Camelia, the Perl 6 bug

IRC log for #mojo, 2010-08-20

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

All times shown according to UTC.

Time Nick Message
00:26 fod joined #mojo
00:52 geekus joined #mojo
02:58 MojoGuest355 joined #mojo
03:01 yakudzo joined #mojo
03:19 janus joined #mojo
03:24 keith joined #mojo
03:25 keith left #mojo
04:04 tholen42 joined #mojo
05:10 koban joined #mojo
05:44 ltriant joined #mojo
05:54 ysyrota joined #mojo
05:57 koban left #mojo
06:03 koban joined #mojo
06:03 koban left #mojo
06:31 nuclon joined #mojo
06:41 koban joined #mojo
06:45 vti sri: oO
06:54 marcus heh
06:54 marcus breaking it *again* ?
07:14 jest joined #mojo
07:16 dotan joined #mojo
07:24 janus breaking to get rid of breaks... i like that :)
07:25 MojoGuest285 joined #mojo
07:25 janus i'd hav esaid brakes with enought coffee
07:26 Foxcool joined #mojo
07:45 Htbaa joined #mojo
07:57 yko joined #mojo
08:22 * yko drops box with a totally broken big red button
08:23 yko I think it's broken :(
08:23 dotan joined #mojo
08:40 stephen joined #mojo
09:14 ilbot2 joined #mojo
09:14 Topic for #mojo is now ♥ Join the revolution! ♥ | http://mojolicio.us | http://irclog.perlgeek.de/mojo/today
09:19 koban joined #mojo
09:20 Foxcool left #mojo
09:21 Foxcool joined #mojo
09:32 baton8 left #mojo
09:32 yko left #mojo
09:40 Foxcool left #mojo
09:41 Foxcool joined #mojo
09:52 sri marcus: it's related to the first
09:53 sri right now we can't optimize long poll with the daemon
09:54 baton8 joined #mojo
10:04 baton8 left #mojo
10:04 baton8 joined #mojo
10:06 Alias_ left #mojo
10:08 sri is anyone actually using body_cb?
10:09 sri or ->res->body(sub {...}) to generate dynamic content?
10:10 * DaTa
10:10 vti i have mp3 streaming client, but i don't use it :D
10:10 sri not for reading
10:10 sri that just keeps working
10:10 DaTa sri: do you want to change the api? :)
10:10 sri i mean specifically to generate chunked responses
10:10 DaTa sri: that's what i'm doing
10:11 sri yes, i want to fix something that requires an api change
10:11 DaTa sri: feel free to do it, it's not productive yet :)
10:11 vti then i don't use it
10:11 sri i want to replace it with ->res->write('foo!') and ->res->finish;
10:12 sri ->res->write('foo!', sub {...}) using the usual drain callback
10:12 vti so chunks are created automatically?
10:12 sri no
10:12 DaTa i guess i can live with that :)
10:13 sri write_chunk() is a possibility though
10:13 DaTa and "" still marks the end of the body?
10:13 sri nope ->res->finish
10:14 sri if you're generating chunks "" is still the 0 chunk
10:14 sri if thats what you mean
10:15 DaTa ok, i call ->res->finish inside the callback, that's fine too :)
10:15 marcus sri: http://www.google.com/codesearch?as_q=body_cb&​amp;btnG=Search+Code&hl=en&as_lang=per​l&as_license_restrict=i&as_license=&am​p;as_package=&as_filename=&as_case=
10:15 sri you can just call it together with the last write, it only sets a flag
10:15 marcus seems github is using it.
10:16 marcus I mean joyent
10:16 sri they don't anymore
10:16 sri they are now a node.js company
10:16 marcus are you sure?
10:17 sri or actually *the* node.js company
10:17 sri marcus: absolutely, they make their perl hackers switch
10:17 marcus My experience is that even if companies go out and say we are a $language company, they still have a lot of code in other languages.
10:17 marcus for instance, Yahoo and PHP
10:18 sri marcus: smart platform is dead since april
10:18 marcus sri: okies. No worries then I guess.
