The web in a box - a next generation web framework for the Perl programming language

IRC log for #mojo, 2014-07-04

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

All times shown according to UTC.

Time Nick Message
00:12 mattastrophe joined #mojo
00:28 howitdo joined #mojo
00:29 mattastrophe joined #mojo
00:53 shmuel joined #mojo
01:15 shmuel left #mojo
01:39 mire joined #mojo
01:56 klapperl_ joined #mojo
02:03 sri hmm, i guess the Role::Tiny way of applying changes to objects is also a pretty efficient way to do singleton methods
02:04 sri with this patch http://pastie.org/9352412 you can do this http://pastie.org/9352413
02:04 sri fun little hack
02:12 crab that looks interesting
02:13 sri as long as you refine your object in the same way over and over you get the same result and no leak, without modifying the original class
02:17 crab i like it
02:18 crab not sure i understand where i'd use it, but i like it
02:18 sri http://pastie.org/9352438 # more relaistic patch
02:20 sri you use it every time you feel like adding a singleton method to some object and monkey patching feels dirty
02:22 sri say you want a sugar method on $c->tx->req->url or some other layer you have little control over
02:23 sri you $c->tx->req->url->refine(some_sugar => sub { ... }) in a hook and get it without any fuzz
02:24 sri no subclassing, no dependency injection...
02:25 sri and you're not monkey patching the original class, Mojo::URL stays clean
02:25 sri you just rebless to the same class over and over
02:38 crab $blah->refine(some_sugar => ...)->some_sugar, whee
02:56 mattastrophe joined #mojo
02:59 KCL joined #mojo
03:08 gatitskiy joined #mojo
03:11 basic6 joined #mojo
03:18 mattastrophe joined #mojo
03:35 sri wow, the author of express.js (and pretty much all the cool node modules) is leaving node for go :o https://medium.com/code-adventures/4ba9e7f3e52b
03:45 noidd always the new shiny
04:37 KindOne joined #mojo
05:13 chankey joined #mojo
05:14 chankey exit
05:15 chankey left #mojo
05:16 cfedde joined #mojo
05:17 xtetsuji joined #mojo
05:30 chankey joined #mojo
05:31 chankey joined #mojo
05:53 mattastrophe joined #mojo
05:59 sujithm joined #mojo
06:00 gatitskiy joined #mojo
06:11 Ptolemarch joined #mojo
06:15 mattastrophe joined #mojo
06:36 mattastrophe joined #mojo
06:45 dod joined #mojo
06:53 batman sri: i would like if you kept the original classname. that Mojo::Base::Refinement:: namespace is not very descriptive. i would rather do something like this: https://gist.github.com/jhthorsen/aa3622ff774d501bbd12
06:55 condor66 left #mojo
07:01 dod joined #mojo
07:02 dpetrov_ joined #mojo
07:03 batman this looks like prototyping ;)
07:13 Insane batman, could you advice? is there anything like sessionId in $c->session?
07:14 trone joined #mojo
07:15 igorsutton joined #mojo
07:17 Insane in case if I have one user logged in from 2+ devices, I want to know which action he did from which device
07:17 jnbek joined #mojo
07:20 gatitskiy joined #mojo
07:21 batman just add one. $c->session(id => Mojo::Util::md5_sum(join '', Mojo::Util::steady_time, $uid, $c->req->headers->user_agent // '')); # something like that.
07:21 batman $c->session->{id} ||= # do this instead so you don't create a new id on each request :P
07:22 batman throw in $$ as well, if you're afraid of conflicts
07:24 Vandal joined #mojo
07:28 KCL_ joined #mojo
07:33 KCL joined #mojo
07:34 condor66 joined #mojo
07:37 punter joined #mojo
07:39 condor66 joined #mojo
07:41 mr-foobar joined #mojo
08:16 john joined #mojo
08:30 batman Insane: ^
08:42 denis_boyun joined #mojo
08:44 batman sri: cool! didn't know perl -le'sub foo { return foo => sub {1} }; print foo for 1..10' resultet in the same code ref...
