Camelia, the Perl 6 bug

IRC log for #mojo, 2012-02-27

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

All times shown according to UTC.

Time Nick Message
00:15 tempire I have a first revision of deploy heroku ready for testing
00:16 jnap joined #mojo
00:18 GitHub159 joined #mojo
00:18 GitHub159 [mojo] kraih pushed 1 new commit to master: http://git.io/1OflZA
00:18 GitHub159 [mojo/master] improved resilience of Mojo::IOLoop::Stream slightly - Sebastian Riedel
00:18 GitHub159 left #mojo
00:19 sri \o/
00:21 tempire only 2 deps
00:22 tempire hmm.  won't require heroku tools.
00:22 tempire not sure how windows works, though, since it uses ssh
00:22 tempire oh, requires git too
00:22 Alias joined #mojo
00:23 tempire I'll have to look into Git::PurePerl
00:28 black joined #mojo
00:41 tempire ok who wants to be a beta tester!
00:41 tempire sri?
00:41 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 making Mojolicious delicious
00:41 sri i'm not the target audience
00:42 tempire you're supposed to say I WANT TO SEE HOW IT WORKS!
00:42 sri you want a beginner, or someone who can break everything like marcus :)
00:42 tempire marcus.  wake up and break stuff.
00:42 sri I WANT TO SEE HOW IT WORKS!
00:42 tempire now I don't believe you.
00:43 sri :/
00:43 * tempire takes his modules and goes home.
00:43 tempire ok.  don't tweet about it, though.
00:43 tempire cpanm https://github.com/tempire/mojolici​ous-command-deploy/raw/master/Mojol​icious-Command-deploy-0.01.tar.gz https://github.com/tempire/mojolicious-​command-deploy-heroku/raw/master/Mojoli​cious-Command-deploy-heroku-0.01.tar.gz
00:44 tempire and then the workflow section of the readme: https://github.com/tempire/moj​olicious-command-deploy-heroku
00:44 tempire I need a windows tester.
00:44 * tempire pokes windows users
00:45 sri you can always bail on windows like hypnotoad does
00:46 tempire so I can make it work using pure perl git, I think.
00:46 sri doesn't strawberry ship with git?
00:46 sri or was that just cygwin?
00:48 sri workflow looks quite pleasant
00:48 Alias Strawberry doesn't ship with git
00:48 tempire I can get away without git...but ssh.
00:48 Alias But it is trivially installable, so as long as your module can auto-detect the git binary, you're sweet
00:49 sri does the heroku installer bundle git/ssh?
00:49 tempire It does with git.  I'm guessing it does ssh as well.  I don't see how else it could work.
00:49 Alias Something other than the core Strawberry install might do so,
00:49 tempire I'd like to make this work without needing the heroku download.
00:50 sri tempire: your perloku generator should set --mode production
01:01 d4rkie joined #mojo
01:02 bird joined #mojo
01:04 tempire I'm really liking this github edit file feature.
01:13 scott joined #mojo
01:14 tempire Alias: thanks
01:16 mire joined #mojo
01:35 jnap joined #mojo
02:44 hshong joined #mojo
02:47 sri hmm, i might have accidentally fixed a bug :o
03:00 * tempire feels like sri has given up all hope on a sleep schedule
03:00 sri but i can't prove it :/
03:00 * sri has given up all hope on a sleep schedule
03:07 tempire *of
03:15 vel joined #mojo
03:16 chansen joined #mojo
03:41 arpadszasz joined #mojo
04:08 mire joined #mojo
04:55 hshong joined #mojo
05:04 Foxcool joined #mojo
05:05 hshong joined #mojo
05:11 xaka joined #mojo
05:15 baton8 joined #mojo
05:54 tempire Searching for mojolicious on twitter is almost useless if you can't read japanese.
05:55 tempire we need perlrocks back.
