Camelia, the Perl 6 bug

IRC log for #mojo, 2011-10-04

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

All times shown according to UTC.

Time Nick Message
00:01 SSergey joined #mojo
00:10 ispy_ joined #mojo
00:12 metaperl joined #mojo
00:12 d4rkie joined #mojo
00:13 perlite joined #mojo
00:13 alnewkirk joined #mojo
00:14 marty joined #mojo
00:14 ispy_ joined #mojo
00:14 bpaluch joined #mojo
00:15 LordVorp joined #mojo
00:18 memowe joined #mojo
00:18 noganex joined #mojo
00:19 alnewkirk joined #mojo
01:21 ispy_ joined #mojo
01:25 gshank joined #mojo
01:40 Foxcool joined #mojo
01:59 mateu joined #mojo
03:31 shonorio_ joined #mojo
04:42 Foxcool joined #mojo
05:24 MojoGuest944 joined #mojo
05:24 MojoGuest944 From: http://www.google.co.kr/url?sa=t&source=web​&cd=7&ved=0CHEQFjAG&url=http%3A%2F%​2Fdev.xantus.org%2F&rct=j&q=websocket%2​0demo&ei=iZiKToboBOqPiAeN6vT-Aw&usg=AFQ​jCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rjt (1 hits)
05:29 SSergey joined #mojo
05:31 ysyrota joined #mojo
05:37 Eugene joined #mojo
05:52 Eugene joined #mojo
06:13 sromanov joined #mojo
07:10 marcus sri: you can serve static files from __DATA__, right?
07:14 marcus found it in the docs
07:14 d4rkie joined #mojo
07:20 cosmincx joined #mojo
07:20 judofyr joined #mojo
07:21 Vandal joined #mojo
07:24 sugar joined #mojo
07:30 Sjors joined #mojo
07:31 Sjors Hi
07:31 Sjors I'm running into some problems where some requests in "script/mojo_app daemon" take a few seconds, freezing the application for other requests, causing a queue of requests to build up etc. causing long response times
07:32 Sjors so I'm wondering: what is the recommended way to get by this problem? FastCGI, CGI, PSGI, running multiple daemons and having nginx/apache/... as a load balancer?
08:00 sugar joined #mojo
08:06 sugar joined #mojo
08:08 sugar_ joined #mojo
08:16 sugar joined #mojo
08:18 sugar_ joined #mojo
08:39 jkitazawa joined #mojo
08:53 marcus sri: http://nordaaker.com/docs/skitch/​Server_Error-20111004-105315.jpg I done fuck up?
08:53 marcus also, lovely colors...
08:57 andrefs joined #mojo
09:08 sugar joined #mojo
09:27 marcus HALP
09:32 marcus sri: https://gist.github.com/1261232 looks a lot like your pod example to me.
09:35 * marcus wonders if this happens because I already have an IOLoop, since I am running inside a Mojolicious app
09:48 trone joined #mojo
10:13 arthas joined #mojo
10:19 spleenjack joined #mojo
10:20 AmeliePoulain joined #mojo
10:25 trone joined #mojo
10:39 sromanov joined #mojo
11:20 perlrocks Twitter: "Oh, Mojolicious, forget everything. And over again." (ja) --tisihara http://twitter.com/tisihara​/status/121182994416406528
11:43 shonorio joined #mojo
11:51 Trelane joined #mojo
12:01 spleenjack joined #mojo
12:06 alnewkirk joined #mojo
12:48 sri marcus: DON'T EVER START THE IOLOPP TWICE!!!
