[tm] (other than this HUB issue)
13:03 mst HUBBA HUBBA?
13:05 avar Ranguard: In my case the speed of the httpd isn't going to be the bottleneck, so having support for e.g. by-date logging might be a bigger factor in deciding.
13:05 avar But I think we're just trying logging via syslog
13:11 Ranguard avar: We log through Plack::Middleware::AccessLog - so doesn't matter what we deploy under
13:16 avar how about for error logging?
13:20 Ranguard Catalyst::Plugin::Log::Dispatch
13:21 Ranguard obviously we use Catalyst :)
13:37 mike So, the release appears (https://metacpan.org/release/DO​HERTY/WWW-3172-Crawler-0.001/) but it isn't listed on my author page (https://metacpan.org/author/DOHERTY)...
13:37 dipsy [ WWW-3172-Crawler-0.001 - metacpan.org ] [ Mike Doherty (DOHERTY) - metacpan.org ]
13:43 rwstauner joined #metacpan
13:43 mo mike: give it some time to be indexed as latest
14:00 rwstauner joined #metacpan
15:04 daxim joined #metacpan
15:17 virtualsue joined #metacpan
16:00 clintongormley joined #metacpan
16:29 virtualsue joined #metacpan
17:20 virtualsue joined #metacpan
17:21 theory joined #metacpan
17:31 rwstaune1_ joined #metacpan
17:34 frew_ joined #metacpan
17:41 frew joined #metacpan
17:56 ag4ve_ joined #metacpan
18:12 castaway joined #metacpan
18:31 Seveas joined #metacpan
18:58 am0c joined #metacpan
19:19 jsut_ joined #metacpan
19:22 clintongormley1 joined #metacpan
19:25 metacpan joined #metacpan
19:25 metacpan [metacpan-web] rwstauner pushed 5 new commits to rwstauner/release-info: http://git.io/eh6lVA
19:25 metacpan [metacpan-web/rwstauner/release-info] Add test for favorites (++) link - Randy Stauner
19:25 metacpan [metacpan-web/rwstauner/release-info] Close broken span tag in favorites link - Randy Stauner
19:25 metacpan [metacpan-web/rwstauner/release-info] Move duplicate release-info api calls to role - Randy Stauner
19:25 metacpan left #metacpan
19:25 dipsy [ Comparing ce1517c...9e826c8 for CPAN-API's metacpan-web - GitHub ]
19:41 alnewkirk joined #metacpan
19:49 jsut joined #metacpan
20:02 migo joined #metacpan
21:19 smpb joined #metacpan
21:33 semifor joined #metacpan
21:36 apeiron joined #metacpan
23:12 ingy joined #metacpan
23:13 DrForr joined #metacpan
23:16 apeiron joined #metacpan
23:17 frew joined #metacpan
23:18 jsut_ joined #metacpan
23:29 dhoss joined #metacpan
23:43 alnewkirk joined #metacpan
23:45 dhoss joined #metacpan

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