Camelia, the Perl 6 bug

IRC log for #mojo, 2011-01-14

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

All times shown according to UTC.

Time Nick Message
00:07 verrens left #mojo
00:17 tempire_ left #mojo
00:30 GabrielVieira left #mojo
01:05 ashleydev joined #mojo
01:06 memememomo_ is now known as memememomo
01:08 memememomo is now known as memememomo_
01:29 ashleydev left #mojo
01:34 daviddelikat left #mojo
01:35 daviddelikat joined #mojo
01:46 ashleydev joined #mojo
01:56 megatron left #mojo
01:56 alnewkirk left #mojo
01:58 megatron joined #mojo
01:58 alnewkirk joined #mojo
01:59 josh left #mojo
03:16 ryan left #mojo
04:05 ashleydev left #mojo
04:09 daviddelikat1 joined #mojo
04:12 avkhozov joined #mojo
04:12 daviddelikat left #mojo
04:19 pid1 joined #mojo
04:23 pid left #mojo
04:35 ispy_ joined #mojo
04:57 ashleydev joined #mojo
05:00 daviddelikat joined #mojo
05:04 josh joined #mojo
05:04 daviddelikat1 left #mojo
05:19 ashleydev left #mojo
06:20 ysyrota joined #mojo
06:32 su-bzero1 joined #mojo
06:32 ltriant left #mojo
06:37 su-bzero_ left #mojo
06:51 su-bzero is now known as su-bzero[A]
06:58 kaare joined #mojo
07:02 kaare_ joined #mojo
07:03 kaare left #mojo
07:42 Christian joined #mojo
07:42 esskar chansen, sorry!
07:42 Christian good morning
07:42 purl For you maybe.
07:45 koban joined #mojo
07:48 zipkid hoi
08:00 marcus hey
08:00 marcus is client process deprecated?
08:04 pid1 left #mojo
08:10 yko marcus: afaik its not deprecated anymore. it's removed in 1.0
08:12 breaker313 joined #mojo
08:16 su-bzero[A] is now known as su-bzero-su
08:24 otaviof left #mojo
08:27 esskar sri, i tested the ...upload.t on windows yesterday; works!
08:32 yko perldoc--
08:35 pid joined #mojo
08:35 yko use Module '3.14' does not the same as use Module 3.14, but no any warnings with that
08:36 breaker313 left #mojo
08:41 su-bzero-su is now known as su-bzero[A]
08:51 pid left #mojo
09:12 su-bzero[A] is now known as su-bzero
09:13 kaare_ left #mojo
09:13 kaare_ joined #mojo
09:18 pid joined #mojo
09:20 MojoGuest129 joined #mojo
09:20 MojoGuest129 From: http://news.ycombinator.com/item?id=1277067 (157 hits)
09:21 MojoGuest129 left #mojo
09:22 megatron left #mojo
09:22 alnewkirk left #mojo
09:39 kaare_ is now known as kaare
09:55 MojoGuest269 joined #mojo
09:55 MojoGuest269 hallo
09:56 MojoGuest269 left #mojo
09:57 vaneska joined #mojo
10:04 spleenjack joined #mojo
10:06 breaker313 joined #mojo
10:31 otaviof joined #mojo
10:32 sri esskar: it failed a lot on cpantesters
10:33 sri marcus: removed after being deprecated for 6 months+
10:33 sri moin
10:34 * yko waves
10:47 * koban waves hello
10:53 marcus sri: mmm. 6month sure flies by :)
10:53 marcus sri: btw, next time you deprecate something, maybe make it give deprecation warnings?
10:54 marcus warn "This method is deprecated and will be removed in a future release";
10:57 * yko warns marcus: Renderer->helper and Renderer->handler are deprecated now
10:57 marcus yko: yay. I use those in plugins. :)
10:57 yko same as Mojolicious->session
10:58 marcus 13     $app->renderer->add_helper($
10:58 marcus I guess you mean those are deprecated?
10:58 yko exactly
10:58 marcus should be  13     $app->helper($ instead?