05:55 * tempire feels disconnected from the japanese brethren
06:01 Cyrus Coming in from Rails, I'm pretty excited to start out with mojolicious. Just wanted to say thanks for all the hard work guys. <3
06:03 preflex_ joined #mojo
06:15 noganex joined #mojo
06:24 marcus Moin
06:39 spleenjack joined #mojo
06:50 grim_fandango joined #mojo
07:23 mire joined #mojo
07:50 espent joined #mojo
08:19 tempire marcus marcus marcus
08:53 berov joined #mojo
08:56 cosmincx joined #mojo
09:00 batman joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:18 jamesw joined #mojo
09:30 olav joined #mojo
09:41 black joined #mojo
09:48 judofyr joined #mojo
09:53 daxim joined #mojo
10:00 black joined #mojo
10:03 judofyr sri: btw, I've pretty much given up on the idea of implementing Mojo on actors in Ruby…
10:04 judofyr sri: but I still think Mojo + co-routines in Ruby is a viable approach
10:45 black joined #mojo
11:12 black joined #mojo
11:17 AmeliePoulain joined #mojo
11:28 jamesw joined #mojo
11:33 crab implementing mojo on actors in ruby? what does that mean?
11:34 black joined #mojo
11:37 judofyr crab: played with Erlang?
11:37 judofyr crab: https://github.com/tarcieri/celluloid is Erlang-inspired actors implemented in Ruby
11:39 crab judofyr: so you are evaluating various different ways to implement a mojo-like API for ruby?
11:39 judofyr crab: yup
11:39 judofyr because async web stuff in Ruby is a mess
11:40 Netfeed even with eventmachine?
11:41 zpmorgan this may be a stupid question. how should I access session data in a template?
11:41 judofyr especially with eventmachine
11:42 judofyr first of all: Rack + async is a mess. there's a somewhat standard of async.callback, but that pretty much only works in Thin
11:42 judofyr and then you need AsyncRack (which monkey patches Rack) and an async framework (e.g. async-sinatra which monkey patches Sinatra)
11:42 judofyr zpmorgan: there's a session-helper, no?
11:43 zpmorgan durr, yes, thanks judofyr
11:43 vervain zpmorgan $self->session('foo');
11:44 zpmorgan I wasn't aware that there was a $self within a template.
11:44 zpmorgan would that be the controller?
11:44 vervain or even % session 'foo' or maybe even % session('foo')
11:44 vervain Yes
11:45 vervain It in M::G::Rendering: "tash values that don't have invalid characters in their name get automatically initialized as normal variables in the template, and the controller instance as $self."
11:45 vervain One if the harder sentences to parse... but it's there.
11:46 zpmorgan okay, I'll check that out. thanks vervain
11:46 vervain np
11:46 judofyr Netfeed: you know Ruby?
11:47 black joined #mojo
11:48 Netfeed yeah
11:49 crab judofyr: so rails can't do all this stuff?
11:50 black joined #mojo
11:50 Netfeed rails is terrible
11:51 judofyr crab: it's not supported at all.
11:51 judofyr crab: I guess you can get it working: https://github.com/igrigorik/async-rails
11:54 black joined #mojo
11:57 berov joined #mojo
11:58 judofyr Netfeed: what are your preferred stack? (if you do web stuff in Ruby)
11:58 judofyr is*
11:58 Netfeed i don't have one honestly, i think that most actually blows
11:59 Netfeed i kinda liked the idea of cuba.is though
11:59 Netfeed but there's never any good documentation imho
12:00 judofyr Netfeed: you should help me with Zoid! (Mojo in Ruby)
12:00 judofyr although I need to start it first…
12:01 Netfeed heh, i actually prefer to learn mojo better, i bet i only use a very small portion of it's power
12:04 judofyr sounds like a good idea
12:07 preflex_ joined #mojo
12:19 judofyr Netfeed: although Zoid is going to be pretty cool because of Ruby's co-routines (fibers)
12:20 judofyr so a lot of async IO is going to look like blocking code
12:20 judofyr (without actually blocking anything)
12:20 Netfeed as long as there's good documentation :)
12:23 Netfeed i'm not sure what it is with the ruby webframeworks, but there's always something that bugs me with them
12:27 Netfeed i kinda like Flask for python, but it's python.
