Camelia, the Perl 6 bug

IRC log for #mojo, 2011-08-08

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

All times shown according to UTC.

Time Nick Message
00:27 preflex joined #mojo
00:28 Kulag joined #mojo
00:33 Alias joined #mojo
00:35 preflex joined #mojo
03:54 jwang joined #mojo
04:29 Foxcool joined #mojo
04:57 sri good morning fellow pancake enthusiasts
05:23 jwang morning sri
05:23 jwang finished my port, using mojo template
05:25 sri jwang: hope it was fun :)
05:27 jwang it was :) being able to drop in perl code was great
05:27 tempire ok
05:27 tempire I'm ready for tomorrow
05:27 * tempire rides on a rainbow into next week
05:28 tempire ping elb0w_
05:28 Skylos yay forrainbows.
05:28 tempire ping alnewkirk
05:29 alnewkirk pong tempire
05:29 Skylos I am wearing my attack of the care were shirt today.
05:29 Skylos Its got a rainbow, in a three color screen print sort of way.
05:29 tempire alnewkirk: going to release mojocast #2 tomorrow.  You're invited to the hackernews upvote party.
05:30 alnewkirk cool, where and when
05:30 alnewkirk ... and whats the cast about, sneakpeek
05:30 tempire placeholders, http methods, and formats
05:31 sri you're spoiling it
05:31 Skylos lol
05:32 tempire alnewkirk: round about 9am pst
05:32 tempire maybe a little before
05:33 alnewkirk okay, i'll stay tuned for the link
05:33 sri that's in 10.5 hours?
05:33 tempire yes
05:33 alnewkirk im building a pretty robust helpdesk app using Mojolicious, etc
05:34 alnewkirk major learning curve, alotta trial and error
05:35 alnewkirk this is what RoR alleviates, ... with its coc lol ... no pun intended
05:35 alnewkirk once I have a winning formula, i think i'll right a rad cli to streamline it
05:36 sri wonder if a little spoiler tweet for the upcoming mojocast would be a good idea
05:38 tempire lull.  Now you're just messing with me.
05:39 Skylos keep it on the tight.
05:39 Skylos wow them with thefinished product in the morning.
05:42 sri oh, falling skies season finale is also out, this looks like a great week :)
06:01 perlrocks Twitter: "Mojolicious:: Lite + OAuth:: I'm trying to use Twitter OAuth in Lite, just work out a 401 Unauthorized ... ..." (ja) --nyarla http://twitter.com/nyarla/​status/100446448163889152
06:19 koban joined #mojo
06:23 perlrocks Twitter: "Library alrighty, Mojolicious finally:: Lite + OAuth:: Twitter can be authenticated by the amount of what has become Lite." (ja) --nyarla http://twitter.com/nyarla/​status/100451953670230016
06:38 perlrocks Twitter: "What tastes better Mojolicious or Dancer?)" (ru) --Greyhard http://twitter.com/Greyhard​/status/100455728300560385
06:47 perlrocks Twitter: "Mojolicious:: To get your own domain I have running in Dahlonega Lite is not I do?" (ja) --nyarla http://twitter.com/nyarla/​status/100458122975522816
06:58 perlrocks Twitter: "Mojolicious:: to take the hostname running the application on the Lite is I, `$ self-> req-> url-> ish I use the base`." (ja) --nyarla http://twitter.com/nyarla/​status/100460808005365760
07:14 arthas joined #mojo
07:25 GitHub52 joined #mojo
07:25 GitHub52 [mojo] kraih pushed 1 new commit to master: http://bit.ly/qt7fpe
07:25 GitHub52 [mojo/master] improved camelize - Sebastian Riedel
07:25 GitHub52 left #mojo
07:35 * sri wonders if it's time to use perl 5.10+ features in documentation
07:35 sri i'm so sick of print "\n"
07:37 jwang go for it. if print "\n" still works, people should 'get it'
07:39 sri might seem like an easy decision for you, but we've tried it a few times before with horrible backlash :S
07:40 jwang is the backlash valid?
07:40 jwang just put a disclaimer on the docs
07:40 jwang then point people to it
07:40 sri hehe… already doing that ;p
07:41 sri http://mojolicio.us/perldoc/Mojolicious/Guides/​FAQ#Why_is_using_Perl_52E82Ex_such_a_bad_idea3F
07:41 sri https://github.com/kraih/mojo​/blob/master/Makefile.PL#L87
07:42 jwang what year did it EOL?
07:42 sri a few ago
07:42 sri heck, 5.10 is now EOL
07:42 jwang might not be a bad idea to put in the actual year
07:42 jwang to make people feel more guilty for complaining ;)
07:45 jwang what's the deal between mojo and dancer anyway. I've just been cruzing some old HN/PM/UP posts
07:46 cosmincx joined #mojo
07:46 jwang was just reminded of that when I saw the 'competitive' comment regarding 5.8 support in the FAQ
07:46 sri dunno what you mean
07:47 jwang in the faq it says, "While we have considered officially deprecating Perl 5.8.x support in Mojolicious altogether, it would sadly be too much of a competitive disadvantage at this point in time."
07:47 jwang I took that to mean competitive with other frameworks
07:47 sri i don't care about dancer, since they don't invent
07:47 sri was talking about catalyst
07:48 sri and LWP
07:49 sri only reason i poke the dancer folks every now and then is because sukria accused me of stealing ideas from him once, which is ridiculous
07:50 sri otherwise i wouldn't even look
07:51 sri last time we tried to move away from 5.8 it was still the default perl on centos
07:52 jwang yeah, lots of people use rhel/centos
07:52 sri that's the real problem
07:52 jwang it's the official distro for zynga
07:53 jwang I'm surprised more people don't build their own perl
07:54 Alias I build my own, but then I distribute it to 50,000 people :)
07:54 Alias It's just an efficiency thing
07:54 Alias It would take those 50k people 50,000 times longer to all build their own perls
07:54 Alias Just not efficient
07:55 sri windows is no big deal, people are used to stuff not working there and having to update every other day :D
07:55 jwang Alias: was just reading your review of mojo vs dancer and how dancer wouldn't install on windows
08:03 perlrocks Twitter: "Mojolicious Google Group – Morbo from Mojolicious 1.72 still doesn’t wish to reload… http://goo.gl/fb/rMlES" --geekspot_ru http://twitter.com/geekspot_​ru/status/100477123898249217
08:04 Alias jwang: There's a long time ago
08:13 spleenjack joined #mojo
08:51 jwang joined #mojo
08:53 sri hmm… maybe a $self->respond_to would make sense for proper REST support
08:54 sri in addition to .html it would also watch for Accept: text/html
08:55 jwang mojo port of robohash is now on github: https://github.com/grokify/Rob​ohash/blob/master/webfront.pl
08:56 sri that looks funny
08:56 jwang what part?
