Camelia, the Perl 6 bug

IRC log for #mojo, 2012-01-08

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

All times shown according to UTC.

Time Nick Message
00:07 mire joined #mojo
00:09 bvierra joined #mojo
00:28 marcus https://dl.dropbox.com/u/621​3850/WebGL/nyanCat/nyan.html
00:31 curt joined #mojo
00:32 memowe sri: https://github.com/memowe/mo​jolicious-plugin-subdispatch - first step.
00:32 mire joined #mojo
00:42 tempire oh
00:42 tempire my
00:42 tempire goodness
00:42 purl Gracious me!
00:42 tempire marcus, you just made my decade
00:43 tempire I counter with this: http://fc02.deviantart.net/fs70/f/2011/25​5/e/b/bat_cat_by_neoslashott-d49iof2.swf
00:51 marcus http://dagobah.net/flash/Madness.swf
01:22 mojobot2572 joined #mojo
02:08 j3nnn1 joined #mojo
02:30 bvierra joined #mojo
02:47 sri this is madness
02:49 sri memowe: looks like a good start
03:01 xaka joined #mojo
03:02 sri marcus: didn't want to make the generated apps look like they require too much boilerplate
03:33 d4rkie joined #mojo
03:38 Psyche^ joined #mojo
03:43 c joined #mojo
03:51 GitHub186 joined #mojo
03:51 GitHub186 [mojo] kraih pushed 1 new commit to master: http://git.io/1WuuCw
03:51 GitHub186 [mojo/master] added description to command example - Sebastian Riedel
03:51 GitHub186 left #mojo
03:56 crab memowe: nice
03:57 GitHub152 joined #mojo
03:57 GitHub152 [mojo] kraih pushed 1 new commit to master: http://git.io/ZaUvbQ
03:57 GitHub152 [mojo/master] added usage to command example - Sebastian Riedel
03:57 GitHub152 left #mojo
04:02 sri hmm
04:02 sri i've just tried an automatically generated secret, it's quite ugly
04:05 sri http://pastie.org/3146617 # this is how it would look
04:23 xxtjaxx There is a possibillity to do templates for json ?
04:30 bvierra joined #mojo
04:40 crab sri: what is so ugly about that?
04:41 sri crab: it's distrating boilerplate
05:56 sri crab: you don't agree?
05:56 sri everything else in the generated apps shows a different aspect of mojolicious
05:58 sri app->start shows how to access the application instance, app->secret seems redundant
06:09 nicomen joined #mojo
06:17 xaka what the reason of asking user enter the secret? i mean mojo could generate it automatically and transparently with no warnings
06:20 sri a secure secret would change between restarts, invalidating all signed cookies
06:22 xaka oh, it should be constant value. got it
06:27 rsimoes I want to memoize some of my controllers
06:27 rsimoes What's the best way to return a data structure and automatically defer rendering?
06:27 alnewkirk :/
06:27 rsimoes use stash, have a chained sub do rendering?
06:27 alnewkirk woah
06:28 rsimoes you and your emoticons, mr. alnewkirk
06:28 alnewkirk arent controllers (route-handlers) already memoized?
06:29 Vandal joined #mojo
06:29 rsimoes route maps are cached
06:29 alnewkirk the difference being?
06:29 rsimoes the controller itself isn't memoized
06:29 rsimoes because it's usually fucking with state and not returning anything useful
06:30 alnewkirk i think that depends on how you set it up
06:32 alnewkirk now ... if you're talking about memoizing the route-handler by parameters ... that seems relatively easy
06:32 xaka do you want to memoize it forever?
06:33 sri there can't be a one size fits all solution, controllers are too dynamic
06:34 alnewkirk use base 'SilverBullet';
06:34 xaka oh, one question sri. does mojo creates new instance of controller's class for every request?
06:34 sri if you just want to cache the rendered result for a method+path combo you can do that easily with two hooks
06:34 sri xaka: yes
06:46 GitHub5 joined #mojo
06:46 GitHub5 [mojo] kraih pushed 1 new commit to master: http://git.io/wnihuA
06:46 GitHub5 [mojo/master] a few more controller examples - Sebastian Riedel
06:46 GitHub5 left #mojo
06:56 GitHub77 joined #mojo
06:56 GitHub77 [mojo] kraih pushed 1 new commit to master: http://git.io/Fv0eug
06:56 GitHub77 [mojo/master] fixed Mojo::ByteStream, Mojo::Collection and Mojo::DOM to not be subclasses of Mojo::Base - Sebastian Riedel
06:56 GitHub77 left #mojo
07:07 d4rkie joined #mojo
07:21 Vandal joined #mojo
08:49 bvierra joined #mojo
08:55 sherr joined #mojo
09:54 preflex_ joined #mojo
11:30 * sri yawns
11:48 sri memowe: btw. maybe ->subreq is a better name than ->subdispatch
11:49 sri at least that's what i called it in catalyst
11:49 memowe Good point. ;)
11:49 sri or marcus did
11:49 memowe Is it right to return the tx? I think for most applications, tx->res would be enough.
11:49 * sri doesn't remember
11:49 sri http://search.cpan.org/~bobtfish/Catalyst-Plugin-​SubRequest-0.17/lib/Catalyst/Plugin/SubRequest.pm
11:50 sri memowe: i have no idea, maybe have two helpers
11:51 memowe Two helpers, one with tx, one with tx->res??
11:51 * sri nods
11:51 sri no idea how useful it is
11:52 memowe maybe the method methods could return tx->res while the subdispatch method could return tx.
11:53 * sri shrugs
11:54 sri SubRequest was never really successful though, no idea if SubDispatch is a better name :)
