Camelia, the Perl 6 bug

IRC log for #mojo, 2010-08-16

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

All times shown according to UTC.

Time Nick Message
00:13 ka2u joined #mojo
01:11 Alias "Bonjour" ?
01:11 Alias How does a web app add Bonjour support, and what for?
01:17 MojoGuest607 joined #mojo
01:17 MojoGuest607 From: http://www.google.com/search?ie=UTF​-8&oe=UTF-8&sourceid=navcli​ent&gfns=1&q=websocket+demo (1 hits)
01:18 MojoGuest607 see me
01:18 Alias Welcome to websockets
01:18 MojoGuest590 joined #mojo
01:20 MojoGuest607 by
01:34 blinky joined #mojo
02:48 janus joined #mojo
03:03 achromic joined #mojo
03:11 ask_ joined #mojo
04:00 dynax60 joined #mojo
04:02 yakudzo joined #mojo
04:04 tl joined #mojo
04:38 xantus___ joined #mojo
04:40 xantus__1 joined #mojo
05:02 [2]awnstudio joined #mojo
05:05 perletc joined #mojo
05:06 [1]awnstudio joined #mojo
05:10 [1]awnstudio joined #mojo
05:39 ask_ joined #mojo
05:50 yko joined #mojo
06:08 dynax60 joined #mojo
06:23 ysyrota joined #mojo
06:27 * sri yawns
06:28 ysyrota sri: good morning
06:28 sri moin
06:30 sri oh xantus irc client is the first hit for websocket demo :o
06:30 sri that explains a lot :)
06:31 koban joined #mojo
06:34 memowe o hai
06:37 vti sri: websocket shell is the first hit for one mojolicious too... :)
06:37 sri :)
06:40 xantus__1 sri: sweet
06:48 xantus I need a new job
06:48 perletc joined #mojo
06:48 sri tempire++ # http://cpanratings.perl.org/dist/Mojolicious
06:48 sri need more!
06:48 sri xantus_: what happened? Oo
06:49 xantus they're trying to send me to st louis
06:50 xantus I can't do this on site contracting for them
06:54 MojoGuest207 joined #mojo
06:54 MojoGuest207 From: http://www.google.com/url?sa=t&source=we​b&cd=1&ved=0CBUQFjAA&url=http%3A​%2F%2Fdev.xantus.org%2F&rct=j&q=webs​ocket%20demo&ei=vOBoTIb9LYzUtQPN3Pi5DQ&a​mp;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (1 hits)
06:55 MojoGuest676 joined #mojo
06:55 MojoGuest676 From: http://www.google.com/url?sa=t&source=we​b&cd=1&ved=0CBUQFjAA&url=http%3A​%2F%2Fdev.xantus.org%2F&rct=j&q=webs​ocket%20demo&ei=vOBoTIb9LYzUtQPN3Pi5DQ&a​mp;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (2 hits)
06:57 sri hmm, i only know st. louis from the news, when it gets hit by hurricanes
07:03 marcus xantus: those bastards
07:06 [2]awnstudio joined #mojo
07:08 marcus xantus: they steal your IP and send you to the south? Not cool.
07:13 sri those bastards indeed
07:22 perlrocks Twitter: "RT @kraih: Released #mojolicious 0.999927 (Comet), the changes file is pretty epic this time. :) http://bit.ly/6v9jOt #perl" --chernomyrdin http://twitter.com/chernomyrdin/status/21296102957
07:24 sri is companies going evil a trend again?
07:30 MojoGuest188 joined #mojo
07:30 MojoGuest188 From: http://www.google.com/url?sa=t&source=we​b&cd=1&ved=0CBUQFjAA&url=http%3A​%2F%2Fdev.xantus.org%2F&rct=j&q=webs​ocket%20demo&ei=vOBoTIb9LYzUtQPN3Pi5DQ&a​mp;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (4 hits)
07:30 [3]awnstudio joined #mojo
07:53 und3f sri, fastcgi works again :)
07:54 sri und3f: i told you two days ago ;p
07:54 und3f what?
07:54 und3f you found a bug?
07:55 sri ye
07:55 sri normal memory leak
07:55 und3f yey
08:01 arthas joined #mojo
08:13 xantus heh, I can use my ipad to rotate my compiz desktop
08:14 xantus with a little js, ajax and wmctrl
08:15 yko wow
08:15 xantus next I will put boxes up that represent the windows locations and sizes
08:16 xantus this is very powerful
08:17 xantus move windows, resize, switch desktops, get the window titles, dimensions, move windows to different desktops
08:17 xantus change the viewport, number of desktops
08:18 GitHub154 joined #mojo
08:18 GitHub154 mojo: master Yaroslav Korshak * ced615e (2 files in 2 dirs): fixed xml empty params bug
08:18 GitHub154 mojo: master Yaroslav Korshak * 4511c16 (2 files in 2 dirs): fixed dom bug
08:18 GitHub154 mojo: master Yaroslav Korshak * 37a568f (1 files in 1 dirs): cleanup
08:18 GitHub154 mojo: master Sebastian Riedel * a1a8934 (3 files in 3 dirs): cleanup
08:18 GitHub154 mojo: master commits 231cc7f...a1a8934 - http://bit.ly/asLdJ5
08:18 GitHub154 left #mojo
08:19 xantus betime
08:19 sri breakfast time :)
08:24 marcus coffee time
08:26 yko time to walk with dog. damn it's hot outside :\
08:33 und3f mojo time
08:34 vti cookies time
08:43 memowe but don't stringify them! #cookies ;)
08:54 marcus #cookiegate ?