08:56 purl well, part is to move as much of the "how much input string for unpacking do we need?" logic into 1 place
08:58 jwang I'm all for improvement :)
08:58 sri not funny, funny funny
08:59 * sri likes robots
08:59 jwang yeah, like Bender
09:00 jwang I want to use them on grokbase
09:00 jwang which is why I ported it
09:04 Alias joined #mojo
09:21 ajgb joined #mojo
09:27 sri http://pastie.org/2338425 # a first try
09:28 sri it would understand the Accept header
09:28 * tempire supports the idea
09:29 sri i actually have the implementation, just not sure about syntax
09:29 tempire seems like it should accept a mime type as well
09:30 sri ?
09:30 tempire $self->respond_to( 'myawesome/mimetype' => …)
09:30 sri eww
09:30 sri it automatically matches mime types to extensions with app->types
09:31 tempire I guess you want people to define app->types elsewhere in the app
09:31 * sri doesn't follow
09:31 tempire never mind, not important
09:32 tempire does it have to be a sub?
09:33 * tempire is off to bed.
09:33 sri STOP!
09:33 sri you're not getting off that easily now ;p
09:33 sri i expect a proposal
09:34 sri how else could it be done?
09:34 sri i don't see a way
09:37 sri grrr
10:03 perlrocks Twitter: "Export entries from LiveJournal, the transition to Posteroushttp: / / post.ly/2o1bT # api # mojolicious # mongodb # perl # posterous" (ru) --sharifulin http://twitter.com/sharifuli​n/status/100507370400120832
10:38 arpadszasz joined #mojo
10:40 spleenjack1 joined #mojo
10:55 fmerges joined #mojo
11:13 GitHub73 joined #mojo
11:13 GitHub73 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oEhi06
11:13 GitHub73 [mojo/master] added experimental extended REST support to Mojolicious - Sebastian Riedel
11:13 GitHub73 left #mojo
11:14 perlrocks Twitter: "Lets handle #rest a little more according to spec in #mojolicious. :) http://t.co/zcCTtAw #perl" --kraih http://twitter.com/kraih/status/100525394121334785
11:16 sri there are a few restrictions
11:16 sri it will only pick up a format from the Accept header when there is only a single one present
11:17 sri but then it gets higher priority than the format in the route
11:22 GitHub45 joined #mojo
11:22 GitHub45 [mojo] kraih pushed 1 new commit to master: http://bit.ly/pvsmkQ
11:22 GitHub45 [mojo/master] fixed typo - Sebastian Riedel
11:22 GitHub45 left #mojo
11:25 sri Mojolicious::Types is possibly a bit too limited, i'd be interested in proposals for making it more flexible
11:25 perlrocks Twitter: "Great! Now I don't need to do it myself! RT @kraih: Lets handle #rest a little more according to spec in #mojolicious. http://t.co/MQ9V9DC" --ollej http://twitter.com/ollej/status/100528143907037184
11:28 diegok Hi!, json content-type response is not setting encoding, it's that correct and I should add "charset=utf-8" or it should be added by default?
11:28 sri why would you ever set that?
11:29 sri charset for json makes absolutely no sense
11:29 diegok ^ chrome is not showing correctly
11:29 diegok sri: oh!, why?
11:29 sri then chrome is broken
11:29 sri the json spec contains mechanisms for detecting the charset
11:29 diegok sri: json is "by default" utf8 or what?
11:29 sri and it can only be 5
11:30 sri it's part of the spec
11:30 diegok oh, ok... so, I'll wait then :)
11:30 diegok sri: thank you!
11:31 sri what could be happening is that chrome thinks your json is text
11:31 sri which json simply isn't
11:31 sri people get that wrong all the time
11:31 diegok I'll wait till a real problem happen to do anything. A real problem not being me looking broken chars when debuguing api with my browser :)
11:32 sri well, you can set the charset manually of course, but it's wrong ;p
11:32 diegok sri: hm, it it thing its text is more broken than I thought...
11:32 arpadszasz joined #mojo
11:32 diegok as the header says application/json
11:32 diegok sri: I don't want to do wrong things if there is no real need for it.
11:33 sri wonder who will be the first to ask for quality support in respond_to :)
11:33 diegok and me having problem when looking at json with my browser isn't a real one :p
11:33 diegok quality support?
11:34 sri application/json;q=9, text/html;q=0.9
11:34 diegok oh!
11:34 sri we won't do it!
11:35 sri because according to my research browsers get quality wrong all the time
11:35 diegok well, and the options to be a list so there is a winner when several accept matches.
11:35 sri we only support Accept with one value, multiple will be ignored :)
11:35 sri so we are sure the useragent is sure what he wants
11:35 diegok do browsers support that?, I thought it was only for custom agents (aka api clients)
11:36 sri browsers send q=x crap all the time
11:36 diegok sri: it looks right for my future needs.
11:36 diegok :)
11:36 sri why would an api client allow multiple formats?
11:37 sri it really doesn't make much sense
11:37 diegok sri: well, in a past project I had api responses in json for js usage and xml for old fashion clients :p
11:37 diegok (but, was 5 years ago)
11:38 sri that's fine, so none of the two needed to support both formats at the same time?
11:38 diegok today, there are only two options for me: html and json
11:38 sri Accept: application/json *or* Accept: text/xml is perfectly fine
11:38 sri just not Accept; application/json, text/xml
11:39 diegok well, the way I did on that ocation was just serializing the same structure to both formats on the view part of the cat-app
11:39 diegok right
11:40 diegok we did that using file extension instead of accept header, but worked the same, only one option: /method.json OR /method.xml :)
11:40 sri yea, that's how it worked in mojolicious so far
11:41 sri REST purists will yell at you for that though ;p
11:42 diegok REST purist yell for everything... I had one on my last talk about mojolicious yelling at me about the 4xx and 5xx codes my api example was throwing
11:42 sri haha
11:43 diegok but, not his fault, to much java in his daily job :p
11:49 sri :)
11:53 GitHub141 joined #mojo
11:53 GitHub141 [mojo] kraih pushed 1 new commit to master: http://bit.ly/plDWgC
11:53 GitHub141 [mojo/master] better REST example - Sebastian Riedel
11:53 GitHub141 left #mojo
11:54 sri that looks better
11:54 sri not sure yet if respond_to should to that automatically though
11:54 sri s/t/d/
11:55 perlrocks Twitter: "kraih Lets handle #rest a little more according to spec in #mojolicious. :) http://t.co/zcCTtAw #perl" --doubious http://twitter.com/doubious​/status/100535671827070976
11:55 perlrocks Twitter: "kraih Lets handle #rest a little more according to spec in #mojolicious. :) http://t.co/zcCTtAw #perl" --doubious_code http://twitter.com/doubious_c​ode/status/100535669859958784
12:52 sri hmm
12:52 sri it would deactivate auto rendering
12:52 sri then again, not sure anyone would want to use auto rendering as a fallback for respond_to
12:54 sri is "or $self->rendered(204);" acceptable?
