Camelia, the Perl 6 bug

IRC log for #mojo, 2011-08-23

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

All times shown according to UTC.

Time Nick Message
00:10 mattastrophe joined #mojo
00:28 tempire !
00:28 tempire mojocast-inspired cookies
00:28 tempire https://skitch.com/tempire/f1215/img-2659
00:31 Alias_ joined #mojo
00:36 sri *drool*
00:39 xaka glass of milk? wtf!
00:46 GitHub80 joined #mojo
00:46 GitHub80 [mojo] kraih pushed 1 new commit to master: http://git.io/lF32tQ
00:46 GitHub80 [mojo/master] release preparations - Sebastian Riedel
00:46 GitHub80 left #mojo
00:47 crab what does the experimental t helper do? fetch tea?
00:47 sri serves crab salad
00:49 sri now please stand over there on the huge X mark on the ground
00:49 perlrocks Twitter: "Mojolicious 1.87 by SRI - http://t.co/OMRyiGl" (ms) --cpan_new http://twitter.com/cpan_new​/status/105803902678597632
00:51 perlrocks Twitter: "Released #mojolicious 1.87 with a few usability improvements, enjoy! http://t.co/rigdhmi #perl" --kraih http://twitter.com/kraih/status/105804425670574080
01:00 perlrocks Twitter: "12k SLOC in libraries and 14k SLOC in tests. :) #mojolicious #perl" --kraih http://twitter.com/kraih/status/105806465759059969
01:03 GabrielVieira joined #mojo
01:05 noganex joined #mojo
01:10 Akron sri: I'm missing a split method in Mojo::ByteStream, that turns everything in a Mojo::Collection!
01:13 xaka joined #mojo
01:17 sri Akron: interesting idea!
01:20 Akron sri: Would allow really funny oneliners!
01:21 sri indeed
01:22 Akron b('  Hello World ')->trim->split(/\s+/)->reverse()->join(' ')->quote ...
01:27 tempire holy chaining objects batman!
01:31 d4rkie joined #mojo
01:34 GitHub35 joined #mojo
01:34 GitHub35 [mojo] kraih pushed 1 new commit to master: http://git.io/mgVkqA
01:34 GitHub35 [mojo/master] added experimental split method to Mojo::ByteStream - Sebastian Riedel
01:34 GitHub35 left #mojo
01:42 sri Akron: ;)
01:42 GitHub17 joined #mojo
01:42 GitHub17 [mojo] kraih pushed 1 new commit to master: http://git.io/LmPWIg
01:42 GitHub17 [mojo/master] less magic - Sebastian Riedel
01:42 GitHub17 left #mojo
01:46 Akron \o/
01:49 sri perl -Mojo -E'b("12345")->split("")->​shuffle->join("\n")->say'
01:49 sri :D
01:49 sri perl really needs proper autoboxing in core -.-
01:56 elb0w__ joined #mojo
02:27 xaka ok, now what about j for json in templates? :) useful for javascript when you just want to convert perl to js
02:28 xaka b(j(object)) :) hoho
02:29 * xaka wonders when it'd be possible to write a(b(c(d(...(object))))
02:34 vel joined #mojo
02:38 ispy_ joined #mojo
02:41 sri perl -Mojo -E'a(b(c(d("mojolicio.us")->dom->ht​ml->body->a->img->{id}))->to_string => {text => "LULZ"})->start' get /noraptor
02:44 sri or maybe
02:44 sri perl -Mojo -e'a(b(c(d("mojolicio.us")->dom->ht​ml->body->a->img->{id}))->to_string => {data => f("mojolicio.us")->dom->html-​>head->title->text})->start' get /noraptor
02:44 sri a b c d e and f! (kinda)
02:45 Alias_ left #mojo
03:12 * crab shudders
03:21 tempire b(crab)->dom->body->split
03:21 ispy_ joined #mojo
03:34 xaka render_to(nice_food => crab)
03:34 xaka s/render/respond/
03:34 elb0w__ joined #mojo
03:34 alnewkirk ?
03:35 xaka !
03:38 tempire .
03:40 Akron ; What am I doing wrong here? Only the post way is present. https://gist.github.com/1164293
03:43 alnewkirk Akron, UMMMM
03:43 alnewkirk add_shortcut is a route method
03:43 Akron alnewkirk: Yes ... so ... ?
03:44 alnewkirk ... and look at the docs
03:44 alnewkirk https://metacpan.org/module/Mo​jolicious::Routes#add_shortcut
03:45 Akron Well - its a routes method and I'm applying it to a route. Looks fine to me.
03:47 Akron alnewkirk: I can't see what I'm doing wrong.
03:48 alnewkirk Akron, https://metacpan.org/module/M​ojolicious::Routes#SHORTCUTS
03:58 Foxcool joined #mojo
04:01 Akron alnewkirk: But what's the difference of what I am doing? I still can't see your point.
04:03 alnewkirk Akron, you're chaining the routes wrong
04:03 alnewkirk this $r->route('/') is what's being passed to the shortcut
04:04 sri Akron: alnewkirk is just wrong
04:04 d4rkie joined #mojo
04:04 alnewkirk I am?
04:04 sri via just doesn't generate a route
04:04 sri thinko
04:07 sri you want ->route->via->to or ->get->to
04:07 alnewkirk sri, which is what he has
04:07 sri inside the shortcut!
04:08 sri he calls ->to twice, which only makes sense if you want new endpoints
04:08 sri everything else is perfectly fine
04:08 alnewkirk which is waht I just said five minutes ago that you said was wrong
04:09 sri no you did not
04:09 alnewkirk yes I did
04:09 alnewkirk [00:04] <alnewkirk> Akron, you're chaining the routes wrong
04:09 sri …i'm done talking to you
04:09 alnewkirk why would you need to declare that? lame
04:11 alnewkirk Akron, essentially you need to create a new route for each, one way to write it would be to create the route in the shortcut instead of the other way around
04:12 alnewkirk think, $r->do_via('/');
04:12 Akron Hm.
04:13 * tempire sprinkles glitter on miscommunications
04:13 * alnewkirk melts
04:14 alnewkirk anyone interested in helping me get the kinks out, feel free to test and feedback, https://github.com/alnewkirk/Mojo​licious-Command-generate-rest_app
04:14 perlrocks Twitter: "Mojolicious a pound upgrade is ..." (ja) --oyomeni_onihime http://twitter.com/oyomeni_oni​hime/status/105855369330434049
04:15 * sri goes to sleep
04:15 sri nn
04:17 Akron So there is no way to get what I want just by changing the shortcut? I have to change the routing in advance? There's no wai to introduce new null endpoints?
04:18 alnewkirk say wah?
04:19 alnewkirk did you do $r->do_via('/')?
04:19 * tempire serenades sri to sleep with the crunch of peanut butter cookies
04:20 Akron alnewkirk: That changes the api of the shortcut - I want to change it inside the shortcut.
04:20 Eugene joined #mojo
04:21 alnewkirk Akron, how so?