08:45 batman perl -le'sub r{@a=@_;sub{@a}};print $_=r($_),$_->() for 1..10' # maybe a better illustration
08:46 batman oh crap. no, it had a bug: perl -le'sub r{my @a=@_;sub{@a}};print $_=r($_),$_->() for 1..5'
08:48 batman sri: i'm not sure if i understand "leak" and "same way", but i think leaking classes will be rather simple because of that last one-liner :/
08:48 batman maybe i'm thinking too complicated, or maybe it's even correct way of leaking
08:49 batman after all, the refined class does have private variables (!)
08:49 batman this might be super cool actually. can i make a standalone module from the code above?
08:51 batman also, my $new = ... code is a mess. but i still think the super class should be part of the $new classname
08:52 tempire so it's adding a method to the object, not the class
08:55 batman tempire: everything is a class
08:55 batman or package.
08:55 purl package is perldoc -f package or lexicaly scoped
09:00 mire joined #mojo
09:05 edestler joined #mojo
09:27 mattastrophe joined #mojo
09:30 root joined #mojo
09:33 callumgare joined #mojo
09:39 callumgare joined #mojo
09:51 basiliscos joined #mojo
10:09 mire joined #mojo
10:10 rem_lex| joined #mojo
10:16 icjs joined #mojo
10:17 icjs hi, i'm trying to test my JSON response for integer value but when using Test::Mojo it converts all values to strings - any way around that?
10:18 denis_boyun joined #mojo
10:21 irq joined #mojo
10:28 sujithm joined #mojo
10:28 mire joined #mojo
10:42 gatitskiy joined #mojo
10:42 callumgare joined #mojo
10:43 batman icjs: $data = $t->get_ok('/json')->tx->res->json; and then test $data maybe..?
10:49 batman you can try to see what's going on with Devel::Peek maybe
10:49 callumgare joined #mojo
10:49 batman https://github.com/jhthorsen/refine # not sure where i go with this
10:52 icjs batman, thanks , yep tx->res is helping us
10:53 batman cool
10:54 iakuf quit
10:54 iakuf
10:54 iakuf
10:58 gatitskiy joined #mojo
10:58 punter joined #mojo
11:07 meshl joined #mojo
11:10 callumgare joined #mojo
11:12 neyasov_____ joined #mojo
11:24 john left #mojo
11:25 john joined #mojo
11:28 callumgare joined #mojo
11:29 john Is this possible? :
11:30 john I'd like the browser to be notified via server side events when a different user GETs or POSTs to a particular uri.
11:30 john I am trying to prevent the app from having to poll a database for when a job has finished running
11:31 john ie a task finishes on a remote machine nd I'd like to notify the user that submitted the task.
11:32 bc547 john: you could look at websockets
11:32 john that handles the notification between user and webserver
11:32 john left #mojo
11:32 john joined #mojo
11:34 Kripton joined #mojo
11:34 john joined #mojo
11:35 john but I need the task to notify the webserver that in turn notifies the browser
11:35 john notifying the webser is easy with a get
11:35 john notifying the browser is easy, with sse or websockets etc
11:36 john but linking the two is maybe not so easy
11:36 john I'm using hypnotoad so the browser and task may link to different processes
11:36 john I could update a db and poll that, but that is kludgy
11:37 john I'm open to suggestions
11:37 john :)
11:41 john Hows about the browser request opens an unnamed pipe and listens on that async, and the task GEts the server which writes to the other end of the pipe?
11:42 john is that doable?
11:51 sujithm_ joined #mojo
11:55 crab use webrtc and have one browser notify the other directly
11:57 punter Use a MQ
12:01 john webrtc?
12:01 purl i guess webrtc is Real-time communication without plugins http://www.html5rocks.com/en/tutorials/webrtc/basics/
12:03 john crab: ah, no the "other" is not a browser.
12:03 john I'm trying to lose the polling
12:04 john hence my pipe idea above
12:10 dod joined #mojo
12:15 mr-fooba_ joined #mojo
12:21 chankey joined #mojo
12:21 callumgare joined #mojo
12:42 d4rkie joined #mojo
12:54 ryozi joined #mojo
13:01 john how do I get at the attributes of Mojo::IOLoop?
13:03 jberger attributes?
