Perl 6 - the future is here, just unevenly distributed

IRC log for #metacpan, 2014-02-04

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

All times shown according to UTC.

Time Nick Message
00:33 castaway joined #metacpan
00:42 SineSwiper joined #metacpan
01:28 jnbek joined #metacpan
01:40 omega ohh, bootstrap got released?
02:23 klapperl_ joined #metacpan
03:11 omega hmm, why are there two different date formats on author pages? https://metacpan.org/author/TURNSTEP vs https://metacpan.org/author/SZABGAB
03:11 dipsy [ Greg Sabino Mullane (TURNSTEP) - metacpan.org ] [ ????? ??? - Gábor Szabó (SZABGAB) - metacpan.org ]
03:21 oalders omega: not sure.  wonder if this hand-rolled datetime parsing is to blame https://github.com/CPAN-API/metacpan-web/blob/master/lib/MetaCPAN/Web/View/HTML.pm
03:21 dipsy [ metacpan-web/lib/MetaCPAN/Web/View/HTML.pm at master · CPAN-API/metacpan-web · GitHub ]
03:21 oalders i'm sure we can simplify that a lot -- unless it's there for a very good reason
03:23 omega haha, no, it gets parsed in javascript
03:23 omega and changed
03:23 omega the html is the same format
03:48 oalders ah, ok.  saw that ticket.
03:48 oalders having said that, in root/inc/release-latest.html there is latest.date.dt_http
03:48 oalders so it's using that vmethod, but then using JS to transform it again, it seems
03:48 omega yeah
03:49 omega to get relative dates, while maintaining cachability
03:49 oalders that vmethod gets called in a bunch of other places, but the logic could be stripped down a lot
03:57 metacpan joined #metacpan
03:57 metacpan [metacpan-web01] oalders pushed 2 new commits to master: http://git.io/P6_jeA
03:57 metacpan metacpan-web/master 0183d04 Andreas Marienborg: Not show favorites table if no favorites...
03:57 metacpan metacpan-web/master 9d22a0c Olaf Alders: Merge pull request #1062 from omega/1061-remove-horiz-scroll...
03:57 metacpan left #metacpan
03:57 dipsy [ Comparing 8997d071fb19...9d22a0cac99e · CPAN-API/metacpan-web · GitHub ]
03:57 oalders omega++
03:57 oalders changes deployed
04:03 metacpan joined #metacpan
04:03 metacpan [metacpan-web01] oalders pushed 2 new commits to master: http://git.io/mWWJbA
04:03 metacpan metacpan-web/master 5df85fd Andreas Marienborg: Change CPANTS to Kwalitee...
04:03 metacpan metacpan-web/master 1a5893a Olaf Alders: Merge pull request #1063 from omega/1053-cpants-to-kwalitee...
04:03 metacpan left #metacpan
04:03 dipsy [ Comparing 9d22a0cac99e...1a5893a5d02b · CPAN-API/metacpan-web · GitHub ]
04:03 oalders also deployed :)
04:04 omega :)
04:04 omega just fixing some low hanging fruits
05:05 metacpan joined #metacpan
05:05 metacpan [metacpan-web01] oalders pushed 2 new commits to master: http://git.io/70rKSQ
05:05 metacpan metacpan-web/master dd1306d Andreas Marienborg: Remove inline-block on an authors-results...
05:05 metacpan metacpan-web/master 09be601 Olaf Alders: Merge pull request #1065 from omega/fix-bad-whitespace...
05:05 metacpan left #metacpan
05:05 dipsy [ Comparing 1a5893a5d02b...09be60161769 · CPAN-API/metacpan-web · GitHub ]
05:05 metacpan joined #metacpan
05:05 metacpan [metacpan-developer01] rwstauner pushed 1 new commit to master: http://git.io/gIBdLw
05:05 metacpan metacpan-developer/master 19bd219 Randy Stauner: Make metacpan user provision work across destroy/up
05:05 metacpan left #metacpan
05:06 dipsy [ Make metacpan user provision work across destroy/up · 19bd219 · CPAN-API/metacpan-developer · GitHub ]
05:06 oalders and the commits keep on coming :)
05:16 metacpan joined #metacpan
05:16 metacpan [metacpan-web01] monken pushed 2 new commits to master: http://git.io/62y9UQ
05:16 metacpan metacpan-web/master d18bfa7 Andreas Marienborg: Move toplists from inline styles to less-file...
