Camelia, the Perl 6 bug

IRC log for #mojo, 2012-01-02

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

All times shown according to UTC.

Time Nick Message
00:12 MojoGuest524 joined #mojo
00:12 MojoGuest524 From: http://www.google.ru/url?sa=t&rct=j&q=mojo​licious%20websocket&source=web&cd=10&s​qi=2&ved=0CG4QFjAJ&url=http%3A%2F%2Fdev.xa​ntus.org%2F&ei=Z-sAT8XzJ9GM-waYmLyuAQ&usg=​AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rja (1 hits)
00:12 MojoGuest524 left #mojo
00:12 MojoGuest70 joined #mojo
00:12 MojoGuest70 From: http://www.google.ru/url?sa=t&rct=j&q=mojo​licious%20websocket&source=web&cd=10&s​qi=2&ved=0CG4QFjAJ&url=http%3A%2F%2Fdev.xa​ntus.org%2F&ei=Z-sAT8XzJ9GM-waYmLyuAQ&usg=​AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rja (2 hits)
00:59 arpadszasz joined #mojo
02:06 d4rkie joined #mojo
03:00 jwang joined #mojo
03:28 MojoGuest226 joined #mojo
03:29 MojoGuest226 foo
03:29 purl bar
03:29 MojoGuest226 zoid
03:29 MojoGuest226 foo
03:29 purl bar
03:30 MojoGuest528 joined #mojo
03:31 MojoGuest528 left #mojo
03:38 Psyche^ joined #mojo
03:40 rsimoes hm, anyone notice their access log doesn't have bytes-downloaded as of a recent version of Mojo?
03:44 * rsimoes wonders if that broke due to mojo or a plack something...
03:45 MojoGuest929 joined #mojo
03:45 MojoGuest929 From: http://www.google.co.uk/url?sa=t&rct=j&q=​html5%20irc%20client&source=web&cd=3&​sqi=2&ved=0CC8QFjAC&url=http%3A%2F%2Fdev.​xantus.org%2F&ei=MSgBT4flPIGC8gOqiMj6BA&u​sg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rja (1 hits)
03:45 sri what access log?
03:45 purl somebody said access log was scrolling by about a page per second
03:45 hshong joined #mojo
03:46 rsimoes I guess it's the plack middleware that handles the access log, yeah?
03:46 sri you tell me
03:46 MojoGuest929 jmm
03:46 MojoGuest929 kbn
03:46 MojoGuest929 /me thinks this is pretty cool.
03:51 rsimoes Oh, wups
03:51 rsimoes I'm just being dumb
03:52 rsimoes gg!
04:02 rsimoes These 206 responses rock for bandwidth savings, sri
04:02 rsimoes <3
04:04 rsimoes "GET /snd/Hop-Frog.mp3 HTTP/1.1" 206 806752
04:04 rsimoes of a 4057824B file
04:04 rsimoes the visitor then switched to another song before finishing
04:05 rsimoes so they didn't get much more than they heard
04:38 driller_work joined #mojo
05:29 alnewkirk joined #mojo
06:00 ear4004 joined #mojo
06:05 buu So um
06:06 buu How do I get from ..
06:11 buu Oh
06:11 buu I was overthinking things.
06:14 inokenty joined #mojo
07:16 d4rkie joined #mojo
07:20 marcus don't do that!
07:23 Vandal joined #mojo
07:32 crab hi marcus.
07:32 purl hi marcus. are you just playing with dotcloud, or did you sign up for paid hosting?
07:32 crab no purl, hi marcus is <reply>
07:33 marcus hi crab.
07:34 marcus crab: recovering from vaction time ? :)
07:34 * marcus has a easy start because of day care planning day.
07:35 crab i think i have recovered by diving in at the deep end.
07:36 marcus crab: sounds good :) I actually worked a few hours yesterday as well, and spent most of new years writing that 2011 recap =)
07:36 marcus planning to bring Eva to Nordaaker offices for lunch today.
07:37 marcus Hmm, Impress.js is really quite shiny.
07:38 marcus would be cool to make a presentation maker for it using Mojolicious...
07:40 crab i tried one of those css transformations/3d/whatnot slideshow thingies (was it reveal.js?) and it looked cool, but sucked to use.
07:40 crab so i've gone back to slidy, which is old and ugly and just works.
07:40 marcus crab: http://bartaz.github.com/impress.js/
07:40 marcus I've been writing my presentations in keynote. mostly because it's really good for editing
07:41 marcus writing my presenttions as markup isn't very appealing to me.
07:42 marcus writing an impress slide stack seems like a painful excersise in guessing content coordinates.