13:03 purl attributes is http://perlmonks.org/?node_id=631931 or working in roles with MooseX::MethodAttributes::Role or how Catalyst assigns Action types
13:06 john reactor
13:06 john The doc says $loop->rector
13:06 jberger ah
13:07 jberger many of the Mojo::IOLoop (class) attributes really delegate to the singleton instance
13:07 john Tho having rad the purl recommendation above, my head has exploded
13:08 jberger so Mojo::IOLoop->start is synonymous with Mojo::IOLoop->singleton->start
13:08 john Ah so via the singleton is fine then? I was worried it was a different loop or somesuch (I'm not au fait with event loops tbh)
13:08 john Thank you, sorted!
13:08 jberger no worries
13:18 sri thinking some more about the singleton method idea, it has to work like this http://pastie.org/9354710 and not use the coderef for the name http://pastie.org/9354723
13:18 sri batman: the coderef changes, so it would leak
13:20 jberger sri: remember than money_patch can accept a list of pairs
13:20 jberger *that
13:23 sri oh right, so the patch becomes http://pastie.org/9354749
13:23 sri :)
13:24 sri btw. i'm not actually proposing this... hacking just for fun
13:27 sri the other thing i talked about yesterday was more of a proposal http://irclog.perlgeek.de/mojo/2014-07-03#i_8970735
13:31 jberger could the syntax be json pointers?
13:33 sri no
13:33 neyasov_____ joined #mojo
13:34 neilhwatson joined #mojo
13:40 KCL_ joined #mojo
13:44 sri the convention used by many frameworks is foo[bar][baz]
13:47 dod joined #mojo
14:00 dod joined #mojo
14:01 callumgare_ joined #mojo
14:06 callumgare joined #mojo
14:10 callumgare_ joined #mojo
14:11 neyasov_____ joined #mojo
14:33 jberger carton install seems super fast, I wonder if it sets multiple cores?
14:38 btyler joined #mojo
14:38 callumgare joined #mojo
14:39 john I have a watcher on a pipe, in its callback I read from the pipe if it's readable.
14:39 john Trouble is, the watcher then fires endlessly
14:39 john DO i have to close and reopen the pipe or something?
14:40 hesperaux_ joined #mojo
14:48 callumgare joined #mojo
14:51 jberger john: might http://mojolicio.us/perldoc/Mojo/IOLoop/Stream be more appropriate?
14:53 callumgare joined #mojo
14:57 john shit hot!!
14:57 john That would be exactly what I'm trying to write :)
14:57 john Thank you again!
14:59 KCL joined #mojo
15:02 sh4 joined #mojo
15:07 davido_ joined #mojo
15:07 callumgare joined #mojo
15:08 jberger sri: https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/lib/Mojo/IOLoop/ForkCall.pm#L39
15:08 jberger is that really all I need to do? it feels like that's too easy
15:08 sri :)
15:09 sri actually, i think it should be Mojo::IOLoop->reset
15:09 sri because who cares about non-singleton ioloops
15:09 jberger but I do allow for it, should I not?
15:10 sri normal ioloops can just be thrown away... delete $self->{ioloop} would be enough
15:15 callumgare joined #mojo
15:22 jberger sri: so this then? https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/lib/Mojo/IOLoop/ForkCall.pm#L38
15:23 jberger and if they happen to carry in a third ioloop into the child, then a pox on their house, right?
15:23 callumgare joined #mojo
15:25 sri jberger: the point is to keep non-blocking operations working
15:25 sri i don't care about non-singleton ioloops really
15:25 sri don't guess, add tests for this stuff
15:28 jberger I thought the fear was cloning the acceptors into the child
15:28 jberger I would like to test, but I guess I still don't know enough to do so :-(
15:34 john I can't get the IOLoop::Stream to work. It won't read.
15:34 john http://p.thorsen.pm/663e76d8f692
15:35 john Anyone any ideas, please?
15:35 mire joined #mojo
15:35 sri doesn't $stream go out of scope?
15:36 john I don't think so.
15:36 john ell its declared out of the eval.
15:37 john *well
15:37 john Should it be a package var?
15:38 john http://p.thorsen.pm/148a933f85f4
15:39 john Is what I had before, but the watcher fires continually after I've read some data.
15:39 * sri wonders what would be the most efficient way to translate {'foo[bar][baz]' => 'yada'} into {foo => {bar => {baz => 'yada'}}}
15:40 crab mechanical turk?