08:55 vti /join cookiegate && get a cookie
08:56 memowe don't forget the tea: http://www.netzverwaltung.info/tea.pl/
09:09 Htbaa joined #mojo
09:33 crab i have a question
09:33 crab i have a layout named default, which is just a simple html wrapper
09:33 crab now, some of my pages need to have some <script> stuff in the <head>
09:34 crab what's a good way to handle this? it's not nice to put the script text into the stash and then include it in the layout template
09:34 yko content
09:35 crab explain?
09:36 yko http://github.com/kraih/mojo/blob/master/lib/M​ojolicious/Guides/Rendering.pod#Content_Blocks
09:37 yko but, actually, i'm doing such things with linked_content plugin
09:37 crab woo
09:38 crab that content thing will do for now, thanks
09:38 crab what happens if i say <%= content 'blah' %> and there is no blah defined?
09:39 yko i don't remember if there's fixup for undef
09:39 yko usually i just do <%== content('blah') || '' %>
09:39 crab looks like it's fine
09:39 yko but i can be wrong
09:41 crab that's not allowed.
09:42 yko ?
09:43 crab you're not allowed to be wrong.
09:43 crab ;-)
09:43 yko :\
09:43 vti kick hime!
09:43 vti *him
09:43 vti :)
09:43 yko gg
09:44 yko hmm... strange...
09:56 yko hmm
10:00 crab what?
10:03 yko nevermind
10:03 yko just /mode #mojo -v crab doesn't work if you are not +v
10:07 crab and it doesn't make any difference if the channel isn't +m anyway
10:08 yko yes, but it was fun to -v someone just for fun
10:09 crab i feel so oppressed
10:09 yko with mirc it works (at least worked  4 years ago )
10:09 yko hehe
10:09 jest joined #mojo
10:09 yko op'pressed that's correct word in this case, ye
10:10 marcus hmm
10:10 marcus seems the redirect body thing is still broken in trunk
10:15 memowe Perhaps this is a solution:
10:15 memowe http://gist.github.com/526737
10:16 crab what is this csrf_id thing that github sends me in a cookie?
10:17 memowe no problem to overwrite the response body after this redirect_to thing.
10:17 koban joined #mojo
10:17 memowe crab: I think, it's something about _c_ool _s_oftware _r_ules.
10:17 vti crab: Cross-site request forgery signature
10:18 vti for your protection :)
10:18 memowe no cool software on github?
10:18 memowe :(
10:19 sri marcus: got a test case?
10:23 memowe sri: perl -Mojo -e 'a("/" => sub { shift->redirect_to("http://localhost:3000/foo") }); a("/foo" => {text => "OK"})->start' daemon
10:23 memowe It's hard to test, the delay appears in firefox, not in the mojo client
10:24 memowe When requesting http://localhost:3000/, I have to wait about 15 seconds to get "OK".
10:24 memowe http://gist.github.com/526737
10:24 memowe I don't know if we want this, but it solves the problem.
10:29 crab vti: how does it work?
10:31 GitHub163 joined #mojo
10:31 GitHub163 mojo: master Sebastian Riedel * 7037a2f (1 files in 1 dirs): fix for quirky browser behavior - http://bit.ly/bUTMgF
10:31 GitHub163 left #mojo
10:35 memowe content_length(0), cool :)
10:35 vti crab: http://en.wikipedia.org/wiki​/Cross-site_request_forgery
10:39 marcus sri: sweet, fixes curl too
10:39 yko hmm... i think that can solve one old error
10:40 marcus memowe: you should probably add a test for content length on redirect.
10:40 marcus since sri didn't ;)
10:40 sri actually i just did
10:41 marcus oki
10:41 marcus not in 7037a2f tho
10:43 GitHub58 joined #mojo
10:43 GitHub58 mojo: master Sebastian Riedel * c0765bb (1 files in 1 dirs): add tests - http://bit.ly/cUNYzo
10:43 GitHub58 left #mojo
10:43 sri github is slow today ;p
10:52 yko joined #mojo
11:22 yko Im thinking if my last patches made DOM slow :\
11:25 yko Strange, looks like no.
11:26 yko I swear, two weeks ago dom was woking much faster
11:38 sri yko: maybe the longer regex we added recently?
11:39 yko sri. actually no. i checked i, speed is the same
11:39 sri for the xml tokenizer
11:39 sri ah good
11:39 yko maybe it's just something in my eye, just ignore me until ill write the test :)
11:40 yko or a good benchmark, or anything else :)
11:40 yko anyway, someone should take care about profiling and optimizing Mojo::DOM... in far far future
11:40 sri someone should profile and optimize *everything* :D
11:41 yko ye... in far far future
11:41 sri or right now
11:42 yko or right now... it it's not me who will care about it in nearest month at least :p
11:42 sri actually, scratch that, we'll just port to perl6 for performance :D
11:43 yko emm. you mean, to show that at the moment Mojo on Perl5 is flash-speed?
11:43 sri where perl5 tasks that take 2 seconds grow to 2 minutes
11:43 yko ahahaha ye
11:46 yko anyway, Mojo::DOM is a cool thing, i'm ashamed that i didn't approve it's idea from the beginning
11:46 sri shame on you!
11:46 yko ye, shame on me
11:47 dotan joined #mojo
12:01 crab hm
12:01 crab i hate having to return json with a text/plain content-type
12:03 crab in a Controller, how do i find a ref to the Base object?
12:03 crab oh, that's what ->app is. duh.