10:59 yko yea
10:59 yko http://mojolicio.us/perldoc?Mojolicious#handler
10:59 yko http://mojolicio.us/perldoc?Mojolicious#helper
10:59 marcus mm
10:59 marcus would probably be smart to add deprecation warnings to those too
10:59 marcus so people update their plugins
10:59 yko well, they will be deprecated another 6 month
10:59 marcus sure
10:59 marcus but people don't read docs
10:59 yko but... you know
11:00 marcus warn "UR METHOD IS GOIN AWAY"
11:00 yko better to do that now, than get fucked after update to 2.0
11:00 marcus :D
11:00 yko and then came here and cry 'oooh, you broke my app! i hate you!'
11:00 marcus I could do both :)
11:01 marcus this is why I think deprecated methods should warn once they are deprecated
11:01 yko or do thomething like http://korshak.name/ra/awlwrong.png
11:01 * marcus gets the gas mask
11:02 yko marcus: well, meybe it have sense to turn such warnings on month or two before removal
11:02 marcus why not when the method is deprecated?
11:04 yko well, session does so
11:04 yko https://github.com/kraih/mojo/bl​ob/master/lib/Mojolicious.pm#L33
11:05 yko but i think that is about another reason
11:05 marcus for maximum sexy time?
11:05 yko no, people was mixed $app->session and $c->session
11:06 marcus aha
11:07 marcus yay
11:07 marcus proxy fails lots of tests now
11:07 marcus sweet
11:10 crab yko: (reading fatapp) __PACKAGE__->attr(config => {x=>42}) means $app->config will be {x=>42}?
11:10 yko crab yes
11:10 crab sweet
11:10 yko crab: as i told, it's for personal use
11:10 crab yko: well, i'm stealing some bits from it :-)
11:11 yko i'm not completely satisfied with it
11:11 yko crab: ill be happy if you do
11:11 yko btw, dont forget to inform about improvements, so i can steal code from you back
11:12 crab yko: sure.
11:14 s0va help me
11:14 s0va with pod_renderer plugin
11:14 s0va i added plugin 'pod_renderer' in my mojolicious app
11:15 s0va i get 404 when i try to access /perldoc location
11:17 yko marcus: i think it's really be nice if deprecated subs behave like that: https://gist.github.com/779486
11:18 maez left #mojo
11:19 yko s0va: try mojo generate lite_app perldoc_app; ./perldoc_app get /perldoc
11:19 yko if it's work than something in your app breakes perldoc plugin
11:22 ok22 left #mojo
11:27 bc5042 joined #mojo
11:28 kaare left #mojo
11:28 kaare joined #mojo
11:29 bc5042 what is a way to decorate all controllers actions with some another actions - checking session for example?
11:30 yko use base 'Mojolicious::Controller'; sub session { my $result = shift->SUPER::session(@_); PROFIT }
11:31 sri marcus/yko: renderer->add_helper() is not deprecated
11:31 sri renderer->helper/handler have just been renamed to helpers/handlers
11:32 yko oh. marcus, sorry. i'm so inattentive
11:32 bc5042 and make my all Controllers use as a base MyController? cool, thx
11:32 yko i was talking exactly about Mojolicious::Renderer->helper https://github.com/kraih/mojo/blob/ma​ster/lib/Mojolicious/Renderer.pm#L23
11:33 yko bc5042: that's way i use
11:33 sri yko: marcus asked about add_helper
11:33 yko sri, yes. i see now :(
11:33 yko sorry
11:34 sri yko++ # for trying
11:34 yko sri: that doesn't removes my proposition about deprecated subs
11:34 yko have you seen gist?
11:35 sri sure
11:35 sri we can do that
11:36 yko that would be really nice
11:36 yko i heared too much bad words about unexpected (for inattentive user) removal of deprecations
11:36 yko sure we always can say "damn, where was your eyes! it's deprecated for more than year, see changelog!"
11:37 yko but truth is user never do so :)
11:37 * sri wonders how many deprecated methods/attributes there are
11:37 yko 3
11:37 sri lol
11:38 yko i'm talking about to make it practice for future
11:38 sri sure
11:38 yko thanks :)
11:38 sri ah, and one of those 3 already warns :)
11:39 yko ye :))
11:39 yko not because of that reason, but because of other. anyway :))
11:48 marcus process didnt warn before it was removed tho :)
11:57 crab if i say ->attr(x=>sub ...), is the sub called as a method, i.e. get $app in @_?