04:23 alnewkirk Akron, https://gist.github.com/1164293#comments
04:24 * alnewkirk is off to bed as well, 2 days no sleep, ... wife is all grumpy n $#!++
04:24 Akron alnewkirk: Yes, thank you. But this is not really elegant when you want to separate routing from crud.
04:24 Akron alnewkirk: Good night and thank you very much!
04:25 Akron sri: nn!
04:25 alnewkirk Akron, why not, add more args to the shortcut
04:25 Akron alnewkirk: This would make it even worse!
04:26 alnewkirk fearless leader has given you all the neccessary tool Akron, see http://mojolicio.us/perldoc/Mojo​licious/Guides/Routing#Shortcuts
04:26 alnewkirk what should give you the cruddy goodness you seek
04:26 alnewkirk that*
04:27 * alnewkirk out
04:28 Akron Ah .... !
04:28 Akron Right! Yes!!!
04:28 Akron Thank you ... I totally missed that.
04:35 sri sorry for being a bit rude… little sleep deprived
04:35 sri now really Zzz
04:35 Akron sri: sleep well! And thank you!
04:42 Firkin joined #mojo
05:10 plu joined #mojo
05:36 alpha6 joined #mojo
05:40 plu joined #mojo
05:41 noganex_ joined #mojo
05:53 crab it's hard to be polite on irc all the time.
05:53 crab but it's always good to try hard anyway.
05:54 crab akron: how did you solve your problem?
05:54 yko joined #mojo
05:55 crab hi yko.
05:55 yko hello crab
05:55 crab what are you up to today?
05:55 sromanov joined #mojo
06:13 perlrocks Twitter: "Mojolicious Web Framework - Join the revolution! Http://t.co/M8D2gdP" (ja) --fifnel http://twitter.com/fifnel/​status/105885305218924544
06:21 crab http://usenix.org/events/hotos03/t​ech/full_papers/candea/candea.pdf <- crash-only software
06:35 alpha6 joined #mojo
06:52 Foxcool joined #mojo
07:06 judofyr joined #mojo
07:11 lammel joined #mojo
07:11 cosmincx joined #mojo
07:12 rhaen good morning
07:12 purl For you maybe.
07:13 preflex joined #mojo
07:14 marcus oh hai
07:15 rhaen hai!
07:15 plu fish!
07:16 marcus phish?
07:16 purl i think phish is at http://www.phish.net or at http://www.phish.com/
07:16 rhaen puh!
07:16 rhaen I don't get some of the concepts of Mojo :(
07:17 rhaen like - the client part.
07:17 marcus the client part is the best part!
07:17 marcus what don't you understand about it?
07:18 xxtjaxx Is it possibly to forward routes into a submodule/plugin in Mojolicious::Lite?
07:18 marcus xxtjaxx: yes, see mount plugin
07:18 rhaen if I have a form on a webserver like <input type=text name=a /> <input type=text name=b />
07:18 rhaen what's the way to send a get req?
07:18 rhaen like:
07:19 rhaen perl -Mojo -e "g('server/form', { a => 'value for a', b => 'value for b'});'
07:20 marcus rhaen: if you want it in the url, construct the url with Mojo::URL
07:20 marcus rhaen: $url->params->append(..) ?
07:20 xxtjaxx get '/form' => sub { my $self = shift; $self->render(text => $self->param('a'); }
07:21 rhaen marcus: ah, so - my example is useful for forms with POST only, right?
07:21 marcus rhaen: post_form is only useful with post.
07:21 rhaen xxtjaxx: it should be a oneline client which req something
07:21 marcus rhaen: f() is the shortcut for post_form in ojo, iirc
07:22 arthas joined #mojo
07:23 rhaen marcus: seeing things from curl is like: curl --get --data "a=value for a" --data "b=value for b" <serveradress>
07:23 rhaen so there is no form with POST method and I have to use a GET req aka put everything in the url.
07:23 marcus rhaen: perl -Mojo -E'say g(Mojo::URL->new("http://foo.com/form")->append(a=>"bar"))->dom(…)'
07:23 marcus something like that I would say
07:24 rhaen ah!
07:25 rhaen marcus: so if I need to encode a value like (a) I would use Mojo::Util::url_escape(value), right?
07:25 xxtjaxx rhaen: Mount is (if I understand it correctly only helpfull if you want to shove the requests into a seperate Lite app no?
07:26 xxtjaxx s/correctly/correctly)/
07:26 rhaen xxtjaxx: sorry, I messed up my question with yours :)
07:26 rhaen err vice versa
07:26 marcus rhaen: perl -Mojo -E'say Mojo::URL->new("foo.com")->query({bar=>"baz"})'
07:26 xxtjaxx marcus: ^
07:27 marcus rhaen: I think Mojo::URL  should handle the escaping for you.
07:27 marcus but try it to be sure :)
07:27 judofyr I certainly hope Mojo::URL handles escaping for me
07:27 xxtjaxx marcus: The goal was to route all requests for a route base like '/blog/*' into a plugin.
07:27 marcus xxtjaxx: yes, mount is for mounting another lite app, I took that as your your 'submodule' bit. For a plugin, look at pod renderer
07:28 marcus it's in the main distro and does what you want
07:28 marcus (Catching /perldoc)
07:30 xxtjaxx marcus: okay... I'll just shill this out into a seperate lite app best to do it this way :)
07:36 cosmincx joined #mojo
07:38 rhaen marcus: huh, that's awesome. Should be something for a nice wiki article.
07:38 rhaen I will work on one.
07:50 plu joined #mojo
07:50 rhaen huh!
07:50 rhaen there is an error :)
07:53 marcus rhaen: huh?
07:53 rhaen sri! HELP!
07:53 rhaen na, there is an error in 1.87
07:54 rhaen sri included a README.pod in the MANIFEST but the file is missing in the distribution
07:54 rhaen nothing serious.
07:54 rhaen already fixed on github
07:55 plu_ joined #mojo
07:56 rhaen strange.
07:56 purl But true.
07:56 rhaen it's on CPAN - but not in the archive? I'll check.
08:04 sugar joined #mojo
08:09 perlrocks Twitter: "Mojolicious, I think you specify the method and reach does not exist error Tetara action." (ja) --hisaichi5518 http://twitter.com/hisaichi55​18/status/105914475667927040
08:11 spleenjack joined #mojo
08:17 perlrocks Twitter: "A couple of video tutorials for beginners on Mojolicious. It is a pity that had not previously found, when he mastered. http://t.co/X66v6Ys" (ru) --akamrv http://twitter.com/akamrv/​status/105916520445640705
08:18 xxtjaxx Am I getting this right then: You setup apache with mod_proxy, put it in sites-enabled and then start the perlscript by hand in order to listen on 3000 to proxy the requests toward port 80?
08:25 alpha6 joined #mojo
08:29 noganex_ xxtjaxx: i didn't feel like doing this, too.
08:32 xxtjaxx noganex: how do you deploy this otherwise?
08:32 xxtjaxx smells like a shitburger to me.
08:35 noganex xxtjaxx: it now runs as a cgi...