10:18 marcus except for darkpan. <threatening_sound_effect/>
10:18 sri as usual
10:19 sri in the end it will likely just affect the person that proposed it ;p
10:19 * sri pokes xantus
10:21 sri the big problem in this design is that we need to work in sync and async environments
10:22 sri marcus: you were the one pushing most for making direct writing easier if i remember right :)
10:23 sri this api would allow us optimizations to write directly to the kernel send buffer
10:26 marcus sri: yeah, I'm for that.
10:27 vti sri: can i write to the brain?
10:27 marcus o/' Cause if you like it then you shoulda put a ring on it o/'
10:28 yko joined #mojo
10:32 sri vti: can you?
10:32 vti i can't, but maybe you can extend api...
10:33 sri only if you give me a use case and a test ;p
10:35 vti use case: world domination!
10:35 vti no test so far =/
10:36 marcus rfc?
10:37 sri too bad i like the use case
10:38 yko $self->stash->{'mojo.rendered'} = 1; should prevent from rendering template?
10:38 sri ye
10:39 yko yes
10:39 yko thank you, that was my mistake :\
10:44 vti but that's hacky
10:44 vti dont' rely on it like i did in bootylicious!
10:51 dotan joined #mojo
11:01 yakudzo left #mojo
11:01 sri write_chunked also added :)
11:01 sri or actually write_chunk
11:04 vti that's really useful ;p
11:04 sri actually i think it is
11:15 perlrocks Twitter: "Mojolicious Web Framework - Join the revolution! http://goo.gl/fb/Ffa7z" --readelicious http://twitter.com/readelicious/status/21656109220
11:16 vti sri: sounds like your last name :D
11:17 sri haha
11:17 vti Delicious Over 50
11:17 sri "Recent Bookmarks Saved by at least 50 People on Delicious.com"
11:17 sri interesting
11:33 sri ok, here's a new idea for a sweet mojolicious level api
11:33 sri $self->start_chunked; $self->write_chunk('hi!'); $self->write_chunk('there!', sub {...}); $self->finish;
11:35 vti can we somehow eliminate start_chunked? maybe on first write_chunk turn it on?
11:35 sri how?
11:36 sri we need to tell the server binding that content generation is delayed somehow
11:36 sri first write_chunk only works if you do it from the handler directly
11:38 sri an alternative might be to introduce a general purpose ->res->finish that always needs to be called (possibly by mojolicious automatically on rendering and stuff)
11:38 sri to signal that the response is ready
11:44 vti on the first write_chunk check if we are already in chunked mode.. or i don't understand smth
11:44 sri you don't understand something ;p
11:44 sri currently we default to static responses
11:45 sri stuff that gets directly rendered completely as soon as the handler finishes
11:45 sri for long poll chunked responses we might not want to start right away
11:46 sri for body_cb the ioloop will atm simply keep polling
11:47 sri the socket stays in the writing queue
11:47 sri there is no way to signal that we want to wait
11:47 sri damn... this is complicated :S
11:49 sri basically, i want automatic pause/resume
11:52 dotan left #mojo
11:54 vti sri: won't it be too complicated?
11:54 sri vti: in what sense?
11:55 vti in sense of more bugs...
11:55 sri i don't think you understand the problem
11:55 dotan joined #mojo
11:56 sri right now, when you delay a chunk by say 1 second, you have maybe 50 polls from the ioloop doing nothing
11:57 sri pol/epoll/kqueue would always keep firing writable events
11:57 sri thats what we have right now
11:58 sri bottom line, our long poll support right now is rather underwhelming
12:00 sri so i'm fixing bugs right now
12:00 vti ok :)
12:02 sri http://pastie.org/1103963 # here's one of my new long poll test cases
12:06 yko left #mojo
12:08 vti sri: option listen is now removed from daemon?
12:08 sri no
12:09 vti https://gist.github.com/46cd75b70784f4adfde3
12:09 yakudza joined #mojo
12:10 sri damn
12:10 sri bug
12:11 fish_ vti: is there still hope for including my comment patch in bootylicious? i so, i would 1) merge it again 2) add some cheapo captcha (my blog is full of spam :D) and 3) write some tests
12:13 vti fish_: there is always hope... keep bugging me :)
12:27 vti fish_: i am trying to find some time for that
12:28 fish_ i belive you. but the question is whether you will find it :D
12:30 vti fish_: do you have them wroking on your blog?