12:50 vervain Is there a way to 'clear cookies' with Test::Mojo or should I just create a new Test::Mojo instance (which works)?
12:52 vervain ^ Ignore that ^  I don't really want to do what I thought I wanted to do. :-)
13:02 abra__ joined #mojo
13:44 vervain I wonder how may HR people cringe at Dancer or Mojolicious as required skills. :-)
13:48 * sri wonders how we could fail two tests on cpantesters
13:48 sri http://www.cpantesters.org/distro​/M/Mojolicious.html#Mojolicious-2.53
13:48 sri at least one could be fixed in master, but no way to prove it :(
13:49 sri i hate to say it... but maybe another release might be a good idea
13:49 judofyr are you trying to release less often?
13:49 sri yea
13:49 vervain Doesn't span have the same automated testing for development versions?
13:50 vervain s/span/cpan/
13:50 sri the harder i try the more releases we end up with...
13:50 judofyr sri: why? (fewer releases)
13:50 sri what's the point of a dev release when master is in good shape?
13:51 sri judofyr: because perl people
13:51 purl because perl people are very conservative when it comes to syntax duh!
13:51 judofyr ah
13:51 judofyr damn perl people
13:51 sri they get all scared when you release too often
13:52 vervain Or too little. :-)
13:52 * sri nods
13:53 * sri wants a release anyway
13:53 * sri cracks the whip
13:53 sri FOR SCIENCE!
13:53 purl Yes, I will kiss the girl from Venus... for science!
13:53 Netfeed perl programmers gets nervous if you release more then once ever two yaer
13:54 Netfeed year*
13:59 vervain Can I get at url_for in tests?  I'm wanting to do ->header_is( Location => $t->app->url_for('/') );
14:03 jnap joined #mojo
14:06 jnap joined #mojo
14:49 vel joined #mojo
14:58 amoore joined #mojo
14:59 daxim http://narf-archive.com/pix/d8b9613a​5ca1090f6263f1e936aa658acce977f3.png
15:18 d4rkie joined #mojo
15:24 arpadszasz joined #mojo
15:28 inokenty joined #mojo
15:32 kaare joined #mojo
15:33 GitHub10 joined #mojo
15:33 GitHub10 [mojo] amenonsen pushed 1 new commit to master: http://git.io/8-Em4w
15:33 GitHub10 [mojo/master] Bump VERSION after release - Abhijit Menon-Sen
15:33 GitHub10 left #mojo
15:42 sri crab++
15:52 GitHub42 joined #mojo
15:52 GitHub42 [mojo] kraih pushed 1 new commit to master: http://git.io/dMd9jg
15:52 GitHub42 [mojo/master] better descriptions for methods in Mojo::Home - Sebastian Riedel
15:52 GitHub42 left #mojo
15:54 marcus so, what happened to slower release rate? ;)
15:57 omf_ left #mojo
16:02 sri marcus: i think the only way to make it work is giving up on trying ;p
16:02 sri there is some serious reverse psychology at play here
16:03 judofyr marcus: the full moon went away and the normal sri is back.
16:03 sri :D
16:04 marcus sri is just trying to deprecate his mistakes in a hurry :-p
16:05 sri HIDE ALL THE MISTAKES
16:06 judofyr here comes "git push -f"
16:06 vervain Is this an acceptable way to get a 'string if matches foo otherwise otherstring' in templates? http://pastie.org/3472781
16:06 vervain Really I'm asking about the way I beat the pattern match into list context.
16:12 crab it's perl, do whatever you want
16:13 vervain If that were true there wouldn't be books called 'Modern Perl' and 'PBP' :-)
16:14 vervain I know that it works... so that means I can get away with it... but is it a decent idiom... and it's relevant to the recent change to force scalar context... or whatever was done.