11:54 memowe great... :D
11:55 sri tempire is good with names
11:55 memowe tempire: help! :)
11:56 memowe I think bypass is misleading.
12:02 * sri also wonders how far tempire got with learning how Mojo::UserAgent works
12:03 sri we need more core devs who know their way around there
12:26 alnewkirk joined #mojo
12:38 rhaen oh boys.
12:42 rhaen just another week - and time for mojo's wiki
12:42 rhaen looking forward to it
13:08 rhaen Yeah, this server will be reinstalled - NOW!
13:08 rhaen cya later!
15:16 d4rkie joined #mojo
15:21 noganex_ joined #mojo
15:35 d4rkie joined #mojo
15:43 mire joined #mojo
15:51 GitHub71 joined #mojo
15:51 GitHub71 [mojo] kraih pushed 1 new commit to master: http://git.io/qliwSA
15:51 GitHub71 [mojo/master] optimized cookie jar and improved tests slightly - Sebastian Riedel
15:51 GitHub71 left #mojo
15:57 ki0 joined #mojo
16:37 abra_ joined #mojo
16:54 yko joined #mojo
17:04 yko joined #mojo
17:34 marcus http://www.charlesproxy.com/
17:35 tempire memowe: I think subroute makes the most sense within the context of mojolicious naming scheme.
17:35 marcus I figure something like this could be made easily using mojolicious.
17:35 tempire with subrequest being 2nd place
17:35 * marcus looks at sri's socks example.
17:36 tempire I think dispatch implies that control of the flow is leaving the current context permanently
17:36 tempire though the examples make it clear, the word subdispatch by itself was confusing to me.
17:36 sri tempire: subroute sounds routes specific
17:37 sri but it's really a full redispatch
17:38 tempire I've read through mojo::useragent
17:38 tempire seems pretty self-evident to me.
17:39 sri marcus: only hard part is intercepting ssl
17:39 marcus sri: but you just told me there's at least 5 ways to do that ;-)
17:40 sri yes, i never said it's easy
17:41 sri you'll have to mess with the non-blocking IO::Socket::SSL api, which most people don't even know about
17:42 * tempire wonders what example is being talked about
17:42 sri don't expect me to be able to talk you through it, never done the server side part myself either ;p
17:43 sri https://github.com/kraih/mojo/blob​/master/lib/Mojo/UserAgent.pm#L236
17:43 sri he wants that for the server side
17:43 sri which is quite a bit harder
17:45 marcus actually, I was referring to examples/connect-proxy.pl
17:46 sri that's just sockets, no tls
17:47 marcus guess I was confused by # Minimal connect proxy server to test TLS tunneling
17:47 sri tls tunneling != tls intercepting
17:47 marcus right, but it's the first step, isn't it? :)
17:47 sri one forwards arbitrary bytes, the other has to actually deal with tls
17:48 sri i guess :)
17:50 * sri predicts marcus will get a headache once he starts researching
17:50 sri digging into IO::Socket::SSL will suck, a lot ;p
17:51 marcus going to start by testing charlesproxy to see what it actually dos
17:52 marcus does
17:57 * tempire believes in marcus
17:57 * tempire sprinkles an extra cloud of fairy dust to help him fly
18:00 marcus hah, charles distributes a CA certificate.
18:04 sri btw. i would use a lite app and the daemon as a foundation, since you'll mostly be dealing with http
18:04 sri it's very different from the connect proxy
18:17 bvierra joined #mojo
18:39 abra__ joined #mojo
18:50 rhaen joined #mojo
18:51 rhaen hello world.
18:51 purl hello world. are you here?
18:51 rhaen just back with the new server :) - well, fresh installed one
19:00 kaare joined #mojo
19:44 kaare joined #mojo
19:49 xaka joined #mojo
20:04 netfeed i hate perl sometimes, why doesn't "my $var => 1" give an error?
20:07 marcus netfeed: because it's valid perl? --- => is just a fancy way of writing ','
20:07 kaare joined #mojo
20:08 marcus netfeed: http://perldoc.perl.org/perlop.html#Comma-Operator
20:08 netfeed really? so i'm writing "my $var, 1"?
20:09 marcus yes
20:10 marcus http://www.rocketaware.com/pe​rl/perlop/Comma_Operator.htm better explanation of how it works.
20:11 netfeed see, this is why some people doesn't like perl :/
20:11 netfeed thank you for the explanation
20:11 marcus or c? :)
20:12 netfeed you say potato :)
20:12 marcus I mostly say tomato
20:12 kaare_ joined #mojo
20:12 netfeed so, "int i => 1" would work in C too?
20:14 netfeed i have no real problem with the glorified , but i'd would like to have a warning or something if it's used instead of = in a my or our statement
20:14 marcus I don't think c has =>, just the comma operator.
20:22 kaare_ joined #mojo
20:28 perlite joined #mojo
20:30 troydm joined #mojo
20:30 kaare__ joined #mojo
20:31 mmcleric joined #mojo
20:33 troydm left #mojo
20:50 Therstrium joined #mojo
20:51 marcus joined #mojo
22:10 ki0 joined #mojo
22:20 marcus ooh, new sherlock is out, guess it will be a late night :/
22:49 memowe tempire: thanks! #subdispatch #subroute
23:08 hshong joined #mojo
23:27 marcus #submarine !
23:36 xaka joined #mojo
23:56 j3nnn1 joined #mojo

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