13:23 dotan left #mojo
13:24 sri We need more ratings! http://cpanratings.perl.org/dist/Mojolicious
13:24 * sri cracks the whip
13:24 * lbr gives sri an F
13:24 marcus WORST DISTRIBUTION EVER!</comic_book_guy>
13:24 * sri gives lbr pain
13:26 * sri sets marcus on fire
13:26 * lbr gives sri an FFFFFFFFFUUUUUUUUUUUU
13:26 * sri whips the burning marcus
13:26 sri overkill is underrated
13:26 vti !tweet Please, rate #Mojolicous distribution http://cpanratings.perl.org/dist/Mojolicious #Perl
13:26 perlrocks vti: http://twitter.com/perlrocks/status/21314641834
13:26 perlrocks Twitter: "Please, rate #Mojolicous distribution http://cpanratings.perl.org/dist/Mojolicious #Perl --vti" --perlrocks http://twitter.com/perlrocks/status/21314641834
13:26 vti shit
13:26 vti remote it
13:27 vti *remove
13:27 perlrocks Twitter: "Please, rate #Mojolicious distribution http://cpanratings.perl.org/dist/Mojolicious #Perl" --vtivti http://twitter.com/vtivti/status/21314690444
13:27 sri Oo
13:27 vti there is a typo =/
13:27 * sri goes to remove
13:28 vti implement 'undo' feature :)
13:29 janus `unsee' while you're at it, please ;)
13:29 sri in soviet russia features undo you!
13:30 * vti runs in tears
13:33 keith joined #mojo
13:33 sri http://www.cpantesters.org/cpan/report​/4cf0a426-6bfc-1014-a35d-d539733f4a4a # wtf?
13:34 Akron joined #mojo
13:34 vti it's windows
13:34 sri the actual hard tests even passed
13:35 marcus vti: o/' Boys don't cry o/'
13:35 * marcus rolls around on his back to put out the fire
13:36 marcus sri: seems you could not connect a socket
13:36 * sri cracks the whip
13:36 * marcus whips the crack
13:37 sri marcus: something liek that, very strange
13:37 marcus maybe a spurious error
13:37 sri windows, who knows
13:37 sri everywhere else we pass so far
13:51 yko joined #mojo
14:10 yko how to bring Mojolicious logs to console? app->log->path(undef) ?
14:11 sri no log directory is enough
14:11 crab oh!
14:11 crab i didn't realise that
14:12 yko hmm. strange. app->log->path(undef) doesn't bring log to console :\ perhaps i'm doing something wrong
14:19 janus your remove/rename the log dir ;)
14:20 janus erm...
14:20 yko janus : already :)
14:20 janus :)
14:20 yko but i thought log-.path(undef) should also do the same
14:21 janus haven't tried that
14:22 memowe 15:34:05 -!- Akron [~chatzilla@Placebo.dhcp.uni-bielefeld.de] has joined #mojo
14:22 sri app->log->handle(\*STDERR) might work
14:22 memowe They are here!!!
14:22 yko hehe :)
14:22 Akron We are not!
14:23 yko great, thank you sri
14:38 Alias joined #mojo
14:48 sri Alias: our built in http/websocket server publishes itself via bonjour, it's very convenient with browsers like safari that have bonjour bookmarks :)
14:48 Alias I really don't even know what that means
14:49 sri second
14:49 Alias TBH, I really just don't get the point of bonjour for the most part
14:50 sri http://twitpic.com/2aa0jb # here's a picture
14:50 sri say you are developing in a wireless network with dhcp, with bonjour you don't need to give people your ip address just to show them something
14:51 marcus macs use bonjour to share afp sharing points, for instance
14:51 Alias So it's the whole windows network share stuff, but more dedicated?
14:51 marcus similar to windows network neighbourhoods, but sane
14:51 marcus Alias: more open
14:52 sri http://www.multicastdns.org/
14:56 xantus joined #mojo
14:56 sri it's only about discovery, no strings attached
15:00 fod joined #mojo
15:23 koban left #mojo
15:25 yko joined #mojo
15:28 crab sri: do you live in bielefeld?
15:29 vti he lives in a beerfield
15:47 xantus__ joined #mojo
16:05 al crab: impossible, bielefeld doesn't exist
16:14 sri crab: nope
16:14 sri mmmmm....beerfield...
16:21 mohawx joined #mojo
16:22 xantus joined #mojo
16:23 maettu joined #mojo
16:28 yko joined #mojo
16:31 wsirc_4542301 joined #mojo
16:38 maettu I POST a JSON-Object to mojolicious by a qooxdoo application. I can see in the mojo-server-console that the request comes in, but how can I access the JSON object / where is it?
16:39 sri maettu: the json object is the body?
16:39 maettu dunno
16:40 sri do you want it encoded or decoded?
16:40 maettu sri: decoded
16:40 sri my $object = $self->req->json
16:40 sri that might just work
16:42 wsirc_7056775 joined #mojo
16:43 maettu sri: Use of uninitialized value $object in print at.. :-(
16:44 sri then it's likely more complicated, possibly urlencoded
16:45 sri then you'll have to get the encoded json via ->param and decode it yourself with Mojo::JSON
16:45 xantus_ yo
16:45 sri but i'm just guessing here
16:45 * sri waves
16:46 maettu I should see everything when I dump $self?
16:46 sri you can print $self->req to get a picture of what you're dealing with
16:46 sri yes, but thats a huge structure to dump
16:47 sri content type would be a good hint i guess
16:51 sekimura joined #mojo
16:53 maettu sri: print $self->req http://nopaste.info/9dcdf6f05d.html
16:54 sri thats not even a post request
16:54 fod joined #mojo
16:55 maettu sri: huu?
