Camelia, the Perl 6 bug

IRC log for #mojo, 2011-06-25

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

All times shown according to UTC.

Time Nick Message
00:27 mattp left #mojo
00:29 D4RK-PH0ENiX joined #mojo
00:32 d4rkie left #mojo
00:33 grim_fandango joined #mojo
00:35 grim_fandango left #mojo
00:35 grim_fandango joined #mojo
00:36 d4rkie joined #mojo
00:38 D4RK-PH0ENiX left #mojo
00:41 grim_fandango left #mojo
00:42 D4RK-PH0ENiX joined #mojo
00:45 d4rkie left #mojo
00:45 d4rkie joined #mojo
00:48 D4RK-PH0ENiX left #mojo
00:58 metaperl left #mojo
01:12 bosphorus left #mojo
01:30 d4rkie left #mojo
01:31 d4rkie joined #mojo
02:13 lammel left #mojo
04:30 D4RK-PH0ENiX joined #mojo
04:36 d4rkie left #mojo
04:59 GitHub22 joined #mojo
04:59 GitHub22 [mojo] kraih pushed 1 new commit to master: http://bit.ly/kZjsCW
04:59 GitHub22 [mojo/master] added more url_for examples - Sebastian Riedel
04:59 GitHub22 left #mojo
05:02 d4rkie joined #mojo
05:05 D4RK-PH0ENiX left #mojo
05:07 GitHub154 joined #mojo
05:07 GitHub154 [mojo] kraih pushed 1 new commit to master: http://bit.ly/igXS50
05:07 GitHub154 [mojo/master] small documentation tweaks - Sebastian Riedel
05:07 GitHub154 left #mojo
05:07 * sri yawns
05:08 D4RK-PH0ENiX joined #mojo
05:11 d4rkie left #mojo
05:12 kaare joined #mojo
05:17 trone_ joined #mojo
05:21 jwang left #mojo
05:22 trone left #mojo
05:29 xantus[] hm, websocket isn't working for me
05:29 xantus[] chrome 12
05:33 sri you slacker
05:33 sri didn't follow the recent development?
05:34 sri chrome 12 is still draft 76, we are ietf-09
05:34 xantus[] bummer :(
05:34 xantus[] I figured it was at least ietf-06
05:34 sri only FF Aurora is ietf-09 atm
05:34 sri chrome has only internal builds for ietf-09 i think
05:34 xantus[] I got server sent events working btw
05:35 xantus[] thats super easy
05:35 sri neat
05:36 sri not sure why the chrome folks would wait so long with ietf-09 support, considering how involved they are
05:36 sri especially now that the spec is in last call
05:36 sri maybe because of the client side api spec
05:37 xantus[] maybe chrome beta has it
05:38 sri last i tried it did not
05:39 xantus[] so yeah, sse
05:39 xantus[] you pull the Last-Event-ID header
05:39 xantus[] set the content type to text/event-stream
05:40 d4rkie joined #mojo
05:41 xantus[] then $self->write("id:123\ndata:foo\n\n");
05:41 sri easy
05:41 xantus[] id is the event id
05:41 xantus[] so when it reconnects it will be in the header
05:41 xantus[] of course it would be nice to have it be a signed value
05:42 sri why?
05:43 D4RK-PH0ENiX left #mojo
05:43 sri can't you just use cookies?
05:43 xantus[] paranoia
05:43 purl Just because the worst scenario isn't the most likely doesn't mean you have to ignore it.