08:35 noganex but i'm not entirely sure if this is a smart thing to do... :/
08:42 nickl joined #mojo
08:42 xxtjaxx noganex: paste config :]
08:46 xxtjaxx noganex: please :) *withbigpuppyeyes*
08:47 noganex you could do something like AddHandler cgi-script .yourextension and Options +ExecCGI...
08:48 noganex can't access my box from here.
08:50 xxtjaxx noganex: that and app->start('cgi'); right?
09:02 xxtjaxx fairly funny...http://paste.debian.net/127109/
09:03 geira joined #mojo
09:07 koban` joined #mojo
09:10 xxtjaxx anyone care to explain what went wrong there?
09:14 sromanov what made you think you can append titles this way?
09:24 xxtjaxx it is not possible to do it in templates?
09:25 xxtjaxx sromanov: http://mojolicious.org/per​ldoc/Mojolicious/Lite#toc
09:25 xxtjaxx sromanov: http://mojolicious.org/perld​oc/Mojolicious/Lite#Layouts
09:26 xxtjaxx Or not?
09:29 sromanov xxtjaxx, ideed
09:30 xxtjaxx sromanov: :) One of the few things you can do in the template.
09:31 sromanov try put it before layout)
09:36 xxtjaxx ah found the issue version of mojo on the server is not equal local version -,-
10:15 gigigi joined #mojo
10:30 xxtjaxx Is there no better way?
10:33 xxtjaxx is there no way I can deploy this prettyly using mod_perl
10:35 judofyr you can't deploy prettyly with mod_*
10:50 marcus judofyr++
10:52 marcus xxtjaxx: http://mojolicio.us/perldoc/Mojolici​ous/Guides/Cookbook#Apache2FFastCGI
11:04 rhaen marcus: have you expericences with Mojo::UserAgent in async mode?
11:05 marcus rhaen: yes
11:05 rhaen I want to use async calls from a lite app - like someone requests a URI from my server, inside the lite app I am polling data from a different server, and return it to the customer.
11:05 rhaen that's basically the scenario.
11:06 marcus rhaen: render_later and add sub to your get request
11:06 marcus rhaen: then render the response inside the callback
11:06 rhaen ah, ok - understood.
11:07 rhaen I'll try - thanks.
11:18 rhaen next question :)
11:19 rhaen I use render_later to disable auto-rendering. k - got it.
11:21 rhaen Then I am doing the mojo::useragent req with a Mojo::IOLoop-> ...(sub { UserAgentRequest in here; $self->render( "value of UAReq"); $self->stop(); }) to stop the IOLoop.
11:21 rhaen is the way of thinking correct?
11:25 rhaen hang on.
11:25 purl it has been said that hang on. is this actually "session is still there but user has been deleted" ?
11:27 marcus self->ua->get('foo',sub {… ;$self->render('bar')});
11:29 xxtjaxx NOw it works
11:29 rhaen ok, so if I understand correctly: this fires a req. and sends the answer to the render methods when it's available.
11:29 rhaen it won't block, due to the callback sub "sub {⦠;$self->render('bar')}"
11:30 rhaen buh, takes a while to get this.
11:31 Kulag joined #mojo
11:33 rhaen marcus: thanks for the "cold" start to get into things.
11:34 marcus rhaen: yepp, sounds right
11:35 marcus async is hard, let's go shopping ;)
11:35 marcus but it's easier with UserAgent than it was with Client
11:35 rhaen true. I looked into it a few month back and didn't get it.
11:36 rhaen hm, interesting. If you run with the $self->render_later; the logfile says:
11:36 rhaen Nothing has been rendered, assuming delayed r                                            esponse.
11:36 rhaen very clever :)
11:47 marcus yupp, that's just a debug output
11:51 geira great talk on mojo at yapc, marcus
11:51 gizzlon link?
11:51 purl i guess link is awesome.
11:52 geira any chance of your slides coming online rsn?
11:52 gizzlon purl: stfu
11:52 purl stfu is DCC SEND "FURBLELOGGER" 0 0 0 or eval: join"",map uc,map$_ eq"i"?"u":$_,("fist"=~/./sg)[2,3,0,1] or "Show Them Fury Unleashed" or St. Thomas Franciscan University or shut the fuck up or subsiste sermonem statim
11:52 rhaen basically it works.
11:53 marcus geira: they are online
11:53 Akron joined #mojo
11:53 marcus geira: nordaaker.com/docs/Mojolic​ious_-_A_new_hope_YAPC.pdf
11:53 geira would be nice if the yapc site would link to them
11:53 marcus agree, it would
11:53 geira didn't they do this previously?
11:54 marcus they had a wiki page
11:54 geira thanks for the link, i'll resend to rest of dept
11:55 rhaen marcus: hu, last question :) So it looks like $self->render_later; my $data = $self->ua->get....->body; $self->render(data => $data);
11:55 rhaen right?
11:55 marcus geira++
11:55 rhaen or is the rendering being done in the callback sub,
11:55 Akron crab: "how did you solve your problem?" -> I used get and post instead of the via restrictions.
11:56 rhaen oh, forget it.
11:56 purl rhaen: I forgot it
11:56 rhaen sorry.
11:56 rhaen purl: thanks :)
11:57 Kulag joined #mojo
11:57 * rhaen slaps himself.
11:57 rhaen <--- gets a coffee (should be better then)
11:57 marcus rhaen: no, that's sync, not a sync.
11:57 rhaen marcus: yup. got it. I already got it...*ugh*
11:57 rhaen :)
12:04 perlrocks Twitter: "Having much fun plugging #plack middlewares in a #mojolicious app and seeing things *just work*. #perl" --satanpenguin http://twitter.com/satanpengu​in/status/105973642231414784
12:19 Foxcool joined #mojo
12:40 ispy_ joined #mojo
12:53 shonorio joined #mojo
13:05 al is there a way to have a certain template or part of a template evaluated last?
13:06 al i.e. to be able to process some form data and render the result in the same go
13:12 al oh, probably shouldn't have put that route at a bridge on /
13:12 al never mind
13:13 rhaen hmpf!
13:15 rhaen marcus: waaaa!
13:17 al although something like the opposite of a bridge would come in handy now
13:18 al well not quit the opposite, it'd add another dispatch cycle, too, but only be called after all other things
13:23 marcus rhaen: sup
13:23 rhaen I got it working - part of.
13:23 rhaen looks very confusing :(
13:24 marcus heh
13:24 rhaen right now I am fighting the rendering part.
13:24 rhaen render(data => <png-image>); won't work.
13:25 rhaen there is something like magic matching in the render view, right?
13:25 rhaen I should learn Mojo from scratch
13:31 marcus huh, magic?
13:31 purl Sam sat on the ground and put his head in his hands.  `I wish I had never come here, and I don't want to see no more magic,' he said, and fell silent.
13:32 rhaen When I want to call something like $self->render(<png image data>); how do I set the content type?
13:32 rhaen bah, it's way to hot. hmpf.
13:33 rhaen nm. got it.
13:38 vel joined #mojo
13:38 rhaen oh boy.
13:38 rhaen marcus: ok, back to where we started.