12:30 fish_ vti: yep
12:30 fish_ ...and its full of spam :D
12:31 fish_ so i think its not usefull unless i add some cheap captcha function to it
12:31 fish_ and atm something else is broken, but i'll have a look and ping you again if its fixed
12:34 jest left #mojo
12:34 jest joined #mojo
12:38 sri ok, i think i have a rather pleasant to use solution :)
12:38 sri at least for mojolicious
12:38 Htbaa left #mojo
12:52 dotan left #mojo
12:54 fod joined #mojo
12:57 GitHub82 joined #mojo
12:57 GitHub82 mojo: master Sebastian Riedel * 343e489 (15 files in 9 dirs): fixed a serious design flaw in Mojo::Message and made long poll much easier - http://bit.ly/bKMrmt
12:57 GitHub82 left #mojo
12:57 sri please test!
12:58 sri xantus: you are the master of long poll, let me know what you think :)
12:58 sri dunno if there are any bigger mojolicious long poll apps out there, but performance increase should be very noticeable
12:59 sri especially with a big number of idle connections
12:59 sri it will take load off poll/epoll/kqueue
13:00 sri vti: --listen is also fixed btw
13:04 crab sri: you should write better commit messages!
13:04 sri crab: like?
13:04 crab what is the serious design flaw?
13:04 crab what was the problem with https support for cgi environments?
13:05 sri it didn't work
13:05 sri simple as that
13:06 crab ok, so perhaps that wasn't the best example.
13:06 sri :)
13:06 crab anyway, i'm just trying to follow along with developments to git head.
13:14 baton8 left #mojo
13:14 baton8 joined #mojo
13:16 koban sri: i got warnings while run tests http://gist.github.com/540287
13:18 sri koban: thanks, seems easy to fix
13:18 koban =)
13:29 sri ok, ->finish is also optional now :)
13:29 sri since ->write_chunk('') already implies ->finish
13:29 sri now you can stream chunked simply by calling ->write_chunk
13:29 dotan joined #mojo
13:31 GitHub36 joined #mojo
13:31 GitHub36 mojo: master Sebastian Riedel * b7ec972 (4 files in 3 dirs): cleanup - http://bit.ly/cL21pV
13:31 GitHub36 left #mojo
13:37 perlrocks Twitter: "Fixed a serious design flaw in #mojolicious and made long poll much easier today. :) http://bit.ly/a5Ds8U #perl" --kraih http://twitter.com/kraih/status/21664837857
13:39 koban sri: one warning remains http://gist.github.com/540325
13:41 GitHub151 joined #mojo
13:41 GitHub151 mojo: master Sebastian Riedel * 1ffca57 (1 files in 1 dirs): cleanup - http://bit.ly/c5Odj9
13:41 GitHub151 left #mojo
13:41 sri koban: should be fixed too now :)
13:41 koban :)
13:44 sri ok, chunked responses should be painless now
13:45 sri so if you want to optimize by sending the head section of your html in advance you basically start with a $self->write_chunk($head_section, sub {...}) and put all remaining logic into the callback
13:46 sri that should guarantee that your head section has been written completely before it continues processing
13:49 perlrocks Twitter: "Here are some examples for how easy long poll with automatic chunking has become in #mojolicious. http://bit.ly/axGeK2 #perl" --kraih http://twitter.com/kraih/status/21665731106
13:56 GitHub64 joined #mojo
13:56 GitHub64 mojo: master Sebastian Riedel * fcaf4e0 (1 files in 1 dirs): cleanup - http://bit.ly/aNjHeK
13:56 GitHub64 left #mojo
14:00 yakudza left #mojo
14:04 vel joined #mojo
14:09 perlrocks Twitter: "Here's an example showing how to optimize by sending the html head section in advance with #mojolicious. http://pastie.org/1104151 #perl" --kraih http://twitter.com/kraih/status/21667220691
14:10 sri that should easier for people to grok :)
14:11 sri the timer is used to demonstrate backend activity
14:12 sri it also illustrates the tradeoffs
14:12 sri like not being able to use layouts
14:16 perlrocks Twitter: "RT: @kraih: Here's an example showing how to optimize by sending the html head section #mojolicious. http://pastie.org/1104151 #perl" --sharifulin http://twitter.com/sharifulin/status/21667731995
14:19 yko joined #mojo
14:20 perlrocks Twitter: "Same example, but a little easier to grok. :) http://pastie.org/1104168 #mojolicious #perl" --kraih http://twitter.com/kraih/status/21668009374
14:21 sri ok, thats about as easy as it gets
14:23 sri marcus: happy? :)
14:25 yko oh... with that it's possible to realize partial rendering of templates
14:25 yko tha'ts terrible...
