Camelia, the Perl 6 bug

IRC log for #metacpan, 2011-06-09

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

All times shown according to UTC.

Time Nick Message
00:03 theory joined #metacpan
00:07 theory left #metacpan
01:22 klapperl_ joined #metacpan
01:27 klapperl left #metacpan
03:00 frew mo: btw, I think you are the most successful GSOC I've seen ever, and it's only june
03:01 frew mo: that's pretty awesome :D
03:09 oalders frew: mo hit the ground running for sure :)
03:10 frew I'm pretty impressed by the codebase
03:10 frew like, plack + ae basically, but not a nightmare
03:10 oalders yeah, when he said he was doing that, i thought it might be messy to sort through, but it's pretty succinct
03:11 frew AE is much neater than I thought it would be
03:11 frew plus this syntax is awesome: ($cv1 & $cv2)->(sub { my ($ret1, $ret2) = ... }
03:12 oalders :)
03:12 frew oh I see that the & is not AE, but his own construction
03:13 oalders which file is that in?
03:13 frew Model.pm
03:13 frew package MyCondVar
03:16 oalders i wouldn't mind seeing a few comments in there.
03:16 oalders actually, maybe even *one* comment :)
03:17 frew what's confusing?
03:18 oalders well, i've never used AE, so i'd probably have to do some digging to get a better idea of what's going on in there
03:18 frew sure
03:19 frew but you'd agree that comments aren't the place to document the libraries used, right?
03:19 oalders true
03:20 oalders but even just a quick description of what the file is there for helps with getting an overview
03:20 frew did you look at the pod for that file?
03:20 frew I didn't but it might have what you are looking for, and there is sme
03:20 frew some*
03:20 frew ok, looks like just a bunch of examples, but still
03:21 oalders in Model.pm?
03:21 frew yah
03:21 frew after __END__
03:23 oalders ah, yes, that's helpful
03:54 hoelzro joined #metacpan
03:56 hoelzro so I'm trying to hack on MetaCPAN, but I'm a little confused as to how it gets its data...
03:56 hoelzro I assume I need to be running ElasticSearch?
03:56 oalders hoelzro: what in particular are you looking at?
03:56 oalders oh, not necessarily
03:57 hoelzro well, if you remember, I'm trying to add annotations
03:57 oalders right. so you need write access
03:57 hoelzro is METACPAN_API supposed to be a thin frontend to an ElasticSearch DB?
03:57 oalders that's one aspect of it
03:58 oalders sorry, you're talking about the variable in the command line
03:58 hoelzro yes
03:58 hoelzro so metacpan-web is the frontend
03:58 oalders yes
03:58 oalders basically, if you want to hack on just the web interface, you can point that at the live API
03:58 hoelzro and completely decoupled from the API server?
03:59 oalders it is
03:59 hoelzro ah ha
03:59 hoelzro *now* I get it
03:59 oalders :)
03:59 oalders originally, it was just the API
03:59 hoelzro ok
03:59 oalders and then the front end came much later
03:59 hoelzro I think the api is where I'd like to start
03:59 oalders sure. in that case you'll need to install cpan-api
03:59 hoelzro or maybe I *should* start at the frontend, and deliver fake annotations
03:59 hoelzro got it
04:00 hoelzro I assume running cpan-api means I need an ElasticSearch instance?
04:00 oalders fake annotations works too, i think
04:00 hoelzro if so, can I point it at my minicpan?
04:00 oalders yes
04:00 oalders let me point you at the docs...
04:00 hoelzro =)
04:00 hoelzro actually, I just found them
04:00 oalders https://github.com/CPAN-API​/cpan-api/wiki/Installation
04:00 dipsy [ Installation - GitHub ]
04:01 oalders please feel free to improve them
04:01 hoelzro =)
04:02 oalders the cool thing is you can point it at your minicpan and then only index part of it, if you want to
04:02 oalders that will get you up and running much faster
04:02 oalders you don't want to index the whole thing if you intend to look at it this evening
04:02 hoelzro heh
04:02 hoelzro I'm guessing it'd take a few hours?
04:02 oalders yep
04:03 oalders a lot of unpacking and parsing
04:03 oalders bin/metacpan release /path/to/cpan/authors/id/
04:04 oalders you can narrow down the path to your cpan so that you get a subset of the folders
04:04 hoelzro I see
04:08 hoelzro grr
04:08 hoelzro elastic search is giving me crap =/
04:08 oalders what's the error?
04:09 hoelzro it's permissions for the logs and data dires
04:09 hoelzro *dirs
04:09 hoelzro I'm reading their docs now
04:09 hoelzro I'm also working on a small PSGI app for my girlfriend
04:09 hoelzro so my attention's a little divided =P
04:10 oalders :)
04:11 oalders the ES docs take some getting used to
04:25 hoelzro ok, that wasn't too bad
04:25 oalders what was the fix?
04:25 hoelzro elasticsearch -f -Des.config=myconf.yml
04:26 oalders ok.  to set the memory, you can also do something like:
04:26 oalders ES_MAX_MEM=2G ES_MAX_MEM=2G bin/elasticsearch -f
04:27 oalders er, one of those should be ES_MIN_MEM
04:27 oalders depends on how much of a pig you want it to be
04:27 oalders i think default min is 256MB and default max is 1GB
04:27 hoelzro I'll probably just index like 8 modules
04:28 oalders go with the minimum :)
04:28 hoelzro hmm...now git submodule update is failing.
04:29 oalders what's the error?
04:29 hoelzro https://gist.github.com/1016060
04:29 dipsy [ hoelzro's gist: 1016060 ? Gist ]
04:31 oalders i think i had a similar error when mo hadn't pushed all of his commits for one of the submodules
04:31 hoelzro =/
04:31 hoelzro I can probably wait till tomorrow
04:31 hoelzro my night's drawing to a close
04:32 oalders sure thing. at least you have ES up and running :)
04:32 hoelzro and I now know the difference between the API server and the frontend =)
04:33 oalders yeah, that should be documented more clearly
04:46 hoelzro left #metacpan
06:07 castaway joined #metacpan
07:11 sheepx left #metacpan
07:24 woldrich left #metacpan
07:30 * mo wonders if we should index the whole cpanstats database
07:31 mo it's 5gb in size and provides everything but the actual report
07:31 mo contains 12.5 million reports
07:31 mo would be awesome to have this available via the metacpan api
07:31 mo opinions?
07:31 dipsy opinions are like assholes
07:32 apeiron left #metacpan
07:32 apeiron joined #metacpan
07:36 Ranguard cpanstats = the cpantesters db?
07:36 mo Ranguard: yes, right
07:36 mo should've mentioned that :)
07:39 Ranguard that has a lot going for it
07:39 Ranguard http://beta.metacpan.org/release​/CPAN-Testers-Reports-Query-JSON <- I created this to access the data
07:39 dipsy [ CPAN-Testers-Reports-Query-JSON-0.04 - beta.metacpan.org ]
07:40 mo so that just downloads the .json files?
07:40 Ranguard yea - and does some parsing
07:40 sheepx joined #metacpan
07:40 Ranguard what I wanted was a "works on windows" , "works on osx" type report
07:41 mo right, so that would be possible if we index the cpanstats database
07:41 mo and then the user can choose which one he wants to see
07:43 Ranguard woooo!!!!!!.....