13:38 rhaen I request an image using Mojo::UA
13:38 rhaen the call is:
13:39 rhaen $self->ua->get('url', sub {$self->render(data => pop->res->body, format=> 'image/png')})
13:39 rhaen this works great!  I request the image from 'url' and return it.
13:40 rhaen however the browser send me application/octet instead of image/png. What do I miss here?
13:42 marcus rhaen: http://mojolicio.us/perldoc/Mojolicious/G​uides/Rendering#Content_Type_28_format_29
13:42 marcus rhaen: notice how you specify the format
13:44 rhaen *BAM*
13:44 rhaen marcus: I'll owe you a beer - or maybe 2 - 3 -4
13:44 rhaen omg!
13:44 rhaen How stupid can one be.
13:44 alpha61 joined #mojo
13:44 rhaen thanks for your time.
13:45 marcus no problem :)
13:48 sri rhaen: README.pod is in the distribution on cpan
13:48 SmokeMachine joined #mojo
13:49 rhaen sri: yup
13:50 rhaen sri: I am not sure why I've lost it in the process of RPM packaging - very strange
13:51 rhaen I'll have a closer look into it - I guess it's there but being removed by something in the %prep process.
13:51 sri what's really strange is that it actually instealls README.pod -.-
13:51 sri "perldoc README"
13:51 rhaen (which would be my fault, too)
13:51 sri is there a way to prevent that?
13:51 noganex lol
13:51 rhaen would you mind to include a Mojolicious.spec somewhere?
13:52 sri a what?
13:52 purl a is b, b=>a => what it's testing
13:52 rhaen like in 'examples'?
13:52 noganex a buildfile for rpm based systems?
13:52 sri if it's for packaging it doesn't go into core
13:52 d4rkie joined #mojo
13:52 rhaen noganex: yep.
13:53 rhaen sri: k. Maybe a place inside the wiki?
13:53 sri sure
13:53 rhaen sri: k
13:53 marcus it will rub the lotion on it's skin, or it will get the hose again.
13:53 noganex rhaen: iirc there are countless tools like cpan2rpm
13:53 noganex or the stuff from cpanplus...
13:54 rhaen noganex: true, however, they don't have proper dependcies in redhat style
13:55 rhaen we need to run prepackaged software here as our config management relies on rpm packages
13:55 rhaen e.g. with proper chkconfig scripts to test the services
13:55 rhaen hmpf. meeting. brb
13:57 noganex lucky me. i'm allowed to use cfengine on my linux and openbsd boxes. :)
13:58 mspo sounds good to me
13:58 sri http://groups.google.com/group/mojolicio​us/browse_thread/thread/7617c01ea2fb36ec # :S
14:07 GitHub163 joined #mojo
14:07 GitHub163 [mojo] kraih pushed 1 new commit to master: http://git.io/g6BVdw
14:07 GitHub163 [mojo/master] fixed small bug in Mojo::JSON - Sebastian Riedel
14:07 GitHub163 left #mojo
14:13 koban` left #mojo
14:16 noganex 34 DEG C... i'll stop coding and get some mochi ice cream... %D
14:17 cosmincx joined #mojo
14:17 sugar joined #mojo
14:27 alnewkirk morning all
14:27 purl evening, alnewkirk
14:27 alnewkirk ummm
14:33 elb0w http://www.addedbytes.com/​blog/if-php-were-british/#.TlO2J6IhRa5.twitter
14:33 ispy_ joined #mojo
14:35 mspo don't use php 5.3.7, by the way
14:35 mspo https://bugs.php.net/bug.php?id=55439
14:36 elb0w Yeah
14:36 elb0w er
14:36 mspo or at all if you can avoid it ;)
14:36 elb0w It was only md5 right?
14:37 elb0w my girl lived there for 16 years
14:37 elb0w whoops
14:45 omega sri: do you never accept patched without tests?
14:45 omega sri: because I have a crazy situation where the call to File::Find::find in Mojo::Home tries to chdir to /root
14:46 omega but only if started via init scripts etc
14:46 omega sri: https://gist.github.com/2d2ddb6546830b89cab9 this patch fixes it, all other tests pass
14:47 omega but I honestly have no idea how to convert our situation to a test case :/
14:51 abra joined #mojo
14:54 mspo omega: I was using File::Find in a mojo helper the other day and didn't experience anything like that
14:58 marcus omega: the home stuff is effing tricky
14:58 marcus omega: it needs tests
14:59 * sri nods
15:01 sri we've had a lot of trouble getting that code to run portably, many test fails only showing up on cpantesters
15:04 sri besides typo fixes there is not much i would accept without tests these days
15:07 noganex mh... File::Find would try to chdir into a directory to which it has no rights? o_O
15:10 kaare joined #mojo
15:15 alnewkirk I can see the need for a simple way to define a catch all render_not_found, etc (other exceptions)
15:17 clintongormley joined #mojo
15:18 clintongormley hiya - any users of WWW::Curl here?
15:18 clintongormley i'm wanting to know if anybody has tied WWW::Curl into AnyEvent
15:19 jamesw yes
15:20 jamesw well libcurl
15:20 clintongormley jamesw: yes to what? being a user, or tieing it to anyevent?
15:20 * sri wrote WWW::Curl::Multi back in the days
15:20 clintongormley hiya sri - i saw that
15:20 jamesw tieing it, libcurl, not WWW::Curl
15:21 jamesw AnyEvent::Curl::Multi
15:21 clintongormley i'm in the process of adding a www::curl backend to elasticsearch - it's about 60% faster than HTTP::Tiny
15:21 clintongormley jamesw: thanks - i'll take a look
15:23 sri just don't forget that it will prolly get *a lot* slower inside anyevent
15:23 clintongormley jamesw: looks like AnyEvent::Curl::Multi uses WWW::Curl under the covers
15:23 sri depending on how much back and forth there will be between perl and xs code
15:23 clintongormley sri: sure - understood. but using this an AE is useful for a different purpose
15:24 clintongormley the things we discover when we look :)
15:24 clintongormley YAPC::EU++
15:25 sri yea, it's all about latency
15:25 sri not sure why you care about parser performance for crawling
15:25 clintongormley parser performance? crawling?
15:25 clintongormley no, this is for talking to an ElasticSearch server over HTTP
15:25 sri ah
15:26 sri ok, i guess then a fast blocking xs client makes sense
15:26 clintongormley also metacpan frontend will call the backend with several independent requests, so doing those in parallel is a win
15:27 sri doubt parser performance will make much of a difference there
15:28 sri since queries usually mean latency
15:28 clintongormley sure
15:28 clintongormley sri: one other benefit that it brings is the persistent connections
15:28 clintongormley LWP has them, but is dog slow
15:30 sri compared to what?
15:30 sri blocking libcurl?
15:31 clintongormley compared to HTTP::Lite or HTTP::Tiny (which don't have persistent connections) and WWW::Curl is about 60% faster than HTTP::Tiny (in my benchmark when indexing data into elasticsearch)
15:31 clintongormley and yes, that is blocking
15:32 clintongormley i tried out AE::HTTP ages ago, and got similar performance to LWP at the time
15:34 sri blocking should always be faster for single requests
15:35 sri not sure why LWP would be that much slower, did you have html parsing activated?