16:55 sri thats an OPTIONS request
16:55 sri without a body
16:56 sri no json in there
16:56 maettu hmm, I see, but ist says: Access-Control-Request-Method: POST
16:59 maettu and its the same .js which works nicely with JSONRpc.pm (from the qooxdoo-docs)
17:01 sri it possibly expects a specific response after which it might sends the json in a followup request
17:07 maettu I don't think so. It instantiates a CGI and CGI::Session then prints a header, checks various things and sends a JSON object back
17:11 tempire maettu: A post request would have "POST /json?nocache=1281977219130 HTTP/1.1" on the first line
17:12 vti tempire: did you get my vim line?
17:12 tempire yes, thanks.
17:13 vti sri: hm.. where is donation button on github?
17:14 sri maettu: well, obviously there is no json object in the request
17:14 sri vti: it's only on mojolicious.org
17:15 xantus_ sri: websockets don't work in chrome 5
17:15 sri xantus_: os x?
17:15 maettu sri&tempire: thank you so far, I'm working on it..
17:16 xantus_ seen it on osx, and linux
17:16 xantus_ gettting unexpected response 101
17:16 sri i think they have not updated to draft 76 yet
17:17 xantus_ thought we had fallback
17:17 sri i removed it
17:17 xantus_ !!
17:17 sri (from official release)
17:17 xantus_ I used github
17:17 sri there were bugs i couldn't fix
17:18 xantus_ hmm
17:18 sri maintaining untested stuff sucks
17:19 sri i would love to keep raft 75 fallback... but someone has to invest some time into it
17:19 sri *+d
17:19 xantus_ crap, I have to figure out how to test for 75/76 client side
17:19 sri *+the
17:19 xantus_ I could probably do that
17:19 xantus_ but then again, I may have to quit this week
17:20 xantus_ I won't sign this non compete, and they want to send me to st louis
17:20 sri bastards
17:21 xantus_ I have a gf and a life.  I can't do services/travel
17:21 vti xantus_: ye, an 2k+ projects :)
17:21 vti *and
17:21 xantus_ and since they do services, if I sign a non compete, technically I can't work doing extjs stuff if I sign this
17:21 xantus_ vti: lol
17:22 vti sri: what beer do you like?
17:22 sri vti: you still have to ask? :O
17:22 vti was it becks?
17:22 tempire non-competes are ridiculous.  It's a tough thing, but (standing your ground)++
17:23 sri of course!
17:23 sri :)
17:24 vti how much does it cost?
17:24 sri xantus_: btw. safari 5 websockets are broken, no matter if draft 75 or 76
17:25 xantus_ good job apple!
17:25 xantus_ rtards
17:25 sri vti: http://www.google.com/products?q=becks
17:25 sri webkit nightly websockets are fine
17:26 crab is extjs like jquery?
17:26 sri damn, i need one of those beer buckets
17:26 xantus_ ext core is like jquery
17:26 xantus_ extjs is like jquery ui but better
17:28 Akron left #mojo
17:31 perlrocks Twitter: "Just donated 4 packs (6 bottles each) of #beer to the the awesome #Mojolicious Web Framework! Retweet and join me!" --vtivti http://twitter.com/vtivti/status/21331973614
17:31 * sri goes to get some becks on vtis tab
17:33 sri !tweet @vtivti cheers!
17:33 perlrocks sri: http://twitter.com/perlrocks/status/21332100551
17:37 perlrocks Twitter: "just donated 2 packs (4 bottles each) of #izze to the the awesome #mojolicious web framework." --tempiretech http://twitter.com/tempiretech/status/21332399543
17:37 tempire I hate beer.
17:38 sri whats izze?
17:38 tempire http://www.izze.com/
17:38 tempire sparkling juice
17:39 sri !tweet @tempiretech cheers!
17:39 perlrocks sri: http://twitter.com/perlrocks/status/21332481846
17:39 sri qute stylish
17:39 sri *+i
17:41 sri wait, thats healthy
17:42 vti never seen it =/
17:50 sri i miss the days when websockets were easy :/
17:53 sri on os x you can pretty much only use webkit nightly right now
17:59 maettu how do I get mojo on https:// ? It says it should work out of the box but does not as it needs IO::Socket::SSL. Which fails to install as make Net::SSLeay fails
18:00 sri you need IO::Socket::SSL sadly as perl has no ssl support in core :(
18:01 maettu sure, I could compile it even, but Net::SSLeay refuses to make
18:01 maettu which seems to be a dependency of IO::Socket::SSL
18:02 sri it is the openssl binding
18:02 maettu it says it is unsupported..
18:03 maettu and it could possibly compile if I used same compiler for perl, openssl and Net::SSLeay
18:04 maettu no https:// then on a cheap hosting plan
18:05 petrus_a joined #mojo
18:09 vti maettu: what errors do you get while compiling?
18:10 yko maettu, if your hosting has cpanel or other similar stuff.. it usually have interface to install perl modules just into your homedir on hosting
18:11 yko in no, you can use stuff like pps (cgi shell) but it depends on hosting configuration...
18:11 yko i'm glad that IO::Socket::SSL is up to date on hosting i use
18:13 maettu vti: uhh, many many messages..
18:13 vti maettu: could you nopaste them?
18:13 maettu yko: happy you :-)
18:14 yko you have no any config panel on your hosting?
18:15 maettu http://nopaste.info/cfddd81b9b.html
18:15 maettu that's just the end..
18:15 vti looks like there is no openssl installed
18:15 maettu they don't host perl modules..
18:16 maettu vti: ahh, how can you see this?