12:49 sri http://mojolicio.us/perldoc​/Mojolicious/Controller#ua
12:49 sri the real example is there
12:50 sri normal useragent examples are for standalone
13:02 * sri makes ->start marcus proof
13:04 metaperl lol
13:06 judofyr MOJOLICIOUS, THERE CAN ONLY BE ONE IOLOPP
13:06 judofyr ONE IOLOPP TO RULE THEM ALL
13:07 metaperl s/lopp/loop/i
13:07 judofyr metaperl: tell sri
13:07 metaperl s/lopp/loop/ig # no need :)
13:12 perlrocks Twitter: "Mojolicious-Plugin-Recaptcha 0.3 by DMITRYNOD - http://t.co/YlXP42m1" (lv) --cpan_new http://twitter.com/cpan_new​/status/121211014455894017
13:12 GitHub118 joined #mojo
13:12 GitHub118 [mojo] kraih pushed 1 new commit to master: http://git.io/9qkYpQ
13:12 GitHub118 [mojo/master] made Mojo::IOLoop->start marcus-proof :) - Sebastian Riedel
13:12 GitHub118 left #mojo
13:12 sri \o/
13:14 judofyr \o\
13:14 sri /o/
13:14 judofyr |o|
13:15 alnewkirk joined #mojo
13:18 marcus you can never make anything marcus-proof!
13:19 sri :S
13:53 GitHub133 joined #mojo
13:53 GitHub133 [mojo] kraih pushed 1 new commit to master: http://git.io/OqTQZg
13:53 GitHub133 [mojo/master] updated changes - Sebastian Riedel
13:53 GitHub133 left #mojo
14:18 KayT3 joined #mojo
14:38 perlrocks Twitter: "Blog wrote. / "To another character only when the output code Mojolicious - aftertaste" http://t.co/3Schyuay" (ja) --taiju http://twitter.com/taiju/status/121232671929155585
15:02 metaperl joined #mojo
15:06 Foxcool joined #mojo
15:13 Vandal can any one tell me transcription of word Mojolicious?
15:14 GitHub86 joined #mojo
15:14 GitHub86 [mojo] kraih pushed 1 new commit to master: http://git.io/1JcFwg
15:14 GitHub86 [mojo/master] added more tests for different plugin options - Sebastian Riedel
15:14 GitHub86 left #mojo
15:21 MojoGuest109 joined #mojo
15:21 MojoGuest109 From: http://www.google.co.uk/url?sa=t&source=web&a​mp;cd=9&sqi=2&ved=0CFwQFjAI&url=http%​3A%2F%2Fdev.xantus.org%2F&rct=j&q=websock​et%20demo&ei=tCOLTvyRPNKY1AWAsqzJBQ&usg=A​FQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rja (1 hits)
15:22 spleenjack joined #mojo
15:22 MojoGuest109 123 test
15:23 MojoGuest109 123 testing in fiddler
15:25 MojoGuest390 joined #mojo
15:25 MojoGuest390 From: http://www.google.co.uk/url?sa=t&source=web&a​mp;cd=9&sqi=2&ved=0CFwQFjAI&url=http%​3A%2F%2Fdev.xantus.org%2F&rct=j&q=websock​et%20demo&ei=tCOLTvyRPNKY1AWAsqzJBQ&usg=A​FQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg&cad=rja (2 hits)
15:36 perlrocks Twitter: "1.99 Mojolicious http://t.co/fVRYLwxF" (sw) --campuscodi http://twitter.com/campuscod​i/status/121247326470815744
15:50 yko joined #mojo
15:52 grim_fandango joined #mojo
16:33 kaare joined #mojo
16:50 trone_ joined #mojo
16:56 sri https://github.com/kraih/mojo/issues/227
16:56 sri i need help
16:56 sri they don't get it
16:59 kjeldahl joined #mojo
17:02 mspo sri: I don't see the mojo code they're using in the bug
17:13 noganex_ joined #mojo
17:14 GitHub125 joined #mojo
17:14 GitHub125 [mojo] kraih pushed 1 new commit to master: http://git.io/tmH7Fg
17:14 GitHub125 [mojo/master] added more tests for content negotiation with broken Accept headers - Sebastian Riedel
17:14 GitHub125 left #mojo
17:17 GitHub110 joined #mojo
17:17 GitHub110 [mojo] kraih pushed 1 new commit to master: http://git.io/uuze8Q
17:17 GitHub110 [mojo/master] fixed typo - Sebastian Riedel
17:17 GitHub110 left #mojo
17:20 sri yay, they got it :)
17:46 grim_fandango I'm having a bit of trouble with bridge callbacks.  For some reason I can't just write cb => $self->require_session because &require_session includes $self->redirect_to(...) which can't be located (but works just fine when it's an anonymous sub).  What am I missing here?