11:57 crab hmm, looks like
12:01 forwardever joined #mojo
12:01 sri oh funny, we were using the deprecated attributes too :)
12:07 GitHub109 joined #mojo
12:07 GitHub109 mojo: master Sebastian Riedel * 6d7b2a4 (6 files in 4 dirs): warn about deprecated attributes - http://bit.ly/hH2Vw1
12:07 GitHub109 left #mojo
12:08 sri and a typo in the warning :S
12:09 GitHub174 joined #mojo
12:09 GitHub174 mojo: master Sebastian Riedel * 56cb826 (1 files in 1 dirs): fixed typo - http://bit.ly/i3AGN9
12:09 GitHub174 left #mojo
12:18 Christian nice weekend to you all
12:19 Christian left #mojo
12:31 bc5042 my question again: what method in controller should i override (or how do it in other way) if i want to do some checks before any controller actions, and for example redirect to login page if someone made a request to resource without being logged in?
12:33 bc5042 my controller has few actions and i don't want to implement checks in all of them
12:33 sri bc5042: see bridges
12:34 sri you don't overload, you route differently
12:35 bc5042 so info about bridges is in routes guide?
12:35 sri http://mojolicio.us/perldoc?Moj​olicious/Guides/Routing#Bridges
12:35 esskar left #mojo
12:35 bc5042 cool, thx again ;)
12:43 avkhozov left #mojo
12:43 s0va yko: thanks.
12:43 s0va yko: works with mojo 1.0
12:43 s0va yko: doesn't work with 0.999936 :)
12:44 s0va now i'm relaxed :)
12:45 snearch joined #mojo
12:47 esskar joined #mojo
12:50 yko s0va: are you crazy? 0.999936 was released last year! lol, year ago :)
12:51 sri crazy!
12:57 esskar left #mojo
12:57 s0va :)
12:57 esskar joined #mojo
13:00 esskar left #mojo
13:01 s0va just upgraded bundled version of mojo to 1.0.1
13:01 s0va works everything except the part of application witch uses static to send normal files.
13:01 s0va removed ->prefix() method call.
13:01 s0va great!
13:01 s0va well, now there is a problem...
13:02 s0va perldoc renderer sets base href="http://localhost:3002/"
13:02 s0va but application is behind nginx ssl proxy
13:03 s0va is there any way to get rid of this base href in output of pod renderer?
13:05 yko umpfhh...
13:05 s0va hm... looks like that this one is Pod::Simple::HTML
13:07 esskar joined #mojo
13:07 sri there will surely be more plugins in the future that expect a sensible url base
13:07 sri rewriting it in a before_dispatch hook might be best
13:08 yko sri: mb after static dispatch?
13:08 sri didn't we have documentation for this already?
13:08 yko i think no
13:08 sri of course
13:08 sri http://mojolicio.us/perldoc?Mojol​icious/Guides/Cookbook#Rewriting
13:08 yko oooh
13:09 yko you are so big :)
13:09 sri there is no way to detect these reverse proxy setups and do the right thing
13:09 yko damn, that example is really amazing
13:09 daviddelikat1 joined #mojo
13:09 s0va /usr/lib/perl5/vendor_perl/5.12.1/i586-​linux-thread-multi/HTML/HeadParser.pm: $p->header('Content-Base')   # to access <base href="http://...">
13:10 sri you can even make it depend on production mode
13:10 s0va ... If i set Content-Base it should work...
13:10 s0va lemme se
13:10 yko s0va: mojo native way much better
13:11 sri lol Content-Base
13:11 s0va nope, this not it...
13:11 s0va wrong class...
13:12 daviddelikat left #mojo
13:12 sri nobody uses Content-Base
13:12 sri it is garbage some vendor made up
13:14 s0va looks like this could be a problem
13:14 s0va https://github.com/kraih/mojo/blob/master/​lib/Mojolicious/Plugin/PodRenderer.pm#L275
13:14 sri s0va: are you listenting?
13:14 s0va how could i set this tag?
13:14 sri *-t
13:15 s0va sri: yes, i'm listening
13:15 sri see above
13:15 s0va --mode=production could help?
13:15 s0va lemme see
13:15 sri NO!
13:15 sri ...
13:15 * sri goes to write more documentation instead
13:16 s0va i don't want to dispose mojo daemon directly to clients... even if i would like to do that i cant.
13:19 s0va oh, i found it...