18:16 yko basterds! they don't respect Perl!
18:17 vti it reports many 'unknown' functions, that usually means there is no backend installed
18:17 vti it has problems with XS module
18:17 maettu yko: as long as Perl respects bastards..
18:17 maettu openssl is installed (says apt-get)
18:18 vti oh
18:18 sri also the dev package?
18:19 vti maettu: if you are on debian, try installing libcrypt-ssleay-perl
18:19 maettu I'm on ubuntu.. :-(
18:19 vti libio-socket-ssl-perl
18:19 vti try the last one :)
18:19 sri yea, it's prolly already packaged
18:20 vti and -dev package too as sri suggested
18:20 sri http://packages.ubuntu.com/d​apper/libio-socket-ssl-perl
18:20 sri it's in there
18:22 maettu THANK YOU!
18:22 yko :))
18:22 sri :)
18:23 maettu could I in some way help you writing such simple things in GUIDES/ ?
18:23 * marcus ponders porting mojomojo to mojo, in order to maximime confusion
18:23 vti marcus: oO
18:23 marcus maximize
18:24 yko lol
18:24 yko mojomojomojo!
18:24 sri maettu: http://github.com/kraih/mojo/wiki # the wiki is a good place for such hints
18:24 sri we could generally use more content in the wiki :)
18:25 maettu o.k, I'll see
18:25 maettu thank you for directing me to the wiki, I didn't realize it existed..
18:25 sri i promise to shamelessly steal everything usable for the cookbook :D
18:25 maettu very much welcome, I'm your reader
18:26 * vti is sri's fan
18:26 vti :D
18:26 vti join the fanclub!
18:26 maettu I'm in
18:26 * sri is filled with happiness
18:26 maettu ehem, there is precisely 1 (one) page in the wiki?
18:27 sri it's new
18:27 sri github just made the wiki cool
18:27 sri few days ago
18:28 vti ph
18:28 vti oh
18:28 vti that's true
18:28 maettu please don't take offence, I just wondered
18:28 vti he has no emotions, he's a robot
18:28 crab jeez. there are 1e666 javascript libraries/frameworks/thingies
18:28 maettu I will fetch miself a github account then and start editing the wiki
18:28 sri robots have no emotions, sometimes that makes me very sad :,(
18:29 maettu booo!
18:29 vti sri: ye, i remembered that quote too :)
18:30 sri i've used that quote like 4 times now? :)
18:31 maettu hihi, I've already had a github account.. And it is sad I lost my password..
18:31 maettu how sad may it be to realize I lost my account..
18:31 maettu (it's a robot, see)
18:31 yko vti: http://korshak.name/sri-fan.jpg -- remember that t-shirt?
18:32 sri rofl
18:32 yko bgg
18:32 vti :D
18:32 vti aweseom
18:32 * yko giggles
18:33 vti i have to tweet
18:33 sri the photoshop skills are strong with this one
18:33 sri *jedi gesture*
18:35 perlrocks Twitter: "Join the dark side, we have cookies! RT @vtivti: Join @kraih's fanclub! http://korshak.name/sri-fan.jpg #mojolicious" --kraih http://twitter.com/kraih/status/21335952317
18:35 crab what does "kraih" mean?
18:36 sri crow in a northern german dialect
18:36 crab cool.
18:37 memowe Everytime I see it I must think of http://de.wikipedia.org/w/index.p​hp?title=Datei:Cray_2_in_rot.jpg
18:38 yko Answer on that question should be written down in mojo-faq
18:39 sri some things should stay mysterious
18:39 vti sri is CRAY
18:39 vti that's more true
18:39 sri pssssst!
18:39 sri nobody needs to know...
18:40 crab sri makes people cr^Wkraih
18:41 sri those cray people sure made awesome furniture http://www.chilton-computing.​org.uk/ccd/jpgs/fig5p1_86.jpg
18:41 maettu sri: You don't want to specify the dialect, will you?
18:41 vti The page cannot be found
18:42 marcus I'm afraid I can't do that, Dave
18:42 sri maettu: low german (plattdeutsch)
18:43 maettu sri: uuh, platt, versteh ich schlecht :-S
18:43 vti deutscher? :)
18:43 maettu worse
18:43 maettu (these days)
18:44 vti russian? :D
18:44 maettu swiss
18:44 sri :O
18:44 maettu höhö
18:44 perlrocks Twitter: ""those cray people sure made awesome furniture http://www.chilton-computing.​org.uk/ccd/jpgs/fig5p1_86.jpg " (via sri on irc.perl.org #mojo)" --paulhenrich http://twitter.com/paulhenrich/status/21336520483
18:45 sri heh
18:45 petrus_a nifty bot.
18:46 sri it's even mojo powered
18:46 petrus_a cool
18:46 petrus_a can I get the source somewhere?
18:46 vti stealing is bad
18:47 vti just joking :)
18:47 sri http://gist.github.com/507388 # but be warned, it's a huge hack i did one evening (drunk)
18:47 vti sri: what?!
18:47 petrus_a hahaha. I was assuming it was OS :D.
18:47 sri it does oauth and irc manually, very hackish proof of concept
18:48 vti oauth...
18:48 sri 2.0 can't come soon enough...
18:48 petrus_a very cool. I've been thinking about consuming the twitter streaming api with perl.
18:48 sri thats very easy
18:49 vti thanks to mojo
18:49 sri except for oauth
18:49 sri which is always painful, by design
18:49 petrus_a yeah. I'll miss http auth until, as you said 2.0 comes along
18:49 sri oh, did they turn of basic auth today at twitter?