12:55 sri i think the big questions will be what's more common
12:55 sri 204 or auto rendering as fallback
13:18 GitHub39 joined #mojo
13:18 GitHub39 [mojo] kraih pushed 1 new commit to master: http://bit.ly/qjpZOg
13:18 GitHub39 [mojo/master] better description - Sebastian Riedel
13:18 GitHub39 left #mojo
13:20 tabbi joined #mojo
13:20 tabbi left #mojo
13:21 SmokeMachine joined #mojo
13:21 sromanov joined #mojo
13:46 tabbi1 joined #mojo
13:46 cosmincx joined #mojo
13:54 tabbi1 left #mojo
13:56 Skylos joined #mojo
14:06 xaka joined #mojo
14:10 abra joined #mojo
14:13 kaare joined #mojo
14:16 sr joined #mojo
14:19 amoore joined #mojo
14:22 sri maybe ->respond_to(xml => {data => '<hello>world</hello>'}) would be a good idea too
14:23 sri you just pass a bunch of values for an implicid render call
14:26 kappataumu joined #mojo
14:37 ndiewald joined #mojo
14:40 perlrocks Twitter: "OOP in time came to bless and write a package,, Class already: the mainstream was Accessor. Class finally: When I first started using the Accessor, everyone was crazy Moose. When you install the Moose finally, the signs of the epidemic in certain Mojo # perl" (ja) --DQNEO http://twitter.com/DQNEO/status/100577017652723715
14:43 koban left #mojo
14:46 tempire Wow.
14:46 tempire That tweet is amazing.
14:46 sri the japanese make all the best tweets
14:46 tempire It's like the moral of a story told in mojo-speak.
14:47 tempire sri: ->respond_to(xml => {data => '<hello>world</hello>'}) was what I was thinking before I disappeared
14:47 sri ah
14:48 sri on it :)
14:48 marcus epidemic in certain mojo sounds scary.
14:48 sri reminds of a tweet few months back
14:48 sri i tweeted that tweet
14:48 tempire tweet
14:48 sri umm… like tweeted a screenshot of that tweet being tweeted into this channel
14:49 sri it's a miracle we don't have avian flu in here with all the tweeting
14:49 ndiewald Everyone was crazy moose back then.
14:50 Akron Hm ... how do I know, which language a returned l()-Value from M::P::I18n came? And is there a best practice for I18n in Plugins?
14:51 marcus Akron: A Møøse once bit my sister ...
14:53 SmokeMachine_ joined #mojo
14:56 kappataumu Hi all, i haven't been able to pass ioloop.t for quite some time, any idea why? W2k3, Strawberry 5.12.1
14:56 kappataumu The most recent Mojolicious hangs at 8/11...
14:56 sri kappataumu: never been reported
14:57 sri kappataumu: have you tried the latest release?
14:57 kappataumu 1.72 as of now hangs still.
14:57 sri no idea then i'm afraid
14:57 kappataumu Tried to upgrade 1.55 which too was failing at 9/11
14:57 tempire Mojolicious hangs 3 days in advance.  It's the future!
14:57 sri wait
14:57 sri 1.55 did hang too?
14:58 marcus sri: is it ok to do something like $r->route('/info/:action')​->to(controller=>'info'); if I just want to render a bunch of templates?
14:58 kappataumu yes, i forced notest to install.
14:58 sri that is weird
14:58 kappataumu This also happens on a Windows 7 64bit machine
14:58 sri must be something very sepcific to your windows or perl version
14:59 sri because we've had successful test reports
14:59 kappataumu Two different physical machines, Perls.
14:59 kappataumu Can i do something to debug this? Help you out maybe?
14:59 sri maybe it's strawberry specific
14:59 sri i think recently we've had more activeperl users again
15:00 sri because of 5.14
15:00 kappataumu I can try activeperl again, but i prefer Strawberry
15:01 sri well, you can always try to fix it :)
15:01 sri but i won't be of much help
15:02 sri if it's only the ioloop.t test it can't be something serious though
15:02 kappataumu some cursory googling earlier showed at least one other person hitting the same thing
15:03 sri now that you say it, i think there was something
15:03 kappataumu under cygwin, but still
15:03 kappataumu see here:
15:03 kappataumu http://groups.google.com/group/mojolicio​us/browse_thread/thread/6ad3898e8a96198f
15:03 sri right
15:03 crab marcus: yes, it is
15:04 sri that propsed change was very harsh
15:04 sri *+o
15:05 sri but if you can verify that it works we can talk about it
15:06 sri the code is now in Mojo::IOLoop::Client
15:06 kappataumu ok, i will monkeypatch it and see if it works. I'll also try with activestate, just for kicks..
15:14 arthas joined #mojo
15:30 GitHub163 joined #mojo
15:30 GitHub163 [mojo] kraih pushed 1 new commit to master: http://bit.ly/mSrehe
15:30 GitHub163 [mojo/master] added many more features to respond_to - Sebastian Riedel
15:30 GitHub163 left #mojo
15:31 sri ok, that looks a lot better :)
15:31 sri http://mojolicio.us/perldoc/Mojo​licious/Lite#Content_Negotiation
15:32 sri isn't it cute?
15:33 sri doesn't get much more restful than that :)
15:33 tempire ok, who's ready for upvoting?
15:33 sri ohoh
15:33 sri the channel was a bit silent today
15:34 Skylos youthink
15:34 Skylos Give me that link
15:34 * sri pokes crab and marcus
15:35 marcus upvote what where?
15:35 sri mojocast!
15:35 sri i think it is release time soonish
15:35 marcus ooh
15:35 tempire sri: 5 up votes needed to stay on the first page, right?
15:35 marcus hn?
15:35 purl hmmm... hn is Hacker News (http://news.ycombinator.com)
15:35 sri tempire: i have no idea, it changes a lot
15:36 tempire xantus[]: ?
15:36 marcus to the trollmobile!
15:37 alnewkirk tempire, ... its up?
15:37 alnewkirk just got back
15:37 tempire momentarily...
15:38 * sri pokes the crazy germans (DaTa, rhaen, Akron, fhelmber_, plu)
15:39 * sri warms up the trollmobile
15:40 sri now you're losing momentum
15:41 tempire go go go!
15:41 marcus tick tock
15:41 tempire http://hackerne.ws/item?id=2860415
15:42 kappataumu :)
15:42 kappataumu First upvote was mine ;)
15:43 tempire 5 upvotes!
15:43 sri where is the tweet?
15:43 purl somebody said the tweet was out there
15:43 sri i need to retweet before i can tweet myself...
15:43 Skylos how do you upvote?
