Camelia, the Perl 6 bug

IRC log for #mojo, 2011-06-06

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

All times shown according to UTC.

Time Nick Message
01:21 bosphorus left #mojo
01:35 abqar left #mojo
02:26 arpadszasz joined #mojo
04:10 kmx left #mojo
04:13 kmx joined #mojo
04:42 sh9 joined #mojo
04:58 sh9 left #mojo
04:58 koban` joined #mojo
04:59 D4RK-PH0ENiX joined #mojo
05:02 d4rkie left #mojo
05:30 perlrocks Twitter: "See jsonselect.org. It might be fun to add some Mojo::JSON::select() along with cool Mojo::DOM CSS3 selectors. /cc @kraih #mojolicious #perl" --mvuets http://twitter.com/mvuets/status/77608342327009280
05:35 koban` left #mojo
05:41 kaare joined #mojo
05:54 perlrocks Twitter: "...sort of perl -Mojo -E 'my $trends = ("http://api.twitter.com/1/trends.json")->res->json->find(".name")' /cc @kraih #mojolicious #perl" --mvuets http://twitter.com/mvuets/status/77614243251355649
05:56 D4RK-PH0ENiX left #mojo
05:56 ysyrota joined #mojo
06:19 Drossel joined #mojo
06:19 Kulag left #mojo
06:38 d4rkie joined #mojo
06:40 fhelmber_ joined #mojo
06:56 esskar joined #mojo
06:59 esskar left #mojo
06:59 tabbi joined #mojo
06:59 tabbi left #mojo
07:29 koban` joined #mojo
07:30 cosmincx joined #mojo
07:39 arpadszasz left #mojo
08:00 Alias left #mojo
08:11 * sri doesn't like json select
08:12 sri good morning fellow apple zealots
08:13 * moritz doesn't feel greeted :-)
08:18 D4RK-PH0ENiX joined #mojo
08:18 GitHub81 joined #mojo
08:18 GitHub81 mojo: master Sebastian Riedel * 345dff7 (1 files in 1 dirs): updated user agent example - http://bit.ly/lX3XTm
08:18 GitHub81 left #mojo
08:21 d4rkie left #mojo
08:42 Foxcool left #mojo
08:44 perlrocks Twitter: "@mvuets Looks more like general purpose selectors for Perl data structures, don't think that would fit into Mojo::JSON." --kraih http://twitter.com/kraih/status/77656981946638336
08:45 Sugar joined #mojo
09:07 pp joined #mojo
09:10 spleenjack joined #mojo
09:12 Foxcool joined #mojo
09:48 bosphorus joined #mojo
10:39 sri wonder is they will release lion during wwdc
10:48 Foxcool left #mojo
10:50 AmeliePoulain joined #mojo
10:54 crab wow, that would be cruel. won't it eat the attendees?
10:58 al what do i do if i want to flash() more than 4k of data?
10:58 al i.e. is there a server sided session persistency?
11:01 Foxcool joined #mojo
11:02 sri al: then you use a database
11:02 sri crab: haha
11:04 al hmkay
11:06 sri anyway, why would you need to store more than 4k?
11:07 sri al: don't abuse sessions for caching
11:07 Akron joined #mojo
11:14 sri if this is about caching you should instead use memcached for storage and just put an identifier into the flash
11:18 gshank left #mojo
11:28 Akron left #mojo
11:37 al sri, it's kind of caching
11:38 gshank joined #mojo
11:38 al user pastes data into form, sends it, gets syntax check result, chooses where to send the data
11:40 al between syntax check and actually storing the data somewhere there's no real need to send the data back and forth
11:40 al so i'd like to skip that
11:40 sri why not just a hidden form element then?
11:41 sri wait
11:41 sri you do realize the flash uses cookies right?
11:41 sri so there is data sent back and forth
11:41 al yea, sure
11:41 al the initial question was a bit inaccruate ;)
11:41 sri ye
11:42 sh9 joined #mojo
11:42 sri anyway, the solution seems to be a database/memcached
11:43 al is either one of them better suited for storing serialized objects?
11:43 al i guess i'll just D::D Dump it somewhere
11:44 sri you would have to manually clean up
11:44 sri memcached will invalidate data automatically for you
11:45 sri redis would work too i guess
11:45 diegok al: memcached/redis is what you want :)
11:46 al diegok, ok, which one exactly? :D
11:47 diegok al: if you only need that, I would go with memcached.
11:48 diegok al: if you aren't planning to scale, D::D would work, but you need to cleanup as sri have said.