15:36 rhaen 16:07 < GitHub163> [mojo] kraih pushed 1 new commit to master: http://git.io/g6BVdw
15:36 sri it parses the <head> section by default and whatnot...
15:36 clintongormley sri: no - i used NYTProf on it, and a huge amount of time was spent parsing headers, of which there are very few indeed :)
15:37 clintongormley with ElasticSearch, i'm sending and received JSON in the body, so no HTML
15:37 rhaen noganex: hm, well - I've implemented a cfengine system here which was in production for 7 years
15:37 rhaen huh.
15:37 rhaen noganex: now we are running puppet
15:37 sri i see, makes sense, header parsing has been the bottleneck in the Mojolicious HTTP parser too when i last profiled
15:39 sri but still, once you have some latency i doubt parser performance will make much of a difference
15:39 sri you could prolly just as well use the LWP parser for non-blocking requests
15:41 clintongormley yeah - i think it might even make sense to use different backends for different use cases
15:42 sri indeed
15:43 sri omega: for the no_chdir case i think i would also accept a very good explanation
15:43 noganex rhaen: mh.. puppet is nice, too.
15:43 sri and some proof that it should work cross platform
15:43 noganex although i've never had the chance to test it on larger systems
15:44 sri omega: googling didn't turn up much information, so convincing us will be up to you
15:47 rhaen noganex: it is nice, but it requires a strong data model behind.
15:47 clintongormley left #mojo
15:47 rhaen noganex: like, this is our base server. class webserver inherits base { ...install apache...}
15:47 rhaen noganex: class cool_webserver inherits webserver { ....}
15:48 rhaen noganex: so you have do it in an oop style which has to be consistent for all your servers.
15:48 rhaen sri: ?
15:48 rhaen sri: marcus brought me to async things with Mojo::UA. That's quite cool.
15:48 rhaen sri: thanks for this framework.
15:49 noganex rhaen: i've used to write java and even worked a bit on hibernate. i guess a bit of that ruby stuff won't hurt me. ;)
15:50 rhaen noganex: nah, it's not the stuff itself - make a concept and structure your servers. That's all.
15:50 rhaen noganex: Ah! And refactor things frequently.
15:50 rhaen noganex: that's a must with puppet.
15:51 sri rhaen: yw :)
15:53 noganex rhaen: what i've tried with puppet wasn't too bad, but somehow i felt happier with cfengine and git.
15:54 rhaen sri, we have the abbrevs for ojo like g,p,f...etc. Can we have them in Mojo::UserAgent as well?
15:55 rhaen sri: could be overengineered.
15:55 rhaen noganex: true. I loved cfengine in vers.2 We looked into vers. 3 and I don't like it anymore
15:55 rhaen noganex: what I really hated about vers2 were the changes of Mark which broke other parts for features.
15:56 rhaen noganex: we had one version and we pinned the repos to that version. This was tested and worked for us.
15:56 sri rhaen: might not be such a good idea
15:56 rhaen sri: yep, you are right probably not.
15:57 crab those abbreviations are good for one-liners, but god, i'd hate to try to understand "real" code written that way.
15:57 * sri nods
15:58 noganex rhaen: well.. i'd rather have a bad config management than no management at all. ;)
15:58 * sri wonders if Mojo::ByteStream->split should generate collections of Mojo::ByteStream objects
15:58 * sri pokes Akron
15:58 noganex when i joined six years ago, there were about 50 systems... ;)
15:59 rhaen crab: it took me a while to understand the following: $self->ua-get('http://foo', sub { $self->render( data => pop->res->body, format => 'png')});
15:59 rhaen where did the pop came from.
15:59 rhaen yep, I understand it, but it took a while
16:00 Akron sri: Here!
16:01 sri rhaen: did you see the examples at http://mojolicio.us/perldoc​/Mojolicious/Controller#ua ?
16:01 sri Akron: collections of bytestream objects?
16:01 sri or just collections of plain old scalars
16:01 rhaen sri: I did. You mean the first non-blocking example, right?
16:02 sri ye
16:02 rhaen looks nicer :)
16:02 rhaen but does the same :)
16:02 Akron sri: They are overloaded by default, right? So most actions of Mojo::Collection would still work ... hm ,,, pretty nice idea.
16:06 Akron sri: So something like that would work: c("Hello Mojo fellows!")->split(" ")->each(sub {$_->quote})->join(" ")->say;
16:06 Akron Damn s/c/b/o;
16:07 sri yes it would
16:07 Akron Vote!
16:07 sri :)
16:08 Akron Maybe not the best balance between funny and useful, but hey: More fun is always useful!
16:08 SmokeMachine joined #mojo
16:13 al is it not ok to use flash across a redirect?
16:13 al cause my flash() is empty in the first bridge route right after a 302
16:14 GitHub41 joined #mojo
16:14 GitHub41 [mojo] kraih pushed 1 new commit to master: http://git.io/XJ0IXw
16:14 GitHub41 [mojo/master] made split method in Mojo::ByteStream a little smarter - Sebastian Riedel
16:14 GitHub41 left #mojo
16:15 sri that's fun :)
16:21 sri http://37signals.com/svn/posts/2991-beh​ind-the-scenes-ab-testing-part-3-final # interesting
16:22 alnewkirk anyone feel like hacking on the new command or scaffolding https://github.com/alnewkirk​/Mojolicious-RestApp-Example
16:35 xaka joined #mojo
16:35 GitHub109 joined #mojo
16:35 GitHub109 [mojo] kraih pushed 1 new commit to master: http://git.io/LrODKg
16:35 GitHub109 [mojo/master] simplified examples - Sebastian Riedel
16:35 GitHub109 left #mojo
16:36 sri lets try some a/b testing :)
16:37 baton8 joined #mojo
16:37 sri wanna see if the bounce rate changes with more simple examples on the frontpage and readme
16:37 Eugene joined #mojo
16:41 vel joined #mojo
16:49 perlrocks Twitter: "Simplified the #mojolicious frontpage a bit, lets see if it changes the bounce rate. :) http://t.co/Fv7oqWX #perl" --kraih http://twitter.com/kraih/status/106045315903270912
16:49 sri might have been a bit overwhelming before
16:53 GitHub29 joined #mojo
16:53 GitHub29 [mojo] kraih pushed 1 new commit to master: http://git.io/rGsTVA
16:53 GitHub29 [mojo/master] documentation tweaks - Sebastian Riedel
16:53 GitHub29 left #mojo
16:53 judofyr joined #mojo
16:58 GitHub37 joined #mojo
16:58 GitHub37 [mojo] kraih pushed 1 new commit to master: http://git.io/w2ESyA
16:58 GitHub37 [mojo/master] removed unnecessary title - Sebastian Riedel
16:58 GitHub37 left #mojo
17:00 ispy_ joined #mojo
17:05 diegok sri: I think bounce rate is higher than usual (if it is like it) because menu links are mostly to external sites, so following one of those count as bouncing...
17:06 sri bounce rate is not very high, i'm just wondering if a more simplistic page would get it even lower
17:07 diegok ok :)
17:18 alpha6 joined #mojo
17:29 crab what is bounce rate?