15:44 sri login and hit the little up arrow
15:44 perlrocks Twitter: "New MojoCast is out! Upvote on hacker news then go watch! Or the other way around :) http://j.mp/qAeSR8 #mojolicious" --marcusramberg http://twitter.com/marcusramb​erg/status/100593199621472257
15:44 Skylos ah, there.
15:44 sri tempire: and now marcus beat you to it!
15:45 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework:  http://mojocasts.com/e2" --hnfirehose http://twitter.com/hnfirehos​e/status/100593373177581569
15:45 perlrocks Twitter: "Mojocast #2: Placeholders, methods, and formats http://t.co/SNPRvmn #perl #mojolicious" --mojocasts http://twitter.com/mojocast​s/status/100593439820881920
15:46 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://bit.ly/rjm0b5" --HNComments http://twitter.com/HNComment​s/status/100593618825379840
15:47 marcus it's on the frontpage now.
15:47 sri haha
15:47 tempire \o/
15:47 tempire \o
15:47 tempire o/
15:48 perlrocks Twitter: "And don't forget to upvote the latest #mojocast over on hackernews! ;) http://t.co/RKxX8H5 #mojolicious #perl" --kraih http://twitter.com/kraih/status/100594180811796480
15:50 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: http://bit.ly/nVxXBa Comments: http://bit.ly/rqp6yy" --HNTweets http://twitter.com/HNTweets​/status/100594639748345856
15:50 xaka joined #mojo
15:51 tempire my delay in tweeting was my moment of panic when someone asked me, "what's the vimeo password to watch the video?"
15:51 marcus lol
15:52 sri lol
15:53 Skylos lamar is thorough that way.
15:54 Skylos he asked me in kanban if I had actually viewed the video before Ivoted it up.
15:54 Skylos I said "naw, I trust Glen more thant hat."
15:55 marty joined #mojo
15:55 * marcus had the password :p
15:55 sri don't forget to submit it to reddit
15:56 sri http://www.reddit.com/r/perl/
15:56 tempire I've got a man for that.  Should be posting it anytime now...
15:56 sri :o
15:56 marcus why not the programming subrredit?
15:56 tempire Skylos: tell theo to get on that.
15:56 alnewkirk shit i missed the ball-drop :\
15:56 * marcus is ignorant on the ways of reddit
15:56 marcus alnewkirk: naah, not to late to upvote
15:57 tempire alnewkirk: it's ok, you're good, go go upvote!
15:57 Skylos theo is sitting in thekanban looking bored, I see him on the cam without his notebook.
15:57 perlrocks Twitter: "Sun eggs Lighthouse: Mojocast # 2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments http://mojocasts.com/e2" (zh-CN) --jjie http://twitter.com/jjie/status/100596418456207360
16:00 perlrocks Twitter: "HNews: Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://bit.ly/nHufQB" --tek_news http://twitter.com/tek_news​/status/100597168699748355
16:00 tempire always good to see support from the Zuid-Holland Province
16:01 sri and you're #1
16:01 sri :)
16:01 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments http://mojocasts.com/e2" --AllSocialThings http://twitter.com/AllSocialTh​ings/status/100597571331960833
16:01 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments http://mojocasts.com/e2" --mytechfeeds http://twitter.com/mytechfee​ds/status/100597572242128896
16:01 perlrocks Twitter: "@2740perday Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments http://mojocasts.com/e2 #autofollow" --sztejk http://twitter.com/sztejk/​status/100597572602826752
16:01 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://mojocasts.com/e2" --hackernws http://twitter.com/hackernw​s/status/100597570782507010
16:01 sri eep
16:02 alnewkirk nice
16:05 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://dlvr.it/fMrdY" --FrikiFeeds http://twitter.com/FrikiFeed​s/status/100598390785712128
16:05 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://j.mp/qrlPd4" --newsycombinator http://twitter.com/newsycombin​ator/status/100598408372432897
16:05 tempire hmm.  the +1's from #1 don't register now that the url for it changed.
16:05 tempire unfortunate.
16:05 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments http://mojocasts.com/e2" --rlxim http://twitter.com/rlxim/status/100598515897602048
16:06 sri huh?
16:06 marcus woot, we're #1 on HN
16:06 tempire used to be mojocasts.com, now #1 is at mojocasts.com/e1
16:06 tempire didn't really think that through initially, I guess
16:06 sri ah
16:07 alnewkirk that is just fucking awesome
16:08 tempire now where are all the other people.
16:09 * tempire lets the power go to his head
16:09 * tempire laughs maniacally
16:09 sri nothing on reddit yet!
16:09 * sri pokes tempire
16:10 * marcus needs to add a slide about mojocasts to his YAPC::EU presentation next week.
16:10 tempire my reddit guy is in a meeting
16:10 tempire sigh
16:11 perlrocks Twitter: "Mojolicious - Thee Web Application Framework for me - … I now present to you … Mojolicious. The Perl Web... http://tumblr.com/xo43z0irv9" --alnewkirkassoc http://twitter.com/alnewkirkas​soc/status/100599913557131265
16:14 perlrocks Twitter: "@Cynicalgrinch RT @newsycombinator Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://j.mp/qrlPd4" --mithunkalan http://twitter.com/mithunkal​an/status/100600796277776384
16:15 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://goo.gl/fb/Qijee #hackernews #kalimba" --kalimbahn http://twitter.com/kalimbah​n/status/100601000569733120
16:17 sri those bastards!
16:17 sri they changed the title
16:17 sri because it has to be equal to the title on the linked page...
16:18 alnewkirk tempire, i want to join the crew, let me know howto
16:19 tempire alnewkirk: upvoted on hacker news, right?
16:19 Skylos I'm trying to use mojolicious to display the results of parsing an XML file generated by Freemind.
16:19 alnewkirk tempire, yup
16:19 tempire then you're part of the crew
16:20 tempire oh, and repost on google+ to all your hip tech friends
16:20 alnewkirk i want to submit a video
16:20 sri aww, no secret decoder rings? :/
16:20 tempire ah
16:20 * alnewkirk doesn't have a g+ invite, .... *hint*
16:20 tempire pm your emailz
16:20 tempire sri: reedit post removed, waiting on karma timeout to repost
16:21 mattastrophe joined #mojo
16:24 Skylos why was reddit removed
16:24 * alnewkirk could kiss sri, ... and all the mojolicious contributers, ... but i wont, i'll just say thanks
16:24 tempire Skylos: 'cause I misunderstood sri's comment.
16:24 tempire but it's being reposted as the original "cloudy goodness" title
16:25 * Skylos ohs.