15:40 purl mechanical turk is probably nothing special. those call distribution systems which autodial and when you pickup connect you to some telemarketer are the same thing. amazon just opened up the api
15:45 mattastrophe joined #mojo
15:45 punter joined #mojo
15:46 KCL_ joined #mojo
15:48 kyshtynbai joined #mojo
15:48 john Package var made it work.
15:50 nemux Hi all!
15:50 purl i heard Hi all! was possible to include comments (like /* comment */ in mysql) in a select statement made using a search()?
15:51 sri john: does a package var really sound like a good idea in an object?
15:52 sri really think about what scope you need
15:53 sri there are many different scopes to choose from, global, application, request, object...
15:54 john It was defined in the sub and didn't work, now it does when defined out of the sub. How else could I do it, please?
15:54 nemux i'm trying to use multiple placeholders in the route but if i use $r->any('/foo/:param/bar')->to('foo#bar'); everything is ok... but if use $r->any('/foo/:param1/:param2/bar')->to('foo#bar'); i get back a page not found error calling -> GET /foo/value1/value2/bar
15:54 nemux any help?
15:54 purl any help is appreciated
15:55 sri john: attributes, stash, ioloop...
15:55 john Where can I read up on attributes and ioloop?
15:57 john Oh, it has to persist across requests
15:57 john btw
15:57 sri nemux: bollocks
15:57 sri perl -Mojo -E 'a("/foo/:param1/:param2/bar" => {inline => q{%= join ":", $param1, $param2 }})->start' get /foo/value1/value2/bar
15:57 john 1 request sets up the read on the pipe, it then waits until a totally diff process writes to the pipe
15:58 sri john: oh that sounds terrible... and i don't feel comfortable helping with that
16:00 john I'm effectively trying to notify a browser when a file changes, without polling with ajax.
16:01 sri just use a database
16:01 john Ten I'm polling the DB
16:01 john *then
16:01 john which is worse
16:02 sri databases these days have pub/sub systems
16:02 sri redis, mongodb, postgresql...
16:02 john Im using mysql
16:03 sri tough luck then
16:03 john :)
16:03 john Besides, I wanted to async it
16:03 john and listening async on a handle seems easy enough
16:04 sri i'm not saying it can't be done, just that i'm not supporting it
16:04 john but I can't tell reactor->io to fire once and then shut up and listen again
16:14 sri http://pastie.org/9355198 # this is what i had in mind for nested parameters
16:15 sri dunno if it's a good idea
16:20 mire joined #mojo
16:28 TitanOfOld joined #mojo
16:28 TitanOfOld So, uh, why would I want to use Mojolicious::Validator instead of just using, ya know, Perl?
16:49 crab to save time and effort, just like when you use any (well, ok, most) perl modules?
16:51 lb TitanOfOld: M::V pays off if you use add_check a lot
16:52 lb add lots of business logic rules in app startup, and much cleaner validation in actions, more DRY
17:08 sri soooo... i guess it's not such a good idea ;p
17:11 TitanOfOld I still don't see the value in it. I can write a check sub, and call that sub on the desired param.
17:12 TitanOfOld I'm convinced I'm missing something seeing as sri is a genius.
17:16 KCL joined #mojo
17:17 sri TitanOfOld: do this without using the validator and then compare the code http://mojolicio.us/perldoc/Mojolicious/Guides/Rendering#Form_validation
17:19 basiliscos joined #mojo
17:19 sri and now stop derailing my topic! if anyone likes nested params please speak up
17:22 crab sri: your patch looks good
17:23 crab but the depth of my not caring about the feature is so deep that i can't even be bothered to describe it entertainingly
17:23 sri it's not done, for now i'm more interested in feedback on the actual feature
17:23 sri i see
17:23 crab who generates such parameters?
17:24 sri people using nosql databases
17:25 sri generate the strucutre you pass to your document database without the {foo => $validation->param([foo]), ...} dance
17:26 basic6_ joined #mojo
17:27 crab i see
17:29 sri or even with postgresql and your json data type
17:32 crab json in 9.4 is going to be so cool
17:32 dod joined #mojo
17:34 denis_boyun joined #mojo
17:40 KCL_ joined #mojo
17:42 sri well, mongodb 2.8 will also kick ass
17:43 crab i'll be sure to use an ass-shield
17:49 * sri pokes jberger, tempire and marcus
17:51 * jberger deflates
18:05 KCL joined #mojo
18:09 jberger sri: your patch can only work on hash-of-hashes?
