Camelia, the Perl 6 bug

IRC log for #mojo, 2010-10-28

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

All times shown according to UTC.

Time Nick Message
00:07 perlrocks Twitter: "RT @kraih: RT @vtivti: How to avoid #Unicode pitfalls in #Mojolicious #perl http://bit.ly/aPSIKS" --xantus http://twitter.com/xantus/status/28938641073
00:07 dotan joined #mojo
00:07 dotan left #mojo
01:12 stephen left #mojo
02:02 * xantus is working on the new mojopaste
02:02 xantus you're going to love it
02:06 kthakore hello
02:06 kthakore does HTLM::Zoom 'plug' into MOjo?
02:09 xantus MY EYES
02:09 kthakore xantus: guess not
02:09 kthakore Z:p
02:11 xantus ;) I suppose it would work if you passed $z->to_html to $c->render_data( )
02:18 perlrocks left #mojo
03:21 alnewkirk left #mojo
03:39 xantus grr, chinese spam on the list
03:41 xantus BANNED, hehe
03:42 xantus and removed
04:04 tholen42 joined #mojo
04:05 alnewkirk joined #mojo
04:06 tl left #mojo
04:32 GabrielVieira joined #mojo
04:35 alnewkirk left #mojo
04:38 xantus Use of uninitialized value $values in join or string at lib/Mojo/Headers.pm line 119,  lib/Mojo/Transaction/WebSocket.pm line 406 is setting the header from the request without checking it
04:39 GabrielVieira left #mojo
04:39 GabrielVieira joined #mojo
04:41 xantus odd thing though, I haven't setup a websocket route yet, so its trying to respond to an upgrade request without checking if it should
04:43 xantus sri: ^^  :)
05:59 ysyrota joined #mojo
06:23 marcus goood morning
06:24 kvorg joined #mojo
06:38 ysyrota left #mojo
06:38 ryanc left #mojo
06:38 fish_ left #mojo
06:38 rhaen left #mojo
06:38 und3f[A] left #mojo
06:38 hidek___ left #mojo
06:38 meraxes left #mojo
06:38 stephan48 left #mojo
06:38 jamesw left #mojo
06:38 crab left #mojo
06:38 kthakore left #mojo
06:38 diegok left #mojo
06:38 charsbr left #mojo
06:38 memowe_ left #mojo
06:38 marcus left #mojo
06:38 Alias_ left #mojo
06:38 chansen left #mojo
06:38 xantus left #mojo
06:38 viking-3 left #mojo
06:38 ashleydev_logbot left #mojo
06:38 go|dfish left #mojo
06:49 arthas joined #mojo
07:19 breaker313 joined #mojo
07:43 MojoGuest702 joined #mojo
07:47 MojoGuest702 left #mojo
07:47 MojoGuest858 joined #mojo
07:48 MojoGuest858 left #mojo
07:51 Alias_ joined #mojo
08:08 Alias_ left #mojo
08:08 memowe joined #mojo
08:08 ysyrota joined #mojo
08:08 marcus joined #mojo
08:08 charsbr joined #mojo
08:08 hidek___ joined #mojo
08:08 stephan48 joined #mojo
08:08 crab joined #mojo
08:08 fish_ joined #mojo
08:08 kthakore joined #mojo
08:08 ryanc joined #mojo
08:08 rhaen joined #mojo
08:08 diegok joined #mojo
08:08 jamesw joined #mojo
08:08 und3f[A] joined #mojo
08:08 meraxes joined #mojo
08:08 Alias_ joined #mojo
08:08 chansen joined #mojo
08:08 xantus joined #mojo
08:08 viking-3 joined #mojo
08:08 go|dfish joined #mojo
08:08 ashleydev_logbot joined #mojo
08:08 Alias__ joined #mojo
08:21 pid joined #mojo
08:38 koban joined #mojo
08:52 DaTa joined #mojo
09:00 MojoGuest605 joined #mojo
09:00 MojoGuest605 From: http://news.ycombinator.com/item?id=1277067 (25 hits)
09:00 MojoGuest207 joined #mojo
09:00 MojoGuest207 From: http://news.ycombinator.com/item?id=1277067 (25 hits)
09:01 MojoGuest605 Hi there
09:01 MojoGuest605 is now known as hacker_man
09:01 hacker_man hi there
09:01 hacker_man anyone here?
09:01 DaTa hi
09:01 MojoGuest207 left #mojo
09:01 hacker_man data
09:02 hacker_man nice to meet you
09:02 hacker_man are you the developer of this application?