05:16 metacpan metacpan-web/master acabc0e Moritz Onken: Merge pull request #1064 from omega/392-fix-styling...
05:16 metacpan left #metacpan
05:16 dipsy [ Comparing 09be60161769...acabc0e97f4c · CPAN-API/metacpan-web · GitHub ]
05:39 metacpan joined #metacpan
05:39 metacpan [cpan-api01] oalders pushed 1 new commit to master: http://git.io/n_o23g
05:39 metacpan cpan-api/master cf2caee Olaf Alders: Don't overwrite author profiles for users without a CPAN author dir.
05:39 metacpan left #metacpan
05:39 dipsy [ Don't overwrite author profiles for users without a CPAN author dir. · cf2caee · CPAN-API/cpan-api · GitHub ]
06:30 [Sno] joined #metacpan
07:05 ether_ joined #metacpan
07:48 neilb joined #metacpan
08:07 neilb joined #metacpan
08:27 neilb joined #metacpan
09:01 neilb joined #metacpan
09:25 neilb joined #metacpan
10:04 mtj joined #metacpan
10:18 dolmen there is a typo in labels for https://github.com/CPAN-API/metacpan-web/issues : « priority:Hight » instead of « priority:High »
10:18 dipsy [ Issues · CPAN-API/metacpan-web · GitHub ]
12:33 metacpan joined #metacpan
12:33 metacpan [metacpan-web01] monken pushed 2 new commits to master: http://git.io/lVw_nA
12:33 metacpan metacpan-web/master 7c3edd8 Andreas Marienborg: Add clearfix class to authors ul...
12:33 metacpan metacpan-web/master 582d63e Moritz Onken: Merge pull request #1067 from omega/1066-add-clear...
12:33 metacpan left #metacpan
12:33 dipsy [ Comparing acabc0e97f4c...582d63ed8f83 · CPAN-API/metacpan-web · GitHub ]
12:45 ranguard dolmen: heh, fixed thanks
13:25 b_jonas { vector<int> s(10, 1); s.clear(); cerr << s.capacity(); }
13:26 b_jonas sorry, wrong channel
14:10 metacpan joined #metacpan
14:10 metacpan [metacpan-web01] rwstauner pushed 1 new commit to master: http://git.io/5aF2Eg
14:10 metacpan metacpan-web/master 5738f6d Randy Stauner: Make last breadcrumb of release not a link...
14:10 metacpan left #metacpan
14:10 dipsy [ Make last breadcrumb of release not a link · 5738f6d · CPAN-API/metacpan-web · GitHub ]
14:45 haarg well that's annoying.  the 0.5.0-TRIAL part of the release name doesn't seem to be available in the api.
14:48 oalders i'm sure that's something we can add
14:58 haarg was trying to fix the cpan testers links for -TRIAL dists.  i have it working, as well as adding links to each of the pass/fail/na counts.
14:58 haarg but right now it's doing release.name.remove('^' _ release.distribution _ '-?');
14:58 haarg in the template.  which is pretty ugly
15:03 oalders there's already some ugly stuff in the templates, but i see what you're saying :)
15:03 oalders if it works, though...
15:09 haarg what would a good name for that type of version be?  full_version?
15:24 oalders isn't it more the name rather than the version?
15:25 oalders (not good at naming things)
15:25 haarg nobody is
15:26 oalders that kind of full_version is something that we couldn't numify, so calling it a version seems a bit confusing
15:26 haarg i'm not sure.  in this case, i'm using it for what other sites think of as the version
15:26 oalders mst: thoughts? ^^^
15:27 dpetrov_ joined #metacpan
15:33 oalders haarg: http://explorer.metacpan.org/?url=/release/ETHER/Moose-2.1108-TRIAL/
15:33 dipsy [ MetaCPANExplorer ]
15:33 oalders "name" : "Moose-2.1108-TRIAL"
15:34 oalders does that not cover it?