07:43 mo I'll just go ahead and do it :)
07:43 Ranguard 'win32.platform.metacpan.org' - only show modules that have passing tests on this platform :)
07:44 mo hahaha, we will see :)
07:45 mo Ranguard: what would be a good name for the endpoint?
07:46 mo not sure
07:46 mo /report? /testing? /cpantesters?
07:46 Ranguard /cpantesters
07:46 mo hm all of the endpoints are singular
07:47 mo I'd like to keep that consistent
07:47 Ranguard but your returning multiple results?
07:47 mo you can also query one
07:47 mo like for /module/
07:47 mo all of them have a _search endpoint
07:47 mo /testreport ?
07:48 Ranguard sounds reasonable
07:52 sheepx mo: think metacpan could be made compatible with urls like these? http://search.cpan.org/~flora/Data-Str​eam-Bulk-0.08/lib/Data/Stream/Bulk.pm
07:52 dipsy [ Data::Stream::Bulk - search.cpan.org ]
07:52 sheepx i did the hosts file remap thing and they just throw up 404 pages :)
07:52 mo hm where does it redirect you
07:52 sheepx http://beta.metacpan.org/module/flora/Data​-Stream-Bulk-0.08/lib/Data/Stream/Bulk.pm
07:52 dipsy urgh. long url. Try http://tinyurl.com/68f8ydb
07:53 sheepx which should be: http://beta.metacpan.org/module/Data::Stream::Bulk
07:53 dipsy [ Data::Stream::Bulk - beta.metacpan.org ]
07:53 mo hm need to uc() flora
07:53 mo http://beta.metacpan.org/module/FLORA/Data​-Stream-Bulk-0.08/lib/Data/Stream/Bulk.pm
07:53 dipsy urgh. long url. Try http://tinyurl.com/5ulmv5c
07:53 dipsy [ Data::Stream::Bulk - beta.metacpan.org ]
07:53 mo works
07:53 mo could you open a ticket? will fix that in a bit
07:53 sheepx ah yes, that works
07:53 sheepx will do :)
07:56 sheepx and added
07:56 mo thanks
07:56 sheepx np, i'm happy to see an open source cpan :)
07:56 sheepx random thought:
07:56 sheepx how well would metacpan work with a minicpan?
07:56 mo very well, in fact, when we started we only indexed minicpan
07:57 sheepx yay \o/
07:57 sheepx i'll have to play with that then
07:57 sheepx hope it copes well with injected dists
07:57 mo ;)
07:57 mo should work, metacpan doesn't even care about 02packages, so just throw a tarball in and it will index it
07:58 mo it must be in an author's dir though
07:58 sheepx sweet :D
07:58 sheepx yeah, cpan::mini::inject does a bunch of mangling
07:58 mo which is not needed for metacpan
07:58 mo maybe at some point when we introduce "authorized" releases
07:59 mo which are obtained from the 02packages file
07:59 * sheepx nods
07:59 sheepx this sounds so neat
08:01 szabgab the new beta is very nice
08:02 mo szabgab: thanks
08:02 szabgab here is a suggestion: when I see beginners they constantly click on "download" to download the package
08:02 szabgab instead of using a cpan client
08:02 szabgab I'd put some explanation or a link to "install"
08:03 mo szabgab: good point. could you open an issue?
08:03 szabgab yeah, maybe a link to "install"  that will show some explanation on how to install
08:03 mo someone mentioned he wanted to install some sort of url handler that would execute cpanm when you hit a cpan:// url
08:03 szabgab oh and the perl keywords in the pod are highlighted in random places
08:04 mo you mean in code blocks that are not perl?
08:04 mo can you give an example?
08:04 szabgab "while I think it is ok"...  and then while is highlighted
08:05 szabgab eg here: http://beta.metacpan.org/module/XML::Stream
08:05 dipsy [ XML::Stream - beta.metacpan.org ]
08:05 szabgab teh word "with", "connect", "send" "format"
08:05 mo szabgab: ah yes, so that's hard to fix, because I cannot tell perl blocks from normal blocks
08:06 mo see, he put all that in a code block
08:06 szabgab in the description?
08:06 szabgab checking it in source now
08:06 mo yes, it's indented
08:06 mo (probably need to toggle pod :))
08:07 szabgab oh you are right
08:07 szabgab I'll patch that in the source
08:07 szabgab I already made some changes to the methods part but it has not been released yet
08:12 szabgab oh and btw the pictures don't always (or never?) show
08:12 szabgab I'll patch that in the source I see the favicon for a sec and then it is gone
08:15 mo the user pictures?
08:16 szabgab yes
08:16 mo that's gravatar's fault I guess, but I noticed that as well
08:16 szabgab on the same link as I gave it
08:16 mo I was thinking about caching the picture on our server
08:30 osfameron hi
08:31 osfameron the link on http://beta.metacpan.org/module/Pod::PseudoPod to Pod::PseudoPod::Tutorial is broken
08:31 dipsy [ Pod::PseudoPod - beta.metacpan.org ]
08:31 osfameron that also happens on search.cpan.org
08:31 osfameron (broken links all over the shop)
08:31 osfameron was wondering if that was something that metacpan had a plan to improve
08:31 mo the problem is that http://api.metacpan.org/source/CHROMATIC/Pod-​PseudoPod-0.17/lib/Pod/PseudoPod/Tutorial.pod is broken to some extend
08:31 dipsy urgh. long url. Try http://tinyurl.com/67v9dm6
08:31 mo it doesn't provide an NAME section
08:32 mo so there is no way to tell what the module/pod name is
08:32 mo all you know is its filename Pod/PseudoPod/Tutorial.pod
08:32 osfameron in fact... http://beta.metacpan.org/release/Pod-PseudoPod doesn't mention that file
08:32 dipsy [ Pod-PseudoPod-0.17 - beta.metacpan.org ]
08:32 mo I'd rather not join them by ::
08:32 osfameron whereas search.cpan does
08:32 mo well it's there http://beta.metacpan.org/module/CHROMATIC/Pod-​PseudoPod-0.17/lib/Pod/PseudoPod/Tutorial.pod
08:32 dipsy urgh. long url. Try http://tinyurl.com/6akaxb5
08:32 dipsy [ Search the CPAN - beta.metacpan.org ]
08:33 mo but not listed, true
08:33 mo not sure what to do about that, it's imho the author's fault
08:33 osfameron yet search.cpan does show it
08:33 castaway so we should smack chromatic?
08:34 osfameron I think it should be listed under its filename
08:34 osfameron but *also* smack chromatic ;-P
08:35 mo open an issue then please :)
08:38 osfameron https://github.com/CPAN-API/metacpan-web/issues/58
08:38 dipsy [ #58: badly formed POD files not indexed - Issues - CPAN-API/metacpan-web - GitHub ]
08:38 mo osfameron: thanks
08:39 castaway osfa++
08:39 clintongormley joined #metacpan
08:40 mo clintongormley: heya!
08:40 clintongormley heya mo
08:40 mo clintongormley: you have some backlog to read ;)
08:40 clintongormley oh dear :)
08:40 mo how was bbuzz?
08:40 castaway hey clinton
08:40 clintongormley it was good - i felt a bit out of my depth though
08:43 clintongormley i'll check out the backlog - where is it?
08:44 mo ^^ topic
08:45 clintongormley i don't see it in the topic
08:45 mo http://irclog.perlgeek.de/metacpan/
08:45 dipsy [ IRC logs - Index for #metacpan ]
08:45 clintongormley ta
08:46 szabgab do you plan supporting ~pauseid   and redirect them to /author/pauseid ?