17:29 purl bounce rate is higher than usual (if it is like it) because menu links are mostly to external sites, so following one of those count as bouncing...
17:31 crab i hate the fact that i have to scroll down to see all the good stuff on mojolicio.us
17:32 sri http://www.blogcdn.com/www.urlesqu​e.com/media/2010/05/hatereagle.jpg
17:35 crab hey, i love giant content-free images that take up half the page!
17:36 crab heh. catalystframework.org has a quote on its frontpage that says "Catalyst brings some serious web development mojo to Perl."
17:37 sri haha
17:37 sri i added those quotes
17:38 crab maybe nobody said anything about catalyst that bears repeating, since then
17:38 sri the whole page has not changed since then
17:38 sri they just added book links
17:44 tempire I like the new frontpage
17:44 tempire Also, I wonder if IBM realizes that amazon is taking over the world.
17:44 tempire They release a game-changing product nearly every month
17:52 sri aws is down so much these days i don't notice new amazon products either -.-
17:55 zakame sup
18:10 GitHub47 joined #mojo
18:10 GitHub47 [mojo] kraih pushed 1 new commit to master: http://git.io/781lhg
18:10 GitHub47 [mojo/master] more verbose examples - Sebastian Riedel
18:10 GitHub47 left #mojo
18:10 sri rhaen: that should help in the future
18:11 Alias joined #mojo
18:11 Foxcool joined #mojo
18:24 vel joined #mojo
18:28 vel joined #mojo
18:29 Firkin left #mojo
18:31 stephen joined #mojo
18:48 perlrocks Twitter: "This should be really fun for oneliners. http://t.co/UjrGt7s #mojolicious #perl" --kraih http://twitter.com/kraih/status/106075269848236032
18:56 sromanov joined #mojo
19:03 tempire turns out mojo::collection was a great idea after all
19:08 abra_ joined #mojo
19:09 sri glad we kept it
19:21 perlrocks Twitter: "Mojolicious 1.88 by SRI - http://t.co/MYZp89Q" (ms) --cpan_new http://twitter.com/cpan_new​/status/106083700005015552
19:21 elb0w Really
19:21 elb0w another update
19:21 elb0w holy crap
19:21 perlrocks Twitter: "This should be really fun for oneliners. http://t.co/bf5yCMm #mojolicious #perl http://t.co/IOAzemv" --PatOnline1 http://twitter.com/PatOnline​1/status/106083756019957761
19:21 elb0w im gonna make a cronjob to update mojo
19:22 elb0w should I do it hourly or daily?
19:22 rhaen puuh. Async stuff is somewhat freaky to understand. Once you've got it - it's fantastic. And it works, too.
19:22 elb0w Read that twisted paper
19:22 elb0w was pretty good, see that one rhaen ?
19:23 rhaen elb0w: what's cool?
19:23 elb0w cool?
19:23 elb0w http://krondo.com/wp-content/upl​oads/2009/08/twisted-intro.html
19:24 rhaen I am looking into it.
19:24 sri most python people i've talked to actually hate twisted
19:24 elb0w http://krondo.com/wp-content/up​loads/2009/08/twisted-intro.pdf if you want as ebook
19:24 elb0w I just read for concepts
19:24 elb0w not to use it
19:24 rhaen that's python.!
19:24 rhaen ah, ok.
19:24 sri twisted is a good example
19:25 elb0w have had a hard time finding good docs on async programming
19:25 sri solid documentation, even an oreilly book
19:25 rhaen sri: thx again.
19:26 rhaen we should spread the word about this perl thing mojo.
19:26 elb0w so 00 16 * * * cpanm Mojolicious
19:26 sri http://www.gevent.org/ # python people these days seemt o be more into this stuff
19:26 elb0w I had a really good python dev here bashing async
19:27 sri i actually don't like async either, but i accept that it's currently the best tool for the job
19:28 tempire what's wrong with async?
19:28 sri ugly code
19:28 purl i think ugly code is push (@gAdjustmentRules, [${$ref}[$i]->[0], ${$ref}[$i]->[1], ${$ref}[$i]->[2], ${$ref}[$i]->[3]]);
19:28 * sri pokes his eyes out
19:28 tempire ah. the concept, though, is a natural development
19:28 sri i really hope at some point threads get cheap and sane
19:29 elb0w werent they reworking threads?
19:29 tempire although event driven is the end game, I think.
19:29 sri async is not the right word though, we mean event loops i guess
19:29 elb0w reactor pattern no?
19:29 tempire all programs want to be event driven when they grow up.
19:29 sri right, reactor pattern makes code rather unpleasant
19:30 elb0w reactor is the event loop no?
19:30 sri it is
19:31 elb0w wish I had dev friends I new irl
19:31 sri i would love an implementation of akka for perl
19:31 elb0w Probably learn a lot easier being able to discuss with people
19:31 sri but we need cheaper threads for that
19:31 tempire elb0w: start a meetup
19:31 elb0w there are perlmongers in nyc
19:31 elb0w but I always feel awkward going there
19:32 tempire awkwardness goes away with familiarity
19:32 elb0w yeah
19:32 tempire same with irc
19:32 sri go to pm meetings if you can
19:32 tempire everyone is a n00b at some point.
19:32 * tempire remembers his irc early days
19:32 elb0w Been using perl close to 3 years now
19:32 elb0w feel like I should be ahead more
19:32 tempire I've been using perl for 10 years
19:32 elb0w I just now got into creating oo modules
19:32 tempire sri?
19:32 purl sri is A PONIE!!1! and a  or no more a javascript noob or evil or a spelling guerilla or the temporary channelclown or a snob now or the perl guerilla or a small mouse with a big head aiming for world domination or the evil sith or has no clue about fuzzy logic or no longer a real hacker :/ or a copycat or pimp to many children or a holy man or looks a lot like pitr or puny or doing ruby now until perl6 or unholy
19:33 tempire 10,11,12?
19:33 sri almost 10 years too
19:33 tempire I should have taken over the world by now
19:33 tempire at least twice
19:33 elb0w did you guys go to school for programming?
19:33 tempire school is for people with patience.
19:33 elb0w I didnt do any programming until 3 years ago
19:33 elb0w at most I would do basic html
19:33 tempire I would have been farther along had I been in the community.
19:34 elb0w ive done perl,python,c++,java,lua since
19:34 tempire sri is a super villain already because he was active from day one
19:34 elb0w I wish someone would of influenced me into it sooner
19:34 * sri is an autodidact, no formal programming education
19:34 tempire so, lesson learned.  GO TO PM meetings.
19:34 elb0w yeah
19:34 elb0w I think I will
19:34 elb0w haha just feel like ill be outcast for not knowing everything
19:34 * elb0w hides
19:35 tempire you will probably be ridiculed for something at some point.
19:35 tempire just accept it.  and keep going.
19:35 elb0w Well im 6'2 250lbs maybe theyll be nice
19:35 elb0w :P
19:35 tempire boosh
19:35 elb0w plus I got a bad ass raptor shirt
19:36 sri that should surely intimidate them
19:36 tempire if you run into the buzzfeed people, tell them to start using an orm already.