18:50 sri *+f
18:50 petrus_a I don't think yet. But they've been doing tests of turning it off
18:50 petrus_a they turned it off for a few minutes a couple days ago
18:50 sri they should just turn it off for non https
18:50 perlrocks Twitter: "@ Vtivti @ kraih # mojolicious found another Feil Mojo, and again kipaliv crap, tomorrow I will test ..." (ru) --sharifulin http://twitter.com/sharifulin/status/21336861497
18:51 sri :o
18:51 vti ;D
18:51 petrus_a sri: for streaming they definitely should. we're not asking people for their creds, so the big benefit of oAuth doesn't matter.
18:52 sri yay for google translate :)
18:52 petrus_a at least I'm not. I guess if you implement user streams.
18:53 sri well, oauth 2.0 will allow direct authentication too
18:54 sri not much of a reason to disallow basic via https
18:54 yko damn, feeling so stupid... i should remember to always \Q\E variables i paste into regexps :(
18:54 yko it's bad when you do s/^$x// and getting '.' into $x :\ very bad
18:56 sri shame on you
18:57 vti sri: why is there a space between '@' and our names?
18:57 sri think it's my second favorite mistake too, right after ||
18:57 sri vti: google translate i guess
18:58 vti ah, rihgt
19:03 petrus_a_ joined #mojo
19:07 korshak joined #mojo
19:25 zakame joined #mojo
19:26 MojoGuest191 joined #mojo
19:26 MojoGuest191 From: http://www.google.com/search?hl=en&amp;&amp;​sa=X&amp;ei=tpBpTN-FFYK78gaD-OiyBA&amp;ved=0​CBwQBSgA&amp;q=websockets+demo&amp;spell=1 (1 hits)
19:26 MojoGuest191 fdsfsd
19:27 yko Most of them coming from google
19:28 sri ye, surprising
19:44 MojoGuest833 joined #mojo
19:44 MojoGuest833 From: http://www.google.com/search?hl=en&amp;&amp;​sa=X&amp;ei=tpBpTN-FFYK78gaD-OiyBA&amp;ved=0​CBwQBSgA&amp;q=websockets+demo&amp;spell=1 (2 hits)
19:45 yko hi MojoGuest833
19:45 yko it works.
19:45 MojoGuest268 joined #mojo
19:45 MojoGuest268 From: http://www.google.com/search?hl=en&amp;&amp;​sa=X&amp;ei=tpBpTN-FFYK78gaD-OiyBA&amp;ved=0​CBwQBSgA&amp;q=websockets+demo&amp;spell=1 (5 hits)
19:46 yko hi MojoGuest268
19:46 yko it works.
19:46 petrus_a I'm mojoguest right now (reminded by the fdsfsd guy), and can't send, but I see your message, yko
19:47 yko hm... strange
19:47 yko i saw same problem on Windows with NOD antivirus
19:47 MojoGuest602 joined #mojo
19:47 MojoGuest602 let's try it in chrome
19:47 yko oh! :)
19:47 yko it really works
19:47 MojoGuest602 hrm. Works in chrome. not safari 5 that time
19:48 sri safari 5 is broken
19:48 yko somebody told that safari has broken websockets
19:48 yko ye, that was sri :D
19:48 sri all versions of safari 5 are broken atm :/
19:48 MojoGuest602 huh. go figure with apple being a big HTML5 participant/pusher
19:49 sri for os x just use webkit nightly
19:49 vti you are using it wrong!
19:49 sri it's even worse, the first safari 5 release used an outdated version of the websocket protocol
19:50 MojoGuest602 would websocket.js be a decent workaround? (not for this proof of concept of course)
19:50 sri maybe
19:51 MojoGuest602 might give it a try some time.
19:53 xantus joined #mojo
19:53 xantus back
19:55 spleenjack joined #mojo
19:57 xantus_ joined #mojo
19:58 vti i have to find a teapot to install mojo on it
19:58 maettu started to edit the wiki, if someone wants to check if it conforms
19:59 yko :) cool
19:59 vti maettu: for debian/ubuntu libio-socket-ssl-perl is enough, it will download needable packages
20:00 maettu vti: is sudo ok for debian?
20:00 yko Strange thing. I have dummy DOM benchmark, and with same version of mojo it shows me ~1.5s on windows and ~22s on linux with ~80k html file :(
20:00 vti maettu: of course
20:01 maettu and because I banged my head against everything in reach before I found the generator script for Mojolicious, I want to add that too. Howto, again?!
20:01 yko sure, that systems have different CPU and other, but... 1.5 vs 22
20:01 marcus agree
20:01 marcus oops, scrolled back
20:02 sri yko: maybe slow io?
20:02 yko no prints at all
20:02 sri i mean read
20:02 yko i have html in .pl file :D
20:02 yko ouch
20:02 sri hmm
20:02 yko i have different perl versions
20:02 yko 5.8.8/510
20:03 yko 5.10 on win
20:03 sri ah that could be it
20:03 sri new regex engine
20:03 sri 5.10 got a brand new one
20:03 yko rly? I didn't know that
20:03 yko actually i'm not watcing new releases, i know, shame on me
20:03 yko thank for info
20:04 sri all the fancy new stuff is in 5.10, recursive regex, grammars *drool*
20:04 yko uhh...  that's cool
20:04 maettu ahh, "mojolicious generate help" stupid me
20:05 marcus sri: when is mojo going 5.10 only?
20:05 sri hehe
20:05 sri you should ask redhat -.-
20:05 marcus redhate
20:05 vti redhat: hey, you!