08:47 clintongormley btw, i was looking at it on my mobile, and it doesn't look too good on the mobile, i think because you have a hard coded left margin of 175px, (not sure if that's the cause)
08:47 mo sco.beta.metacpan.org/~perler/
08:47 apeiron clintongormley, It's in the /topic. Get a better client. :)
08:47 clintongormley pidgin--
08:48 apeiron pidgin--
08:48 mo szabgab: sco.metacpan.org/~perler/
08:48 clintongormley hmm, apeiron if i try /topic it tells me No topic is set. not sure if that is the right command
08:48 apeiron yeah.
08:48 apeiron pidgin has... issues with IRC.
08:49 clintongormley :)
08:49 apeiron People come onto #freenode every now and then complaining about it.
08:49 clintongormley what do you guys use?
08:49 szabgab mo:  what is *this* sco?
08:49 apeiron irssi
08:49 castaway apeiron: no topic here, care to reset it?
08:49 castaway (I think magnet lost it..)
08:49 mo search.cpan.org :)
08:49 apeiron castaway, ah. whoops. yes
08:49 clintongormley hah!
08:49 Topic for #metacpan is now http://metacpan.org | A better search and API for the CPAN | Fork away at https://github.com/CPAN-API/metacpan-web and https://github.com/CPAN-API/cpan-api | http://irclog.perlgeek.de/metacpan/
08:49 apeiron lo siento
08:49 castaway ta
08:49 clintongormley hiya castaway
08:49 szabgab ah :)
08:49 apeiron still, pidgin sucks. :)
08:49 clintongormley what clients do you guys use?
08:49 daxim joined #metacpan
08:50 szabgab mo: so I can assume that will be integrated?
08:50 mo szabgab: right now it's a couple of nginx rewrite rules
08:50 mo but yes, will do that I guess
08:50 szabgab I guess that's enough
08:51 clintongormley mo: should it not be handled by the app so that if the user doesn't exist, it can be handled gracefully? "Did you mean..."
08:51 clintongormley szabgab: you want to open an issue? https://github.com/CPAN-API/cpan-api
08:51 dipsy [ CPAN-API/cpan-api - GitHub ]
08:51 szabgab I can but apparently it is already doe
08:51 mo it redirects to /author/someauthor and if that doesn't exist you get a nice 404
08:51 szabgab done
08:51 apeiron clintongormley, I'm on irssi, btw.
08:51 apeiron scriptable in iperl!
08:52 clintongormley apeiron: will give it a try
08:52 apeiron er. s/iperl/perl/ # o_O
08:52 clintongormley .oO(Apple has released a version of Perl???)
08:52 apeiron OS X does ship with two perls. But you're better served avoiding them; they're used for system stuff.
08:53 clintongormley no apeiron. i was referring to "iperl"
08:53 * apeiron knows
08:54 mo clintongormley: https://gist.github.com/1016369
08:54 dipsy [ monken's gist: 1016369 ? Gist ]
08:54 castaway iperl?
08:54 mo I don't see it
08:54 mo ah got it, the os hash is a string
09:04 clintongormley mo: busy backlogging
09:04 clintongormley re google traffic to beta/search
09:04 clintongormley do you have a google webmasters account?
09:04 mo google analytics?
09:04 dipsy google analytics is pretty, but i don't seem to be getting any actual data
09:04 clintongormley no webmasters
09:04 mo no what's that?
09:04 clintongormley if so you can specify that the domain has changed
09:04 sheepx mo: the dist pages nee a link to http://deps.cpantesters.org/?modu​le=Git%3A%3APurePerl;perl=latest as well
09:04 clintongormley https://www.google.com/webmasters/tools/dashboard
09:05 mo sheepx: that's already a ticket
09:05 sheepx ok :)
09:05 dipsy [ Webmaster Tools ]
09:05 mo sheepx: I think I'll turn the Dependencies header in a link there
09:05 clintongormley mo: add me to analytics please: clinton@iannounce.co.uk
09:05 sheepx mo: yeah, that'd be fine
09:05 mo not using analytics
09:06 clintongormley why not?
09:06 mo not allowed in germany :)
09:06 clintongormley it's not in germany ;)
09:06 clintongormley it's hosted in the UK
09:06 mo and it doesn't give realtime stats
09:06 clintongormley and?
09:06 dipsy and is it still working
09:07 apeiron well, is it?
09:07 clintongormley is it what
09:07 clintongormley working?
09:07 dipsy working is something you feel dipsy's organisation would be prepared to do from a 'management' point of view .
09:07 clintongormley define working
09:07 clintongormley (not you, dipsy)
09:11 clintongormley heh: Bender2 was kicked by apeiron: I don't trust you either
09:35 clintongormley mo: what's happened to the authors search?
09:36 tom_m joined #metacpan
09:37 mo clintongormley: what happened to it?
09:37 mo oh your ticket?
09:38 clintongormley no, there used to be a link on the left to allow you to choose Authors
09:38 clintongormley and it isn't there anymore
09:38 clintongormley neither the tags
09:38 apeiron tom_m, I believe you were saying you wanted better source hilighting?
09:38 mo clintongormley: those were mockup code
09:38 mo no functionality
09:38 clintongormley the tags, but the author search link worked
09:38 clintongormley now there seems to be no way to do an author search
09:39 mo what did it do?
09:39 clintongormley ie if i searched on DRTECH then it showed my modules
09:39 clintongormley but if i clicked on the author search it showed me in the search results
09:39 clintongormley unless i'm remembering incorrectly
09:40 mo will readd it in a bit
09:40 mo can you open a ticket?
09:40 tom_m apeiron: So far the metacpan layout looks great, although personally I find the colours and contrast between bold/normal text harder to read than cpan.org. The links from use XYZ; etc. are a nice touch though.
09:40 clintongormley oh btw, kimchy is planning on adding multi-searches, so you can do multiple searches with one request, and they will be handled asynchronously on the server
09:40 apeiron multiple searches with one request? whaT?
09:40 clintongormley yep
09:40 * apeiron smells premature optimization
09:41 clintongormley why?
09:41 dipsy because
09:41 apeiron well, why does kimchy want to add this feature?
09:41 tom_m Also, the left sidebar has inaccessible content if the browser window is too small (e.g. netbook with any toolbars enabled). A scrollbar or way to disable the floating behaviour would be great, happy to submit a patch if there are no objections.
09:41 clintongormley apeiron: to be clear, kimchy is the author of elasticsearch, and he is planning on adding that feature to elasticsearch
09:42 mo clintongormley: this will especially benefit apps that don't run from the metacpan server
09:42 mo tom_m: I'm planing to add a settings panel where you can swtich off all that javascript magic
09:42 mo there's a ticket for that already
09:42 clintongormley so that, eg, if somebody searches for 'foo' on metacpan, you can do an author search, a module search and a distribution search in one request, which will be done in parallel on the server
09:43 apeiron ah
09:43 apeiron sorry, now that you've clarified it, it makes sens.
09:43 apeiron e
09:44 tom_m mo: thanks - that sounds ideal. github didn't want to show me the existing tickets without logging in :(
09:57 sheepx mo: think you can indent code in module page displays a bit?
09:57 mo I think so :)
09:58 sheepx it'd make things easier to parse if displayed code was over by 1 or 2 characters
09:58 sheepx want a ticket?