11:49 MisterHatt anyone toyed with msgpack at all?
11:49 al yea, i don't like cleaning up
11:49 al thanks, sri, diegok
11:50 al is there a donation thingy somewhere besides buying tshirts?
11:51 sri MisterHatt: yea, it's not bad
11:52 sri but i prefer json
11:52 MisterHatt reasoning?
11:52 sri i don't like binary formats
11:52 MisterHatt ah
11:52 sri well, json is kinda binary too... but you get what i mean :)
11:54 MisterHatt yeah, mostly
11:54 sri messagepack reminds me a lot of bson
11:54 MisterHatt was wondering about using it internally to allow some components of a project I'm admining to communicate with each other
11:55 sri i've been bitten by binary protocols so often... if i have the choice i will always go with what's easiest to debug
11:56 MisterHatt fair enough, thanks for that
11:59 sh9 left #mojo
12:10 fmerges joined #mojo
12:14 Akron joined #mojo
12:30 j3nnn1 left #mojo
12:53 dekimsey joined #mojo
13:07 al any opinions on MojoX::Session?
13:11 Foxcool left #mojo
13:13 Foxcool joined #mojo
13:14 sri al: don't use it
13:15 gbacon joined #mojo
13:16 metaperl does $self->flash do sessioning automatically?
13:17 sri sessioning?
13:17 metaperl s/does/doesnt
13:17 al sri, not the response i was looking for ;)
13:17 metaperl hmm, I think I should but out of this conversation.. i need to run in 10 minutes
13:17 sri al: it predates core session support and is now abandonware
13:18 al sri, oh, ok
13:18 * marty waves good morning to everyone  o/
13:18 sh9 joined #mojo
13:18 sri happy apple day to you too :)
13:19 sri metaperl: you make very little sense atm
13:20 marty It's a good day.  I've finally a working, general purpose, mongodb editing framework on top of Mojo.
13:23 marty I found a bug in the perl mongodb driver.   I'm planning on switching to the native driver for Mojo when it's available.  Then, easy mongodb editing with no perl dependencies.  :)
13:26 koban` left #mojo
13:26 sri \o/
13:35 SmokeMachine joined #mojo
13:36 metaperl left #mojo
13:36 crab apple day?
13:36 sri it's wwdc!
13:36 SmokeMachine Hi, is there any known problem with Mojolicious::Lite using websocket on dotcloud?
13:36 sri SmokeMachine: it doesn't work
13:36 purl It's a Y2K error!  Panic!  Sue!
13:37 sri it's PSGI, which doesn't do websockets
13:37 crab that's as concise and to-the-point a response as you will ever see.
13:37 SmokeMachine sri: thare are no way to do it work?
13:37 sri not yet
13:37 sri they will have perl worker support with port forwarding at some point
13:38 SmokeMachine sri: so, everything that I can do is to wait?
13:39 sri well, to be fair, websockets are not production ready anyway
13:39 sri the protocol is not finished
13:40 sri with a bit of luck mozilla could be the first to ship the final protocol version in firefox 6
13:41 SmokeMachine sri: isnt the specification complete?
13:41 sri no
13:42 sri ietf-07 looks really good, but it is not frozen yet
13:42 sri chrome and safari don't even ship that version yet
13:42 sri not even in nightly builds
13:43 sri firefox aurora is the first
13:43 sri btw. i'm not sure we'll ever setup websocket applications like that (dotcloud)
13:44 sri that infrastructure is very easy to abstract away http://pusher.com/
13:44 sri dynamic content can just be broadcastet from some background worker process
13:44 SmokeMachine thanks god I am just studding it yet... and my app isnt going to production... :)
13:46 sri in the not too distant future i would expect a new kind of websocket message passing server
13:46 sri a message bus where you subscribe to channels
13:48 arthas joined #mojo
13:49 MisterHatt anyone able to give me a quick rundown of when to use the task or app namespaces instead of starting a new one for an application?
13:49 SmokeMachine sri: what do you mean?
13:50 MisterHatt sri: what do you mean perl worker support, as in ruby's background jobs equivalent or child processes or what?
13:50 SmokeMachine a websocket connected to many diferent servers?
13:51 sri SmokeMachine: no, see http://pusher.com
13:51 sri an nginx extension could pretty much provide the same service
13:52 sri MisterHatt: http://docs.dotcloud.com/components/python-worker/
13:53 MisterHatt and it works the same way for perl?