07:44 marcus but I guess also, more importantly, these kinds of slides are too distracting to watchers of your presentations.
07:44 marcus I have a friend who's a phd in psychology, and he says he always gets compliements about his shiny prezi slides, but that nobody has a clue what he spoke about :)
07:46 crab heh
07:46 crab that definitely isn't what i want.
07:47 crab my talk is difficult enough without added distractions for the audience.
07:51 newbie joined #mojo
07:55 marcus :)
07:58 bvierra joined #mojo
07:58 bvierra_ joined #mojo
08:15 batman joined #mojo
08:25 arpadszasz joined #mojo
08:38 bvierra_ left #mojo
09:24 tl joined #mojo
09:50 abra joined #mojo
10:34 abra_ joined #mojo
10:35 MojoGuest496 joined #mojo
10:35 MojoGuest496 From: http://networkedblogs.com/33Knp (39 hits)
10:35 MojoGuest496 hi guys!
10:36 purl OI! i'm a girl ye monkey!
11:08 abra__ joined #mojo
11:26 MojoGuest294 joined #mojo
11:26 MojoGuest294 From: http://blog.kraih.com/perl-is-ready-for-html5 (99 hits)
11:36 Averna joined #mojo
11:37 Averna left #mojo
12:26 tl joined #mojo
12:42 marcus CLOUD CASTLE!
12:45 marcus hi!
13:01 tholen42 joined #mojo
13:09 tl joined #mojo
13:19 tholen42 joined #mojo
13:27 rhaen yeah! Happy 2012
13:27 rhaen hej.
13:28 rhaen sri?
13:28 purl sri is A PONIE!!1! and a  or no more a javascript noob or evil or a spelling guerilla or the temporary channelclown or a snob now or the perl guerilla or a small mouse with a big head aiming for world domination or the evil sith or has no clue about fuzzy logic or no longer a real hacker :/ or a copycat or pimp to many children or a holy man or looks a lot like pitr or puny or making Mojolicious delicious
13:42 memowe Happy new year, rhaen!
13:56 espent joined #mojo
14:02 sri rhaen?
14:02 purl it has been said that rhaen is in desperate need of help
14:03 hshong joined #mojo
14:24 hshong joined #mojo
14:25 d4rkie joined #mojo
14:44 espent left #mojo
14:57 baton8 joined #mojo
15:00 sri seems like everyone is watching sherlock
15:00 sri except for tempire, who is watching community
15:10 kaare joined #mojo
15:15 kaare_ joined #mojo
15:16 kaare__ joined #mojo
15:20 noganex joined #mojo
15:33 Foxcool joined #mojo
15:34 Britzel_ joined #mojo
15:35 GitHub190 joined #mojo
15:35 GitHub190 [mojo] kraih pushed 1 new commit to master: http://git.io/8xVgyg
15:35 GitHub190 [mojo/master] better documentation for alias methods - Sebastian Riedel
15:35 GitHub190 left #mojo
15:35 sri btw. i think it might be time for the first release of the year
15:42 MojoGuest852 joined #mojo
15:42 MojoGuest852 From: http://www.google.de/url?sa=t&amp;rct=j&amp;​q=extjs%20websocket&amp;source=web&amp;cd=6&​amp;ved=0CFgQFjAF&amp;url=http%3A%2F%2Fdev.x​antus.org%2F&amp;ei=T9ABT9jKI4fe4QTzm8yNCA&a​mp;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (1 hits)
15:43 MojoGuest852 safdsa
15:43 MojoGuest852 fdsa
15:43 MojoGuest852 fdsa
15:43 MojoGuest852 fdsa
16:12 ear4004 joined #mojo
16:15 sri hmm, guess it's my turn to release if nobody else is around ;p
17:13 mercutioviz joined #mojo
17:14 Britzel_ HI. I can't figure out how to use helpers like url_for in Text::Xslate. Any hint? Do the need to be registered as functions in T::X?
17:14 Britzel_ +y
17:16 sri that is renderer specific and should be documented in the plugin
17:17 Britzel_ sri: Right. Unfortunately the docs on T::X are not perfect in that respect. Ok, will investigate further with T::X.
17:18 Britzel_ anyway, congrats on mojolicious, really love it. Much easier (for me) than catalyst.