17:48 grim_fandango Duh, \&require_session.
18:11 sri hmm, boring apple event
18:11 tempire everyone knew what was coming
18:11 tempire inevitable
18:11 tempire "I don't know what Droid Bionic users need to do between pictures, maybe go get coffee."
18:15 jwang I hope this isn't a sign of things in the post-Jobs era
18:16 sri well, the new voice technology will be exciting
18:16 tempire most of apple's events aren't in the amazing category.  they're still far ahead of anyone else.
18:17 sri the stuff they bought from siri
18:17 tempire apparently all the processing is done on the device
18:18 sri yes, i've seen some leaked information a few weeks ago, very scifi
18:19 tempire dude.
18:19 tempire timed questions…now that's cool.
18:35 sugar joined #mojo
18:54 MojoGuest908 joined #mojo
18:54 MojoGuest908 hi
19:19 marcus siri is neat
19:19 marcus I'll get the new iphone for the camera
19:19 marcus the iphone 4 is already my most used camera
19:19 marcus but yeah, kinda boring
19:24 tempire fickle fanboys
19:24 tempire APPLE IS THE BEST EVER OMIGOODNESS
19:25 tempire there.  balance is restored.
19:26 tempire woah
19:26 tempire apple.com fail
19:29 marcus whois fail actually
19:29 marcus oh, you're right
19:29 marcus permissions fail
19:34 zenbitz joined #mojo
19:35 zenbitz is there some trick to adding a method to a Mojolicious::LIte app?
19:36 zenbitz I just add a sub foo {} but $self->foo no work.
19:37 tempire app->helper(name => sub { });
19:37 zenbitz ah.
19:38 tempire http://mojocasts.com/e3#Helpers
19:38 zenbitz I knew there had to be a trick
19:38 tempire er, my bad.  no app.
19:38 tempire just helper
19:38 zenbitz I SO wish I had time to watch those casts...
19:39 tempire they're like 5 minutes
19:39 tempire you have time
19:39 sri you better rethink your priorities
19:40 tempire these node.js arguments are tiresome.
19:40 sri every minute of watching those screencast will prolly save you hours of bug hunting
19:40 sromanov joined #mojo
19:40 zenbitz noted
19:40 zenbitz I will carve some time in
19:41 sri tempire: i sense an evented vs threads war coming up
19:41 zenbitz does startup have to be registered in a helper or is it auto?
19:41 tempire startup is only for full mojolicious apps, so you don't have to worry about it in lite
19:42 zenbitz well that explains why M:Plugin::Database doesn't work!
19:42 sri that node.x project is quite interesting
19:42 sri by targeting the jvm they get so many languages for free
19:43 zenbitz I really feel bad for pestering you guys instead of reading docs
19:43 zenbitz normally I am not under this kind of st000pid time pressure
19:45 tempire I don't really understand the responses to ryan dahl's original post.  They reveal repressed bitterness of the posters' than anything about dahl.
19:45 tempire or maybe I just agree with dahl.  his point has nothing to do with node.js.
19:46 sri what i personally find most enetertaining about the whole situation is that it will only end up with more people being aware of node.js and in turn a lot of new users
19:47 sri the irc channel has grown from 600 to 700 people since the troll posts happened
19:47 tempire lulz
19:47 tempire I find it strange that even 'experienced' programmers don't know what the solution is to current software development.
19:47 sri proves again that there is no such thing as bad publicity
19:47 * tempire knows
19:49 sri some arguments against node.js are very valid though, like v8 being optimized for the client side
19:49 marcus you need threads as well as evented tho, evented doesn't solve multicore
19:49 marcus which even the iphone has now
19:49 sri it crashes *a lot*
19:49 sri marcus: for now processes + evented is ok
19:50 marcus sri: for web apps, yeah.
19:50 sri with a bit of luck everything will be a web app in a few years ;p
19:51 marcus would be nice to have gcd in perl
19:51 sri MacPerl
19:51 marcus gcd exists on linux and freebsd too
19:54 tempire it's true, there's really no reason to choose either threads or events
19:54 tempire they both work together pretty well
19:56 trone joined #mojo
20:09 sri tempire: but you have to solve the interpreted language + threads problem first
20:09 tempire meaning thread support in each language?