15:34 haarg i need just the 2.1108-TRIAL part
15:35 oalders ah, right
15:36 ribasushi bleh
15:36 ribasushi why are you guys still dropping the author part
15:36 ribasushi https://metacpan.org/release/VPARSEVAL/List-MoreUtils-0.24 vs https://metacpan.org/release/ADAMK/List-MoreUtils-0.24
15:37 dipsy [ List-MoreUtils-0.24 - Provide the stuff missing in List::Util - metacpan.org ] [ List-MoreUtils-0.24 - Provide the stuff missing in List::Util - metacpan.org ]
15:37 ribasushi in fact I can not navigate to the VPARSEVAL version at all via the dropdown, if I didn't know it's there
15:37 ribasushi this used to work on the old site iirc
15:38 ribasushi ah no it does show up in the backpan area now
15:38 ribasushi strange, it didn't a minute ago...
15:39 * ribasushi can't rebreak it... wtf
15:41 oalders :)
15:41 rwstauner if( ribasushi ){ delete unless $seen++ }
16:12 metacpan joined #metacpan
16:12 metacpan [metacpan-web01] oalders pushed 1 new commit to master: http://git.io/zlI_KA
16:12 metacpan metacpan-web/master 873aaab Olaf Alders: s/button/submit -- fixes #1076.
16:12 metacpan left #metacpan
16:12 dipsy [ s/button/submit -- fixes #1076. · 873aaab · CPAN-API/metacpan-web · GitHub ]
16:44 ether you can't make assumptions about distribution names that aren't enforced by PAUSE
16:44 ether it would be nice if we could, because there's a lot of crappy dist names, but we can't
16:44 ether all that we can count on is that "AUTHOR/Dist-Name-moar-stuff" is a unique string
16:45 ether I'm not even sure if the extension, e.g. tar.gz vs tgz, is taken into consideration
16:45 ether xdg and andk can advise
16:47 ribasushi ether: they are not, you can upload ETHER/Moose-666.tar.gz and ETHER/Moose-666.zip
16:47 ribasushi the system will allow it
16:47 bowtie_ joined #metacpan
16:48 ether awesmoe (typo intentional)
16:48 haarg any assumptions you make will be inaccurate
16:48 haarg but lots of sites already made those assumptions
16:50 BinGOs assumptions make an ass out of 'u' and mption
17:03 rwstauner everybody knows that
17:09 ether not everyone knows what assumptions are just umptions though ;)
17:09 ether sometimes they look like hard and fast rules but they actually aren't
17:11 rwstauner :-)
17:16 ribasushi anyway, even if this stuff gets solved in PAUSE looking forward
17:17 ribasushi it is still much nicer to have the historic stuff show up correctly (see List::Moreutils example)
17:17 ribasushi having a point with a high degree of truthiness and all that
17:34 mudler joined #metacpan
17:35 metacpan joined #metacpan
17:35 metacpan [metacpan-web01] oalders pushed 2 new commits to master: http://git.io/OgNFrg
17:35 metacpan metacpan-web/master 7ee2a2f Graham Knop: fix cpan testers links for -TRIAL dists, and link pass/fail/na counts
17:35 metacpan metacpan-web/master c87b6f6 Olaf Alders: Merge pull request #1077 from haarg/testers-links...
17:35 metacpan left #metacpan
17:35 dipsy [ Comparing 873aaabd7456...c87b6f673eca · CPAN-API/metacpan-web · GitHub ]
17:35 [Sno] joined #metacpan
17:35 oalders haarg++
17:40 punter joined #metacpan
17:41 punter metacpan seems quite slow to me (this page for example takes 3.5 seconds to start loading)
17:41 punter https://metacpan.org/pod/release/ETHER/Moose-2.1202/lib/Moose/Manual.pod
17:41 dipsy [ Moose::Manual - What is Moose, and how do I use it? - metacpan.org ]
17:41 punter not always
17:41 oalders punter: https://github.com/CPAN-API/cpan-api/issues/293
17:41 dipsy [ Slow performance · Issue #293 · CPAN-API/cpan-api · GitHub ]
17:42 punter Can I donate to metacpan?