16:26 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework: Comments:" --YCHackerNews http://twitter.com/YCHackerNe​ws/status/100603834807750656
16:26 perlrocks Twitter: "Mojocast #2 - More cloudy goodness from Perl's Mojolicious Web Framework http://j.mp/rqI1GR" --jinnli http://twitter.com/jinnli/​status/100603886259277824
16:27 perlrocks Twitter: "Mojocasts are screencasts to teach you how to set up Mojolicious applications http://mojocasts.com" --perlbuzz http://twitter.com/perlbuzz​/status/100603959361810434
16:27 sri really unfortunate that they changed the hackernews title, web framework makes more people look and upvote
16:27 tempire there you go, sri.  perlbuzz picked it up
16:27 sri lol!
16:27 tempire yeah.  will have to think about that for the next one.
16:28 tempire There's no reason every episode can't have a cute title
16:29 tempire I wonder if it's an automated change.
16:29 tempire I could change the title on the page.
16:29 sri doubt it
16:29 sri but someone will check the title every now and then i bet
16:30 alnewkirk that is just fucking scary how much information Google has on me
16:30 tempire http://www.reddit.com/r/perl/comments/jcis0/​mojocast_2_more_cloudy_goodness_from_perls/
16:31 sri upvoted!
16:31 sri and commented
16:33 tempire everyone!  storm the reedit!  TO THE MOJOLICIOUS!
16:34 sri and you had it all planned out so well ;p
16:34 alnewkirk do you need an acount
16:34 tempire it was better.  I still go side tracked by excitement.
16:34 tempire I had a list of things to do, but next time I'll have it on my iPad so it doesn't get lost in all the screens.
16:35 sri well, reddit doesn't have to be quick, so it's all fine
16:35 tempire sweet, it's battling for the top spot right now
16:35 sri quite interesting how the hackernews post lost momentum when the title changed
16:36 tempire two more up votes in the past 10 secs
16:36 sri i think a descriptive title is very important
16:36 * sri takes notes
16:37 sri oh, and #1 again :)
16:38 tempire episode one got another disqus like & 3 +1's
16:39 sri oh right, i didn't +1 it before either
16:39 sri wait, that button must be new
16:39 tempire yeah
16:39 tempire well, because the url changed from / to /e1
16:40 perlrocks Twitter: "Mojolicious - Thee Web Application Framework for me http://p.ost.im/p/eRSaAM" --alnewkirkassoc http://twitter.com/alnewkirkas​soc/status/100607210912104449
16:41 tempire it's nearly 10am in california, and it's #1 on hacker news…will be interesting to see the google analytics
16:50 alnewkirk tempire, https://plus.google.com/u/0/113553301350100146228
16:50 tempire You were added to my circles when I invited you
16:51 yko joined #mojo
16:51 sri and blogged too
16:52 sri tempire: hackernews doesn't actually result in too many hits, but signal to noise is great
16:52 alnewkirk ah
16:52 tempire it's exciting.  that's good enough for me.
16:52 alnewkirk i got 30k hits last time i released scrappy and blog about it on HN
16:52 sri it's great :)
16:53 * tempire waits for google analytics to report this time period
16:53 Skylos there must be test tools that use this selector syntax to allow you to assert your output without any browser involved.
16:53 sri my experience with frontpaged posts is more around 5k hits
16:53 tempire indeed
16:53 sri after analytics though, not raw
16:54 tempire Test::Mojo for .t, and mojo get /appurl 'css selector string'
16:54 Skylos nifty.
16:54 tempire er, myapp get /appurl
16:55 tempire for domain-less typing
16:55 sri Mojo::DOM was originally written specifically for testing ;)
16:55 Skylos and now to modify nimbus backroom.  <thunderclouds and angry lightnings>
16:56 alnewkirk by-the-by, i stumble across this in my travels, http://www.gosquared.com/
16:57 alnewkirk havent tried it yet though
16:57 tempire so the hacker news title change is definitely manual.  the other titles don't match the page title exactly.
16:57 tempire or I suppose it could detect keywords
16:58 tempire "Philips LED bulb Tear-down" vs "Remote Phosphors: Philips LED bulb, Tear-down Part II - PowerSource | Blog on EDN"
16:59 tempire well this was exciting.  what other social media sites can we storm?
16:59 dotan joined #mojo
17:00 Skylos Can you get it on perlmonks lists?
17:00 tempire "You can make up a new title if you want, but if you put gratuitous editorial spin on it, the editors may rewrite it."
17:00 tempire (from news guidelines)
17:00 sri hmm
17:00 sri oh permonks might be a good idea
17:01 sri *+l
17:01 sri i've not been there for years though :o
17:01 tempire me neither
17:01 tempire it doesn't even have any rainbows
17:01 Skylos might have an established mailing list base to distribute to though.
17:01 jwang I was going to suggest ironman but it's lready there
17:01 jwang :p
17:02 tempire ironman picked up the blogs.perl.org post
17:02 sri mine should be on it soon too
17:02 jwang I could have used that mojocast yesterday ;)
17:03 jwang but it looks nice - will avoid confusion with the :placeholder .
17:04 sri i'm ready to upvote on perlmonks though!
17:04 tempire is perlmonks an appropriate place?  I thought it was just for questions
17:05 jwang there are non-question sections
17:05 jwang but they're not at the top
17:06 tempire hmm
17:06 tempire I have no idea how to post
17:07 perlrocks Twitter: "@kraih is there a way to check if a stash variable is defined and avoid "Global symbol "$foo" requires explicit package name" #mojo #perl" --shiftycow http://twitter.com/shiftyco​w/status/100614187918704640
17:09 perlrocks Twitter: "@shiftycow http://t.co/xxThJCO" --kraih http://twitter.com/kraih/status/100614507444977664
17:10 sri oh crap, twitter bought another company, #1 is gone
17:11 Skylos I don't think that doc link reallyanswers the question.
17:12 Skylos none of this says what $c->stash('foo') does to the non-existance of foo.
17:19 dotan left #mojo
17:22 sri but it leads to Mojolicious::Controller
17:22 Skylos brutal
17:22 sri the connection is important
17:22 Skylos oh?
17:22 sri he needs to know that the helper is the same as the stash explained in Mojolicious::Controller
17:23 sri http://mojolicio.us/perldoc/​Mojolicious/Controller#stash
17:23 sri and that knowledge leads to the guides, which contain more information about the stash than you ever wanted to know :)
17:24 Skylos whats brutal is rather than answering thequestion,only providing a source of overly plentiful information
17:25 Skylos I suppose you could think 'this guy merely lacks the concept he can look at a man page'
17:25 Skylos but My thought would be 'this guy doesn't think he should have to wade through four levels of documentation to find out a simple answer.
17:25 Skylos and I tend to agree.
17:26 sri the alternative is that i say nothing
17:26 Skylos you can't say 'merely call ->stash('foo')'?
17:26 sri that usually leads to followup questions and so on
17:27 Skylos those are different questions though.
17:27 sri all i can do is help people find the answers
17:27 sri adn of course write them
17:29 sri Skylos: btw. calling ->stash would be wrong ;p
17:29 Skylos you could be a little more explicit though.  "this page mentions stash is part of controller, read about stash on controller man page"
17:29 sri Skylos: i just don't have the time or patience
17:30 Skylos The question isn't answereed by controller page.