17:18 sri glad you like it
17:23 sri and released 2.42
17:24 ZadYree nice
17:32 GitHub196 joined #mojo
17:32 GitHub196 [mojo] kraih pushed 1 new commit to master: http://git.io/2fP7pQ
17:32 GitHub196 [mojo/master] a few better content method descriptions - Sebastian Riedel
17:32 GitHub196 left #mojo
17:43 gbacon joined #mojo
17:57 MojoGuest27 joined #mojo
17:57 MojoGuest27 From: http://blog.kraih.com/perl-is-ready-for-html5 (100 hits)
17:57 MojoGuest147 joined #mojo
17:57 MojoGuest147 From: http://blog.kraih.com/perl-is-ready-for-html5 (101 hits)
17:59 MojoGuest82 joined #mojo
17:59 MojoGuest147 bla
18:00 MojoGuest147 /whois MojoGuest147
18:00 MojoGuest82 bla ?
18:00 bla left #mojo
18:09 xaka joined #mojo
18:14 MojoGuest239 joined #mojo
18:14 MojoGuest239 From: http://blog.kraih.com/perl-is-ready-for-html5 (102 hits)
18:38 xaka joined #mojo
18:46 preflex_ joined #mojo
18:52 sri btw. the Mojo::Message::Request->clone code could use a review
18:53 batman joined #mojo
18:53 sri we use it for 301 and 307 redirects
18:54 Britzel_ Am I right in thinking that I have to set up a route for static files like css? So, direct a "GET style.css" to a template 'style' which then contains a single line <link... >?
18:55 sri the tutorial explains that
18:55 Britzel_ ok, then I didn't see all tuts. Thx.
18:56 sri http://mojolicio.us/perldoc#TUTORIAL
19:03 Britzel_ ok, I am probably too stupid or too tired to understand that. Thanks anyway. night.
19:04 Britzel_ left #mojo
19:07 marcus yay for the first release of the year
19:10 vel joined #mojo
19:13 sri wonder if app->sessions->secure(1) should be even more secure by not sending session cookie over insecure connections itself
19:13 sri *+s
19:14 sri right now it only sets the secure flag on the cookie
19:15 sri but preventing Set-Cookie over insecure connections would be trivial as well
19:15 * sri pokes crab
19:16 * tempire upgrades to 14.2
19:25 sri i suppose most of the time it won't matter much
19:25 sri but if you're mixing http and https it could make things easier
19:27 xaka joined #mojo
19:31 sri tempire/marcus: thoughts?
19:31 purl "I'm never here, I'm never here, I'm never here  /  never a bird, or a flower in the tree, or the pain of the respect thereof..."
19:34 sri maybe i'm overthinking it
19:34 marcus sri: I'm not sure of the implications
19:35 marcus sri: crab is probably the right person to ask, since he added secure.
19:36 sri my thought was that the point of the secure flag is kinda to prevent the cookie from being seen on the wire
19:36 MojoGuest85 joined #mojo
19:36 MojoGuest85 left #mojo
19:36 sri which Set-Cookie over HTTP would allow
19:37 sri we have a test case where this happens
19:37 sri https://github.com/kraih/mojo/blob/ma​ster/t/mojolicious/tls_lite_app.t#L85
19:37 tempire are you saying you want the cookie to be sent over a separate https request if they're being mixed?
19:38 sri nope
19:38 sri let me just show you the patch
19:38 sri http://pastie.org/3114117
19:39 sri "+  return if (my $secure = $self->secure) && !$c->req->is_secure;"
19:39 sri that's basically it
19:40 sri tempire: basically, Set-Cookie for the session cookie could never happen over HTTP
19:40 sri right now it could
19:41 sri the secure flag only enforces it client side for Cookie
19:42 sri in a hybrid HTTP/HTTPS app you would have to manually take care of it without the patch
19:43 tempire it sounds reasonable to me
19:43 tempire but I don't think about cookies much, so I'm not sure my opinion means anything
19:43 tempire what are the downsides?
19:43 purl the downsides are obvious, if you don't have the PCI slot, or you can't rig the external pad well
19:43 sri none to my knowledge, that's why i'm asking you ;p
19:44 sri of course, by making it harder to be stupid we lower the bar
19:47 tempire I like that ->secure(1) would mean that the https is required both ways.
19:48 tempire I wonder if it would be good to have a warning happen in debug mode.
19:48 tempire "Set-cookie attempted without secure connection"
19:49 sri i don't think so
19:49 marty joined #mojo
19:50 sri maybe it actually encourages bad design… hmm
19:52 sri when you actually hit this security feature you have a hybrid app that allows login information to be sent via HTTP
19:52 sri so you lost anyway
19:53 sri in this case it would actually mask the flaw
19:57 * tempire finally took the time to understand how session cookies work in mojolicious
19:57 * tempire pretends he did it in 2009
19:58 * sri cracks the whip
19:59 tempire crab's post should be required reading
20:10 sugar joined #mojo
20:20 MojoGuest519 joined #mojo
20:20 MojoGuest519 From: http://www.google.co.uk/url?sa=t&amp;rct=j&a​mp;q=websocket%20irc&amp;source=web&amp;cd=3​&amp;ved=0CDAQFjAC&amp;url=http%3A%2F%2Fdev.​xantus.org%2F&amp;ei=QBECT_qEB-Hb4QTUqr1J&am​p;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (1 hits)
20:20 netfeed tempire: link?