13:54 sri not yet
13:55 MisterHatt should be about the same tho if I udnerstood that correctly
13:55 MisterHatt quite a bit different from what I had in mind tho, based on ruby's background jobs
13:56 sri sad thing is node.js will have a huge headstart once websockets become mainstream
13:57 MisterHatt if you've got a moment, able to enlighten me re namespaces?
13:57 sri MisterHatt: i don't use either ;p
13:57 MisterHatt seems google isn't very forthcoming
13:58 MisterHatt so I see
13:58 SmokeMachine sri: that pusher looks good!
13:58 sri i was the first user of the Task namespace though, with Task::Catalyst back in the days
13:58 MisterHatt but mojo and catalyst don't really seem like the sort of things that would use either o them anyway
13:58 sri looking at it should be a good indicator for what it is meant to be used
13:58 MisterHatt from what I could tell, Task seemed to be more for things relating to cpan though
13:58 MisterHatt at least from reading perldoc on it
13:59 * sri sticks to cutesy names for new projects
14:00 MisterHatt catalyst is cute? :P
14:00 MisterHatt miyagawa seems to do that too, hmm
14:00 sri i used to be a fan of categories
14:00 MisterHatt ah
14:00 sri but the problem is that category namespaces discourage diversity
14:01 MisterHatt I think for some things they work tho
14:01 MisterHatt certinaly not globally though
14:01 sri you can also mix cutesy names with categories
14:02 MisterHatt unrelated: just realised cosimo strppone is back in italy, missed the past 2 melbourne.pm meetings and wanted to talk to him :/
14:02 MisterHatt fair enough
14:02 MisterHatt I'm trying to figure out what namespace to use for this small app I wrote for deploying stuff
14:02 MisterHatt sorta like vlad but a bit more feature rich I guess
14:03 sri sounds like App:: would work
14:03 MisterHatt I figure App is probably it but seeing as it processes a bunch of Tasks, that might be better
14:03 MisterHatt yeah
14:03 sri not Task
14:03 sri MisterHatt: see http://search.cpan.org/~flora/Task-​Catalyst-4.01/lib/Task/Catalyst.pm
14:03 sri tasks are collections of related modules
14:05 MisterHatt ah I see
14:05 MisterHatt App is definitely the one to use then I guess, thanks for the clarification
14:06 sri hmm, i've been erased from Task::Catalyst Oo
14:06 sri rafl--
14:06 MisterHatt somebody applied politics.pl to it?
14:07 metaperl joined #mojo
14:08 MisterHatt hurf, I should sleep, starting a new job tomorrow and it takes like an hour to get there :V
14:08 MisterHatt night all, thanks for the help
14:12 perlrocks Twitter: "Mojolicious bloody murder is not" (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/77739713619963905
14:13 sh9 left #mojo
14:16 perlrocks Twitter: "Omotta Mojolicious Q is a great and how can I support? A blog and Twitter success stories can be shared. Tsukemashou attract many people." (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/77740570885361664
14:19 perlrocks Twitter: "This person made a Mojolicious Sebastian Riedel @ kraih # yuruperl" (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/77741386719432704
14:20 marty bloody murder?   classic.
14:20 sri REDRUM!
14:20 marty It sounds like Yoda is reviewing Mojolicious
14:21 sri 5 stars getting it will
14:31 AmeliePoulain left #mojo
14:35 sh9 joined #mojo
14:38 perlrocks Twitter: "I tried to join Mojolicious of Gugurugurupu. International hackers feel like something w http://groups.google.com/group/mojolicious" (ja) --PerlCafe http://twitter.com/PerlCaf​e/status/77746296290689024
14:55 hidek left #mojo
14:57 hidek joined #mojo
14:58 hidek left #mojo
14:58 hidek joined #mojo
14:59 GitHub20 joined #mojo
14:59 GitHub20 mojo: master Sebastian Riedel * 78ca3b8 (4 files in 4 dirs): fixed a few small default status code bugs - http://bit.ly/mt4VTx
14:59 GitHub20 left #mojo
15:05 perlrocks Twitter: "Work hard together! Mojo is hello world sweat RT @ PerlCafe @ umeyuki worked hard to read in English too. . . But I'd love to play with Perl Orimasuru framework. The best regards w" (ja) --umeyuki http://twitter.com/umeyuki​/status/77752853921402880
15:07 sri i will miss translations so much :/
15:20 cosmincx left #mojo
15:20 D4RK-PH0ENiX left #mojo
15:23 ysyrota left #mojo
15:30 Foxcool left #mojo
15:34 GitHub143 joined #mojo
15:34 GitHub143 mojo: master Sebastian Riedel * 90a67e8 (2 files in 2 dirs): added experimental status code support to rendered method in Mojolicious::Controller - http://bit.ly/mCqLu9
15:34 GitHub143 left #mojo
15:43 Foxcool joined #mojo
15:48 bosphorus left #mojo
15:50 al does mojo expose a session id?