17:30 Skylos I had to go to the code itself to find the answer.
17:30 Skylos and then it was only because I understand perl as well as I do.
17:30 Skylos And I find myself wondering
17:30 Skylos Does the formal definition of the ->stash api state that the $foo will automaterialize
17:31 Skylos or was the programming of this to do so in the # Get section an oversight
17:31 Skylos as functioanlly speaking
17:31 Skylos calling an accessor should *NOT* have a side effect.
17:31 Skylos so because of that, it makes me nervous.
17:31 Skylos I think, this might be an undocumented/unsupported feature.
17:32 Skylos Because I learned it through inspection rather than documentation.
17:33 sri i have absolutely no idea what you're taling about
17:33 Skylos That is worrisome.
17:33 sri *+k
17:33 sri and i somehow don't like your attitude
17:34 Skylos I'm a nazi about APIs.
17:34 tempire sri: it's not personal
17:34 Skylos Glen knows.  :/
17:34 tempire he just has an interesting way of communicating.
17:34 tempire Skylos: wall of text
17:34 crab it's not undocumented.
17:35 Skylos link please
17:35 purl it has been said that link is awesome.
17:35 crab i thought links were brutal?
17:35 Skylos no, pointing somebody to not-the-answer is brutal
17:35 Skylos when the actual answer is 2+ steps away from there
17:35 Skylos and not hinting that he has to go on a scavenger hunt.
17:35 sri umm
17:35 sri that's not actually true
17:36 sri i assumed he had that knowledge already from reading the guides
17:36 sri so he could make that connection
17:36 crab well, i don't want to look for a link, but at the very least i think ::Guides::Rendering mentions that behaviour
17:37 * sri still has no idea what behavior you guys are talking about
17:37 * tempire sprinkles glitter on skylos
17:37 tempire functional glittler
17:37 crab sri: stash values turning into $variables
17:37 sri oh, yea, that's all over the guides
17:37 crab it's a bit unexpected, maybe, but certainly not undocumented or unsupported
17:38 sri "Stash values that don't have invalid characters in their name get automatically initialized as normal variables in the template and the controller instance as $self."
17:38 sri with all the bells and whistles
17:38 * Skylos readsrendering
17:39 * Skylos glitters.
17:39 crab i guess i would have answered that question differently too, though. i'd say something like "Use '« if (stash 'x') {»in your template"
17:39 Skylos so a stash key with a value of undef exists
17:40 crab stupid ircii, not getting utf-8 right in the twenty-second century.
17:40 sri "shiftycow: @kraih perfect, thanks :)"
17:40 sri it worked for him
17:40 crab very brutal.
17:41 Skylos and if you get a stash value, it automaterializes with a value of undef.
17:41 crab what does "if you get a stash value" mean?
17:41 jwang joined #mojo
17:41 Skylos $foo = $c->stash('foo');
17:41 Skylos get/accessor activation
17:41 crab you do not have to do that for the variable to exist.
17:42 Skylos this is correct.
17:42 Skylos but if I do it without the variable existing first
17:42 Skylos it then exists.
17:42 SmokeMachine joined #mojo
17:43 crab ...
17:44 crab by magic?
17:44 Skylos by the magic of perl, 'if it is referenced it exists', yes.
17:45 sri must be the magical glitter
17:45 crab wtf are you talking about?
17:45 Skylos you are familiar withthe exists operator, correct?
17:45 sri crab: lol, my thought 10 mins ago!
17:45 crab i'm sorry, i'm really trying to understand, but... what?
17:45 crab yes, i am.
17:46 sri he's misunderstanding something i think
17:46 Skylos State prior to call: $self->{stash} = { bar => 'bang' }; - Call $c->stash('foo') - State after call: $self->{stash} = { foo => undef, bar => 'bang' };
17:46 crab i think maybe he thinks things autovivify in non-lvalue contexts?
17:46 crab Skylos: wrong
17:47 Skylos orly?
17:47 purl YA RLY.
17:47 crab is this what you learned "by inspection", i.e. you haven't actually tried it?
17:47 Skylos 10 years of perl experience.
17:47 sri ohoh
17:47 Skylos I experimented with this stuff along time ago.  Here, I'll verifywitha test.
17:50 sri from my 10 years of irc experience i've learned that once someone uses the years of experience argument they are almost certainly wrong :o
17:50 Skylos lol
17:50 Skylos well, you know, I don't go test EVERY assertion I have about perl everytime I talkabout it.
17:50 Skylos it would rather be a waste of my time.
17:51 crab it's much easier to make assertions without caring whether they're right or wrong?
17:51 sri perl -MData::Dumper -E'my $foo = {youre => "wrong"}; my $bar = $foo->{bar}; warn Dumper $foo'
17:51 sri for closure
17:51 Skylos I care very much.
17:52 Skylos in fact, I'm puzzled as all heckright now.
17:52 Skylos as to how Icould be so wrong.
17:52 * Skylos stares at his test script.
17:52 Skylos it says yo'rue right and I'm wrong, there's not much arguing withthat.
17:52 Skylos I'm trying to figure out why I thought otherwise.
17:52 sri i'm sure there's an xkcd for this
17:53 Skylos But it does mean that calling the accessor isn't going to cause the variable to appear in the template, doesn't it.
17:53 sri http://imgs.xkcd.com/comics/duty_calls.png
17:53 Skylos if its not a key in the hash,then, its not a variable in the template.
17:53 sri nope, you have to initialize it
17:53 Skylos but it might already have a value.
17:53 crab if you use $x{key} as a reference in an lvalue context when it doesn't exist, it'll auto-vivify
17:54 Skylos I see crab, it has to do with sub-hashes, it doesn't apply to the end leaf.  It hought it applied to the end leaf too.
17:54 xantus[] you'll have to check down the tree
17:55 Skylos $a->{b}{c}{d} means $a = { b => { c => {} } }; - d doesn't get involved.
17:56 Skylos so you have to put in your route (or other code), '$c->stash('foo', $c->stash('foo'))' to initialize it?
17:56 Skylos left #mojo
17:56 Skylos joined #mojo
17:56 Skylos curse focus.
17:58 Skylos That pattern of non-lossy initialization is slightly ugly.  :/
17:58 sri there are many ways
17:58 sri ->render(foo => 'bar') is popular
17:59 sri anyway, time to step back and study the tutorial/guides ;)
18:00 Skylos thats destructive of previously created foo.
18:00 Skylos I seriously appreciate the work you've put into the tutorials and guides.
18:00 sri make up your own pattern then, and turn it into a helper
18:01 sri but it is rather common to define values that need to be passed to templates only at one point
18:03 Skylos Hm.