16:15 sri the recent change only means you have to worry less
16:15 sri if it works it just works
16:16 sri (in all variations)
16:18 sri that said... i think ()[0] is more common than []->[0]
16:24 batman joined #mojo
16:26 crab yes, definitely
16:27 vervain Ah... ya... I knew there was something uneasy looking about what I did.. thank you!
17:09 kaare_ joined #mojo
17:24 echoecho joined #mojo
17:26 echoecho Hello! I am trying to get the hang of mojo websockets with this pretty simple chat-room example on Github (http://bit.ly/yOa4cZ), but whenever I send a message, my server fails with "Event "read" failed: Can't locate object method "send_message" via package "Mojo::Transaction::WebSocket"". Any ideas as to what I'm doing wrong?
17:27 sri that example is outdated, some methods have been renamed
17:27 echoecho Ah, that makes sense! Could you help me figure out which, or point to a newer example?
17:27 sri the websocket api is still experimental and changing
17:27 purl i already had it that way, sri.
17:28 judofyr echoecho: https://metacpan.org/module/​Mojolicious::Controller#send
17:28 sri the core documentation tends to have better examples
17:28 echoecho judofyr: Thanks, reading now!
17:28 echoecho sri: Noted, I'll poke around in there as well. I learn best by example.
17:28 sri http://mojolicio.us/perldoc/Mojolicious​/Guides/Cookbook#WebSocket_web_service
17:30 echoecho Very nice, just swapping out the method calls made it work. I'll read both linked docs.
17:31 echoecho I take it, as the API evolves/stabilizes, mojo perldoc will be the smart place to keep up-to-date?
17:31 sri please update the wiki example while you're at it :)
17:32 echoecho I'm not in control of that Github project; I'll send an errata to the guy though.
17:32 sri yes, core docs will always be most up to date
17:32 sri it's a wiki
17:32 sri everybody can change pages
17:33 echoecho In that case I'll just make an edit then :)
17:33 sri echoecho++
17:36 kaare joined #mojo
17:45 bobkare joined #mojo
17:45 sri i've opened a ticket for stabilizing the websocket api https://github.com/kraih/mojo/issues/295
17:45 sri everybody who cares is welcome to comment
17:46 echoecho :)
17:58 batman left #mojo
17:59 batman joined #mojo
18:05 black joined #mojo
18:08 sri hmm, strange that there's already a SPDY book
18:08 sri it's even a few months old, and doesn't cover the spec submitted to the ietf :S
18:09 echoecho Obsolete by the time it gets to the stores, I imagine? :/
18:09 vervain Who's the publisher?
18:09 sri pragmatic
18:09 purl couldn't get the headlines: http://www.inkdroid.org/perl/pragmatic wasn't successful
18:10 sri http://pragprog.com/book/csspdy/the-spdy-book
18:10 vervain Ah yes.. an eBook.
18:11 Cyrus left #mojo
18:11 echoecho I'll start buying ebooks when I can apt-get update
18:11 echoecho as it were
18:11 sri seems very strange considering there's not a single book about websockets, and they have a much wider scope
18:12 echoecho Well, for SPDY I imagine somebody decided to print out the Google Code white papers and add a binding to make some quick cash...
18:13 vervain I'm into radio control stuff... most of the ebooks are hashed together forum posts.
18:13 echoecho "No known errata." - SPDY book site. Well that's good. >< lol
18:13 sri xD
18:13 echoecho No known purpose either, mayhap...
18:13 vervain :-)
18:25 xaka joined #mojo
18:36 GitHub135 joined #mojo
18:36 GitHub135 [mojo] kraih pushed 1 new commit to master: http://git.io/7mN5iA
18:36 GitHub135 [mojo/master] made Mojo::UserAgent::Transactor slightly smaller - Sebastian Riedel
18:36 GitHub135 left #mojo
18:37 metys nd they have a much wider scope
18:38 tempire marcus: your assistance is requested.