09:02 hacker_man left #mojo
09:03 spleenjack joined #mojo
09:13 DaTa impatient
10:29 breaker313 left #mojo
10:39 pid left #mojo
10:43 sri moin
10:44 vti_ moin
10:45 sri xantus: websocket is the constraint, normal get works too for a handshake
10:55 pid joined #mojo
10:57 fish_ sorry for asking again, but maybe there is a mojo based wiki somewhere now? ;)
11:03 vti_ there was mojowka
11:14 fish_ "was"?
11:14 vti_ i am not sure about its state now
11:24 marcus mojohawk
11:24 sri oh cool, you can be on top of planet perl ironman by blogging from the future
11:24 marcus :D
11:27 breaker313 joined #mojo
11:27 breaker313_ joined #mojo
11:27 breaker313_ left #mojo
11:30 vti_ marcus: my rss is not that bad ;p
11:34 sri "not that bad" = still bad?
11:36 vti_ we'll see when i post a new article
11:36 vti_ :D
11:36 sri date it 2 years into the future!
11:37 vti_ and call it 'to mydescendants
11:37 vti_ *to my descendants
11:38 sri haha
11:39 marcus sounds good.
11:48 DaTa left #mojo
11:48 DaTa joined #mojo
11:53 GabrielVieira left #mojo
11:57 DaTa left #mojo
12:01 MojoGuest202 joined #mojo
12:01 MojoGuest202 From: http://news.ycombinator.com/item?id=1277067 (27 hits)
12:03 sri how come nobody ever comments javascript code?
12:09 DaTa joined #mojo
12:11 fish_ uhm.. Thu Oct 28 14:10:34 2010 error MojoX::Dispatcher::Routes:118 [21118]: Can't locate object method "write_cb" via package "Mojo::IOLoop" at ./websocket-irc-example-1 line 79, <DATA> line 1165.
12:11 sri outdated code
12:12 fish_ possibly the websocket-irc-example needs an outdated mojo..?
12:12 sri right
12:12 sri ->write_cb(sub {...}) became ->write('chunk') a few months ago
12:14 fish_ is there a more recent websocket irc client? i know, i know, this is just an example.. but i hoped i could use it for small group of people..
12:15 sri i actually thought xantus had updated his irc client
12:16 fish_ nope.. http://github.com/xantus/mojo-webs​ocket-examples/tree/master/script/
12:17 sri not too hard to update though
12:18 DaTa_ joined #mojo
12:19 sri mojo/examples/connect_proxy.pl contains fresh tcp client code
12:20 fish_ i'll try it
12:21 sri writing/not_writing have become obsolete and write_cb got replaced with write
12:22 sri so the whole code should get much simpler
12:22 sri you just push writes now, instead of schedule a pull
12:25 dotan joined #mojo
12:26 lee7 left #mojo
12:27 fish_ sri: so i can just use write as soon as i want to send out a message, right? (in this case after receiving a message from a websocket client for example?
12:27 sri right
12:42 lee7 joined #mojo
12:46 DaTa left #mojo
12:46 DaTa_ is now known as DaTa
12:47 achromic left #mojo
12:48 achromic joined #mojo
13:05 sri and here i thought stargate universe couldn't get worse...
13:11 perlrocks joined #mojo
13:11 perlrocks Twitter: "RT @ vtivti: ETag caching Adding to your app http://bit.ly/bCCROh Mojolicious # # perl" (vi) --kraih http://twitter.com/kraih/status/28987389381
13:13 vti_ what language do you speak? vi?
13:13 sri hehe
13:13 sri i let google translate detect the language
13:14 sri not the slightest idea what vi is
13:18 sri hmm
13:18 sri wonder if we should add app->hook(after_dispatch => sub {...});
13:18 sri similar to app->helper(...)
13:18 vti_ :)
13:19 vti_ ye, ->plugins doesn't feel intuitive, i had to look into docs
13:20 perlrocks Twitter: "RT @ kraih: RT @ vtivti: ETag caching Adding to your app http://bit.ly/bCCROh Mojolicious # # perl" (vi) --j3nnn1 http://twitter.com/j3nnn1/status/28988110371
13:23 fhelmber_ joined #mojo
13:27 MojoGuest202 left #mojo
13:31 sri damn
13:31 sri i need to break plugin hooks :(
13:32 sri they are not consistent with helpers
13:32 sri @_ contains the plugins object before $c
13:33 vti_ we have to have mojolicious delta posts too :)
13:33 sri i might have to add a very ugly hack for deprecation :S
13:34 vti_ that would make it easier to update apps
13:38 MojoGuest116 joined #mojo
13:38 MojoGuest116 Test
13:38 MojoGuest116 is now known as namealex
13:39 namealex hi
13:39 namealex /help
13:39 namealex /?