05:43 xantus[] sure
05:43 D4RK-PH0ENiX joined #mojo
05:44 sri actually i prefer the cookie solution i think
05:44 sri since it works consistently for everything
05:44 sri will be the same for websockets
05:45 sri ietf-09 handshakes are http compatible :)
05:46 d4rkie left #mojo
05:54 xantus[] nice
06:04 xantus[] grr firefox sucks
06:05 xantus[] its xhr impl in a webworker is incomplete
06:05 xantus[] crashes constantly
06:12 trone_ is now known as trone
06:21 d4rkie joined #mojo
06:22 D4RK-PH0ENiX left #mojo
07:28 lammel joined #mojo
07:37 lammel left #mojo
07:38 D4RK-PH0ENiX joined #mojo
07:40 d4rkie left #mojo
09:03 perlrocks Twitter: "Added some new stuff to the #mojolicious and #perl raptor shop. :) http://t.co/IEWGrP8 http://t.co/xLQGuFG" --kraih http://twitter.com/kraih/status/84547180072935424
09:05 sri thought i would sell more raptor shirts
09:12 sri then again, rare geek shirts are cool i guess :)
09:30 sherr left #mojo
09:30 perlrocks Twitter: "Running Django, Rails and Mojolicious on Windows Server 2008 with IIS 7: http://t.co/Ma73FyK" --govorunov http://twitter.com/govoruno​v/status/84554077115133952
09:33 xaka|2 left #mojo
09:37 d4rkie joined #mojo
09:40 D4RK-PH0ENiX left #mojo
09:52 perlrocks Twitter: "mojolicious-plugin-page-navigator 0.01 by SILVIOQ - http://metacpan.org/release/SILVIOQ/mo​jolicious-plugin-page-navigator-0.01/" --cpan_new http://twitter.com/cpan_ne​w/status/84559602372448257
10:04 Foxcool left #mojo
10:09 D4RK-PH0ENiX joined #mojo
10:11 d4rkie left #mojo
10:15 arpadszasz joined #mojo
10:18 arpadszasz left #mojo
10:18 d4rkie joined #mojo
10:21 D4RK-PH0ENiX left #mojo
10:24 arpadszasz joined #mojo
10:31 sherr joined #mojo
10:33 Drossel joined #mojo
10:38 Kulag left #mojo
11:55 Foxcool joined #mojo
12:08 Foxcool left #mojo
12:19 sri hmm
12:20 sri wonder if there is real interest in lite becoming less verbose
12:21 sri with a 4 line change i could allow helpers without $self
12:23 D4RK-PH0ENiX joined #mojo
12:24 d4rkie left #mojo
12:26 sri oh, i think it can even be a plugin
12:27 yko and create more syntaxes
12:27 yko we already have m::lite and  m with a bit defferent syntax
12:28 yko do you really want to make that difference bigger?
12:28 sri that's the big question
12:28 yko i'm afraid of one thing
12:28 sri it could get in the way of growing
12:29 yko you must been reading those post about ruby on rails and newbies
12:29 sri :)
12:30 sri well, i will hack up a gist and let the community decide what happens with it
12:30 yko i'm really afraid if after 2-3 years mojolicious will grow in the same
12:30 sri that's always a possibility
12:30 sri but you shouldn't forget that we don't bundle a model layer
12:31 sri most of the newbie problems with rails are ActiveRecord problems
12:31 yko well, you are clever enough to make mojolicious difficult without model layer :D
12:32 sri it's funny you bring that up while i'm talking about making lite easier :D
12:32 yko not in one day, but maybe in 2 years hehe
12:32 sri wait till you see the gist
12:53 sri https://gist.github.com/1046452
12:53 sri yko: ;p
12:55 yko it's not perl anymore :D
12:56 yko looks fun though
12:56 yko makes growing a bit harder. but it's only my opinion
12:59 sri updated evil2.pl
13:00 yko heh
13:01 yko it looks pretty damn cool
13:01 yko but makes really hard to understand what's going on
13:03 d4rkie joined #mojo
13:04 perlrocks Twitter: "Built a rather evil plugin for #mojolicious, wonder if there is interest in a more extendable dsl for lite apps. http://t.co/6c8R2oB #perl" --kraih http://twitter.com/kraih/status/84607851124899840
13:06 D4RK-PH0ENiX left #mojo
13:12 D4RK-PH0ENiX joined #mojo
13:13 * sri wonders what marcus and crab think
13:13 sri actually, i know crab will be against it :)
13:14 sri maybe tempire and xantus, they tend to be less conservative ;p
13:14 * sri pokes around