09:58 mo yes please
09:58 sheepx aight :)
10:00 sheepx https://github.com/CPAN-API/metacpan-web/issues/60
10:00 dipsy [ #60: make code blocks easier to recognize - Issues - CPAN-API/metacpan-web - GitHub ]
10:02 sheepx mo: https://github.com/CPAN-API/metacpan-web/issues/61
10:02 dipsy [ #61: have normal text blocks use text-align: justify - Issues - CPAN-API/metacpan-web - GitHub ]
10:07 mo sheepx: I don't really like justified text
10:07 sheepx aw :(
10:07 mo it's imho easier to read if the spaces between words are the same in length
10:07 sheepx well, the thing is
10:08 sheepx http://beta.metacpan.org/module/Pack​age::DeprecationManager#DESCRIPTION <- this would be made MUCH more readable, as an example
10:08 dipsy [ Package::DeprecationManager - beta.metacpan.org ]
10:08 sheepx with justified text you know that when the end of the line is at a certain horizontal level, then there's another line in the paragraph
10:08 mo or the author could structure his documentation properly :)
10:09 sheepx and when it ends before that level, then you can go on and read the next paragraph
10:09 sheepx what's what justify is meant for: to easy recognition of paragraph changes
10:09 sheepx anyhow, your site, i can still hack it in via user css
10:11 mo I'm really open to improvements, but we should discuss them with a broader audience to get the best results for everyone
10:12 * sheepx nods
10:12 sheepx don't worry, i won't be heartbroken over this ;)
10:12 mo cool :)
10:12 rafl there's always user stylesheets
10:14 sheepx yeah, i mentioned that already :)
10:27 * clintongormley has finally finished backlogging
10:27 clintongormley i think i should go away more often!
10:27 mo yeah it was a good time :)
10:27 clintongormley mo: great work
10:27 mo thanks :)
10:28 clintongormley and that blog post was spot on - just what you needed
10:28 clintongormley it's done wonders for getting people involved
10:28 mo so what do you think about indexing cpan testers results?
10:28 clintongormley why not
10:28 clintongormley but...
10:28 mo I'm very happy with how things evolve
10:28 clintongormley that's an extra
10:28 clintongormley focus on the primary deliverables first
10:29 mo but cpan testers is part of my schedule
10:29 mo and doing the partial import is almost as work as doing the full import
10:29 clintongormley ok, then cool
10:29 clintongormley as long as it doesn't introduce a blocker
10:29 clintongormley ie consumes 4*$time that you were expecting
10:29 mo it will require ~7gb
10:30 mo but our box has plenty of free mem left, or do you think this is a problem?
10:30 clintongormley that shouldn't be an issue
10:30 clintongormley no, should be fine
10:30 mo awesome
10:30 mo very excited about this in peticular
10:30 mo let's you do awesome stuff with facets
10:30 rafl what is it you're planning to do? how come presenting aggregated results for dists isn't enough and actually indexing reports is necessary?
10:30 clintongormley yeah :)
10:31 mo aggregated is good enough for sure
10:31 mo but the goal of metacpan was to provide a single api to all cpan related data
10:31 mo and it's really not much of a hassle to do
10:31 clintongormley but not necessarily to replace all cpan infrastructure
10:31 rafl i don't have any opinion on this in particular. i was merely trying to avoid you doing something that doesn't have a clear deliverable
10:31 clintongormley rafl: ++
10:31 mo no but to improve accessibility
10:32 rafl cause, as i understand it, it's actually metabase's responsibility to index its stuff and provide querying
10:32 rafl the way i understood the initial metabase plan anyway
10:33 mo yeah that's true
10:34 rafl i think there might be many interesting things to be done with indexing reports
10:34 rafl but it'd appear like it might be more practical to do after metabase actually has structured cpan reports
10:34 clintongormley i agree with rafl on both accounts
10:35 clintongormley also, as you can see from the responses to your blog post, there are a lot of small things that people would like changed, and these are important for user acceptability.  and they can end up taking a lot of time
10:35 clintongormley so i'd make sure that by the end of GSOC you deliver a project that is acceptable to users, and extra features can be added with time
10:36 rafl fwiw, i agree with that in general, but wouldn't go for lengthy tiny user interface tweaks all that much. while it's important, it's something that seems like people could be easily volunteered for, assuming all the necessary infrastructure exists
10:40 mo later
10:42 sheepx mo: browse link here is broken: http://beta.metacpan.org/release/Git-PurePerl
10:42 dipsy [ Git-PurePerl-0.47 - beta.metacpan.org ]
10:43 rafl heh
10:44 clintongormley sheepx: open an issue :)
10:52 sheepx aight
10:53 sheepx https://github.com/CPAN-API/metacpan-web/issues/63
10:53 dipsy [ #63: browsing a dist doesn't work - Issues - CPAN-API/metacpan-web - GitHub ]
10:53 clintongormley sheepx: ++
10:54 rjh related to https://github.com/CPAN-API/metacpan-web/issues/2
10:54 dipsy [ #2: Fix directory listing - Issues - CPAN-API/metacpan-web - GitHub ]
11:38 szabgab what do you guys think about adding the "outline" of each module to the database?
11:38 szabgab and show it on the GUI as well?
11:39 szabgab that is - names of packages, used modules, subs and maybe even variables - in the given file
11:39 osfameron eeeek, variables?
11:39 rafl that's what abstract in META.* is for
11:39 rafl and everything in META is already indexed.
11:39 rafl +1 one on the displaying, though
11:40 szabgab I am talking about module-wise - for code reading
11:40 rafl i just don't know what you mean then
11:41 szabgab http://cpandigger.org/dist​/Dancer/lib/Dancer.pm.json
11:41 rafl osfameron: seen your fp edits. lovely! :)
11:41 szabgab this is what we have in the "outline" view of padre
11:41 szabgab good for code reading
11:41 osfameron like ctags
11:42 rafl now i know what you mean, but i don't have an opinion on it :)
11:42 szabgab osfameron: sort of, yes
11:42 osfameron what would that look like on the browser page?
11:42 osfameron does it make sense outside an editor?
11:43 osfameron rafl: the pure-fp-book?  ah, you're following it, cool!
11:43 szabgab IMHO it is good for code reading
11:43 szabgab I'd imagine a tree-like thing on the right-hand side
11:44 * rafl suspects all patches are welcome and will be considered
12:13 castaway left #metacpan
12:54 hoelzro joined #metacpan
13:12 sheepx http://beta.metacpan.org/searc​h?q=object+git+pureperl+commit <- no result
13:12 dipsy [ Search the CPAN - beta.metacpan.org ]
13:12 sheepx but
13:12 sheepx http://beta.metacpan.org/source/BROQ/Git-Pur​ePerl-0.47/lib/Git/PurePerl/Object/Commit.pm <- clearly exists
13:12 dipsy urgh. long url. Try http://tinyurl.com/5wtx4k4
13:12 dipsy [ Search the CPAN - beta.metacpan.org ]
13:12 sheepx known bug?
13:43 mo Ticket please
13:46 sheepx aight
13:54 szabgab mo: any plans to add code navigation features, eg. clicking on a method call will let you  view the docs of the method and/or jump to the definition of the method
13:57 clintongormley mo: why have you avoided all <h1-6> tags?