13:39 namealex test 123
13:39 namealex exit
13:39 namealex bye
13:39 namealex logout
13:39 namealex left #mojo
13:40 MojoGuest667 joined #mojo
13:40 MojoGuest667 left #mojo
13:54 breaker313 left #mojo
14:04 GitHub93 joined #mojo
14:04 GitHub93 mojo: master Sebastian Riedel * 14377dc (13 files in 6 dirs): eprecated the old plugin hook calling convention and added experimental hook method to Mojolicious - http://bit.ly/bZN5y3
14:04 GitHub93 left #mojo
14:05 perlrocks Twitter: "Made the #mojolicious plugin system a lot more fun today. :) http://bit.ly/9MyTxc #perl" --kraih http://twitter.com/kraih/status/28992053069
14:11 GitHub34 joined #mojo
14:11 GitHub34 mojo: master Sebastian Riedel * 3f59cda (13 files in 6 dirs): Deprecated the old plugin hook calling convention and added experimental hook method to Mojolicious - http://bit.ly/avDyI6
14:11 GitHub34 left #mojo
14:13 perlrocks Twitter: "Made the #mojolicious plugin system a lot more fun today. :) http://bit.ly/9JR9Gx #perl" --kraih http://twitter.com/kraih/status/28992721429
14:14 sri interesting how the old link still works
14:14 kvorg left #mojo
14:14 sri even though i rewrote history
14:15 DaTa you have to wait for the next git gc run :)
14:19 sri STOP THE WORLD!
14:23 vti_ so my example doesn't work now
14:28 sri ?
14:29 sri vti_: your latest blog post?
14:29 sri that works, but is deprecated
14:29 vti_ oh, you've added a deprecation hack
14:32 sri ye, i try not to break stuff if i don't have to
14:32 GitHub145 joined #mojo
14:32 GitHub145 mojo: master Sebastian Riedel * f8cb91f (1 files in 1 dirs): cleanup - http://bit.ly/cCU66E
14:32 GitHub145 left #mojo
14:33 vti_ but i will update it hehe
14:40 GitHub4 joined #mojo
14:40 GitHub4 mojo: master Sebastian Riedel * e70ad70 (1 files in 1 dirs): cleanup - http://bit.ly/a9z91P
14:40 GitHub4 left #mojo
15:09 koban left #mojo
15:27 kvorg joined #mojo
15:27 perlrocks Twitter: "RT @kraih: Made the #mojolicious plugin system a lot more fun today. :) http://bit.ly/9JR9Gx #perl" --insurgentsoft http://twitter.com/insurge​ntsoft/status/28999236700
15:30 dotan left #mojo
15:30 dotan joined #mojo
15:39 dotan so this change breaks existing plugins?
15:40 vti_ no
15:41 dotan how so? doesn't it now pass the controller as the 1st arg instead of the plugins obj as the 1st arg?
15:42 dotan (except for one hook where it's the transaction object before)
15:43 dotan My plugin sets a before_static_dispatch hook:
15:43 dotan http://github.com/dotandimet/mojol​icious-plugin-alias/blob/master/li​b/Mojolicious/Plugin/Alias.pm#L63
15:44 dotan http://github.com/kraih/mojo/commit/3f59cd​a732c51e6e01ea069bbbc0e83e928bcbb7#diff-1
15:44 dotan It throws away the 1st arg (because plugins is a pretty useless object), but now that's the controller, so it's broken.
15:44 DaTa left #mojo
15:46 vti_ =/
15:49 dotan the only way I see not to break plugins is to call the hooks differently depending on how they were set (plugins->add_hook hooks get run with old-style args, app-hook hooks get run with new-style args).
15:49 dotan Backwards compatibility. It's not pretty.
16:11 perlrocks Twitter: "The company Hakkason html5, I mean WebSocket decided to try to make something with. Possible real-time collaborative editing canvas but with your problem ... What Peintoapuri called to make in the period. Should I Mojolicious server side to implement." (ja) --hndtykz http://twitter.com/hndtykz/status/29002960436
16:11 xantus fish_: doh
16:12 xantus sri: you're talking about that lame ass episode with the plants?
16:14 xantus fish_: did you patch it ok?