15:50 sri not directly
15:50 al or does memcached generate a unique identifier for what i put in it?
15:50 sri our session *is* the id
15:51 al yea.. how do i tie whatever i put into memcached to the session?
15:59 al ok, wrote a dumb 'sessionid' helper
16:01 sri just put the key into the session
16:08 sh9 left #mojo
16:10 al but i need one key per session
16:13 d4rkie joined #mojo
16:16 al cool, it's working
16:18 al i'm using $self->session(id => rand()) in a helper and then $memd->set($self->sessionid.'_data' => $data);
16:21 Akron What is wrong with the session as an id?
16:22 sri it can change
16:41 abra left #mojo
16:44 fmerges left #mojo
16:45 sh9 joined #mojo
16:47 Akron Sure, but I thought for extending the cache ...
16:48 Akron Nevermind. I don't like it, that bugfixing is so easy, but testfinding so hard.
16:52 abra joined #mojo
16:56 GitHub26 joined #mojo
16:56 GitHub26 mojo: master Sebastian Riedel * df377e6 (1 files in 1 dirs): fixed a warning in Mojo::IOLoop - http://bit.ly/mx8kFr
16:56 GitHub26 left #mojo
17:14 kaare left #mojo
17:19 Foxcool left #mojo
17:30 Foxcool joined #mojo
17:37 sri Akron: any is never wrong, i'd rather make it be always right
17:38 Akron Okay.
17:38 marty hmmm
17:38 marty sri:  Is this right?      perl -Mojo -e 'print Mojo::JSON->encode({ 'path' => "/a/b/c" })'
17:38 sri hmm, github bot is slacking again
17:39 sri marty: no
17:39 marty I did'nt think so.  :)
17:39 sri i see two bugs
17:39 sri who can beat that?
17:39 marty hehee.  You are like greased lightning, sri!
17:41 sri perl -Mojo -E'say Mojo::JSON->new->encode({path => "/a/b/c"})'
17:42 marty Hmm.  that gives me the same results
17:43 sri wow, lion will only be available via app store :o
17:44 sri marty: wait... that oneliner actually works Oo
17:44 marty ahhh, ok sri.  Thank you
17:44 sri there is a quoting bug 'path'
17:45 sri ->encode on class will only work if there are no errors
17:45 sri if there are it explodes
17:45 sri marty: anyway, what's wrong with the result?
17:46 marty the escaped forward slashes.
17:46 sri that's json
17:46 marty really?
17:46 sri YA RLY!
17:46 marty no, really?
17:47 marty :)
17:47 sri YA RLY!!!1
17:47 marty ok. Hmmmm.   so if we are storing a path how would be get it into the stash without the backslases?
17:48 * marty needs to learn how to write, read speak english.   geeeeesh
17:49 marty my spelling is horrible.
17:52 marty scratch that last question.  I think I'm looking at the problem wrong.
17:53 sri not sure i understand the problem
17:53 sri perl -Mojo -E'say Mojo::JSON->new->decode(Moj​o::JSON->new->encode({path => "/a/b/c"}))->{path}'
17:54 sri you might want to take a look at the result
17:54 marty I've got a path stored in a mongodb collection.  I want to insert it into a page without the escaped path.
17:54 sri the path is only exscaped in json form
17:55 sri it has roundtrip integrity
17:55 marty Ahh ok.  Makes sense.
17:58 spleenjack left #mojo
18:05 Sugar left #mojo
18:07 sri it's a bit ridiculous what apple is doing
18:07 sri so many new features are actually taken from existing apps
18:11 stephen joined #mojo
18:13 sri ROFL
18:13 sri and it keeps going on
18:14 marty Come to the dark side, sri.  Build your own linux box.  :)
18:15 sri not gonna happen ;p
18:26 tabbi joined #mojo
18:26 tabbi left #mojo
18:33 sri wow, icloud is free :o
18:37 sri haha, i'm sure google will be really pissed now
18:40 moritz w 19
18:40 sri w 20
18:48 fhelmber_ left #mojo
18:48 SmokeMachine left #mojo
18:49 SmokeMachine joined #mojo
18:58 sh9 left #mojo
18:58 AigleNoir left #mojo
18:59 AigleNoir joined #mojo
19:09 marty sri:  perl -Mojo -E'say Mojo::JSON->new->encode({path => true})'
19:10 marty should true be quoted on output?