20:06 sri depends entirely on what rehat enterprise ships
20:06 sri *+d
20:06 marcus http://twitter.com/cjno/status/21340276084 # upgrades are hard?
20:06 sri yea, rails 3 has huge performance problems
20:07 marcus I thought they were switching to a lighter core (merb)
20:07 sri too much syntax sugar makes you fat i guess
20:07 yko :D
20:07 sri one would expect, but it is more bloated than ever
20:08 sri if ruby people consider something too slow...
20:09 quigglez joined #mojo
20:09 sri the whole breaking thing is not too bad imo, you can't replace the whole routing system and keep backwards compatibility
20:11 yko if you are keeping backwards compatibility you can not go forward...
20:11 sri that sounds a bit extreme :)
20:11 yko well, ye
20:11 yko but if you break everything every month - who will use that :)
20:12 vti yko: you? :)
20:12 yko damn, you are right :'(
20:12 yko hate u
20:12 marcus red meat!
20:12 yko gg
20:13 vti marcus: ;p
20:16 marcus http://www.redmeat.com/red​meat/1997-10-13/index.html
20:18 sri lol
20:19 maettu added http://github.com/kraih/mojo/wiki​/Frequently-Encountered-Problems
20:19 maettu and linked to main page
20:19 sri i would rather call it frequently asked questions, friendlier :)
20:21 maettu sri: you are the boss
20:21 vti there is mojo-faq!
20:22 vti http://github.com/vti/mojo-faq
20:22 sri that one is not on the wiki!
20:23 sri wiki is for everyone!
20:25 maettu I see that wiki is for everyone, which scares me a little bit. I could just edit everything.. zzz.
20:25 Yappo joined #mojo
20:25 hidek___ joined #mojo
20:25 sri well, history is preserved
20:26 sri and login requirement keeps most trolls out
20:26 maettu not me!
20:26 sri :D
20:28 maettu I think I once encountered a cool --reload option for development
20:30 xantus fuck, I need to fix this websocket issue some how
20:31 xantus_ joined #mojo
20:38 yko joined #mojo
20:40 xantus ok, well it will be missing the sec header, and maybe I can detect and trigger from there a downgrade to longpoll
20:45 maettu sri: remeber my problem with OPTIONS some hours ago?
20:46 maettu sri: it was because of security restrictions of JavaScript. I need to contact from same domain, https instead of http and it works
20:50 sri xantus: you might be able to detect it by intentionally setting some Sec-* header for an ajax request
20:51 sri Sec-* headers are forbidden for ajax requests in draft 76 compliant browsers
20:51 xantus ah, so it might throw
20:51 xantus good idea
20:51 sri ye
20:51 sri maettu: makes sense :)
20:51 maettu sri: your guides make fast progress!
20:52 maettu sri: with hindsight everything makes sense..
20:56 maettu ahhh, me again: how can I ask Mojolicious to serve file "../source/Application.js" on "/whatever"?
20:58 yko static files are supposed to be in 'public' actually
20:59 xantus app->static->serve( $self, $path ); # relative to public
20:59 xantus app->home->rel_dir( 'public' ); # setting public dir, relative to home
20:59 sri or just "ln -s" for specific files
20:59 xantus er no
20:59 sri (into public)
20:59 xantus thats fetching public path relative from home
21:00 yko or change  $c->app->static->root('your/source');
21:00 yko $c->render_static($path);  and then change $c->app->static->root back :D
21:00 xantus app->static->root( '/whatever' );
21:00 yko *not sure if it works
21:00 MojoGuest785 joined #mojo
21:00 MojoGuest785 From: http://www.google.com/search?clien​t=safari&amp;rls=en&amp;q=websocke​ts+demo&amp;ie=UTF-8&amp;oe=UTF-8 (1 hits)
21:00 yko hi MojoGuest785. it works
21:00 sri or just throw all your stuff into public
21:00 xantus websockets!
21:01 maettu you help faster than I can learn!
21:01 sri learn faster!
21:01 * yko giggles
21:01 * sri cracks the whip
21:01 yko i'm training last year to help faster then somebody learn
21:02 maettu sri: no, go steady, go far
21:02 yko sri had much more practice and whip...
21:02 yko :'(
21:02 apeiron_ joined #mojo
21:02 apeiron_ From: http://www.google.com/search?clien​t=safari&amp;rls=en&amp;q=websocke​ts+demo&amp;ie=UTF-8&amp;oe=UTF-8 (2 hits)
21:02 sri and yko back to fixing Mojo::DOM bugs! :)
21:02 yko hehe ye
21:02 yko or made regexp faster under perl 5.8
21:02 yko *make. whatever.
21:02 * xantus wants to work on mojo full time
21:03 yko xantus++
21:03 apeiron_ left #mojo
21:03 * sri wants xantus to work on mojo full time
21:03 * xantus wonders if someone will pay him his 6 fig salary to do it
21:04 sri sadly companies rarely sponsor frameworks :(
21:04 xantus yeah
21:05 maettu ehm, statically delivering a .js file shows the listing in the Browser..
21:06 sri http://blog.typekit.com/201​0/08/16/typekit-and-adobe/ # wow, looks like typekit has them all now
21:06 sri a new age of web typography
21:07 yko hmm
21:07 sri hmm...but garamond on the screen...i dunno
21:08 xantus sadly, I can't try catch that header set
21:08 sri and of course trajan...
21:08 sri aww
21:08 sri is it only filtered out?