16:14 fish_ xantus: now :/ i realized that i should first really understand who it does and that this would take a little bit more time than i had
16:15 xantus I can update it when I leave work
16:19 xantus sri: What are the writers thinking?  They need a lesson in multiple story arc writing.
16:20 ysyrota left #mojo
16:43 pid left #mojo
16:45 sri dotan: NOTHING BREAKS!
16:47 dotan sri: explain - what's the 1st arg to a before_dispatch_static hook function?
16:47 sri app->hook is a new method
16:47 DaTa joined #mojo
16:47 sri it uses a hack to remove the plugin obejct for now
16:47 sri after the deprecation period your app will break
16:47 sri but not yet
16:48 sri plugins->add_hook() has not changed at all
16:48 sri YET!
16:49 dotan sri: so if I call add_hook directly, I still get that useless first arg, if I call app->hook, I use new style callbacks?
16:49 sri right
16:50 dotan sri: sweet :)
16:50 sri but only until the deprecation period ends
16:50 sri after that it will break
16:50 sri so you have a few months to update your apps to the new style :)
16:51 dotan sri: breaking it seems like gratuitous fastidiousness.
16:52 sri two different styles to do the same thing is wrong
16:53 dotan sri: true.
16:53 xantus I like it!
16:55 sri all documentation and examples already uses the new style exclusively, and two major releases deprecation period seems quite reasonable
16:56 sri the important thing here is that there is clearly added value in the new style, it's not a just for fun renaming :)
16:57 sri is it just me or does the redis cluster design look like a huge fail?
16:58 sri http://bit.ly/aYSLxG
16:59 sri so much logic gets pushed to the client side
17:07 dotan sigh. My mojolicious slideshow app is now usable enough that I actually need to start using it to write my presentation...
17:07 sri oh noes
17:08 sri wonder if ssd will change the way we use memcached today
17:09 spleenjack left #mojo
17:10 xantus dotan: :)
17:13 sri http://www.cpantesters.org​/distro/M/Mojolicious.html # why don't we have any windows tests yet? :(
17:18 vti_ because it's windows
17:21 sri windows doesn't like us?
17:25 vti_ it doesn't like anyone :D
17:25 sri :D
17:32 dotan left #mojo
17:34 sri hmm, i might actually add some crud features to MojoX::Riak :)
17:35 sri riak-search is so awesome
17:36 sri calling riak a key/value store really doesn't do it justice, with riak-search you get automatic free indexing of json/xml/text objects
17:41 vti_ marcus: looks like feed is ok now
17:58 dotan joined #mojo
18:01 stephen joined #mojo
18:13 alnewkirk joined #mojo
18:28 marcus vti_: yepp. nice
18:32 xantus is riak backed by a file store yet?
18:34 xantus or another question...why should I switch from couchdb
19:33 dotan left #mojo
19:57 marcus wow this move is stupid
19:58 marcus d't watch frozen, it's a waste of time.
20:01 MojoGuest706 joined #mojo
20:01 MojoGuest706 From: http://www.google.com/search?hl=en&amp;saf​e=off&amp;q=extjs+websocket&amp;aq=f&amp;a​qi=g1g-sx1&amp;aql=&amp;oq=&amp;gs_rfai= (1 hits)
20:01 MojoGuest706 left #mojo
20:03 sri marcus: at least now we know that jumping is a pretty stupid idea!
20:04 sri and that there are countless wolves out there waiting for their chance to eat us!
20:04 sri xantus: file store?
20:05 sri as in does it store data on disk?
20:05 sri yes it does
20:05 sri it is kinda the opposite of couchdb
20:05 sri made to scale out
20:06 sri you just throw machines at it
20:07 sri that process is pretty much effortless
20:08 marcus sri: jumping could maybe be smarter if you tried minimizing the distance a bit. For instance hang by your arms
20:08 sri riak itself is basically a smart key/value store, where you put data in buckets
20:08 rhaen moin!
20:08 marcus or even better,tie your jacket to the cart.
20:08 sri marcus: haha, my first thought!!!
20:08 sri TIE YOUR DAMN JACKETS TOGETHER!!!
20:08 rhaen damnit!
20:09 sri moin
20:09 marcus sri: they are too dumb to get any sympathy
20:09 sri s/moin/evening/
20:09 rhaen Do what sri said!
20:09 marcus they didn't even close their jackets properly. No wonder they are cold
20:09 rhaen btw, node.js is web scale
20:09 sri umm...no it is not
20:10 sri it has no built in functionality that would make it scale like at all
20:10 rhaen oh, why not.