13:20 s0va app->helper(base_tag => sub {})
13:21 yko s0va: you look stupid now. please read link sri gave to you: http://mojolicio.us/perldoc?Mojol​icious/Guides/Cookbook#Rewriting
13:21 megatron joined #mojo
13:21 alnewkirk joined #mojo
13:21 s0va app->helper(base_tag => sub {'/'});
13:21 yko -_- ok
13:22 * yko goes write some redis models
13:22 s0va yko: you're right.
13:22 s0va now i look stupid :)
13:26 GitHub106 joined #mojo
13:26 GitHub106 mojo: master Sebastian Riedel * ac2a992 (1 files in 1 dirs): fixed typo - http://bit.ly/fl26yA
13:26 GitHub106 left #mojo
13:40 megatron left #mojo
13:41 ashleydev joined #mojo
13:43 alnewkirk left #mojo
13:44 sri hmm
13:45 * sri wonders if examples should keep using "use base" or switch to "use Mojo::Base"
13:48 ashleydev left #mojo
13:49 ashleydev joined #mojo
13:49 yko i'd keep use base for sometime
13:53 GitHub102 joined #mojo
13:53 GitHub102 mojo: master Sebastian Riedel * b8b071e (1 files in 1 dirs): made Mojo::Base behavior more transparent - http://bit.ly/esnyll
13:53 GitHub102 left #mojo
13:54 GitHub118 joined #mojo
13:54 GitHub118 mojo: master Sebastian Riedel * c564d4a (1 files in 1 dirs): pod fix - http://bit.ly/esVYRt
13:54 GitHub118 left #mojo
13:55 sri http://mojolicio.us/perldoc?Mojo/Base#FUNCTIONS # thoughts?
13:57 jfuller Do you think that "has" will confuse Moose people?
13:58 sri does Class::Accessor::has confuse Moose people?
13:58 sri (no i don't think so)
13:58 jwang left #mojo
13:59 sri if you want Moose behavior, you use Moose
13:59 jfuller Actually I think my original confusion was the other way around, since I knew about Class::Accessor::has before Moose
13:59 sri heh
14:00 sri anyway, if you're using Moose for the accessors you're doing it wrong :)
14:01 jfuller Agreed. I do have some cgi apps that use Mouse for the lazy building of attr though
14:02 sri shame on you! :)
14:02 jfuller lol
14:03 sri oh, class::accessor doesn't do lazy building
14:03 jfuller At least not the last time I checked
14:03 sri shame on class::accessor
14:04 sri isa constraints but no lazy building... usability fail
14:05 crab hmm.
14:05 crab can any controller module have its own inline templates?
14:05 kaare left #mojo
14:05 kaare_ joined #mojo
14:06 sri you might have to pass a template_class stash value or so, but yes
14:07 crab and if i put a file in templates/whatever/blah.html.ep, it'll just override that inline template? must do. sweet.
14:07 sri "inline templates" is a little misleading though
14:07 sri better say DATA templates
14:08 crab what is the right ter... ah.
14:08 crab yes, fine.
14:08 sri since we now do ->render(inline => '<%= 1 + 1 %>');
14:10 crab we do? hm.
14:12 crab i have lib/X.pm and a script that does $ENV{MOJO_APP}||='X' before ->start.
14:12 crab i also have lib/Y.pm which has base 'X', and if i do MOJO_APP='Y' ./script daemon, it seems to work right, but still read x.conf instead of y.conf
14:13 crab guess i'd better have a look at the code which decides where the configuration file is.