19:10 sri marty: umm, what the hell is true?
19:10 purl true is http://search.cpan.org/perldoc?true
19:10 marty the opposite of false.  tada!!!  :)
19:11 * sri pats purl
19:11 purl how condescending
19:11 sri marty: there is no true in perl
19:11 marty no. the encoded jason.   it's quoted
19:11 marty ie:  name: "true"     instead of name: true
19:11 sri because it's a string
19:12 sri perl -Mojo -E'use strict; say Mojo::JSON->new->encode({path => true})'
19:12 sri try that
19:13 marty ahhhh.,  I see.  so sowy, my bad
19:14 sri now, what you want is described here http://mojolicio.us/perldoc/Mojo/JSON#DESCRIPTION
19:14 * marty is reading...
19:15 mattastrophe joined #mojo
19:15 marty oooo.  Ok, now starting to make sense.  Thanks again, sri!
19:26 sri http://www.youtube.com/watch?v=FwbA5JUQ3bA # wow, what an awesome cover of immigrant song
19:27 sri they will get a second oscar i bet
20:00 tempire wwdc was pretty standard.  no real surprises
20:00 tempire glad apple finally decided to give icloud services for free, since google has been doing it for ages.
20:01 tempire the cost was too old-world thinking
20:01 tempire the itunes match sets the bar, though.  that's good
20:01 tempire it's funny that it's the same thing mp3.com got pounced on for
20:02 tempire apple put a lot of leg work into making it acceptable for the riaa, though, so it's fitting
20:02 tempire osx looks dated in comparison to new win8
20:03 tempire ios does too, for that matter
20:12 memowe $ perl -Mojolicious
20:12 memowe Viva la revolution!
20:12 memowe (print"Viva la revolution!\n";close STDIN;1
20:13 memowe )
20:13 memowe :)
20:33 sri tempire: the instapaper guy just tweeted "Shit."
20:35 sri google is in trouble
20:38 sri http://oreilly.com/catalog/9780596004927/ # OMG!
20:39 sri new camel book
20:41 yko Covers Perl 5.8
20:41 sri 5.14
20:41 sri 4th edition
20:41 yko http://oreilly.com/catalog/9​780596101053/preview#preview
20:42 sri see my link
20:42 sri i'm talking about *THE CAMEL*
20:42 sri 1000 pages of perl awesomeness
20:42 yko damn. i'm wrong
20:43 yko there's no cover image on your link!
20:43 sri because it is brand new
20:43 yko how can i read book with no images!
20:43 sri they announced it a few mins ago
20:43 marty I was'nt aware that Oreilly did Perl books any more.   wakka wakka wakka
20:43 sri apparently they listen
20:43 marty nice
20:44 sri learning perl will have a new edition out soon too
20:44 sri also covering 5.14
20:58 yko is Mojo::Exception for internal use only or users can use it also in their apps?
20:58 sri it's designed for internal use
20:58 yko thanks
20:59 sri the lack of documentation should be a hint ;)
21:00 sri the api is also very unpolished
21:00 purl okay, sri.
21:00 espent_ joined #mojo
21:00 sri argh
21:00 yko i'd never ask if i don't doubt
21:00 sri purl: the api?
21:00 purl i heard the api was fine.
21:00 sri Oo
21:00 espent left #mojo
21:01 yko lack of pod and unfriendly design was a hint
21:01 sri don't think we have end user classes without good examples
21:01 yko purl, api?
21:01 purl api is probably abstra ct or very unpolished
21:02 sri oh
21:02 yko your definition went there
21:02 * marty just registered mojotopia.com
21:03 marty \o/ I'll be famous for sure!  :)
21:05 dekimsey left #mojo
21:24 sri heh
21:26 Akron left #mojo
21:29 arthas left #mojo
21:35 SmokeMachine left #mojo
22:04 metaperl left #mojo
22:12 Akron joined #mojo
22:33 perlrocks Twitter: "I tried to join Mojolicious of Gugurugurupu. International hackers feel like something w http://groups.google.com/group/mojolicious" (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/77865755034652672
22:35 perlrocks Twitter: "And More Mojolicious something I came from members of the ML is a fellow feeling in the world could" (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/77866120576634881
22:42 metaperl joined #mojo
23:30 tempire http://asset.soup.io/asset/1132/5699_1492_500.jpeg
23:56 Akron left #mojo

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