20:10 rhaen oh. ok.
20:10 sri it is single process
20:10 sri it allows forking, but you have to build the infrastructure yourself
20:11 sri something built on top of node.js might be able to scale
20:11 rhaen right - brr. true. Maybe they shouldn't a browser vm :)
20:11 und3f[A] is now known as und3f
20:12 rhaen I just came home from business travels and I looked into it.
20:12 rhaen back in Mojo land - and it feels good!
20:13 sri http://github.com/alexyoung/i​rcd.js/blob/master/server.js # this is how node.js code looks in the end
20:14 marcus ugly ascii art?
20:14 sri and this is one of the better examples
20:14 sri marcus: javascript kiddies :D
20:15 sri well, this is the next generation of programmers making their first steps
20:15 rhaen hm, I don't like it - brr. Do ctags support javascript?
20:15 sri from the browser to the server side
20:16 sri the big selling point of node.js will be the one language to rule them all argument
20:16 sri even though it is silly
20:18 marcus one silly language to rule them all?
20:19 sri ;p
20:19 rhaen :)
20:20 * sri wonders where yko hides again
20:21 sri xantus: but what makes riak truly awesome is riak-search
20:22 sri like in couchdb you throw json at it, but it gets automatically indexed for fulltext search if activate it for a bucket
20:22 sri *+you
20:22 sri it's also more of a traditional database in the sense that it doesn't want to be exposed to the internet directly
20:23 * marcus exposes himself to the internet.
20:24 xantus nice
20:24 rhaen send pics!
20:24 xantus lol
20:24 * xantus covers marcus with a jacket
20:25 marcus rhaen: http://bit.ly/aeWJw
20:25 rhaen oooold!
20:25 sri think of riak as a cassandra written in erlang with a rest api and a setup process that only involves running two or 3 commands
20:26 marcus what can I say, I'm a dinosaur...
20:26 sri i have a feeling riak is getting big
20:26 sri search is just too cool
20:26 xantus ok, write a ircd in mojo in less than 150 lines .... go!
20:26 * sri exposes marcus to 4chan
20:27 sri xantus: i think there's an app....umm...cpan module for that!
20:27 marcus in fact, I think xantus's previous employer made one
20:27 xantus marcus: hahaha, the corporate filter labed that picture as 'sex' and blocked it
20:27 xantus labeled
20:28 marcus xantus: that's probably good, not very SFW
20:28 xantus great, the security goons are going to come looking for me
20:28 sri hmm, guess i'll not click
20:29 marcus never follow a bit.ly link, you don't know where it'll take you
20:29 marcus (in this case, to tubgirl.ca)
20:29 xantus its probably tubgirl
20:29 xantus lol
20:29 rhaen marcus: na, click on everything!
20:29 xantus ugh
20:30 marcus I used to have two girls working for me, when I was product manager for a search engine
20:30 marcus they stopped clicking on my links after I sent them 2 girls 1 cup...
20:31 sri that would be a very expensive lawsuit around here i guess
20:31 xantus I knew it, marcus is actually tubgirl
20:32 marcus well, if your job is to filter search engine porn queries, you can't really sue for internet links.
20:32 xantus it was research....yeah research
20:33 sri sounds like a job for anonymous
20:33 sri or a new kind of captcha :)
20:34 sri [offensive] [not offensive]
20:34 sri hmm, i better patent that -.-
20:35 marcus seems like a captcha no republican will get past.
20:36 sri lol
20:42 perlrocks Twitter: "http://community.livejournal​.com/shlomif_tech/53479.html - Rehovot #Perl #mongers meeting - #POE and #mojolicious" --shlomif http://twitter.com/shlomif/status/29021561719
20:58 perlite_ joined #mojo
21:02 perlite left #mojo
21:04 xantus hahahah
21:14 xantus Has Node.js got you down?  Did you fall for the web scale lies, and now you can't even read your own code?  Get your Mojo back!  http://mojolicio.us/
21:14 xantus ;)
21:14 sri lol
21:15 xantus Someone HAS to tweet that
21:18 alnewkirk left #mojo
21:25 xantus left #mojo
21:25 xantus joined #mojo
21:32 xantus wth?
21:39 arthas left #mojo
21:43 baton8 left #mojo
21:46 alnewkirk joined #mojo
21:52 baton8 joined #mojo
22:01 stephen left #mojo
22:03 stephen joined #mojo
22:14 und3f is now known as und3f[A]
22:31 fhelmber_ left #mojo

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