21:10 xantus Refused to set unsafe header "Sec-"
21:10 sri ah
21:10 xantus its printed to the console, but not trapped
21:11 sri eww
21:11 sri you could actually perform the request
21:12 xantus yeah, if its same domain
21:12 xantus or use a xdomain xhr
21:13 xantus I'll try a cookie trick
21:14 xantus oh.. Set-Cookie will probaby not work
21:14 xantus_ joined #mojo
21:14 xantus probably
21:18 xantus well, actually doing the ajax req and setting the header Sec-Draft: 75 would work
21:18 xantus for thise case
21:18 xantus /se/s/
21:21 maettu what is $c?
21:22 maettu from $c->render_static($path);
21:23 sri $self
21:23 yko $c is a Controller
21:24 sri we differentiate mostly between $app and $c, for application instance and controller
21:24 yko but most people call it $self :)
21:24 yko sri++
21:24 sri both are usually written as $self
21:24 yko not for me, i preffer $c for controller (it's much shorten)
21:25 sri matter of taste
21:25 yko ye, agree
21:25 maettu ok, thanks, but it doesn't work
21:25 sri it's a relative path
21:25 sri relative to your public directory
21:26 yko there's restriction that does'nt allows somebody to get ../../../../../etc/passwd or something like that i think
21:26 maettu Can't locate object method "app" via package
21:27 yko app? where do you call it? and on which object?
21:28 maettu on the object of the app itself
21:29 maettu I call it in script/[my_app].pm
21:29 maettu in sub startup{}
21:31 yko ohh... so. let me introduce you structure of Mojolicious internal objects
21:31 yko controller has accessor app - that represents 'application' object
21:31 yko so for access application in controller you need to call $c->app->something....
21:32 yko if you are in startup { } you already have aplpication object- your $self
21:32 yko so you should call $self->method instead of $self->app->method
21:32 yko that is one more reason why i use $c/$app instead of $self in both cases.
21:33 xantus http://www.geek.com/articles/geek-cetera/best​-buy-sues-priest-over-god-squad-car-20100816/
21:33 maettu ok, thank you. I'm wondering how I get this into FEP (Frequetly Encountered Problems)
21:34 yko :))
21:36 xantus maybe I'll seek a grant from the gordon and betty moore foundation
21:36 xantus http://www.moore.org/
21:36 xantus doh
21:36 xantus "Please note that the Foundation does not accept unsolicited proposals."
21:37 xantus don't ask us, we'll ask you
21:57 maettu successfully served static pages outside public :-)
22:03 xantus sri: Chrome 5 also refuses to set Sec-* even though its on draft 75
22:03 ltriant joined #mojo
22:03 xantus heh
22:04 xantus my alternative is to repatch 75 into mojo
22:08 sri xantus: isn't the failing handshake actually enough?
22:09 sri maybe we've been thinking too complicated
22:11 sri error event should be enough to fallback no?
22:13 sri but sleep now :)
22:13 sri nn
22:14 xantus it actually gets through to mojo
22:14 sri yea, but gets a 404 or 500
22:14 xantus it gets a 101
22:14 xantus as if its valid
22:15 sri interesting, but if that was changed it might be enough
22:15 sri you just change the response code to 4xx in your handler
22:16 xantus https://gist.github.com/2e90e70595dd139fe0d6
22:16 sri seems to be a side effect of very defensive handshake code
22:18 sri the big question is what the browser does if you change the code in your action
22:19 xantus yeah, more testing
22:20 sri Zzz (rly)
22:25 * xantus taps sri on the nose to check if he's faking
22:40 mudpit joined #mojo
22:49 yko joined #mojo
22:51 spleenjack left #mojo
23:07 mib_up13ms joined #mojo
23:08 petrus_a joined #mojo
23:08 petrus_a_ joined #mojo
23:11 ferreira joined #mojo
23:12 mib_up13ms are there any proxies that can be used to dump websocket traffic from an external site?
23:19 xantus I can show you how to use ngrep
23:19 xantus mib_up13ms: what system are you on
23:19 xantus os
23:21 mib_up13ms OS X
23:22 mib_up13ms (mac)
23:29 xantus can you get ngrep?
23:29 mib_up13ms yes, just installed
23:30 xantus ok, do you know what host/port its connecting to
23:31 mib_up13ms yes, do i need ip or will hostname work?
23:31 xantus also, what interface is your network, eth0?
23:31 mib_up13ms en1
23:31 xantus try this: sudo ngrep -d en1 -W byline port <port you have>
23:32 xantus see if it dumps what you need
23:32 xantus change <port you have> of course to dest port of the ws :)
23:33 mib_up13ms i'm seeing what looks like binary, so it might be other network traffic. where do i specify host?
23:34 xantus alongside port
23:35 xantus replace port $port with   host $host and port $port
23:36 xantus you may need to resolve that host to an ip, but try the host first
23:36 mib_up13ms pcap compile: syntax error
23:37 xantus whats your ngrep line look like
23:37 mib_up13ms ngrep -d en1 -W byline host 78.129.202.38 port 80
23:37 xantus add and between that ip and port
23:37 xantus host 78.129.202.38 and port 80
23:37 mib_up13ms works
23:37 mib_up13ms thank you!
23:38 xantus ok,  you'll see dots in the stream
23:38 xantus those are either \x0A\x0D or the websocket \xFF and \x00
23:38 xantus framing
23:38 mib_up13ms ok
23:38 mib_up13ms and the hashes (#)?
23:39 xantus non maching packets
23:39 xantus usually tcp things
23:39 mib_up13ms ok
23:40 xantus you can use -q to not print the hashes
23:40 xantus before the -d
23:41 mib_up13ms thanks again
23:43 xantus :)
23:43 * ashleydev just tried ngrep --- nice thx

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