20:09 sri look at perl threads
20:10 tempire I'm waiting for someone to write a perl port on top of beam.
20:11 tempire let's all switch to reia!
20:11 sri the erlang vm is really really slow
20:12 tempire fine.  parrot then.
20:12 * tempire crawls up in a fetal position, repeating, "don't make me depend on the jvm, don't make me depend on the jvm…"
20:21 GitHub199 joined #mojo
20:21 GitHub199 [mojo] kraih pushed 1 new commit to master: http://git.io/pvnnuw
20:21 GitHub199 [mojo/master] fixed small detection bug in Mojolicious::Types - Sebastian Riedel
20:21 GitHub199 left #mojo
20:24 sugar installation of mojolicious -1.99 hangs up
20:24 sugar fully
20:24 sri already fixed in master
20:24 sugar on my two desctops
20:24 sugar i have broken AE?
20:24 sugar ok
20:24 sri ALREADY FIXED IN MASTER
20:25 tempire AHHH!
20:25 sri AAAAAAAHHHH!
20:25 sugar when i type i dont read messages
20:25 sugar =)
20:25 * sri rewrites sugar with threads and evented io
20:27 Sjors Hi
20:27 Sjors A friend of mine has some failing tests on his machine
20:27 sri AAAAAAAAAH!
20:27 * tempire hits sri with a beam.  waits for responding message.
20:27 Sjors embedded_lite_app.t; already fixed?
20:27 sugar hmm... master - Failed 66/75 test programs
20:27 sri ALL FIXED!
20:28 metaperl|2 joined #mojo
20:30 sri if anything is still broken in master please send a proper and detailed report
20:31 sugar and you will say "not a bug"?
20:31 sugar no thanks
20:31 sri only if it's not a bug ;)
20:32 sugar ok, thanks,
20:32 sugar oh, installed master
20:34 sri aside from the usual unfixable freak errors on windows i don't expect many fails for master though, the improvements are huge
20:34 sugar you add denpend to Digest::SHA?
20:35 sri it's a core module
20:37 sugar i know
20:37 marcus tempire: afaik parrot does not have threads support yet
20:37 tempire sigh
20:37 sugar and it's has C-backend
20:37 marcus doesn't java as well?
20:38 marcus c++
20:38 marcus \o/
20:40 GitHub198 joined #mojo
20:40 GitHub198 [mojo] kraih pushed 1 new commit to master: http://git.io/9Sly4A
20:40 GitHub198 [mojo/master] updated dependency list - Sebastian Riedel
20:40 GitHub198 left #mojo
20:41 abra joined #mojo
20:48 sri sugar: anyway, what would you rather hear from me if something is not a bug?
20:49 sri *+like to
20:51 sri apparently "it's not a bug" is considered rude among our russian users, why is that?
20:52 sri you're like the third russian complaining
20:55 sri perhaps i should just close issues without comment
20:55 tempire no, that's worse.
20:56 sugar yes, it's worse
20:56 andrefs joined #mojo
20:56 tempire You could say, "This is the intended design"
20:57 sugar but some of users do not understand it's not a bug
20:57 sri github issues are strictly for bugs and patches, not support
20:58 sugar but i don't mean extremal cases =)
20:58 sri i will never explain something there
20:58 sugar ok
20:58 tempire do you include that in response?
20:59 tempire "this is the intended design.  please visit our support channels for further explanation"
20:59 sri hmm
20:59 sri i've never seen anyone do that on github
20:59 tempire well
20:59 tempire welcome to the revolution :)
20:59 sri actually i'm thinking about disabling github issues ;p
21:00 sri so all requests come in here or on the mailing list
21:00 tempire I don't see any problem with that.
21:01 tempire I wonder if you can have issue requests post to the mailing list
21:02 sri hmm, but what do i link to from the makefile?
21:02 sri if i leave it open there will be rt tickets :S
21:02 tempire that's what I was thinking.