17:42 punter I'll google that
17:42 punter is metacpan in need of cash?
17:43 ether or to rephrase: if metacpan received money, how would it spend it?
17:43 punter give it all to the biggest donor
17:43 punter as thanks
17:46 ranguard maybe we could hire an ElasticSearch person to help optomise - if we had cash - think it is more that than hardware
17:53 trs has anyone profiled the slow bits?
17:53 * trs noticed some DB::enable_profiler() droppings from mo in a file the other day
18:04 neilb joined #metacpan
18:39 ranguard trs: it's more someone who knows ElasticSearch having time to look into it, or to answer your question... no :)
18:39 ranguard splitting up the indexes should help a lot apparently
18:54 trs nod
18:55 trs ranguard: I'm wondering if anyone has profiled other places that might be bottlenecks?
18:56 trs i.e. not ES directly, but something around it.  is the www -> api layer slow, for example?
18:56 trs internally
18:57 * trs wonders if metacpan wants a statsd and graphite server which some timing metrics wrapped around low-level hot spots?  or just run nytprof on some small portion of requests ;)
18:57 * trs is thinking aloud
19:04 rwstauner <3 statsd + graphite
19:06 rwstauner something else we could do is cache/index data to speed the website
19:06 rwstauner we're using ES as a datastore
19:06 rwstauner and the docs are presentably for external queries
19:06 rwstauner but if we use half a dozen queries to build the web interface, we could be caching/indexing those together into a different index for our own use
19:07 rwstauner s/presentably/presentable/
19:10 trs yup
19:11 ranguard trs: http://munin.metacpan.org/ might be useful
19:11 dipsy [ Munin :: overview ]
19:13 trs ranguard: I haven't poked too much there, but does it have timing stats beyond "how long does /recent take?"
19:13 rwstauner there's also an es slow queries log on the machine
19:13 rwstauner in case there are queries to optimize
19:13 trs nod
19:13 trs ++ for looking at those
19:13 dipsy Thanks!
19:14 trs as a first order of business.
19:14 trs and I wonder what pressure the ES instance is under. memory?
19:14 rwstauner we've asked for 36GB RAM so that the entire index can be stored in memory
19:14 rwstauner (that's what i've been told)
19:15 trs what do we have now?
19:15 rwstauner that is what we have now
19:15 trs ahh
19:15 rwstauner when we get machines, that's what we ask for
19:15 trs nod
19:16 trs we should be plugging es into munin.
19:17 trs we don't have any memory stats for anything but starman and the /var/cpan watcher
19:17 trs speaking of /var/cpan, it's creeeping up towards ENOSPACE
19:17 rwstauner i see "es open files" but don't know much about it or what else there is
19:18 trs there's a whole bunch of es munin plugins in puppet, but none of them are used.
19:18 rwstauner rats
19:18 trs as far as I can see.
19:18 trs which doesn't appear to be the intent.
19:19 trs ahaha, they're symlinked.
19:19 trs but:
19:20 trs http://zulutango.net/~tom/paste/2014-02-04RfceMgAq
19:20 trs whomp :)
19:20 rwstauner nertz
19:20 rwstauner looks like it needs the perlbrew perl in it's path
19:21 trs good luck. I think you're better off making the munin plugins able to use the system perl.
19:21 rwstauner could do that too
19:21 trs there really shouldn't be too many deps.
19:21 rwstauner could apt-get libjson-perl
19:21 trs but you could change the shebangs in the plugins.
19:21 trs yeah, or that.
19:22 trs I assume the apt-get wants to be added to puppet instead though
19:22 ranguard rwstauner: just add it as a package in puppet
19:22 rwstauner right
19:22 trs or I'd have just done it already :)
19:22 rwstauner package { 'libjson-perl': }
19:22 trs ++
19:22 dipsy Thanks!
19:23 trs I \N{HEAVY BLACK HEART} that munin reports other kinds of plugin errors, but not "this process exited non-zero"
19:23 ranguard trs: feel free to setup statsd + graphite through puppet (then I can nick it for elsewhere) :)
19:23 trs although, it's probably in munin logs, just not mailed?