13:57 clintongormley not good for SEO
14:00 hercynium joined #metacpan
14:00 alnewkirk left #metacpan
14:00 clintongormley also, you're using XHTML-Strict, but your HTML is not even slightly valid, which throws browsers into Quirks mode
14:04 mstratman joined #metacpan
15:15 sheepx hmm
15:15 sheepx metacpan can't find common::sense
15:15 sheepx that might be a good thing though :v
15:16 rafl it *is*
15:36 perigrin sadly JSON::XS deps on common::sense so you'll have a lot of breakage with that
15:38 rafl it's not like you have to look up all deps of a module on a search site in order to install it
15:40 oalders sheepx: can you open an issue for that? coverage isn't 100% yet :)
15:40 sheepx oalders: already is: https://github.com/CPAN-API/metacpan-web/issues/64
15:40 dipsy [ #64: non-PODded modules cannot be searched for - Issues - CPAN-API/metacpan-web - GitHub ]
15:41 sheepx common::sense "installs" itself by dumping sense.pm from the __DATA__ section of a .PL script
15:41 sheepx hmm
15:41 sheepx i guess it's an issue that POD in .PL scripts isn't picked up?
15:41 perigrin rafl: I thought that cpanminus did
15:42 oalders sheepx: in that case, it's maybe not the same issue?
15:42 sheepx yep
15:42 sheepx already preparing issue
15:42 oalders thx!
15:42 rafl perigrin: it looks up the package name in an index, not something containing most of the package name in a search engine :)
15:43 perigrin Ah fair enough
15:44 sheepx https://github.com/CPAN-API/metacpan-web/issues/65
15:44 dipsy [ #65: dist introspection needs to go deeper - Issues - CPAN-API/metacpan-web - GitHub ]
15:45 oalders rafl: but it does screen scrape sco if it has to
15:45 rafl still quite different from using its search
15:46 oalders true
15:52 oalders sheepx: thanks for digging into that. nice to have more eyeballs on this
15:52 sheepx i'm not really digging :)
15:52 sheepx i've just done the hosts thing and am whining when meta doesn't do like search. does :P
15:52 oalders that meets my loose definition of digging :)
15:53 sheepx :D
16:06 hoelzro left #metacpan
16:06 hoelzro joined #metacpan
16:12 autarch joined #metacpan
16:15 autarch left #metacpan
16:18 autarch joined #metacpan
16:21 autarch so if I'm running the web frontend, any idea how I can get it to show errors on the console?
16:21 autarch right now it seems like it just fails with no output to anywhere
16:22 oalders mo: ^^^
16:23 clintongormley autarch: you talking about elasticsearch errors?
16:23 clintongormley if error responses from the ES server?
16:23 autarch clintongormley: I have no clue where the error is ;)
16:23 autarch I'm not getting any output!
16:23 autarch but I'm _guessing_ it's from my template changes
16:23 clintongormley ah ok, then i don't know
16:24 clintongormley if you want to see what is happening with elasticsearch, then just find where the ElasticSearch object is instantiated and add (trace_calls=>1)
16:24 autarch ok, good to know
16:25 autarch basically, I want the equivalent of Catalyst in debug mode
16:25 autarch lots of output on errors
16:25 clintongormley there must be something in Plack surely
16:25 clintongormley apologies, never used it :)
16:26 autarch well, bizarrely, the code only _warns_ on a view error, and then proceeds to return output anyway
16:28 metacpan joined #metacpan
16:28 metacpan metacpan-web: master Olaf Alders * f6cabef (1 files in 1 dirs): Clarifies how to run a local front end, without running a local API - http://bit.ly/jv8vIn
16:28 metacpan left #metacpan
16:28 dipsy [ Commit f6cabeff96cba0d56616c206d1848ef345a8ea30 to CPAN-API/metacpan-web - GitHub ]
16:29 clintongormley oh heya autarch - i've just remembered who uses that nick :)
16:29 clintongormley we met at YAPC last year
16:29 autarch I do, of course
16:29 autarch yes, I remember
16:30 * clintongormley wonders if that is a good thing :)
16:30 autarch I guess that's up to you
16:31 oalders autarch: you were getting output before your changes?
16:31 autarch oalders: yes
16:31 clintongormley :)
16:31 autarch err, no, I didn't test this particular page
16:33 autarch ok, no it doesn't work even without my changes
16:33 autarch this is the module view
16:33 oalders just a blank page?
16:34 autarch aha, I see the problem
16:35 autarch the "-R ." is a big problem, it kills the server when anything in the checkout dir changes, including cached template files
16:35 autarch it should be -R lib, I think
16:36 autarch arg, the templates repeat huge chunks of the same stuff
16:36 autarch includes!
16:36 sheepx left #metacpan
16:37 oalders autarch: can you try Plack::Middleware::StackTrace->wrap($app);
16:37 oalders in lib/MetaCPAN/Web.pm
16:38 autarch I can do it from the CLI too
16:38 autarch it doesn't seem to help with template errors
16:38 oalders ah, right
16:39 autarch there's also the Debug middleware
16:40 Ranguard autarch: http://beta.metacpan.org/module/P​lack::Middleware::Log::Contextual - I've not used it yet - but might help?
16:40 dipsy [ Plack::Middleware::Log::Contextual - beta.metacpan.org ]
16:41 autarch but that's not enough, because I think in some cases you want the code to act differently
16:41 autarch like in dev mode template errors should be spit out to the browser, otherwise to a log
16:41 oalders for sure
16:42 rjh http://beta.metacpan.org/module/​Plack::Middleware::ConsoleLogger woo
16:42 dipsy [ Plack::Middleware::ConsoleLogger - beta.metacpan.org ]
16:50 autarch rjh: that'd be great if this freaking app logged anything!
16:53 autarch ok, why does the app not see template changes?
16:54 autarch (and if _I_ find this so frustrating, do you think other people will contribute?)
16:54 oalders autarch: is there some caching going on?
16:54 oalders var/tmp/templates/templates/
16:55 autarch yes, but I kind of assume that it looks for changes before reading the cached version
16:55 oalders and, yes, we need to get these obstacles out of the way ASAP
16:55 oalders basic debugging should be available without jumping through hoops
16:55 oalders this is the first week that we really are getting a low-level code review from people, so lots of issues are coming up
16:56 oalders mo is very quick about fixing things, though
16:57 autarch it seems like Template::Alloy doesn't detect template changes automatically or something
16:58 autarch is there a good reason to not just use Template Toolkit?
16:59 clintongormley i'm with you autarch
16:59 tom_m left #metacpan
16:59 autarch I hate TT, but at least i know it works
16:59 oalders i think Template::Alloy is supposed to be faster?
16:59 clintongormley T::A is hyped a lot by the author, but I found a number of incompatibilities with TT
16:59 clintongormley oalders: the author claims it is, but given that a huge chunk of TT is in C, i find it difficult to believe
17:00 oalders TT is already pretty fast. i'm OK with switching to TT
17:00 clintongormley either way, template rendering is usually a tiny part of serving a page
17:01 autarch who cares if it's faster, template rendering is almost never the slow path
17:05 sheepx joined #metacpan
17:10 oalders also, as far as code goes, we should get back to using perltidy
17:10 clintongormley oalders: perltidy++
17:10 autarch please dog yes
17:11 oalders -pbp ?
17:16 oalders autarch: are you able to get past that template issue or is that a blocker for you right now?
17:20 autarch no, it's not a blocker
17:21 mo T::A is faster and it still slows down the whole rendering
17:22 autarch mo: well, sure it's slower than not using templates ;)
17:22 autarch the question is whether TT would make a _noticeable_ impact
17:23 mo bit the whole is site is fast enough imho
17:23 autarch is or is not?