20:20 purl link is probably awesome.
20:21 tempire the best part about it is that it's always accessible via google: https://www.google.com/webhp?sourceid=chrome-inst​ant&amp;ix=heb&amp;ie=UTF-8&amp;ion=1#sclient=psy​-ab&amp;hl=en&amp;site=webhp&amp;source=hp&amp;q=​mojolicious%20cookies&amp;pbx=1&amp;oq=&amp;aq=&a​mp;aqi=&amp;aql=&amp;gs_sm=&amp;gs_upl=&amp;fp=91​8a461f06fdbea1&amp;ion=1&amp;ion=1&amp;bav=on.2,or.r_gc.r_pw.,cf.osb&fp=918a46​1f06fdbea1&biw=865&bih=666&ion=1
20:21 tempire "The Advisory Boar: Mojolicious session cookies"
20:22 netfeed thank you
20:22 MojoGuest519 k
20:28 wheelnotkiss joined #mojo
20:29 vel joined #mojo
20:29 wheelnotkiss is there any trick to getting 'morbo' to stop cleanly?
20:29 wheelnotkiss as in, run all END{} blocks as it shuts down?
20:30 wheelnotkiss I tried doing Mojo::IOLoop->singleton->stop (or something like that) but it didn't exit...
20:31 wheelnotkiss (I'm trying to run my app with Devel::Cover loaded, but it only seems to write coverage information at END{} time.)
20:33 mercutioviz joined #mojo
20:34 mercutioviz joined #mojo
20:38 tempire wheelnotkiss: how are you using morbo with your tests?
20:39 wheelnotkiss well, let me be more clear: in my *unit* tests, I'm just using Test::Mojo, and that works fine with Devel::Cover.
20:39 perlite_ joined #mojo
20:39 wheelnotkiss But I'm trying to actually start the app and use it with a browser, while collecting coverage statistics.
20:40 wheelnotkiss The idea, if this works, would be to collect coverage stats while doing end-to-end tests using e.g. selenium, and see what proportion of code I actually cover.
20:41 sri don't even think about changing any of the signal handling in morbo, it took months to get it working on all platforms ;p
20:41 wheelnotkiss So I'm using 'morbo --listen https://localhost:3000 script/app'
20:42 wheelnotkiss and to load Devel::Cover, I just use PERL5OPT=-MDevel::Cover
20:43 wheelnotkiss sri: no, my app checks if $INC{'Devel/Cover.pm'} and adds a route whose action calls Mojo::IOLoop->stop.
20:43 wheelnotkiss … except it doesn't exit :-(
20:43 wheelnotkiss I'm sure I'm just doing something stupid… the docs on Mojo::IOLoop all claim it should stop accepting connections and then exit the loop.
20:44 sri morbo is a separate watcher process, independent of the actual web server
20:46 sri if you want a single process go with --daemon
20:46 sri umm
20:46 sri daemon
20:46 sri ./script/app daemon --listen ...
20:47 grim_fandango joined #mojo
20:48 wheelnotkiss ok, will try!
21:01 bvierra joined #mojo
21:01 bvierra_ joined #mojo
21:11 wheelnotkiss sri: awesome, that worked great.
21:11 wheelnotkiss thanks!
21:41 gbacon joined #mojo
21:48 driller_work joined #mojo
23:19 arpadszasz joined #mojo
23:34 eris joined #mojo
23:59 eris I'm just starting to learn Mojolicious and have a question about specifying routes that I can't find an answer to.  Take this for example:
23:59 eris $r->any('/')->to(controller => 'login', action => 'index')->name('index');
23:59 eris The docs say "The controller value gets camelized" to determine the class name. Why?
23:59 eris From what I can see I'm required to name my classes with CamelCase (which is normal of course), but I must refer to them with lower case when specifying a controller, and that is confusing.  I must be missing something here, but I haven't come across it yet.  I tried the obvious test of just changing the code to be correctly cased to begin with, but it fails:
23:59 eris $r->any('/')->to(controller => 'Login', action => 'index')->name('index');
23:59 eris I would have expected that providing the exact spelling of the class would work. What am I missing?

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