18:10 btyler joined #mojo
18:14 jberger I cannot think of a way to test that the child process is not influenced by the parent's ioloop
18:15 jberger I've been racking my brain
18:20 KCL_ joined #mojo
18:21 crab what does influenced mean?
18:22 crab if you're thinking about connections in the parent remaining open in the child, you can test that directly
18:24 jberger how?
18:24 sri jberger: so far, not going to finish it before there is visible interest
18:24 jberger I'm trying to emulate this: https://github.com/kraih/minion/blob/master/t/file_lite_app.t
18:25 sri like i said before foo[1] could be first array element and foo[] a push on an array
18:25 howitdo joined #mojo
18:26 sri jberger: that's the ioloop reset test case https://github.com/kraih/minion/blob/master/t/file_lite_app.t#L63
18:27 sri which makes this break, because you can't start the event loop twice https://github.com/kraih/minion/blob/master/t/file_lite_app.t#L34
18:28 jberger so it would throw there if that happened?
18:28 sri yes
18:32 basiliscos joined #mojo
18:32 jberger \o/ got it!
18:34 jberger https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/t/loops.t
18:35 rawler joined #mojo
18:44 bjoernfan ... sooo a friend took a photo of my broken car keys on top of my laptop and Google Goggles recognized a little something... https://undef.io/dump/Screenshot_2014-07-04-20-40-20.png
18:44 xevz joined #mojo
18:45 bjoernfan That's... a pretty unique logo you've got there. :)
18:45 jberger bjoernfan++
18:45 jberger hahaha
18:47 xevz TIL that Goggles search your photographs automatically out of the box.
18:47 xevz :)
18:48 jberger sri: so since the child doesn't have access to the $fc object to get a non-singleton loop, and thus since it can't try to start the same loop again (without closing over it) then do I need to delete it in the child still?
18:50 mire joined #mojo
18:51 jberger then again, if I explicitly reset the $fc object's loop it does fix the case where that loop is closed over
18:52 rawler joined #mojo
18:57 jberger does calling Mojo::IOLoop->singleton->reset twice in a row cost much?
19:01 jberger well, now it's protected even if you are calling to the $fc's non-singleton loop: https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/t/loops.t#L31
19:02 sri bjoernfan: lol
19:05 punter could we make a "Made with Mojolicious" mini-banner?
19:05 punter I want to place it on my site
19:05 punter (s)
19:06 punter not so important though, I understand
19:06 jberger maybe tempire has one already?
19:06 punter tempire ^^ ?
19:07 jberger he would probably just have you hot-link to "pink fluffy unicorns"
19:08 punter dancing on a rainbow
19:08 punter I'd do that
19:10 jberger sri: do I still need this warning then? https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/lib/Mojo/IOLoop/ForkCall.pm#L150-L152
19:12 jberger (and that waning was mostly cargo-culted from: https://metacpan.org/pod/AnyEvent::Util#fork_call-CODE-args-cb--res)
19:12 sri jberger: no
19:13 jberger oh and one more thing too, should I fork in a next_tick?
19:13 jberger the fork happens immediately when ->run is called
19:13 sri jberger: no
19:13 jberger which might, in concept, happen before the ioloop watcher is ready
19:13 sri if you invoke some callback... then that should be deferred with a next_tick
19:14 jberger hmmmm, yeah, I'm not sure how that could work anywya
19:14 sri next_tick is just to make sure callbacks don't fire right away
19:14 sri punter: give me a few minutes :)
19:15 punter ok
19:16 jberger sri: ok, but if the $fc's loop is not running, then what happens if the child finishes and this is not called: https://github.com/jberger/Mojo-IOLoop-ForkCall/blob/reset/lib/Mojo/IOLoop/ForkCall.pm#L77
19:16 * jberger checks to see if that is tested
19:16 sri i actually wonder if the small standard mojolicious logo we put at the bottom should have a little "made with" tag line
19:17 punter no smile, rainbow or unicorn?