18:38 tempire also, vervain
18:38 sri ALL HANDS ON DECK!
18:38 echoecho lol
18:38 vervain I hope it involves tasting chocolate.
18:39 * sri too
18:39 tempire almost.  beta tester for heroku deployment.
18:40 echoecho Still poking at the mojo docs, sorry if this should be obvious: how do I stop my clients from timing out? The chat example works perfectly, but as I fiddle I notice that the socket closes after awhile, Firefox Error Console mentioning that "The connection to ws://localhost:3000/echo was interrupted while the page was loading."
18:41 tempire vervain: have you done anything with heroku?
18:42 vervain Not yet.
18:42 tempire excellent.
18:42 purl excellent. are there any japan-natives lurking in here?
18:43 tempire also, anyone else in here who has recently started with mojolicious, I want your feedback as well.
18:43 tempire https://github.com/tempire/moj​olicious-command-deploy-heroku
18:43 sri FOR SCIENCE!
18:43 purl Yes, I will kiss the girl from Venus... for science!
18:45 * sri really likes the simplicity of heroku
18:47 * tempire waits for feedback, questions and problems
18:47 tempire just follow the work flow, and see if you run into any problems.
18:47 * sri pokes the guinea pigs
18:47 tempire also, anyone who has windows, speak up.
18:48 echoecho <--- Windows
18:48 echoecho This heroku stuff looks like cocaine, will fiddle this evening.
18:48 tempire perfect, I want a windows user who is unfamiliar with heroku
18:48 * sri puts echoecho into the cage with the other guinea pigs
18:48 tempire xp/7?
18:48 echoecho 7
18:48 tempire do you have git installed, or ssh?
18:49 echoecho Both
18:49 tempire hmm.
18:49 tempire well that's still ok.
18:49 tempire your feedback is still requested.
18:49 echoecho on the github link above?
18:49 tempire yes.
18:49 tempire follow the workflow
18:49 echoecho Roger
18:49 tempire echoecho: what ssh do you have installed?
18:50 echoecho I use PuTTY atm, know there are alternatives but haven't researched
18:50 tempire oh, that's great.  I was thinking command line.
18:50 echoecho There's some pure-openSSH thing, but putty remembers my keys easier I think
18:50 echoecho forget why I made the original decision
18:50 purl echoecho, I didn't have anything matching why i made the original decision
18:51 echoecho thanks purl!
18:51 tempire doesn't matter to me.  I just want to know how heroku handles it.
18:51 echoecho kk
18:51 tempire the toolbelt part, specifically.
18:51 echoecho I will see if I can poke around with it this evening & report back.
18:51 tempire I would love to eliminate the need for downloading anything from heroku, but the ssh part confuses me.
18:51 tempire what timezone are you in?
18:51 echoecho GMT-6
18:52 tempire is that central?
18:52 echoecho err UTC-6? it's 1pm here
18:52 echoecho yeah
18:52 tempire ok.  I'll be around.
18:52 echoecho time, how does it work
18:53 vervain \o/ finally someone in my tz :)
18:54 echoecho Sorry to repeat myself, but does anyone know how to avoid websocket timeout with mojo? Should I just send keepalives from the client?
18:54 sri there are multiple ways
18:55 echoecho Is there an established best practice / good idea?
18:55 xaka echoecho: 1. keepalive 2. change inactivity timeout for the whole app 3. change the timeout for the websocket route
18:56 echoecho xaka: Noted. Will look into each. Thanks.
18:56 sri http://mojolicio.us/perldoc/Mojolicious/​Guides/Cookbook#EventSource_web_service # here's an example
18:56 echoecho Excellent
18:57 sri wait, that might not actually work
18:57 echoecho So in that case, sending each message would delay closure for 5 minutes past the current time.