19:24 trs yup
19:24 trs don't we have logwatchr unning?
19:25 trs ah, looks like no
19:27 neilb joined #metacpan
19:28 oalders since we already have ES, we could look at kibana http://www.elasticsearch.org/overview/kibana/
19:28 dipsy [ Kibana | Overview | Elasticsearch ]
19:30 trs that might be cool to explore metacpan's data, but I don't know if it helps with log watching and perf/memory tracking without more work to setup than the munin already running?
19:31 trs i.e. something would have to write that data out in a time series, is my undersatnding, so that kibana could process it.
19:31 oalders trs: kibana + logstash
19:31 rwstauner ELK
19:31 oalders http://www.elasticsearch.org/overview/logstash/
19:31 dipsy [ Logstash | Overview | Elasticsearch ]
19:31 trs but it'd be cool to setup a public kibana to let people play with the cpan indexes, if I understand it correctly.
19:32 oalders we're using kibana + logstash + ES at $work
19:32 trs oalders: right, so that just moves your logs from files into es
19:32 trs you still have to have the useful data you want in the logs
19:32 oalders for sure
19:32 oalders which we don't
19:32 trs I don't see what kibana gets us that other perf monitoring tools don't with less work.
19:33 trs but I _do_ think it'd be cool as an explorer as sorts.
19:33 punter left #metacpan
19:33 oalders i'm not saying it's better, but it has a nice UI and works with ES
19:33 * trs has been looking for a justifiable reason to try kibana at work actually, but coming up short.
19:33 trs :)
19:33 trs sure
20:24 mst oalders: I'd probably have distribution_name and distribution_version
20:29 oalders mst: right. i guess that's what we currently have http://api.metacpan.org/release/ETHER/Moose-2.1108-TRIAL/?fields=name,version,version_numified,archive,distribution
20:29 dipsy urgh. long url. Try http://tinyurl.com/pcajbzd
20:29 oalders the question is do we want some other field that would return "2.1108-TRIAL"
20:30 ether also note if you want the actual 'is trial?' boolean flag, you can't depend on the uploaded tarball name at all - you need to look in META.*
20:31 oalders right
20:31 ether don't assume that trial releases end in _\d+ or -TRIAL
20:31 ether although I think PAUSE might, for _\d+ at least, which is another thing to look at (i.e. make metadata the sole canonical source of that bit of info)
20:32 oalders would be nice to clean up moving forwards, but there's so much historical stuff to deal with either way
20:34 ether yeah
20:34 ether the best that can be done on your end is just abstract that logic away into subs that can have their guts change later
20:35 ether e.g. instead of putting heuristics everywhere, use Dist::Metadata or something to provide that info
20:35 ether and then we just fix that as PAUSE gets fixed, etc
20:36 mst ooh, that module looks rather promising
20:36 oalders we're using Module::Metadata, but i don't see that we're using Dist::Metadata
20:36 ether I have plans to make Module::Metadata have an is_trial? flag, but it's not there yet
20:37 ether and is dependent on other fixes first
20:37 oalders would be helpful for sure
20:38 rwstauner D::M could use some work, but it could be a good starting point
20:38 rwstauner could merge some metacpan functionality into it (like the Parse::PMFile business)
20:42 ether I've recently started playing with CPAN::Common::Index as a central API to interrogate PAUSE about things, e.g. "is this module indexed" or "who really has permissions to upload this thing"
20:42 mst what does Parse::PMFile achieve that Module::Metadata doesn't?