19:17 punter :-)
19:18 punter it's not bad
19:19 punter if it had a smile, a rainbow, a unicorn and a cloud, it would be even better
19:19 punter (I think)
19:24 sri yea, doesn't work http://i.imgur.com/VacZfwU.png
19:24 punter not bad at all :-)
19:24 sri DO NOT USE THAT
19:24 punter letters are too thin
19:24 sri the spacing is all wrong
19:24 punter ok, I wont
19:28 sri a bit more tacky http://i.imgur.com/AAL5UN6.png
19:29 jberger I love that Convos shows images inline
19:29 jberger batman++ marcus++
19:29 sri possibly like this http://i.imgur.com/CEFbAIo.png
19:30 jberger I like "and"
19:30 punter Isn't love old-fashioned
19:30 punter ?
19:30 marcus punter: never!
19:30 punter :-)
19:30 marcus punter: love is enduring.
19:31 punter But when you make something out of love, it's usually because there's no money involved.
19:31 punter Or am I wrong?
19:31 punter or not enough money
19:31 punter no, BS
19:31 punter i'm saying BS
19:31 jberger sri: I like it
19:31 marcus cynic.
19:32 sri i like it too, gonna look into the formatting a bit... but i think that's the theme i'm going with
19:32 marcus I like it too, sri.
19:32 sri possibly replacing the bottom logo with it
19:32 marcus it could read like Made with love and clouds - Mojolicious tho
19:32 sri gives a more personal touch
19:32 jberger marcus: that is how I read it with &
19:33 jberger but with "and" it made more sense somewhow
19:33 sri with & you get special character overload
19:33 punter but
19:33 punter can you imagine a business saying in public "made with love"?
19:33 punter some won't want to do that
19:33 punter i think
19:33 sri SHUT UP AND SPREAD THE LOVE
19:33 punter maybe
19:33 punter ok!!!!!! :-_)
19:34 marcus punter: http://madewithlove.be/
19:34 marcus I guess I can.
19:34 jberger I think there should be a series of pics, each with a different unicode character there!
19:34 punter made with love, lots of sex, and a VC
19:34 marcus Made with Pile of Poo and Mojolicious
19:34 jberger make with "EXTENDED MIDDLE FINGER" and mojolicious :-)
19:35 sri that escalated quickly
19:35 punter I agree with series of pics, each with different unicode char
19:35 marcus Made with U+1F494 and Mojolicious
19:35 punter choice is a good thing
19:36 jberger oooh, webapp that generates the pic, with any given unicode character
19:36 sri Made with "that missing glyph symbol" and Mojolicious
19:40 sri some minor formatting changes http://i.imgur.com/1jMbTUA.png
19:43 sri different font weight http://i.imgur.com/xdNTfWd.png
19:44 punter not readable enough (i think)
19:44 cpan_mojo joined #mojo
19:45 punter What if someone didn't make his site with love
19:46 punter But he made it with Mojolicious, nevertheless
19:46 marcus punter: then it’s gonna suck
19:46 punter hehehe
19:46 punter yes
19:46 sri http://i.imgur.com/2bACZGN.png
19:47 punter I was under the impression sri was motivated by excellence and perfection and beauty of code, not by love itself
19:47 sri if you don't love web development this is not the right place for you ;p
19:47 punter was I wrong?
19:48 punter ok
19:48 punter I might make enemies, but I think a heart looks a bit girly
19:49 punter The visitors will think a girl made the site
19:49 punter my site
19:49 sri i'm totally in touch with my feminine side
19:49 punter ok
19:49 jberger if it is axiomatic that you love what is fun, then since Mojolicious is about fun, then you do it with love. qed.
19:50 punter fine
19:50 jberger (doctored)
19:51 Eke- joined #mojo
19:51 sri "when we say made with heart...we mean the hearts of our dead enemies"
19:51 sri masculine enough for you? :)
19:51 neyasov_____ joined #mojo
19:51 punter Yess
19:52 punter Can't you at least make the heart red?
19:52 punter or maybe... I don't know
19:53 punter I'm off to bed. Thanks for the logo.