18:04 sri as with most things in perl, it all depends on your own discipline
18:04 Skylos I have a routing policy in an app I was working on where there are any number of different key/value pairs on the url .../application/key/value/key/value/key/value - each of those keys showed up as available variables in myt emplates.
18:05 Skylos I had to battle a solution around referencing keys in my templates that weren't created either through the url or the parsing prior to the render point.
18:05 Skylos I suspect templates aren't expected to mutate the application in the mojlicious framework though.
18:06 Skylos Hm.
18:06 sri don't forget templates are just perl, you can always go no warnings and no strict if it makes more sense
18:06 sri uninitialized is easily disabled
18:07 Skylos mm, yes, indeed.
18:07 sri actually reminds me that i wanted to try making all warnings fatal by default :)
18:07 Skylos I like it.
18:11 Foxcool joined #mojo
18:21 marcus CPANTS data generated with Perl 5.010001, Module::CPANTS::Analyse 0.82_01 and Module::CPANTS::ProcessCPAN 0.77, at 2008-08-13 10:00:03. wtf? :-o
18:21 marcus cpants is running with 3 years old data? :o
18:21 marcus seen ask?
18:21 purl ask was last seen on #perl 54 days, 20 hours, 50 minutes and 55 seconds ago, saying: ttk, waltman: thanks - http://log.perl.org/2011/06​/faster-perldocperlorg.html  [Jun 14 21:30:53 2011]
18:23 sri only way to reach ask these days seems to be github
18:23 marcus sri: tried twitter
18:23 marcus he posted there 8hrs ago
18:26 sromanov joined #mojo
18:27 mhi Ok, who in here left this comment on HN? - "Very pretty, is this really Perl?" ;-)
18:27 marcus nailed it!
18:27 marcus :I
18:28 mhi :D
18:29 tempire http://www.youtube.com/watch?v=​hvj8b1Ia1hw&amp;feature=related
18:29 perlrocks Twitter: "1.72 Mojolicious http://ow.ly/1e9vlu" (sw) --campuscodi http://twitter.com/campuscod​i/status/100634684739948544
18:30 tempire exactly the response we're going for
18:30 sri \o/
18:31 sri i still think our slogan should be "Perl: It's not PHP"
18:32 jwang is PHP the competition or Python/Ruby?
18:32 sri PHP has the masses
18:32 sri easier target
18:32 metaperl joined #mojo
18:33 jwang the nice thing about PHP is that most things are compiled in with few things via PECL
18:33 jwang no perlbrew to deal with
18:33 Skylos I thought that waswhy it wa so nasty.
18:34 Skylos everytime you want to install something, you have to recompile.
18:34 jwang or pythonbrew
18:34 jwang what if 99% of users don't want to install something?
18:35 Skylos by making it difficult you perpetuate the existing paradigm instead of improve it.  :(
18:35 jwang I think that's what happened to PHP vs Perl in the way beginning. PHP gave good performance out of the box. Perl's response was "use mod_perl" but 99% of the people didn't want to use mod_perl
18:36 Skylos php got their zend thing going with the IDE and the XDEBUG and such which as tools, compensate for a pile of shortcomings.  :(
18:36 jwang I still remember reading the threads on why forum software was moving to PHP. people hosting forums couldn't get mod_perl running and that was the Perl community's response to them.
18:36 sri a php to perl migration path would be cool
18:37 Skylos php is fairly close to perl.
18:37 Skylos syntactically.
18:37 sri not surprising since it has grown out of perl
18:37 Skylos a bunch of bridge work for that nasty massive public namespace of php functions.
18:37 jwang I still wish you could do $hash[keyname] without ['keyname']
18:38 jwang one benefit of having things compiled in for people that don't test on staging servers is that most things just work
18:38 jwang with Perl and CPAN, it's a must to have a staging server b/c upgrading your Perl modules often breaks the app
18:39 Skylos there needs to bea deployment solution that fixes that too.
18:40 Skylos so manythings to create
18:40 Skylos not enough hours in the day.
18:43 SmokeMachine joined #mojo
18:50 jwang I'm starting to think making Perl easy for the PHP masses would require a very minimal CPAN-deployment, i.e. most things you want you can bake in so people don't have to worry about installing CPAN modules or incompatibilities between versions
18:50 * Skylos nods.
18:52 jwang I've actually run into that myself. If I have a few Perl apps running on a server and I upgrade the modules for one, the other app breaks
18:52 jwang one response is perlbrew
18:53 jwang but the vast majority of PHP people probably don't want a diff perl for each app
18:53 Skylos as long asthe deployment is painless
18:53 Skylos taken care of by scripts
18:53 Skylos they're merely being obstinate.
18:53 Skylos they want it to work, in the end.
18:55 jwang it's the same theme with mojolicious, make it painless
18:57 jwang mojolicous <- no deps
19:02 SmokeMachine joined #mojo
19:04 marcus to please the phpers :)
19:05 marcus jwang: hi btw. migrated from catalyst? ;)
19:08 jwang marcus: have you checked out the new grokbase?
19:08 jwang http://grokbase.com/groups
19:08 marcus jwang: not yet.  is it running on mojo ? :)
19:08 marcus looks very nice. and snappy!
19:09 jwang yes! performance was my top priority
19:09 marcus http://grokbase.com/g/lists.scsys.co.uk/mojomojo yay
19:10 marcus jwang++
19:10 marcus jwang: Did you add it to the mojolicious wiki?
19:10 jwang it's not running mojolicious - robohash was actually my first mojolicious app
19:10 marcus ah :)
19:10 marcus but it's super nice.
19:11 jwang thanks :) it's been a ton of work
19:11 marcus jwang: what's the storage backend?
19:11 purl well, the storage backend is irrelevant
19:11 jwang lzf compressed json
19:11 jwang custom NoSQLFS
19:11 jwang which is nosql on the file system
19:12 marcus finally a way to read the xapian ml in style ;)
19:13 marcus jwang: most of my filesystems have no sql
19:13 marcus ;-)
19:13 Skylos left #mojo
19:13 jwang yup, I even found all my posts to the xapian ml from way back when
19:14 jwang I'm still using Xapian but have my own "Xaplicious" wrapper to make it perlish
19:14 jwang I have an aversion to using Search::Xapian ;)
19:14 jwang the related discussions is driven by Xapian
19:15 jwang and I may add a Xapian-driven username search
19:16 jwang lol, good point on the filesystems. mostly I have a datastructure on the filesystem and I have a bunch of tools to query it
19:16 jwang the way I figure it, most sql databases just write data to the filesystem ;)
19:17 marcus jwang: Xapilicious sounds interesting. I've been using Search::Xapian, but it's feels like a too thin wrapper over c++
19:17 marcus jwang: any plans to open source it?