19:36 tempire and a real framework.
19:36 elb0w lol
19:36 elb0w yeah I need to start using DBIx::Class more
19:37 tempire wait a month you'll have something newer and fresher
19:37 elb0w that datamodel?
19:38 tempire seems like akka is just a pub-sub model for events
19:38 sri actor pattern
19:38 elb0w I need to read GoF again
19:38 sri it's cheap threads with dead simple message passing
19:39 sri elb0w: for reading the head first book might be better
19:39 sri GoF is the bible, but it's not so great at explaining stuff
19:40 sri https://metacpan.org/release/Mojolicious/ # btw. don't forget to +1 again!
19:40 sri they lost all +1s a week or so ago
19:40 tempire already done
19:41 tempire are threads really that much cheaper than linux forked processes?
19:42 elb0w Would you think this event loop I wrote is missing anything? http://pastie.org/2418403
19:43 GitHub142 joined #mojo
19:43 GitHub142 [mojo] kraih pushed 1 new commit to master: http://git.io/-vMwKw
19:43 GitHub142 [mojo/master] use slightly more verbose example - Sebastian Riedel
19:43 GitHub142 left #mojo
19:43 sri tempire: *a lot*
19:44 sri take Coro for example, a thread there is like 4kb
19:48 elb0w LOL
19:49 elb0w why have I ever bought a book
19:49 elb0w http://tinyurl.com/3ruks7g
19:50 sri because you want to support the authors and make sure they write more in the future
19:51 * tempire concurs
19:51 rhaen +1
19:51 purl 1
19:52 elb0w Nah, if it comes on kindle I buy it
19:52 elb0w my girl has barred me from buying anymore programming books
19:54 * sri wonders when the next kindle comes out
20:00 elb0w nah I do it in everything
20:00 elb0w whoops
20:13 Akron joined #mojo
20:17 alnewkirk i think we need to expand on the file not found subject (as well as other exceptions)
20:18 alnewkirk its seems that in the Mojolicious source, html is assumed for all exception cases
20:18 alnewkirk .. so you cant have a not_found.html.ep and a not_found.json.ep, etc
20:19 tempire why would you need a not_found.json?
20:19 exc joined #mojo
20:20 perlrocks Twitter: "Mojolicious-Plugin-ISAjax 1.00 by EXC - http://t.co/SRzkcZ0" --cpan_new http://twitter.com/cpan_new​/status/106098445571469313
20:21 alnewkirk tempire, when responding to json requests
20:21 tempire exc: http://mojolicio.us/perldoc/​Mojo/Message/Request#is_xhr
20:22 exc heh, i must be blind this day :), thanks
20:22 alnewkirk why respond to a json request with an html response, amoungst other things
20:23 sri if an endpoint does not exist, it of course doesn't have a representation
20:24 alnewkirk technically the not_found template becomes its representation, else, to be accurrate we should send back a blank screen using the mime type that was requested
20:24 rhaen Hep - the n00b with another question is here...
20:25 rhaen it's about Mojo::Log
20:25 sri https://github.com/kraih/mojo/blob/master/t/m​ojolicious/lib/MojoliciousTest/Exceptional.pm # this however migth interest you
20:25 rhaen https://gist.github.com/1166405
20:25 sri alnewkirk: no it does not!!!
20:25 sri a REST endpoint exists or it doesn't
20:25 sri 404 means it does not exist
20:26 sri it can't have a representation by definition
20:26 alnewkirk if i remember correctly, overriding the functions (render_exception, etc) would have to be placed in all controller classes, correct
20:26 sri also wrong
20:26 stephan48 make an baseclass
20:26 sri what stephan48 said
20:27 Akron If something does not exist, it should at least have the format I expected ... ;)
20:28 alnewkirk i think overridding the standard exceptions should be easier and better documented IMHO
20:30 rhaen ok, well - documentation is always an issue. Feel free to join in and send pullup req. :)
20:31 alnewkirk sri, stephan48, what would creating a custom base class do?
20:31 * sri is busy now
20:31 rhaen I can call $self->ua->get(..) inside a lite app. However calling $self->log->debug(...) won't work. The short question is: why?
20:32 exc rhaen: beucase is has not been set to log (to a file)
20:33 exc on the other hand, app is preconfigured already to log somewhere
20:33 alnewkirk rhaen, because the app logs and the controller doesnt
20:33 exc $self->app->log->debug('Yep - I am here');
20:33 rhaen app->log->debug(..) is short then and works
20:33 alnewkirk as i stated before [16:26] <alnewkirk> if i remember correctly, overriding the functions (render_exception, etc) would have to be placed in all controller classes, correct
20:34 alnewkirk so i can only assume that your recommendation to create a custom base class is your work-around for injecting those functions in each controller class
20:35 rhaen exc: (which is written inside the docs app->log...)
20:36 rhaen exc: but you are right. Given the state that I need to configure the app, $self->app->... is the correct way.
20:36 rhaen exc: so true. Should be in that way.
20:37 exc I mostly do in Mojolciious (not Mojolicious::Lite), so if I log in a controller, I use $self->app->log
20:42 rhaen exc: what confused me is that app->ua->get(..) is fine, $self->ua->get(..) is fine, however $self->log-> won't work.
20:43 rhaen and $self->app->ua->get.. works, too.
20:43 rhaen but it's stated inside the documentation which is the right way so it's fine.
20:44 * alnewkirk off to watch some californication for a bit
20:45 rhaen So, accessing the Mojolicious instance to fire a get request would be not that useful, tho.
20:46 sri alnewkirk: calling it a work-around implies that you have a "proper implementation", so shoot
20:52 alnewkirk sri, I don't, I'm using Mojo alot nowadays and I am simply offering suggestions to make life easier
20:52 sri no you are not
20:53 * tempire puts on his sheriff hat
20:53 sri you're not understanding something and assuming it must be a framework problem
20:53 omega sri: the thing that happens, is that we have a script that drops privs from root to non-root user. However Cwd::getcwd for some reason still returns /root, so $cwd in File::Find ends up as /root
20:53 omega for some reason, if no_chdir is false, it tries to chdir to $cwd trying to find files
20:53 omega which failes with ENOACCESS
20:54 omega but since the usage in Mojo::Home deals with absolute paths anyways, I don't see why it can't do no_chdir
20:54 alnewkirk what am I not understanding?
20:54 tempire sri, alnewkirk : this is all due to a difference in the way you both communicate.
20:54 tempire just let it go.
20:54 alnewkirk okay
20:54 sri alnewkirk: if your intention was to make our lives easier you would read up on how it works now, and present a solid solution afterwards
20:55 tempire alnewkirk: you've rubbed some nerves.  your best bet is to contribute some code and go from there.
20:55 alnewkirk will do, thanks
20:55 tempire sri will appreciate the gesture and you'll have a better foothold.
20:56 sri if there's something that annoys the hell out of me it's hand waving :)
20:57 sri tempire++ # btw
20:57 omega mspo: no, it is a very specific situation :p
20:57 purl okay, omega.