14:26 sri whats terrible? Oo
14:28 jest left #mojo
14:30 perlrocks Twitter: "This is the result for both examples, the first chunk will be sent 3 seconds before the second. http://pastie.org/1104186 #mojolicious #perl" --kraih http://twitter.com/kraih/status/21668718356
14:32 vti sri: that's awesome
14:32 vti maybe write_chunk('') is a bit hacky?
14:32 sri the trick is to get the timing right, the new design guarantees that the first chunk is written before processing continues :)
14:33 sri vti: what makes it hacky?
14:33 vti the look
14:33 sri feel free to improve
14:33 vti write_last_chunk mayber better?
14:33 vti just a shortcut
14:33 sri oh you mean the last chunk
14:33 vti ye
14:33 sri thats in the nature of chunked
14:34 vti i know, but people have to know the specs this way :)
14:34 sri an empty chunk marks the end
14:34 vti they never read
14:34 vti and will ask
14:34 sri sure, but i like to just say "an empty chunk marks the end of the stream"
14:34 fish_ vti: you 'let improve security' patch, broke dozend of my blog posts.. not that dramatic for me - nobody reads my blog - but if you planing to release it to cpan you should warn ppl about that
14:34 sri it's so simple :)
14:35 vti fish_: oO
14:35 vti fish_: are you sure? everything works before it?
14:35 sri fish_: ohoh, then you might have made a mistake
14:35 fish_ vti: (because the _ in some of my alias are not in the white list)
14:35 sri huh?
14:35 fish_ sri: http://github.com/vti/bootylicious/commit​/b30985b86c93eeefaf6488e0ed7cfb01a292e324
14:36 vti fish_: adding them
14:36 fish_ the route only maches if the "alias"-part of the url matched that regexp
14:36 sri oh you mean vti -.-
14:36 vti sri: ;p
14:36 fish_ sri: ;)
14:36 * sri needs to learn to read :(
14:37 fish_ its hard with those 3 letter nicks
14:37 vti sri: in irssi your nickname is highlighted, easy to recognize ;p
14:37 fish_ 4 letter nicks are much better. good as all four letter words :D
14:37 vti fish_: try it
14:38 vti and give your blog's link
14:39 fish_ http://freigeist.org/
14:39 fish_ i added _
14:40 vti that big url looks odd :)
14:41 vti oh.. spam :D
14:41 fish_ vti: yep :D
14:42 yko sri, why you are writing empty chunk at the end?
14:43 sri -.-
14:43 * yko try to look like he did'nt read all above.
14:43 * sri cracks the whip
14:43 sri "an empty chunk marks the end of the stream"
14:44 vti sri: i've told you
14:44 yko ggg
14:44 vti sri: btw, why do you write an empty chunk?
14:45 fish_ vti: and i added some tests for the comment stuff
14:45 GitHub130 joined #mojo
14:45 GitHub130 mojo: master Sebastian Riedel * cf79793 (1 files in 1 dirs): pod update - http://bit.ly/bW4FOC
14:45 GitHub130 left #mojo
14:45 sri vti: ^^^
14:46 yko Who reads pods? :D
14:46 vti fish_: that's good
14:48 GitHub88 joined #mojo
14:48 GitHub88 mojo: master Sebastian Riedel * 645db38 (2 files in 2 dirs): updated changes - http://bit.ly/9dNxef
14:48 GitHub88 left #mojo
14:52 GitHub61 joined #mojo
14:52 GitHub61 mojo: master Sebastian Riedel * 70a2a54 (2 files in 1 dirs): cleanup - http://bit.ly/bTaWgD
14:52 GitHub61 left #mojo
14:52 vti without warning!!!