18:57 sri doh, the websocket timeout overwrites the manual timeout change
18:58 sri since it happens later
18:58 echoecho ic
18:59 xaka sri: why dont use inactivity timeout for all cases?
18:59 sri because it's short?
19:00 echoecho My very limited empirical testing seems to indicate < 5m
19:00 sri that's the long websocket default
19:00 sri 300 seconds
19:00 purl somebody said 300 seconds was a good default
19:00 sri for normal connections it's 20 seconds
19:02 sri i'm actually not a big fan of the websocket timeout value
19:02 sri maybe it should be removed
19:03 echoecho stealing that example seems to be working, but don't quote me on it
19:03 sri that example doesn't work
19:03 sri just checked
19:03 echoecho kk
19:05 * xaka still waiting for separate request/response timeout values...
19:14 sri it's actually just 15 seconds for normal connections
19:23 GitHub81 joined #mojo
19:23 GitHub81 [mojo] kraih pushed 1 new commit to master: http://git.io/UA956w
19:23 GitHub81 [mojo/master] combined WebSocket timeout with normal inactivity timeout - Sebastian Riedel
19:23 GitHub81 left #mojo
19:23 sri new websocket example includes the timeout upgrade, whcih works now
19:24 echoecho http://mojolicio.us/perldoc/Mojolicious/​Guides/Cookbook#EventSource_web_service looks the same to me
19:25 echoecho oh I'll just read the diff
19:26 marcus Cosimo: never heard back from you about your hypnotoad woes?
19:33 GitHub74 joined #mojo
19:33 GitHub74 [mojo] kraih pushed 1 new commit to master: http://git.io/_oBBng
19:33 GitHub74 [mojo/master] also mention how to increase the inactivity timeout in Mojolicious::Controller - Sebastian Riedel
19:33 GitHub74 left #mojo
19:35 tempire marcus: verifying that you saw heroku deployment thing ^
19:36 marcus Broke it already.
19:36 marcus J/k
19:37 * tempire breaks marcus' skis
19:37 marcus /me puts tempire on a snowboard.
19:38 * marcus pushes tempire.
19:38 * tempire slides on marcus misplaced /me
19:41 GitHub86 joined #mojo
19:41 GitHub86 [mojo] kraih pushed 1 new commit to master: http://git.io/Wix_vQ
19:41 GitHub86 [mojo/master] refer consistently to inactivity timeouts - Sebastian Riedel
19:41 GitHub86 left #mojo
19:43 sri ultimately we might have to increase the 15 second default as well
19:44 sri but first lets see how much confusion it causes
19:44 ccushing joined #mojo
19:48 azawawi joined #mojo
19:49 GitHub24 joined #mojo
19:49 GitHub24 [mojo] kraih pushed 1 new commit to master: http://git.io/vcp4PA
19:49 GitHub24 [mojo/master] add forgotten experimental mark to build_websocket_tx method - Sebastian Riedel
19:49 GitHub24 left #mojo
19:53 marcus :-o
19:53 marcus experimentalize?
19:53 marcus I plan to do a websocket app in a couple of weeks.
19:53 sri it should be common sense, but just to be sure :)
19:53 marcus so I might have some more feedback on the api then.
19:53 marcus right now I started on a very tight deadline iphone app :)
19:54 sri right now i'm especially interested in feedback about timeout defaults
19:55 sri and something most people don't care much about yet... protocol selection
19:55 marcus tempire: Net::Heroku fails to install here.
19:55 echoecho sri: Do I need a code update to mojo for the timeout example to work? Because I set a 1 hour timeout and it still closes after 5 min.
19:56 sri echoecho: you do
19:56 echoecho sri: What's the best/smart way to run a Perl library from Github forkage?
19:56 sri i've actually just secretly uploaded a new cpan release... just don't tell anyone!
19:56 echoecho hah!
19:57 * sri tries to hide his mistakes
19:57 * marcus pushes sri into the alligator pit.