19:17 jwang yeah, I've been wanting to do it for a long time
19:17 jwang I always feel like it's good enough for me to use but not good enough code to release
19:18 jwang but that probably shouldn't be an excuse
19:19 marcus probably just needs a bit of cleaning up and some pod ;)
19:19 marcus or just stuff it on github and solicit patches to make it cpan-worthy ;)
19:20 jwang yeah either one, I'm pretty happy with it now actually. it's my 3rd wrapper ;)
19:20 jwang each major Xap object has a Moose wrapper
19:21 * marcus is ruined today. squash for the first time in a month or so.
19:21 marcus jwang: sounds good.
19:28 jwang nice, I haven't played squash in years
19:28 jwang but it was super fun
19:28 jwang hardball or softball?
19:29 jwang I'm guessing soft. reading on wikipedia that hardball has fallen out of favor
19:29 jwang I always thought soft was much more challenging
19:30 jwang marcus: I'll probably clean up Xaplicious and put it on github for comment first. then to CPAN
19:31 marcus dunno what hardball is. we play with a rubber ball with 2 dots :)
19:31 jwang apparently hardball was only popular in north america
19:32 marcus jwang: tell me when it's on github, I'd love to look at it.
19:32 jwang I'm guessing widebody raquets killed it. when tennis raquects started getting wider, so did squash raquets. you could literally just touch the ball and it would go flying
19:33 marcus sounds easier =)
19:35 jwang yes much. I think it became too easy
19:36 jwang marcus: definitely. I've been meaning to open it up for a while so now I have incentive :)
19:42 SmokeMachine joined #mojo
20:02 sri hmm
20:03 sri wonder if we should just make all warnings fatal everywhere
20:03 SmokeMachine joined #mojo
20:08 sri whatever, lets try it :)
20:09 GitHub180 joined #mojo
20:09 GitHub180 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oBFYFW
20:09 GitHub180 [mojo/master] made all warnings fatal - Sebastian Riedel
20:09 GitHub180 left #mojo
20:21 sri wonder if uninitialized should be excluded
20:32 trone_ joined #mojo
20:40 baton8 joined #mojo
20:40 xaka wow, warnings are fatal, holly molly
20:41 xaka wonder if my project won't run anymore for a long time :-D
20:42 sri unless you really screwed up i doubt it
20:43 sri if you used stuff that causes warnings intentionally you would have "no warnings" anyway
20:48 SmokeMachine joined #mojo
20:50 GitHub109 joined #mojo
20:50 GitHub109 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oE6hfT
20:50 GitHub109 [mojo/master] better respond_to examples - Sebastian Riedel
20:50 GitHub109 left #mojo
20:51 xaka hm, about Mojo::Base and "has". It doesn't allow me to use [] or {} as a default value, asks for constant or code ref, but write sub { [] } is really ugly and too many letters
20:52 sri it's not a choice
20:53 sri sub { [] } serves a purpose, it's not for fun
20:54 sri has foo => [1, 2, 3];
20:54 sri that's pretty, but could never work
20:54 xaka [] or {} could be the same constant as number of string, sub {} make sence in case of you need to do calculations
20:55 sri you would end up with a singleton
20:55 sri anyway, feel free to dig into the code, make it work and send patches :)
20:55 xaka oh, ok, i got your thought, it'll share the same ref between all instances
20:55 * sri nods
21:02 perlrocks Twitter: "Selection of materials on Mojolicious (version 2.0, will be added): - Run Mojolicious / PSGI application: min ... http://bit.ly/qvjjAh" (ru) --wujasoxahe http://twitter.com/wujasoxah​e/status/100673194985398272
21:09 tempire last on the front page
21:09 tempire about to be bumped
21:09 * tempire cries
21:09 tempire 38 upvotes, not too shabby
21:13 perlrocks Twitter: "Made some more adjustments to #rest handling in #mojolicious, quite happy with it now, good foundation. :) http://t.co/uXUw2by #perl" --kraih http://twitter.com/kraih/status/100675946444619776
21:14 sri tempire: perl post on #1, that was fantastic! :)
21:14 arthas joined #mojo
21:14 tempire actually, we got two tweets from ycombinator twitter feed
21:14 tempire when the name was changed, it reposted
21:14 sri i think it would have gone higher without the title change
21:15 tempire probably
21:15 tempire well
21:15 tempire there's nothing sexier than placeholders and formats
21:15 tempire those two words make me want to have babies right now
21:15 * tempire makes babies
21:17 sri formats are dead, long live respond_to!
21:20 sri btw. "any" is a fallback that eats everything
21:22 xaka is it possible to move that logic to render? which is common used
21:22 sri no
21:23 perlrocks Twitter: "@kraih ok, sounds great - thanks! btw upcoming webapp running on mojolicious as an api: http://t.co/SI63MVA" --f055 http://twitter.com/f055/status/100678535794012160
21:29 d4rkie joined #mojo
21:30 perlrocks Twitter: "If you have a project or company using #mojolicious, please add links to the wiki, it really means a lot to us. ;) http://t.co/nXNs5vc #perl" --kraih http://twitter.com/kraih/status/100680358722412544
21:43 arpadszasz joined #mojo
21:49 sri hmm, i have a feeling fatal warnings will go horribly wrong somehow
21:53 xaka i think about Mojo::Base and %{} operator. When you apply it on object, it automatically iterates over declared attributes and generate a hash, even for not yet existence attributes (key => undef)
21:53 xaka it'd be helpful when you make your object model based on Mojo::Base and convert such objects => hash => json
21:54 xaka now, if you never call attribute accessfor, attribute doesn't exist so it's not in the hash
21:55 xaka s/accessfor/accessor/
21:58 SmokeMachine joined #mojo
22:11 xantus[] we need to get tempire a mic pop screen
22:11 xantus[] ;)
22:11 xantus[] http://en.wikipedia.org/wiki/Pop_filter
22:20 xantus[] tempire++ # mojocasts are good
22:23 perlrocks Twitter: "Gonna experiment with making all warnings in #mojolicious fatal, lets see how it goes. http://t.co/hA6QJDz #perl" --kraih http://twitter.com/kraih/status/100693572721971200
22:23 xantus[] uh oh
22:23 xantus[] development mode only I hope
22:24 xantus[] sri: ^
22:24 sri nope
22:25 sri well, making warning not fatal again is like a 3 line change
22:25 sri *+s
22:25 sri new -strict flag makes these kinds of experiments rather easy
22:26 xantus[] cool
22:27 sri i'm almost certain the experiment will end horribly, but should be worth trying out :)
22:28 sri maybe it makes sense to keep warnings fatal for templates
22:28 xantus[] ok, I'm sure you're a robot now, and you want to kill all humans
22:28 sri i never denied that
22:28 xantus[] *evil robotic cackle*
23:44 tempire I had never listened to the audio with headphones, or any sort of real bass
23:45 tempire I think a sock over the mic would do the trick
23:52 xaka used sock :)

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