14:53 * sri marks vti as experimental
14:53 * sri changes vti
14:53 fish_ vti: still two issues.. 1) captcha.. should be simple and stateless i guess.. 2) adding a comment doesnt mark the page as modified..
14:54 vti fish_: no captcha, that sucks or recaptcha at least, i have a plugin that prevents your blog from spamming, i will release it some day..
14:54 crab how does this plugin work?
14:54 sri it's surprising how little of our experimental features get changed
14:54 vti it includes many technics like honeypot urls, invisible fields etc
14:55 vti crab: http://github.com/vti/mojox-common
14:55 vti look at readme
14:55 sri *-of
14:56 vti sri: do you consider mark them not experimental? ;p
14:56 vti *marking
14:57 sri over time of course
14:57 vti 10 years?
14:57 sri experimental doesn't mean i plean to change them, just that we are collecting more experience
14:57 sri maybe a month, maybe a year
14:58 sri what do you think?
14:58 vti maybe when you release mojo?
14:58 sri i don't think there should be strings attached for experimental stuff
14:58 yko vti, you still believe in that?
14:58 sri it should give us more freedom to try stuff
14:58 vti yko: in what?
14:59 yko in release
14:59 yko :D
14:59 vti yko: 0.999928 is a release too ;p
14:59 sri most people only try stuff when it's on cpan :/
15:00 sri not releasing experimental features would have prevented Mojo::DOM for example
15:00 sri it would simply not be there
15:00 * yko has two states: living on mojo from github and digging code and living on some stable point and doing something else :)
15:01 sri imo experimental should just mean "when it's ready"
15:03 * yko marks himself as experimental.
15:04 * sri changes yko
15:04 sri http://korshak.name/ra/ooX.jpg
15:04 * sri pokes yko
15:05 yko http://korshak.name/ra/hold-still.jpg
15:05 * yko pokes sri back.
15:05 sri HOLD STILL FOR THE PROBING!
15:05 * yko giggles
15:05 baton8 left #mojo
15:11 koban left #mojo
15:21 baton8 joined #mojo
15:31 tempire mojolicious doesn't seem to recognize that Net::Rendezvous::Publish::Backend::Apple is installed
15:31 tempire Gives 'We won't be doing any rendezvous publishing, please install a Net::Rendezvous::Publish::Backend:: module' warning
15:32 sri thats would be Net::Rendezvous::Publish complaining
15:32 sri i have the apple backend installed btw
15:32 sri *-s
15:32 tempire works fine for you?
15:32 sri ye
15:32 tempire harumph
15:45 yko left #mojo
15:59 vti sri: how about substituting $self->stash->{'mojo.rendered'} = 1; (which i see in a lot places) buy $self->no_auto_rendering or smth like that? people use this hack, and it doesn't look very clean
16:03 crab *shrug* if you introduce something else to do it, then you'll have *both* kinds of hacks scattered everywhere.
16:03 crab and people who want to suppress automatic rendering will invariably start attributing some sort of magic to one of the two ways of doing it.
16:03 vti crab: it will be just a shortcut, that does the same thing
16:04 vti anyway code repetition is not good
16:04 yko joined #mojo
16:05 koban joined #mojo
16:06 Foxcool left #mojo
16:11 sri agreed
16:11 sri $self->rendered
16:11 sri ?
16:12 yko sri++
16:13 vti ye
16:18 GitHub193 joined #mojo
16:18 GitHub193 mojo: master Sebastian Riedel * b5d339f (4 files in 4 dirs): added experimental rendered method to Mojolicious::Controller - http://bit.ly/cw1Cp6
16:18 GitHub193 left #mojo
16:19 GitHub63 joined #mojo
16:19 GitHub63 mojo: master Sebastian Riedel * 71929a4 (1 files in 1 dirs): cleanup - http://bit.ly/d9tm1h
16:19 GitHub63 left #mojo
16:19 sri vti: i even cleanup my own commits ;p
16:20 * sri is a perfectionist and can never be happy :(
16:21 vti that feels good :D
16:23 marcus beh
16:24 marcus Handed in gsoc evaluation.
16:24 vti and? what result?
16:24 marcus failed
16:25 * yko flips the flag.