14:14 vaneska left #mojo
14:16 kaare_ left #mojo
14:22 sri hmm, wonder if we should also enable 5.10 features if possible
14:23 fhelmberger left #mojo
14:52 kaare joined #mojo
15:12 bc5042 left #mojo
15:13 bc5042 joined #mojo
15:25 ashleydev left #mojo
15:34 a|newkirk[assoc] joined #mojo
15:41 fhelmberger joined #mojo
15:44 forwardever Plack-Middleware-Debug is quite impressive, especially when used with Devel::TrackObjects and the Memory tracker to find circular references
15:46 ryan joined #mojo
15:59 otaviof left #mojo
16:11 MojoGuest947 joined #mojo
16:11 MojoGuest947 From: http://www.google.com/search?q=websockets+demo&amp​;ie=UTF-8&amp;oe=UTF-8&amp;hl=en&amp;client=safari (1 hits)
16:11 MojoGuest947 left #mojo
16:12 GabrielVieira joined #mojo
16:16 koban left #mojo
16:28 koban joined #mojo
16:38 forwardever left #mojo
16:38 esskar left #mojo
16:40 breaker313 left #mojo
16:49 forwardever joined #mojo
16:49 ashleydev joined #mojo
16:57 ysyrota left #mojo
17:15 otaviof joined #mojo
17:45 forwardever left #mojo
17:54 Alias__ joined #mojo
17:56 Alias_ left #mojo
17:59 pid left #mojo
18:08 avkhozov joined #mojo
18:49 pid joined #mojo
19:05 spleenjack left #mojo
19:13 a|newkirk[assoc] left #mojo
19:46 warthurt left #mojo
19:56 plu left #mojo
19:57 plu joined #mojo
19:57 warthurt joined #mojo
19:59 perlrocks Twitter: "Using #Perl, #Mojolicious and #Redis in a real world #asynchronous application http://tinyurl.com/4nma58o." --vtivti http://twitter.com/vtivti/status/26005462797262848
20:14 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application:  http://bit.ly/eSdUES" --hnfirehose http://twitter.com/hnfireho​se/status/26009329152823296
20:19 warthurt left #mojo
20:19 warthurt joined #mojo
20:23 ryan left #mojo
20:30 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application: http://bit.ly/eRZLZM Comments: http://bit.ly/fueUv5" --HNTweets http://twitter.com/HNTweet​s/status/26013199077740544
20:34 warthurt left #mojo
20:34 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application: Comments http://digfoc.us/gKL4xv" --bundleOfThought http://twitter.com/bundleOfTho​ught/status/26014321297326080
20:34 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://goo.gl/fb/eOUHW" --stevesandersonf http://twitter.com/stevesander​sonf/status/26014398527049728
20:44 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application: http: Comments: http:" --YCHackerNews http://twitter.com/YCHackerN​ews/status/26016803217997824
20:58 GabrielVieira2 joined #mojo
21:00 perlrocks Twitter: "An interesting looking #modernperl #mvc framework. Mojolicious http://goo.gl/eacCG" --mnology http://twitter.com/mnology​/status/26020928735420416
21:01 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://goo.gl/fb/Kuip0" --HackerNewsYC http://twitter.com/HackerNew​sYC/status/26021070645497856
21:01 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application: Comments http://bit.ly/dGMI2h" --pokkrap http://twitter.com/pokkrap​/status/26021163356389376
21:02 GabrielVieira left #mojo
21:03 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://goo.gl/fb/kadzS" (ja) --myikegami_bot http://twitter.com/myikegami_​bot/status/26021515333992448
21:07 perlrocks Twitter: "Mojolicious - Web framework for Perl http://www.findbestopensour​ce.com/product/mojolicious # opensource" (id) --bestopensource http://twitter.com/bestopenso​urce/status/26022669740675072
21:18 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://dlvr.it/DLW3p" --FrikiFeeds http://twitter.com/FrikiFee​ds/status/26025287460323328
21:19 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application: Comments http://goo.gl/fb/WjXnw" --bartezzini http://twitter.com/bartezzi​ni/status/26025632982892544
21:23 avkhozov left #mojo
21:24 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://ow.ly/1aS5Bw" --LatestPosts http://twitter.com/LatestPos​ts/status/26026870420013056
21:25 jfuller I feel left out, maybe I should "retweet" that message
21:25 tetrad joined #mojo
21:26 perlrocks Twitter: "Using Perl, Mojolicious and Redis in a real world asynchronous application http://hndiary.com/r/Ic4" --hndzerobot http://twitter.com/hndzerob​ot/status/26027433396281344
21:26 sri heh
21:26 * sri wonders why those are not retweets Oo
21:28 sri at least it's weekend, so nobody will notice the spam :)
22:10 kaare left #mojo
22:12 barwin joined #mojo
22:18 barwin I'm still having some trouble getting the Mojo::Client to work with an https proxy.  I've tried with tinyproxy, apache's mod_proxy, and another one called "micro_proxy" and in each case, Mojo::Client gets a successful 200 response from the proxy and then simply hangs.... never seems to initiate the actual request to be proxied.
22:18 barwin however, the unit test that ships with mojo passes
22:19 barwin and if i fire up the websocket_tls_proxy_lite_app, it *does* work with that
22:19 barwin am i nuts?
22:19 sri sounds like you have everything to prepare a new unit test
22:20 barwin regrettfully i'm not very familiar with how the 'CONNECT" protocol works.  Do you know of a good guide/reference to follow?