17:23 mo I start the server with -R templates/ -R lib/
17:23 autarch oalders: I don't really care what style, so long as it's in a tidy file
17:23 mo so that works quite well
17:23 autarch mo: yeah, I changed the README.pod to show that
17:23 mo autarch: sweet :) thanks
17:24 autarch but you're optimizing prematurely here
17:25 autarch I think the #1 goal should be to make it as easy as possible for people to contribute
17:25 mo T::A has more vmethods and stuff, I'm just used to it
17:25 autarch in fact I'd say the only reason I care about this project all is because s.c.o is _not_ open
17:28 autarch I think that's why a lot of people care, and they want to contribute
17:33 oalders autarch++
17:34 oalders the easier it is to contribute to, the faster the pull requests will appear
17:34 clintongormley yep
17:34 mo oalders: METACPAN_API defaults to api.metacpan.org so that is redundant :)
17:34 oalders mo: i know.  i set up the dns :) but it wasn't working for me when i tested it
17:35 mo it does for me
17:35 oalders although eventually that should probably include a version string
17:36 daxim left #metacpan
17:41 tom_m joined #metacpan
17:58 mo sheepx: there is a shaded background on code blocks, how bright is your screen? :)
17:58 sheepx let me look up the thing again
17:58 sheepx i'm fairly sure there was literally none
17:59 mo #F9F9F9
17:59 sheepx jesus christ
18:00 sheepx if i lay my head on the table so the angle is specific i can see there is a SLIGHT difference
18:00 sheepx like, really tiny :D
18:00 mo sheepx: ;)
18:00 sheepx (i'd still really prefer indentation though, since that's what i parse by)
18:00 mo sheepx: will fix that
18:00 sheepx thanks :D
18:01 * mo is still overwhelmed by the feedback!
18:03 sheepx haha
18:03 sheepx well, that blog post got well spread :)
18:03 sheepx and i think many people were waiting for something like this for a long time
18:05 sartak god yes
18:05 clintongormley mo: you made the mistake of asking for opinions :)
18:06 clintongormley rafl: you said that scraping cpan ratings was a no-no. was that just being polite, or has that been an actual response from somebody?
18:08 sartak hmm, I want to add tooltips to relative dates so you can get the exact date
18:09 sartak unfortunately metacpan displays dates using the relatizeDate jquery plugin, which means I can't just go in and change it up :)
18:09 sheepx change the plugin?
18:10 sheepx mo: i'm still amazed that there actually is a block
18:10 sartak I'd have to change it upstream
18:10 mo sartak: add a title attribute to the div that contains the date? (not sure if that works thoguh:))
18:11 sartak yeah, that's how to get the tooltip, but where do I do it? I guess I could do it just before calling relatizeDate, for maximum cleanliness
18:11 mo sartak: that relatizeDate plugin has already been altered by me :)
18:11 sartak ha, ok, then I will do it there
18:13 sartak here's my stab at it: https://github.com/sartak/me​tacpan-web/commit/fc041cebf
18:13 dipsy [ Commit fc041cebfdae701a6dfb79f72fb0707d2600bb6e to sartak/metacpan-web - GitHub ]
18:14 mo and does it work? ;)
18:14 sartak ... yes... !
18:14 mo :D
18:15 sartak I expect that I will have lots more patches for you, so I'll get a test environment set up
18:15 mo sartak++
18:15 mo let me know if you need any help
18:16 oalders clintongormley: i asked ask if we could screen scrape cpanratings at some interval that works for him, but i don't have a response on that
18:16 mo there is almost no documentation, so once you figure something out it'll be great if you could document that in place :)
18:16 clintongormley hmmm, i'm getting the feeling that we're making progress and that we're not getting enough co-operation from the existing infrastructure guys
18:16 clintongormley and they don't own that data
18:17 mo they sure don't
18:17 clintongormley as long as we're polite about how fast we scrape, i don't see the issue
18:17 oalders scraping++
18:17 mo ++
18:17 dipsy Thanks!
18:17 oalders jbl already wrote the scraper
18:17 mo at least they could've said "no"
18:17 mo oalders: yes saw that
18:17 clintongormley there is a *potential* issue where the users haven't agreed to have their data on another site
18:17 mo jbl++
18:17 clintongormley but i doubt that is likely to be a problem for anybody
18:18 sartak have you folks gotten anywhere with pulling bug counts out of rt.cpan?
18:18 mo if it is, we are happy to remove their rating from the site
18:18 mo sartak: yes there is a summary file
18:18 clintongormley perhaps we need to send another email saying "guys, time is short, we're making progress and we really want to include this data.  the plan is to scrape your site slowly.  please reply if you have any objection to this"
18:18 mo but we don't have a script yet that parses it and indexes the content
18:19 sartak alrighty
18:19 clintongormley "assent by omission"
18:19 mo clintongormley: cpanratings is down for 2 weeks now
18:19 clintongormley ???
18:19 sartak I work for the peeps who make rt.cpan, just wanted to make sure you're getting what you need
18:19 clintongormley mo: works for me
18:19 mo you cannot add new ratings
18:19 mo you get a 5000
18:19 mo 500
18:19 clintongormley oh right
18:20 clintongormley well, all the better ;)
18:20 mo sartak: good to know :)
18:20 mo a live feed would be great thoguh ;)
18:20 sartak understood
18:20 sartak unfortunately I just quit, haha
18:21 mo ;)
18:21 sartak so beyond pointing you toward what search.cpan.org was using, I don't think I have time to distribute rt.cpan ponies
18:22 mo well, it's good enough I think
18:22 mo one problem is that github limits the api to 5000 requests a day
18:24 clintongormley mo: you may be able to get special permission from them
18:24 clintongormley what do you need  more than 5000 per day for?
18:25 mo I'd like to pull the issue count once per hour or something
18:25 mo metacpan is really all about real-time data :)
18:25 clintongormley on every module hosted on github?
18:25 mo peeps love the instant recent page
18:25 clintongormley hmm, that could be an issue
18:25 mo sure
18:25 mo it's not that much I think
18:25 clintongormley and their API is pretty limited
18:25 mo actually we could run a query against the api and find out ;)
18:26 mo it's good enough for the issue count
18:26 clintongormley what you *could* do is make that a javascript request from the page, so the user is the requestor
18:27 clintongormley or you want to include that data in your ranking calculation?
18:27 oalders the JS Is a good idea, but it would be great to aggregate some github data in es
18:27 oalders it would be great to bump up pagerank on dists with more watchers/forks etc
18:27 clintongormley yeah++
18:27 * clintongormley wonders if Scott Chacon would be the guy to speak to
18:28 clintongormley he spoke at YAPC::EU last year, selling his book, so he'd probably be amenable to suggestions
18:28 clintongormley mo: you want to put together a list of data you'd like from github, then we can try talking to them
18:28 oalders well, they allotted a lot of disk space for gitpan. they seem like they'd be helpful
18:28 clintongormley yes
18:28 sartak mo: the README.pod tells you to use Twiggy, but it's not a dep. should I add it, or is the plack backend not actually important?
18:30 sartak I expect that you need Twiggy since it's all AE up in yo biznass, but I wanted to confirm
18:30 tom_m how about the user's browser requesting the github data, and reposting it back to metacpan? although it'd be polite to check with github first, of course.
18:32 clintongormley tom_m: too easy to game
18:32 sartak I don't like that idea for several reasons.. not least of which is forgery
18:32 clintongormley how about we write a trojan that will create a botnet and make requests from multiple users?