16:26 sekimura joined #mojo
16:26 marcus In the end, rushing for three days can't make up for a month of missing work. It was just too little, too late.
16:27 yko agree
16:27 * yko flips the flag again
16:28 dotan left #mojo
16:28 sri marcus: try write_chunk!
16:28 * marcus tries write_chuck
16:28 marcus will it write the next season for me?
16:29 sri maybe
16:35 ysyrota left #mojo
16:42 nuclon left #mojo
16:42 nuclon joined #mojo
16:49 nuclon left #mojo
16:50 vti fish_: damn, merging with your branch is not that easy =/
16:53 yko vti what are you merging there?
16:53 vti comments
17:10 yko left #mojo
17:22 perlrocks Twitter: "@runeb Bullet proofing the Mojolicious test suite." --nordmarcus http://twitter.com/nordmarcus/status/21681302435
17:37 keith joined #mojo
17:59 yko joined #mojo
18:08 sri http://www.perl.com/pub/2010​/08/rethinking-perl-web.html # how come perl folks are always so far behind when it comes to web technologies?
18:10 sri "rails is almost a thing of the past now, lets clone it!"
18:14 xantus sri: ok, will do
18:14 sri \o/
18:18 vti sad
18:20 koban left #mojo
18:21 xantus ಠ_ಠ
18:22 sri Oo
18:22 yko gg
18:26 xantus funny how he avoids saying mojo
18:26 sri strange he didn't mention any framework
18:27 sri "but now with modern Perl we may be able to create a postmodern fat-model MVC web framework"
18:27 xantus dancer is mentioned
18:27 sri oh
18:27 xantus "I like the simplicity of Dancer, but anything compatible with Plack will work for web applications."
18:28 sri we need more blogging
18:31 sri the cgi mindset of some people is quite strong
18:33 xantus I shall do more blogging
18:33 sri thou shalt do more blogging!
18:36 perlrocks Twitter: "#perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious" --kraih http://twitter.com/kraih/status/21686042104
18:36 vti sri: http://houston.pm.org/talks/2010ta​lks/1003Talk/perl-web-frameworks/
18:36 marcus crud frameworks are so meh
18:36 vti He has now come back to perl with a new framework based on Sinatra
18:36 vti facepalm
18:37 sri lol
18:37 xantus I will post at xantus.tumblr.com
18:38 sri at version 4 i left?
18:38 marcus sri: did you really leave catalyst at version 4? I thought it was early v5
18:38 sri it was 5.5
18:38 marcus mm
18:38 sri i think
18:38 marcus sounds more right
18:38 sri 4 was the __PACKAGE__->action(...) thingy
18:39 xantus new version of vim is out
18:39 marcus xantus: anything ool?
18:39 marcus cool
18:39 * sri waits for textmate2 :/
18:39 vti marcus: perl6 support
18:39 xantus http://www.pcworld.idg.com.au/article/357630​/vim_editor_updated_modern_language_support/
18:40 sri if macvim had a real project drawer i would consider it again
18:41 vti sri: consider using console vim + plugins ;p
18:41 sri but this nerdtree thing just doesn't work for me
18:42 yko sri: project.tar.gz
18:42 yko but actually i don't think you would like it
18:43 sri unlikely
18:43 sri hmm, looks almost like nerdtree
18:44 xantus http://developers.enormego.com/view/w​hat-happened-common-courtesy-facebook
18:44 xantus I think facebook handled that really well
18:44 perlrocks Twitter: "RT @kraih: #perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious" --rainboxx http://twitter.com/rainboxx/status/21686581534
18:45 perlrocks Twitter: "RT @kraih: #perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious" --xantus http://twitter.com/xantus/status/21686602144
18:45 perlrocks Twitter: "#perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious (via @kraih)" --vtivti http://twitter.com/vtivti/status/21686646189
18:46 yko ?tweet #mojolicious is pretty cool with cgi!
18:46 yko :D
18:46 tempire huh.
18:46 tempire "but now with modern Perl we may be able to create a postmodern fat-model MVC web framework that makes the easy things trivial, the hard things easy, and the annoying things wonderful"
18:46 tempire I've heard that before....
18:46 sri yko: HOLD STILL FOR THE PROBING!