22:21 barwin or at least the appropriate thing to google for ;)
22:23 sri not sure there is much good reading material
22:27 sri MOJO_CLIENT_DEBUG=1 will show you whats happening btw.
22:28 sri wait a minute
22:28 sri barwin: what kinda proxy were those you tried?
22:30 sri from what i can see those are pure http proxies
22:30 sri do they even support CONNECT?
22:31 barwin i'm pretty positive that tinyproxy supports 'CONNECT'
22:31 barwin because I also pointed firefox to tinyproxy, and sniffed the request headers and saw 'CONNECT" all over the place
22:31 sri i would expect a yes or no
22:31 yeo left #mojo
22:32 barwin i go with 'yes' then
22:32 barwin especially because i've had MOJO_CLIENT_DEBUG=1 and i can see that it gets a 200 OK response from the 'CONNECT' line
22:33 sri ok, then it's up to you to get us a proper test case :)
22:34 sri showing me the result of MOJO_CLIENT_DEBUG=1 might also help
22:35 barwin can having tinyproxy installed and running be a pre-requisite for the test case? :)
22:35 sri nope
22:35 barwin hrmm, so i need to provide a public proxy for testing?
22:36 sri not acceptable either
22:36 barwin output from MOJO_CLIENT_DEBUG coming in a sec
22:36 sri you replicate the exact behavior of tinyproxy in a test case similar to the existing test
22:37 sri but maybe the debug output will be enough, i generally find http bugs rather exciting and wouldn't mind investigating myself :)
22:38 sri (as long as a bug has been proven)
22:38 barwin http://pastebin.com/BfX8LzsB
22:39 barwin naturally
22:39 barwin i really don't want to waste your time. I've been hammering at this for the better part of the week and figured it was time to ask for help
22:40 barwin oh hangon i will update the pastebin with also the output from tinyproxy's log
22:40 sri ok, actually i see the problem
22:40 sri (maybe)
22:42 barwin oh fantastic
22:42 sri yea, got it
22:43 barwin I love you.
22:59 GitHub126 joined #mojo
22:59 GitHub126 mojo: master Sebastian Riedel * 81b3f1c (4 files in 4 dirs): fixed a small bug in the relaxed HTTP parser - http://bit.ly/hEwiC8
22:59 GitHub126 left #mojo
22:59 sri barwin: that should do the trick
22:59 ryan joined #mojo
22:59 barwin thank you SO much. I'll give it a shot
23:00 koban left #mojo
23:01 barwin It works perfectly now!
23:01 sri \o/
23:01 barwin you just saved me hours and hours of banging my head against the monitor
23:01 sri :)
23:02 barwin have a good weekend! I'm off
23:02 * sri waves
23:06 barwin left #mojo
23:19 GitHub150 joined #mojo
23:19 GitHub150 mojo: master Sebastian Riedel * 6f2d402 (1 files in 1 dirs): cleanup - http://bit.ly/dISYZ8
23:19 GitHub150 left #mojo
23:20 sri reddit.com, hackerne.ws and yandis.ru are really good test cases
23:20 barwin joined #mojo
23:23 sri wrong russian site... think i lost the good test case :(
23:24 sri ah, yandex.ru
23:28 barwin sri: I should have gone home when I had the chance... I discovered another small http bug, not related to the proxy, but rather with subrequests
23:28 barwin if i hit an SSL URL which is a 302 redirect, the subrequest defaults back to non-SSL
23:29 barwin This doesn't happen if the 302 redirect specifically includes the scheme + host + port info, but most sites simply specify the path and assume the client will keep the original scheme+host+port in the request
23:31 sri relative redirects are invalid
23:34 barwin can i pay you a bounty to add support for relative redirects?
23:34 barwin validity aside, they're very common
23:34 snearch left #mojo
23:35 barwin And I think having support for them in your client would be useful for others as well
23:39 barwin sri I sent you a PM, now I've got to run for real. Either way, thanks much for fixing the earlier bug!
23:39 barwin left #mojo
23:44 GitHub101 joined #mojo
23:44 GitHub101 mojo: master Sebastian Riedel * d11b47a (3 files in 3 dirs): fixed support for broken redirects - http://bit.ly/fED20S
23:44 GitHub101 left #mojo
23:44 sri it was actually an untested feature that didn't work :)

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