18:32 clintongormley botnet++
18:33 sartak mo: hey wow, it all just works. very nice
18:34 sartak and yes my patch seems to work :)
18:46 oalders mo: have you pushed all changes to submodules? hoelzro was having some issues with getting the API running locally
18:58 mo oalders: I recently did
18:58 mo sartak: twiggy is required
18:59 oalders mo: thx
18:59 sartak okay, I'll add a dep
19:05 hoelzro oalders, mo: I haven't tried since late last night; I intend to give it another whirl after work today
19:05 hoelzro so just have it ready by then ;)
19:05 oalders :)
19:13 tom_m left #metacpan
19:15 sartak aww. I'm trying to make the source code view line numbers into links, like github
19:15 sartak but it doesn't seem to work because they're generated too late by javascript
19:17 sartak here's my attempt: http://sartak.org/tmp/paste/2011-06-098de_AlyU
19:31 autarch it'd be nice to use a grid layout, it's _much_ easier to make changes, the downside being it often po
19:32 autarch downside being it pollutes the HTML a little
19:32 sartak for what?
19:32 autarch the upside being you can actually change the layout without wanting to kill yourself
19:32 dipsy for what is life without either
19:32 sartak for all the ui?
19:32 autarch sartak: sorry, I'm talking about in general, not in response to you
19:32 autarch like I'd like to expand the left side "search bar" width but that means changing many other things cuase it's a floated div
19:39 arcanez joined #metacpan
19:39 arcanez hmmm, http://beta.metacpan.org/source/EALLEN​III/MooseX-Role-AttributeOverride-0.0.​6/lib/MooseX/Role/AttributeOverride.pm raw code link doesn't work
19:39 dipsy urgh. long url. Try http://tinyurl.com/3or4w4x
19:39 dipsy [ Search the CPAN - beta.metacpan.org ]
19:40 autarch arcanez: it looks like maybe none of them work?
19:41 arcanez I'd swear I viewed raw code on metacpan before..
19:41 arcanez but who knows how often it gets updated
19:41 arcanez it isn't sco after all
19:42 oalders http://api.beta.metacpan.org/source/EALL​ENIII/MooseX-Role-AttributeOverride-0.0.​6/lib/MooseX/Role/AttributeOverride.pm
19:42 dipsy urgh. long url. Try http://tinyurl.com/44ot2vx
19:42 sartak one thing we do in hiveminder is display the SHA of what's running on the live site. would be nice here
19:42 oalders mo: can you check the nginx conf?
19:42 oalders it looks like links to the api might still need the beta in them?
19:43 oalders that's the problem with the source link
19:43 oalders and it would explain why i updated the README with the beta url
19:43 oalders ah, yes
19:44 autarch I just made a commit to make those links _not_ hard code the domain name like that
19:44 oalders api.metacpan.org redirects to beta.metacpan.org
19:44 autarch how are we supposed to develop locally?
19:44 sartak fix /etc/hosts, obviously
19:44 autarch sartak: that's why you're not seeing the change, I suspect ;)
19:44 oalders autarch: the source links are actually provided by the api, not by the front end
19:45 oalders so, it shouldn't be hard coded
19:45 autarch oalders: it works fine if I use my local server
19:45 autarch well, in general, the generation of URIs in the templates at all is a bad thing
19:45 oalders for sure
19:45 autarch if this is REST, the URI should be associated with the object data
19:45 oalders i guess i was saying that the api url should be whatever was defaulted to when the front end server was started
19:45 autarch the only URIs in templates should be links for front end chrome, etc
19:46 autarch but this works - http://houseabsolute.urth.org:5000/sour​ce/FLORA/Moose-2.0007/lib/Class/MOP.pm
19:46 dipsy [ Search the CPAN - beta.metacpan.org ]
19:46 autarch it's just the raw link that doesn't
19:47 autarch oh crap, now you can see my jenkins cookie!
19:47 autarch I"m gonna be l33t hacked
19:47 autarch oh wait, I'm high, only _I_ can see my cookie
19:47 oalders sorry
19:47 oalders the *raw* source is provided by the api
19:48 oalders the formatted source is provided by the front end
19:48 autarch I think that should be changed
19:48 autarch the raw source should at the very least have some links at the top anyway
19:48 oalders i'm ok with that
19:48 oalders that would be helpful
19:48 autarch that's one thing I hate about s.c.o, the browse/raw links are _too_ raw
19:48 oalders they are :)
19:49 oalders so, that's a good idea
19:49 autarch the raw view should include a download link that really gives you the pure raw source
19:49 oalders that sounds good to me
19:49 confound joined #metacpan
19:49 confound http://beta.metacpan.org/sour​ce/RJBS/HTTP-Throwable-0.014/ gives me syntax highlighted html instead of displaying it
19:49 dipsy [ Search the CPAN - beta.metacpan.org ]
19:50 autarch the raw links are just broken right now
19:50 autarch I think there may be a DNS issue?
19:51 autarch class="brush: pl; class-name: 'highlight'; toolbar: false;"
19:51 oalders the DNS is wildcard right now
19:51 autarch what is that? that's bizarre
19:51 oalders so i think it might be nginx
19:52 sartak yeah I saw that class thing too. it's configuration for the syntax highlighter, but why it's specified there in that way is beyond me
19:52 autarch sartak: yeah, it's truly bizarre
19:53 autarch it does seem to be legal
19:53 hoelzro left #metacpan
19:53 autarch although the page has a _lot_ of validation errors not related to that class attribute
19:58 autarch is it intended that there can really only be one method per controller?
19:58 autarch the URL structure of using the Module as a path is really problematic
19:58 autarch I can't append anything after the module name because it looks like part of the name
19:59 autarch I guess that's just some URLs, like for the sourc
20:03 oalders arcanez: those raw links should be working now
20:03 oalders was the nginx config
20:03 autarch I'm working on making a "real" raw view & download link
20:04 oalders excellent
20:05 autarch pushed
20:06 arcanez it isn't working in Chrome, but I think it's just being a bastard
20:06 arcanez FF works fine
20:06 perigrin chrome does that sometimes
20:06 arcanez I _think_ I turned off the dns prefetching
20:07 arcanez but it likes to cache ot make things faster-er
20:07 arcanez oalders: thanks
20:07 oalders :)
20:09 anelson joined #metacpan
20:11 metacpan joined #metacpan
20:11 metacpan metacpan-web: master Olaf Alders * 1e73e66 (1 files in 1 dirs): Adds api.metacpan.org to api hostnames.
20:11 metacpan metacpan-web: master Olaf Alders * 2491763 (1 files in 1 dirs): Merge branch 'master' of github.com:CPAN-API/metacpan-web
20:11 metacpan metacpan-web: master commits f6cabef...2491763 - http://bit.ly/kTuJh7
20:11 metacpan left #metacpan
20:11 dipsy [ Comparing f6cabef...2491763 for CPAN-API's metacpan-web - GitHub ]
20:13 autarch oalders: want to look at my pull request?
20:13 oalders autarch: it looks good
20:14 sheepx mo, oalders: is it intentional that there's no link to http://matrix.cpantesters.org/?dist=Test-TCP+1.13 ?