13:15 d4rkie left #mojo
13:29 sri http://groups.google.com/group/mojolicio​us/browse_thread/thread/cbe3cca1ff73f743
13:29 sri taking it to the list
13:30 d4rkie joined #mojo
13:33 D4RK-PH0ENiX left #mojo
13:36 perlrocks Twitter: "♺ @kraih Built a rather😈plugin for #mojolicious, wonder…is there interest in a more extendable dsl for lite apps? http://t.co/6c8R2oB #perl" --Metztli_IT http://twitter.com/Metztli_​IT/status/84616017669652481
13:44 fhelmber_ joined #mojo
13:45 D4RK-PH0ENiX joined #mojo
13:47 d4rkie left #mojo
13:55 arpadszasz left #mojo
13:55 sri hmm, guess i tend towards "it's a bad idea" now
13:56 d4rkie joined #mojo
13:59 D4RK-PH0ENiX left #mojo
14:48 alnewkirk|com hypnotoad seems to be putting my server in product-mode without my consent
15:27 arpadszasz joined #mojo
15:42 sri alnewkirk|com: hypnotoad is a production web server
15:43 alnewkirk|com sri, i know
15:44 sri then i can't follow your logic
15:47 alnewkirk|com sir, wiki page updated
15:47 sri alnewkirk|com++
15:48 alnewkirk|com i am setting-up my production environment
15:48 alnewkirk|com i having other services hooked into my mojolicious run mode
15:49 alnewkirk|com e.g. i have my config file split into production and development environments with database, email server, etc configurations
15:49 alnewkirk|com which profile gets used is based on my mojo runmode
15:50 alnewkirk|com which should be development but it seems that hypnotoad is forcing into production
15:54 toebu joined #mojo
15:54 toebu good evening
15:54 purl Ah, evening. The tumultuous mind tarries and contemplates, reveling in the silence afforded by the diurnal proletariat. Good evening, indeed.
15:55 sri i don't see why you would expect a pure production server to default to development mode
15:55 * sri waves
15:55 toebu I played with the Mojo::IOLoop today ... and started to doubt understanding of perl ...
15:56 sri heh
15:56 toebu I am using a timer and a Mojo::UserAgent to pull down webpages and analyze them ...
15:57 sri sounds easy
15:57 toebu yep
15:57 sri what went wrong?
15:58 toebu in the callback I was calling timelocal without loading Time::Local ...
15:58 sri that does not sound like such a good idea
15:58 alnewkirk|com its not yet obvious to me why the server would/should care about which mode the script is running in
15:59 toebu If the IOLoop had said anything about the callback failing this would have been trivial to find :-)
16:00 toebu but the program worked seemingly fine except that it did not properly analyze the incoming data but stopped in the inner most loop (where the timelocal statement was) ...
16:01 sri there should have been a warning on stderr
16:02 toebu I am using the IOLoop in my own program ... so maybe I have to configure it somehow ?
16:02 toebu a warning would be great ...
16:02 toebu Do I need to provide a logger instance to the IOLoop
16:03 sri perl -MMojo::IOLoop -E'Mojo::IOLoop->timer(1 => sub {die "works!"}); Mojo::IOLoop->start'
16:03 sri there it is
16:05 toebu it was the callback of the useragent that failed ... sort of a callback inside the callback
16:07 sri nope it's not, useragent with callback will reuse the outer ioloop and perform the request non-blocking
16:07 sri shouldn't be any nesting
16:07 sri (not like that would prevent errors on stderr)
16:08 toebu thats what I thought ... trying to come up with a simple example
16:10 toebu you are right ... so ... maybe the problem is that I am doing this inside a new Command I wrote ... which is started via mojo
16:10 toebu Mojolicious::Commands
16:12 sri perl -MMojo::IOLoop -MMojo::UserAgent -E'my $ua = Mojo::UserAgent->new; Mojo::IOLoop->timer(1 => sub { $ua->get("mojolicio.us" => sub { die "works too!" })}); Mojo::IOLoop->start'
16:12 sri that works too
16:12 sri MOJO_USERAGENT_DEBUG=1 for more details
16:12 toebu yes ... will trimm down my code to something manageable ...