19:53 jberger punter++, good idea
19:53 sri doesn't work so well then
19:54 sri same way an emoji image wouldn't work... it removes this "this is still text, but with an odd symbol" charme
19:56 cpan_mojo Mojo-IOLoop-ForkCall 0.10 by Joel Berger - http://metacpan.org/release/JBERGER/Mojo-IOLoop-ForkCall-0.10
19:56 jberger \o/
19:56 sri but of course i'm willing to experiment http://i.imgur.com/rEa5yJg.png
20:07 sri i think this version looks pretty decent http://i.imgur.com/7RvArSx.png
20:08 jberger +1
20:08 purl 1
20:08 jberger silly purl, you don't get a vote
20:08 * jberger takes purl's voter registration
20:09 jberger see N.B.: https://metacpan.org/pod/release/JBERGER/Mojo-IOLoop-ForkCall-0.10/README.pod#WARNINGS
20:09 sri purl: what do you think?
20:09 purl sri: wish i knew
20:23 sri hehe, with the emoji heart http://i.imgur.com/EiCdosZ.png
20:25 sh4|2 joined #mojo
20:33 sri haha
20:33 sri look at the bottom of the atom page for a different variant of the concept https://atom.io/
20:38 denis_boyun joined #mojo
20:56 Eke- joined #mojo
20:57 marcus I like the emoji heart
20:58 marcus But of course, I’m a girly man.
20:58 marcus I mean https://twitter.com/marcusramberg/status/483872106669563904
20:59 marcus ^^ this is my favorite convos embed, jberger.
20:59 sri getting closer :) http://i.imgur.com/LR4XA2J.png
20:59 sri proper heart makes a big difference
21:00 marcus sri: Can we make it align with the i and remove the dot? ;-O
21:00 marcus keming
21:00 sri been trying... but it just won't look good
21:01 jberger marcus: yeah, twitter inline is cool too
21:01 jberger gists aren't sized right, but then again, I'm not sure what size they should be
21:04 Adura https://twitter.com/hashtag/mojolicious+smores
21:04 Adura If only that could be a codename.
21:05 marcus our dear friend Deborah.
21:05 Adura Unfortunately her site isn't powered by Mojolicious.
21:13 denis_boyun joined #mojo
21:18 mire joined #mojo
21:43 sri still tweaking little details... http://i.imgur.com/mutgx21.png
21:44 Phil21 hrm, having a little trouble with a "basic" client JSON call, are there any real-world examples other than the docs anyone can point me to?
21:44 Phil21 just trying to basically do this w/ mojo: curl -k  -X POST https://x.x.x.x:8457/rest/nms -H "Content-Type: application/json" -H "Authorization: Basic SECRET==" -d '{"object": "folder", "params": [""], "method": "get_names"}'
22:06 davido__ joined #mojo
22:12 neyasov_____ joined #mojo
22:17 cpan_mojo Ado 0.50 by Krasimir Berov - http://metacpan.org/release/BEROV/Ado-0.50 (depends on Mojolicious::Plugin::DSC)
22:23 go|dfish Phil21: Are you using 'mojo get' from the command-line? http://mojolicio.us/perldoc/Mojolicious/Command/get
22:30 meshl joined #mojo
22:32 good_news_everyon joined #mojo
22:32 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/4_3x_Q
22:32 good_news_everyon mojo/master 83744b2 Sebastian Riedel: ALL GLORY TO THE HYPNOTOAD
22:32 good_news_everyon left #mojo
22:35 tempire nested params: interesting
22:35 tempire I've always just used the mojolicious logo for a "made by" thing
22:36 sri gonna test the made with heart version for a bit, and then decide if it stays... i'm not 100% sure yet
22:37 sri it's now live on mojolicio.us
22:38 tempire I like the idea
22:38 tempire I think I like the original better, though
22:39 berov joined #mojo
23:05 good_news_everyon joined #mojo
23:05 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/JQRgGQ
23:05 good_news_everyon mojo/master 7174bc3 Sebastian Riedel: switch back to original logo
23:05 good_news_everyon left #mojo
23:05 sri guess i'll listen to tempire
23:30 xevz batman: /whois batman
23:31 xevz Bah
23:31 xevz :)
23:31 xevz batman: I took matter into my own hands, regarding those Convos notifications I pinged you about yesterday, I now have a Greasemonkey script that send Pushover notifications. :)
23:48 xevz batman: Pushover is probably bad to use at work, but it works as a concept. :)

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