20:14 dipsy [ CPAN Testers Matrix: Test-TCP 1.13 ]
20:14 oalders sheepx: i don't *think* so
20:15 sheepx ok, ticket inbound
20:16 sheepx and it's up
20:16 arcanez when in doubt, ticket
20:16 autarch oalders: by look I really meant, "look, pull, and release" ;)
20:16 oalders autarch: noted :)
20:16 sheepx arcanez: i worked for two years as mmo support
20:16 oalders and initially understood.  am in process of looking still
20:16 arcanez sheepx: that sounds horrific
20:16 sheepx avoiding duplicate tickets is in my blood and bone :)
20:16 sheepx oh it was fun actually
20:17 metacpan joined #metacpan
20:17 metacpan metacpan-web: master Olaf Alders * e7b20cf (6 files in 4 dirs): Merge pull request #66 from autarch/master ... - http://bit.ly/kMtttx
20:17 metacpan left #metacpan
20:17 dipsy [ Commit e7b20cf544979a331a21717c0bfd72f38acdc66c to CPAN-API/metacpan-web - GitHub ]
20:17 arcanez sheepx: you've seen http://www.youtube.com/watch?v=1Q4Ut761FQE then?
20:17 autarch cool, thanks ;)
20:17 sheepx running around with gm chars set to invisible and invinicble, but still attracting mob aggro makes for fun impromptu events
20:17 dipsy [ YouTube - ?WoW Freakout - Password Change?? ]
20:17 arcanez I'm guessing that's your average customer
20:17 sheepx arcanez: no, watching
20:17 oalders autarch: pulled to live site
20:18 sheepx arcanez: due to the nature of the game the average customer i had were people who made stuff like this: http://www.youtube.com/watch?v=8RiXc3XFTu0
20:18 autarch oalders: do you need to restart the live server or something? I don't see my changes there
20:18 dipsy [ YouTube - ?Restless Child?? ]
20:18 sheepx (i'm still a bit in love with that game)
20:19 oalders i had hoped it would restart on its own by seeing changes. not 100% sure how mo deployed it. let me see
20:19 autarch sheepx: what game?
20:19 sheepx ryzom
20:19 sheepx arcanez: that guy is amazing :D
20:22 oalders autarch: how about now?
20:23 autarch yep, now I see it
20:23 oalders ok. it was running in a screen session
20:23 autarch heh
20:23 autarch that's not how you deploy to production ;)
20:23 oalders yeah :)
20:27 confound browse is stil broken
20:27 autarch I wasn't trying to fix that
20:28 autarch but it should probably get the same treatment as I just did for raw
20:28 autarch in fact, the raw view should probably be connected to the browse view
20:30 arcanez o/` the raw view is connected to the..browse view, the browse view is connected to the .. o/`
20:31 oalders the browse can be cleaned up a bit for sure.
20:32 oalders autarch: "Plack::Request::params is deprecated. Use 'parameters' instead. at /home/metacpan/metacpan.org/lib/MetaCP​AN/..//MetaCPAN/Web/Controller/Raw.pm line 18"
20:32 autarch doh, soryr
20:32 autarch easy to fix, I guess
20:33 oalders :)
21:06 mmcleric joined #metacpan
21:23 hoelzro joined #metacpan
21:23 hoelzro left #metacpan
21:31 hercynium left #metacpan
21:37 hoelzro joined #metacpan
21:41 mstratman left #metacpan
22:04 hoelzro left #metacpan
22:08 hoelzro joined #metacpan
22:14 hoelzro left #metacpan
22:22 clintongormley left #metacpan
22:22 metacpan joined #metacpan
22:22 metacpan metacpan-web: master Olaf Alders * cbff8ec (0 files in 0 dirs): Adds .perltidyrc
22:22 metacpan metacpan-web: master Olaf Alders * 551d968 (7 files in 5 dirs): Merge branch 'master' of github.com:CPAN-API/metacpan-web
22:22 metacpan metacpan-web: master commits e7b20cf...551d968 - http://bit.ly/lz7kmT
22:22 metacpan left #metacpan
22:22 dipsy [ Comparing e7b20cf...551d968 for CPAN-API's metacpan-web - GitHub ]
22:27 metacpan joined #metacpan
22:27 metacpan metacpan-web: master Olaf Alders * f286955 (2 files in 2 dirs): Merge https://github.com/dgl/metacpan-web ... - http://bit.ly/mL0Xti
22:27 metacpan left #metacpan
22:28 dipsy [ dgl/metacpan-web - GitHub ] [ Commit f286955b98652fec130052e8eb0c984512267403 to CPAN-API/metacpan-web - GitHub ]
22:28 autarch oalders: maybe that should be just "perltidyrc", people might not notice it as a dot file
22:28 autarch and it's not like they're going to replace their own ~/.perltidyrc
22:28 oalders autarch: thanks. i thought it had to be a dot file
22:29 autarch perltidy -pro=path/to/rcfile
22:30 metacpan joined #metacpan
22:30 metacpan metacpan-web: master Olaf Alders * 0fd38c0 (0 files in 0 dirs): Renames perltidyrc - http://bit.ly/jjOdpj
22:30 metacpan left #metacpan
22:30 dipsy [ Commit 0fd38c0a834eb5354ed081bdc615effc9c8da8a1 to CPAN-API/metacpan-web - GitHub ]
22:32 metacpan joined #metacpan
22:32 metacpan cpan-api: master Moritz Onken * 0190716 (1 files in 1 dirs): fix mime type for scripts without file extension, requires reindex
22:32 metacpan cpan-api: master Moritz Onken * f1ea88b (2 files in 1 dirs): fixes #92
22:32 metacpan cpan-api: master Moritz Onken * ff7109e (1 files in 1 dirs): cpan ratings type / indexer / endpoint
22:32 metacpan cpan-api: master commits eae9841...ff7109e - http://bit.ly/k6DCBS
22:32 metacpan left #metacpan
22:32 dipsy [ Comparing eae9841...ff7109e for CPAN-API's cpan-api - GitHub ]
22:33 oalders autarch: that should be fixed now
22:34 metacpan joined #metacpan
22:34 metacpan cpan-api: master Olaf Alders * 4c43097 (0 files in 0 dirs): Adds perltidyrc - http://bit.ly/mcaaJP
22:34 metacpan left #metacpan
22:34 dipsy [ Commit 4c43097e7a1120fd1348d64ff3e61d89b1027d3a to CPAN-API/cpan-api - GitHub ]
22:38 hoelzro joined #metacpan
22:38 hoelzro hey guys, I'm back
22:38 hoelzro and ready to hack on cpan-api =)
22:38 oalders excellent! you should be able to update your submodules now
22:39 hoelzro unfortunately, I keep getting this error when I try 'git submodule update': https://gist.github.com/1017945
22:39 dipsy [ hoelzro's gist: 1017945 ? Gist ]
22:39 hoelzro =(
22:39 oalders you're still getting that... you've pulled in all the updates?
22:39 hoelzro ugh
22:40 hoelzro I just realized...
22:40 hoelzro I've been pulling from my fork =P
22:40 oalders :)
22:41 hoelzro grrrr
22:41 hoelzro same error!
22:41 oalders let me try
22:42 oalders ok. i get that too
22:44 hoelzro I think I'm just not going to worry about it for now; I'll just use fake annotation data
22:45 oalders that's probably not a bad thing anyway
22:45 oalders gets you thinking about UI
22:45 hoelzro yeah, top-down design
22:45 oalders mo: can you check to see if you've pushed all changes for p5-elasticsearch-model?
23:43 osfameron woot.  metacpan gets better in realtime
23:44 osfameron I think 2 of the things I vaguely mentioned irked me are already fixed!

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