18:47 vti maybe they have another internet?
18:47 yko http://korshak.name/ra/radiation.jpg
18:48 perlrocks Twitter: "RT @kraih: #perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious" --tempiretech http://twitter.com/tempiretech/status/21686795521
18:48 xantus tempire: I said to myself, that last part IS mojo
18:49 dotan joined #mojo
18:53 perlrocks Twitter: "#ModernPerl IS #Mojolicious! #perl" --vtivti http://twitter.com/vtivti/status/21687079737
18:53 perlrocks Twitter: "RT @vtivti: #ModernPerl IS #Mojolicious! #perl" --ykorshak http://twitter.com/ykorshak/status/21687100902
18:54 perlrocks Twitter: "RT @ykorshak: RT @vtivti: #ModernPerl IS #Mojolicious! #perl" --kraih http://twitter.com/kraih/status/21687129272
18:55 xantus http://code.google.com/apis/predict/
18:55 xantus wow
18:56 perlrocks Twitter: "RT @vtivti: #ModernPerl IS #Mojolicious! #perl" --xantus http://twitter.com/xantus/status/21687243607
18:56 sri combobreaker!
18:57 perlrocks Twitter: "RT @vtivti: #ModernPerl IS #Mojolicious! #perl" --tempiretech http://twitter.com/tempiretech/status/21687324384
18:57 tempire I feel like a lemming.
18:57 sri stop talking and jump!
18:57 * yko jumps.
18:58 sri \o/
18:58 * tempire jumps uniquely
19:01 tempire vti: Have you used vim+terminal with any large projects?
19:02 tempire I had to move to macvim 'cause 50+ folders is just too much without being able to use a mouse with nerdtree.
19:02 vti tempire: i use it for all projects...
19:02 yko tempire: how large?  I remember, when vti worked on Microsoft, he complained that it's not everything ok...
19:02 * yko giggles.
19:03 vti LIAR!
19:03 yko yes
19:10 perlrocks Twitter: "One word: "Mojo-licious": @ItaliaTrent #Film #Writer #Mojolicious" --danperezfilms http://twitter.com/danpere​zfilms/status/21688089765
19:10 vti wtf is that
19:14 ferreira joined #mojo
19:15 sri somebody making a film called mojolicious?
19:15 ferreira left #mojo
19:25 tempire It's the follow-up to "The Social Network"
19:25 tempire documents in an overly dramatic way how Mojolicious is taking over the Perl world.
19:26 tempire Guillermo del Toro is directing.  There's a fight scene between sri & vti, I believe.
19:27 sri :D
19:27 * sri cracks the whip
19:27 sri harrison ford should play me
19:28 tempire I vote Dolph Lundgren for vti
19:28 vti chuck norris! and you are really in trouble
19:28 sri :O
19:30 xantus I just called chunk, he said he's going to round house kick both of you in the face
19:31 tempire Chuck Norris wrote Mojolicious without Perl as a dependency, using only one assembly instruction.
19:32 xantus lol
19:32 vti Chuck Norris wrote sri!
19:33 sri $sri->roundhousekick();
19:33 tempire Chuck Norris doesn't write tests, his code writes it for fear that it will fail him.
19:34 crab tempire: hah
19:37 fod left #mojo
19:37 koban joined #mojo
19:37 xantus haha
19:38 xantus http://xantus.vox.com/library/post/c​huck-norris-facts-rssatom-feed.html
19:38 xantus wow, nearly 4 years old
19:56 dotan left #mojo
20:02 sri i'm going to allow $self->write('foo!') in controllers too
20:02 sri (including drain callback)
20:03 sri it will behave just like chunked, but you need to set a content_length header
20:11 koban left #mojo
20:33 sri i love this api
20:33 sri sooo simple
20:35 GitHub49 joined #mojo
20:35 GitHub49 mojo: master Sebastian Riedel * ac12d80 (7 files in 7 dirs): added experimental write method to Mojolicious::Controller - http://bit.ly/aRsBk8
20:35 GitHub49 left #mojo
20:35 sri xantus: please review :)
20:37 sri now we can finally guarantee that everything has been put into the kernel send buffer before the drain callback gets invoked
20:40 sri btw. the behavior is just like websockets
20:40 sri writing will not start before the handler finished
20:41 sri so everything is perfectly flat
20:43 sri if i'm not mistaken all big design flaws are gone now
20:44 sri pretty much all imaginable optimizations should be possible now
20:51 perlite left #mojo
20:56 perlite joined #mojo
21:18 yko left #mojo
21:24 didls left #mojo
21:26 arthas left #mojo
21:29 GitHub106 joined #mojo
21:29 GitHub106 mojo: master Sebastian Riedel * f6652d8 (1 files in 1 dirs): cleanup - http://bit.ly/ap4krE
21:29 GitHub106 left #mojo
21:34 vel left #mojo
21:46 xantus sweet
21:46 keith left #mojo
22:06 janus hmm, just having finished my cocktail i wonder how ``mojo on perls'' might look and taste :>
22:07 janus or something like that...