20:42 mst other than "supports the broken hack that common::sense uses"
20:43 oalders :)
20:43 rwstauner that's it
20:44 oalders i'm pretty sure that's exactly why we use Parse::PMFile
20:44 rwstauner and that's all we use it for... if /\.pm\.PL/
20:45 rwstauner there's other data metacpan extracts that D::M doesn't, but we could certainly add stuff to it
20:46 oalders https://github.com/CPAN-API/metacpan-web/issues/625
20:46 dipsy [ no common::sense on metacpan ? · Issue #625 · CPAN-API/metacpan-web · GitHub ]
20:47 mst right. personally, I prefer the approach of "ignore it, it's broken"
20:49 rwstauner we did that for a long time
20:49 oalders that was fine before cpanm started using the api
20:50 mst yeah, so at that point you file a bug against common/sense.pm telling it to use 'provides' or something else that actually works properly
20:50 mst intentionally breaking metacpan to work around a stupid dist doesn't seem like a feature to me
20:51 mst so, you now owe me a patch to Module::Metadata to allow it to *optionally* allow that
20:51 mst and then you can remove Parse::PMFile
20:51 oalders can you actually file a bug against common::sense? :)
20:51 mst given you already broke metacpan, there's no point trying to keep Module::Metadata intact anymore
20:51 mst since now it's just confusing, because things don't consistently fail
20:52 oalders would be something worth discussing at the qa hackathon
20:52 ether schmorp would ignore the ticket anyway, and just make the submitter cry, so I wouldn't bother
20:53 ether oalders: you're coming to Lyon right?
20:53 oalders ether: i am. just have to make my travel arrangements
20:53 ether cool! I need to buy tickets this week as well
20:53 mst ether: the whole point of not making Module::Metadata accept the hack was to put social/technical pressure on him to not ignore it
20:54 mst since apparently the metacpan team have decided to side with schmorp, it's basically academic now
20:54 ether I'm not convinced they have knowingly done so
20:54 oalders well, i wouldn't put it quite that way and i'd be willing to revert that patch
20:54 oalders but it fixed a real problem at the time
20:55 mst no, it worked around a problem in common::sense
20:55 mst if everybody works around the problems schmorp causes, what incentive is there for him to fix anything?
20:56 oalders it's a valid point.  so, if there's a united effort to get this fixed, we can bring MetaCPAN in line with that
20:56 oalders why does common::sense even behave this way?
20:57 mst because he generates the file per-perl-version as amicro-optimisation to avoid loading strict.pm and warnings.pm
20:57 mst and so the dist code only exists in __DATA__ of the .PL script
20:57 mst this is why Cpanel::JSON::XS doesn't use common::sense
20:58 mst hence why my JSON::MaybeXS uses that instead of JSON::XS
20:58 ether ++
20:58 dipsy Thanks!
20:58 rwstauner :-)
20:58 alnewkirk to whom it may concern ... I really dig the metacpan enhancements
20:58 oalders alnewkirk: thanks!
20:59 ether I have a dist that ships a module in a different filename than what it eventually gets installed as, but it's a nasty hack (and I set 'provides' data properly to indicate this)
20:59 rwstauner sadly, that leaves AnyEvent, which is also something i wish less dists depended on
20:59 mst AnyEvent doesn't depend on common::sense
20:59 mst it inlines its own version
20:59 rwstauner it doesn't?
20:59 rwstauner oh
20:59 alnewkirk oalders: thank you
20:59 mst hence why I was confident we didn't need to allow for common::sense
20:59 mst hence why last time we discussed this we agreed to not break metacpan
20:59 rwstauner hmm
20:59 mst apparently later somebody broke it anyway without talking to anybody
21:00 mst which is why I might be being sliughtly cranky here
21:00 mst because I spent about an hour discussing this and the impact last time
21:00 oalders alnewkirk: big team effort :) mo spent a lot of time on it
21:01 oalders it would help if we documented stuff we all agree on
21:01 oalders it's essentially mostly IRC logs
21:01 alnewkirk ... in opensource you rarely get a thanks let alone acknowledgement so I figured I'd buck the system and thanks for all the hard works guys
21:02 oalders yeah, it's mostly "please fix this broken thing"
21:02 mst notice that nobody except miyagawa spoke in favour of the nasty hack
21:02 mst and he isn't competent to have an opinion on toolchain code
21:02 oalders well, tokuhirom was asking for a workaround as well
21:02 mst cpanm only works at all because I spent days guiding him through unfucking it, and carton only works at all because doy did the same thing
21:03 mst right, to which the response is "file a bug"
21:03 mst also, I was planning to work on a patch to fix PAUSE
21:03 ether it would be great to get PAUSE finally sorted
21:03 ether there's that big patch from rjbs last year that hasn't gone in yet
21:03 mst a hardcoded exception for common::sense would be reasonable
21:04 mst much more reasonable than the ugly shit you've done instead
21:04 mst now you're encouraging people to make more similarly broken dists
21:05 oalders ok, so we can agree that it's ugly. what's the best thing to do moving forward?