20:57 mattp joined #mojo
20:58 tempire alnewkirk: a pull request and a use case as a test would put you beyond the rest of the suggestions, and will sooth the nerves.
20:58 tempire you might still get shot down, but it will be to merit.
20:58 tempire *due to merit
20:58 tempire and really, sri tends to think about it for several hours and come up with a compromise.
20:58 tempire so sometimes you get what you want if you just let it stew for a bit.
20:59 sri tempire: in this case there is not even a visible problem though
20:59 tempire general advice.  the specifics are irrelevant if the general advice is followed
21:00 tempire 'cause it works itself out in the correspondence.
21:01 sri SRI: The Missing Manual
21:01 * tempire sprinkles glitter to restore the rainbow
21:02 alpha6 joined #mojo
21:04 stephen joined #mojo
21:04 omega sri: how is that for explanation?
21:05 arpadszasz joined #mojo
21:05 sri omega: hmm, little worried
21:05 omega about what part?
21:06 sri if it affects portability
21:06 sri only way to find out is to push a release and wait :S
21:07 sri omega: if you find me a windows tester, we might be able to try it
21:07 omega to make matters worse, I don't think list_files from Mojo::Home is ever tested :p
21:08 sri not directly, but indirectly all over the test suite
21:08 omega sri: what do you want tested on windows?
21:08 omega that the test-suite runs?
21:08 sri yes
21:09 omega Alias: didn't you have some windows test machines or blog about that some time ago?
21:10 GitHub92 joined #mojo
21:10 GitHub92 [mojo] kraih pushed 1 new commit to master: http://git.io/vkZ4NQ
21:10 GitHub92 [mojo/master] improved Mojo::Home portability - Sebastian Riedel
21:10 GitHub92 left #mojo
21:12 sri hopefully that's true
21:12 omega hehe
21:16 perlrocks Twitter: "Mojolicious 1.89 by SRI - http://t.co/sWLcTmS" (ms) --cpan_new http://twitter.com/cpan_new​/status/106112667147706368
21:16 sri only way to really find out is cpantesters
21:18 exc sri: what brings you back to Perl from RoR? just asking :)
21:18 y0j_ joined #mojo
21:18 sri nothing specific, i just like perl
21:20 exc well, Ruby seems to have better OO, at least better that Perl, but I am not a Ruby dev anyway
21:21 sri yes, but it sucks at other stuff
21:21 sri they all have advantages and disadvantages
21:21 omega sri: thanks for applying and testing it!
21:22 sri cpantesters++
21:22 sri omega: if it doesn't work out you better run though ;)
21:23 omega well, I have only tested os x and linux :)
21:23 omega but you didn't give me much time to find a windows
21:24 sri testing on one windows box doesn't mean much anyway… sadly
21:24 sri they are all broken differently
21:25 omega ahh
21:25 exc sri: sorry for being annoying, i promise this is my last question about this ruby topic, but what was the "most" problematic part of Ruby (comparing to Perl)?
21:26 sri exc: don't worry, you're not annoying
21:26 exc ;)
21:26 sri hmm… i might have to think about that for a bit
21:26 sri (wrong mindset atm)
21:26 omega oh wel, bed time
21:26 omega good night
21:29 sri nn
21:30 sri exc: lack of cpan
21:30 sri yea
21:31 sri and that includes everything around the cpan
21:31 sri cpantesters, metacpan….and and and
21:32 sri while some gems are really exceptionally good the majority is just plain awful
21:32 sri especially documentation is a huge problem
21:33 sri there is just no documentation culture
21:33 exc :) i see, well sometimes i have feelings that cpan packages and their doc are in similar situation, but i have never rubygems for real
21:34 sri try it
21:34 sri the simple fact that you can find so much documentation just by searching metacpan is *HUGE*
21:35 sri you are basically forced to have documentation, or you just don't exist on those platforms :)
21:37 exc that's true. but what sometimes drives ma angry about cpan is that sometimes a package method returns an object and i need to search the code to find out what it is.
21:37 sri same goes for TDD, which is supposedly so big in the ruby world, the reality is different
21:38 sri second on my list would prolly be oneliners
21:39 sri i just love perl oneliners
21:40 exc :) yes, that is really addictive
21:40 exc instead of bask/sed/awk just simple perl $F[1] - for example ;)
21:41 memowe joined #mojo
21:41 Alias_ joined #mojo
21:47 GitHub67 joined #mojo
21:47 GitHub67 [mojo] kraih pushed 1 new commit to master: http://git.io/3UStSg
21:47 GitHub67 [mojo/master] documentation tweaks - Sebastian Riedel
21:47 GitHub67 left #mojo
22:10 jwang joined #mojo
22:12 sri alnewkirk: the pull request is completely broken, but that is something we can work with
22:14 alnewkirk yeah im still trying to wrap my head around git, ... when i started it said I was 367 commit behind your latest master, i thought i cleaned everything up but ...
22:15 sri always work on feature branches
22:15 sri you can only drop and refork your master rbanch
22:15 alnewkirk oh right, sorry about that, that makes sense
22:48 GitHub93 joined #mojo
22:48 GitHub93 [mojo] kraih pushed 1 new commit to master: http://git.io/kOOGUA
22:48 GitHub93 [mojo/master] improved render_exception and render_not_found to use the current format if available - Sebastian Riedel
22:48 GitHub93 left #mojo
22:48 sri alnewkirk: there is a lot more to it, but that might work
22:51 GitHub168 joined #mojo
22:51 GitHub168 [mojo] kraih pushed 1 new commit to master: http://git.io/E1fQDQ
22:51 GitHub168 [mojo/master] fixed typo - Sebastian Riedel
22:51 GitHub168 left #mojo
22:54 sri the chain is now exception.development.xml.*, exception.xml.*, exception.development.html.*, exception.html.*, built in html template
22:55 alnewkirk sri, still looking at the code
22:56 alnewkirk sri, https://github.com/kraih/mojo/commit/a7a24​83ca3884936215b4e76ce62859046a7fdf9#diff-0
22:56 alnewkirk ^^ should the content type be html, or application/json or ...
22:57 alnewkirk oh
22:57 alnewkirk sorry about that, didn't look at the whole file
23:08 tempire I need a design inspiration for mojocasts.com
23:09 sri http://dribbble.com # browse this
23:09 * tempire is accepting ideas and/or hot girls for muses
23:11 tempire The problem with receiving cookies in the mail is that I eat them all
23:12 sri i hope the special cookie she promised was in the box? :)
23:13 tempire I don't think so…unless I ate it without noticing its specialness
23:14 sri :o
23:15 tempire hmm
23:15 tempire I'd like to know how to make this sort of icon: http://dribbble.com/shots/244382-Cut-th​e-cloud-icon?list=popular&amp;offset=19
23:17 sri it actually doesn't look that complicated
23:19 tempire I don't know how to get the shine on he 3rd dimension
23:21 sri gradients and transparency ;p
23:26 * sri just remembered he had cookie flavoured ice cream in the fridge!
23:52 tempire sketch's gradient is bobo
23:53 tempire *gradient tool

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