22:08 sri !tweet Mojo on Perls! #perl
22:08 perlrocks Twitter: "Mojo on Perls! #perl --sri" --perlrocks http://twitter.com/perlrocks/status/21698759448
22:08 perlrocks sri: http://twitter.com/perlrocks/status/21698759448
22:10 janus sweet and fruity :)
22:13 * janus likes sex on the beach
22:21 janus !tweet Perl Sensitive Mojolicous has been specially designed to help people develop a relaxed attitude to the web.
22:21 perlrocks janus: http://twitter.com/perlrocks/status/21699578319
22:21 janus couldn't resist that one :)
22:22 janus the special fun at this one is the full original name
22:22 janus ``Joo Janta 200 Super-Chromatic Peril Sensitive Sunglasses''
22:23 perlrocks Twitter: "@kraih What use cases is #Mojolicious for? Able to replace Catalyst for large scale apps?" --rainboxx http://twitter.com/rainboxx/status/21699746032
22:25 janus what's the large there, the moose overhead? *scnr*
22:39 GitHub86 joined #mojo
22:39 GitHub86 mojo: master Sebastian Riedel * 30df485 (3 files in 3 dirs): cleanup - http://bit.ly/cF6rmg
22:39 GitHub86 left #mojo
22:40 * sri wants to be drunk too
22:43 janus what do you drink? :)
22:43 sri nothing :(
22:44 janus anti-alcoholic or something medical?
22:45 sri i mean right now :)
22:45 janus oh, i meant what you drinks you like
22:49 GitHub31 joined #mojo
22:49 GitHub31 mojo: master Sebastian Riedel * df90f8a (1 files in 1 dirs): cleanup - http://bit.ly/d05Rtc
22:49 GitHub31 left #mojo
22:49 sri becks beer and tequilla
22:53 geekus left #mojo
22:53 geekus joined #mojo
22:56 stephan48 left #mojo
22:56 xantus tequilla++
22:56 xantus patron++
22:57 janus can't help out there... only with brandy, gin, vodka and other cocktail ingridients
22:59 janus still, becks is alright... a good standard so to say :)
23:00 sri it's the local beer, the brewery is not far away :)
23:01 stephan48 joined #mojo
23:03 janus in that regard i currently would have to promote astra, but my taste is stronger than my patriotism ;)
23:03 sri haha
23:04 DaTa Augustiner! :)
23:08 su-bzero joined #mojo
23:12 su-bzero_ left #mojo
23:22 perlrocks Twitter: "@rainboxx Absolutely, #mojolicious is Catalyst2." --kraih http://twitter.com/kraih/status/21703103939
23:28 sekimura left #mojo
23:28 [1]awnstudio joined #mojo
23:37 [1]awnstudio is now known as alnewkirk
23:38 ryanc left #mojo
23:40 sekimura joined #mojo
23:47 ryanc joined #mojo
23:50 perlrocks Twitter: "RT @kraih: #perl developers really need to get rid of the cgi mindset, this is the #html5 decade. #mojolicious" --insurgentsoft http://twitter.com/insurge​ntsoft/status/21704688729
23:52 xantus http://bit.ly/bWpaDT
23:54 xantus I really wish draft 75 support wasn't removed
23:55 xantus detecting and dealing with it is difficult
23:55 xantus I have to get a browser version sample or something so I can detect it client side

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