16:13 toebu It seems to be connected to the while thing running inside mojo ... will get back to you
16:13 toebu got to dinner now ... wave
16:13 * sri waves
16:14 * sri goes to get some icecream
16:24 alnewkirk joined #mojo
16:27 GitHub150 joined #mojo
16:27 GitHub150 [mojo] kraih pushed 1 new commit to master: http://bit.ly/lLePI9
16:27 GitHub150 [mojo/master] mention that hypnotoad defaults to production mode - Sebastian Riedel
16:27 GitHub150 left #mojo
16:38 Foxcool joined #mojo
16:51 Akron joined #mojo
16:55 alnewkirk left #mojo
16:58 fhelmber_ left #mojo
17:11 sri hmm, maybe not the best day to propose big changes
17:12 alnewkirk joined #mojo
17:17 sri http://pastie.org/2121328 # i do like the simplicity
17:17 sri and moving from functions to methods when growing should be a familiar concept for our users
17:17 sri due to get/post... and friends
17:20 fhelmber_ joined #mojo
17:20 fhelmber_ left #mojo
17:31 sri it feels consistent
17:35 sri get '/list/:offset' => sub { render json => {list => [0 .. param 'offset']} };
17:39 alnewkirk left #mojo
17:42 sri https://gist.github.com/1046696 # comparing directly
18:04 D4RK-PH0ENiX joined #mojo
18:06 d4rkie left #mojo
18:25 toebu the problem ocures with this url ... it is very slow as it comes over a gprs link (do NOT publish)
18:26 toebu if I GET it with the useragent and then die in the callback, the error does not show
18:28 baton8 left #mojo
18:29 sri toebu: that's because your server is borked
18:30 toebu hmmm this does not realy help me all that much ... the server is not under my control
18:31 toebu it claims http/1.1 and does not specify a content length ...
18:32 sri i'll take a look if there's something we can handle better
18:32 toebu that would be cool
18:44 sri ok, i know how to make it work
18:44 sri but not how to test it :/
19:03 GitHub44 joined #mojo
19:03 GitHub44 [mojo] kraih pushed 1 new commit to master: http://bit.ly/m30Z1Y
19:03 GitHub44 [mojo/master] fixed a few small error reporting bugs around Mojo::UserAgent - Sebastian Riedel
19:03 GitHub44 left #mojo
19:03 sri toebu: ok, that should cover it
19:04 D4RK-PH0ENiX left #mojo
19:04 sri we could in fact already handle the web server, problem was a small bug in the connection drop code
19:07 sri i'd really like to test it better, but no idea yet how
20:03 bosphorus joined #mojo
20:35 d4rkie joined #mojo
20:36 Foxcool left #mojo
20:38 toebu sri: thanks will test
20:38 j3nnn1 left #mojo
20:42 kaare left #mojo
20:44 GitHub70 joined #mojo
20:44 GitHub70 [mojo] kraih pushed 1 new commit to master: http://bit.ly/iuNPIh
20:44 GitHub70 [mojo/master] fixed small bug in previous commit - Sebastian Riedel
20:44 GitHub70 left #mojo
20:44 Akron left #mojo
20:55 Foxcool joined #mojo
20:57 toebu cool, now the errors get displayed ... thank you!
20:57 sri \o/
20:58 toebu when I want the IO loop to run a task repeatedly, I re-insert it with a timer as soon as it got called ?
20:58 fhelmber_ joined #mojo
20:59 sri or use a recurring timer
21:00 sri oh, the description is not correct
21:01 D4RK-PH0ENiX joined #mojo
21:03 d4rkie left #mojo
21:04 GitHub133 joined #mojo
21:04 GitHub133 [mojo] kraih pushed 1 new commit to master: http://bit.ly/k6ZqKy
21:04 GitHub133 [mojo/master] fixed recurring timer description - Sebastian Riedel
21:04 GitHub133 left #mojo
21:04 sri that's better
21:40 spleenjack joined #mojo
22:01 Foxcool left #mojo
22:35 spleenjack left #mojo
22:42 toebu thanks
23:01 mattp joined #mojo

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