21:03 tempire groups.google.com/group/mojolicious
21:03 sri oh, disabling them is really easy
21:06 GitHub109 joined #mojo
21:06 GitHub109 [mojo] kraih pushed 1 new commit to master: http://git.io/TX9jmQ
21:06 GitHub109 [mojo/master] disabled GitHub issues - Sebastian Riedel
21:06 GitHub109 left #mojo
21:13 GitHub193 joined #mojo
21:13 GitHub193 [mojo] kraih pushed 1 new commit to master: http://git.io/cwIuFw
21:13 GitHub193 [mojo/master] fixed typo - Sebastian Riedel
21:13 GitHub193 left #mojo
21:15 sri that's good
21:15 sri pull requests still get through, and design discussions are better done on the list or here anyway
21:16 sri haha, wonder if i could use an irc:// link in Makefile.PL
21:16 sri :D
21:17 sri or i could link to http://dev.xantus.org ;p
21:18 sri and we make it open a query window to tempire!
21:19 tempire that sounds amazing.
21:20 tempire but even better, have it be a bit.ly link to a futurama episode
21:25 sri rickroll!!!
21:25 purl i think rickroll is "to post a misleading link with a subject that promises to be exciting or interesting" or getting 0wned by Rick Astley
21:50 rsimoes joined #mojo
21:56 Sjors what's the recommended way to make it possible to handle multiple requests at the same time?
22:02 D4RK-PH0ENiX joined #mojo
22:48 tempire Sjors: what's stopping you from doing that now?
22:48 zenbitz at what point do you switch from a ::Lite to a full?
22:48 Sjors tempire: "morbo" and Mojolicious::Commands::daemon simply don't
22:49 tempire hypnotoad is the included production server
22:49 Sjors aaaah
22:49 Sjors great
22:49 * Sjors looks into that
22:49 zenbitz I guess when I want to wright a unittest for a method in the lite app script
22:49 tempire zenbitz: when it becomes easier to do something in full
22:50 sri if you're blocking morbo so badly you might have a problem in your app though
22:50 zenbitz btw the mojocast #1 was good
22:50 zenbitz hopefully I will catch up tonight on them
22:50 sri morbo is evented, so it can handle thousands of parallel requests as long as nothing blocks for too long
22:51 tempire EVENTED!!?
22:51 tempire unix is hard
22:51 sri hypnotoad is only a pool of morbos
22:51 tempire profound words
22:52 sri so blocking is not as bad
22:52 sri but if your app depends on really bad blocking calls you will have to configure hypnotoad for that
22:53 sri as in limit client connections from 1000 to 1 perl worker
22:53 sri s/perl/per/
22:53 * sri sends SIGKILLs at a-pool-of-morbos
22:54 * a-pool-of-morbos captures and forks
23:00 metaperl joined #mojo
23:04 Sjors hmm
23:04 Sjors I'm not sure if hypnotoad is totally safe
23:04 Sjors <!-- PID: 16816, id: uid=0(root) gid=0(root) euid=33(www-data) egid=33(www-data) groups=33(www-data),0(root) -->
23:04 Sjors effective UID is www-data, but real UID is still root
23:04 Sjors AFAIK, you can elevate your privileges again as long as your real uid is unchanged
23:04 Sjors so one could still elevate to root in, for example, a template...
23:05 d4rkie joined #mojo
23:10 Kallisti joined #mojo
23:35 sri Sjors: seems broken on many platforms anyway :/
23:36 sri just been playing around with it on os x, and i seem to always be able to switch back to root
23:46 grim_fandango How do I get at the session from the template?
23:47 grim_fandango $self->session(...)
23:47 grim_fandango Boy oh boy
23:47 GitHub143 joined #mojo
23:47 GitHub143 [mojo] kraih pushed 1 new commit to master: http://git.io/DD-ZQQ
23:47 GitHub143 [mojo/master] improved setuidgid in Mojo::Server::Daemon slightly - Sebastian Riedel
23:47 GitHub143 left #mojo
23:48 sri Sjors: would be nice if you could try that on linux
23:54 sri by any chance a windows perl user here?
23:54 sri i need someone to run a oneliner for me
23:55 sri hmm
23:56 sri i'll just break master, if it breaks people will surely complain ;p

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