21:05 mst back out the fix, add a hard-coded exception for common-sense, add a warning on the page that it's only indexed via a hard-coded hack to encourage people to complain about it
21:05 mst which is what I said was the fall back plan last time we discussed this
21:06 mst ether: right, but sorting PAUSE requires the ability to show that the change is correct
21:06 mst which requires other things to consistently do it the right way
21:11 rwstauner mst: why does cpanel-json-xs "recommend" common::sense?
21:11 oalders mst: https://github.com/CPAN-API/cpan-api/issues/295
21:11 dipsy [ Don't encourage badly formed dists in the style of common::sense · Issue #295 · CPAN-API/cpan-api · GitHub ]
21:12 mst rwstauner: if memory serves, because some of the tests require it, and reini preferred to skip those if it isn't present
21:13 rwstauner hmm... ok
21:14 mst y'know what'd be interesting - doing a scan across metacpan of what other dists use .pm.PL with __DATA__
21:14 mst see if there's anything using it for good
21:16 trs mst: cpangrep --color file:.pm.PL __DATA__  -> http://zulutango.net/~tom/paste/2014-02-047RCKsca9
21:17 trs that's grep.cpan.me, not metacpan
21:17 mst so basically almost nothing
21:17 trs yeah
21:17 trs short enough to go audit visually :)
21:17 mst and short enough to be able to send patches
21:17 trs yup
21:18 trs grep.cpan.me's index was last updated  Sun, 08 Dec 2013 20:29:02 GMT
21:18 trs so not even that out of date
21:22 trs looking for __END__ turns up LDS/GD-2.50/GD/Image.pm.PLS
21:22 trs and another hit from MONS/uni-perl, whatever the hell that is.
21:23 trs oh, but that __END__ is embedded in a heredoc
21:23 trs which I didn't know you could safely do.
21:32 rwstauner it's a token
21:33 rwstauner perl -le 'print 1, __END__'
21:35 trs yeah, I see that now.  but I've also seen capital-P Perl which naively scans source until /^__END__/
21:38 rwstauner :-)
23:07 punter joined #metacpan
23:07 punter I found a bug in the website - should I say it here or in the issue tracker?
23:09 punter reported: https://github.com/CPAN-API/metacpan-web/issues/1078
23:09 dipsy [ ++ button appears un-checked at a certain page · Issue #1078 · CPAN-API/metacpan-web · GitHub ]
23:11 oalders punter++
23:11 oalders trs: maybe something related to the changes you made?
23:20 trs oalders: no, git blame points to the original bootstrap refactoring
23:20 oalders ok :)
23:20 trs the problem is that release used to pass module = release to inc/favorite and module passed module = module
23:20 trs now inc/breadcrumbs only passes module = module
23:21 trs so you favorite the empty string.
23:21 trs :)
23:21 trs which returns success from the api!
23:21 trs so the JS believes it.
23:21 trs it's handy that you can unfavorite the empty string too
23:21 trs and you'll see an empty row on your author page.
23:21 * trs <3's software
23:25 punter trs: How can I un-favorite the empty string? Is it possible by clicking somewhere?
23:25 trs punter: click the same button :)
23:25 oalders nice
23:26 punter right!
23:26 punter you saved me!
23:29 trs pushed fixed.
23:29 trs someone deploy?
23:29 metacpan joined #metacpan
23:29 metacpan [metacpan-web01] tsibley pushed 1 new commit to master: http://git.io/HZnZ9g
23:29 metacpan metacpan-web/master 08b9494 Thomas Sibley: Not all pages which include breadcrumbs define "module", such as the release page...
23:29 metacpan left #metacpan
23:29 dipsy [ Not all pages which include breadcrumbs define "module", such as the rel... · 08b9494 · CPAN-API/metacpan-web · GitHub ]
23:30 oalders trs++
23:37 neilb I like the new MetaCPAN look (felt I should say that, and not just submit nitpicky bug reports :-)

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