19:57 marcus with a snowboard!
19:57 * sri falls into empty pit
19:57 marcus bottomless, empty pit
19:57 echoecho "Mojolicious is up to date (2.53)" - wat do?
19:57 sri :/
19:57 sri wait for mirrors to update
19:58 sri or download from metacpan
19:58 echoecho kk (sorry, I am teh bad at perl)
19:58 marcus cpanm lets you use an url
19:58 marcus so you could go to metacpan and get the url, then cpanm it
20:00 echoecho On Windows with Strawberry Perl, may be easily simply to wait
20:00 sri if you have git just checkout master
20:01 sri umm... clone master
20:02 echoecho right
20:03 * sri clones marcus
20:04 sri hmm... i did not think that through...
20:04 * marcus forks himself
20:04 echoecho well, I could clone the git repos and run "make install" I assume
20:04 echoecho On Windows, that may result in insanity though
20:04 echoecho I like letting Strawberry Perl handle that
20:04 marcus echoecho: Strawberry perl should be low on insanity.
20:05 echoecho indeed
20:05 marcus echoecho: I think it just installs modules normally using perl Makefile.PL && make && make install
20:05 echoecho well, wherever it has its make tool located... yes
20:06 echoecho I've never busted open the strawberry folder to see how it organizes
20:15 marcus tempire: Weird, Net::Heroku seems to depend on Devel::Dwarn, which I cannot find on the cpans
20:17 GitHub163 joined #mojo
20:17 GitHub163 [mojo] kraih pushed 1 new commit to master: http://git.io/84jRLA
20:17 GitHub163 [mojo/master] more Mojo::IOLoop documentation tweaks - Sebastian Riedel
20:17 GitHub163 left #mojo
20:17 marcus Sri: I see you are not banned for life from Perl Weekly :)
20:18 marcus szabgab++
20:18 sri strange
20:18 sri i see you lost your ipad keyboard ;p
20:19 ccushing I'm confused, I have an around BUILDARGS that modifies an object into a hashref, and it works just fine, but when I try to create the actual BUILDARGS it still see's the original arguments in @_ (e.g the object)
20:19 ccushing oops
20:19 ccushing wrong channel
20:20 ccushing my bad
20:20 sri ccushing: you should answer or close your ticket
20:20 marcus sri: actually just disconnected it to lie on the sofa. Used it to test tempire's module.
20:20 ccushing sri which ticket?
20:21 sri https://github.com/kraih/mojo/issues/294 # i'm assuming that's you
20:21 * marcus ponders starting to use SRI exclusively.
20:21 ccushing yeah that's me, I have gods knows how many tickets open on gods knows how many projects
20:23 ccushing I still don't see how that helps me with the problem of having named parameters in a mock received request object
20:23 sri apparently nobody understands your problem then
20:25 ccushing that's probably true
20:26 ccushing I want to mock a request object as if it had come through routes and a controller
20:26 ccushing e.g. $c->req
20:26 ccushing without actually sending it through one
20:27 ccushing I don't understand why this concept is difficult
20:27 ccushing I have another object which takes a request object as input
20:28 ccushing so I want to isolate testing it
20:28 sri that sounds exactly like what the last comment answers
20:29 ccushing maybe I don't understand the answer
20:31 ccushing I don't see how build_tx is going to result in a request object with named parameters pulled from the uri
20:36 sri it doesn't do that, only the basic grunt work
20:36 sri there is no complete solution for what you want
20:36 sri and i don't think there should be in core
20:45 sri the closest you can get to an example is prolly https://github.com/kraih/mojo/blo​b/master/t/mojolicious/dispatch.t
20:46 sri or even https://github.com/kraih/mojo/bl​ob/master/t/mojolicious/routes.t
20:48 lukep joined #mojo
21:07 kaare_ joined #mojo
21:31 GitHub39 joined #mojo
21:31 GitHub39 [mojo] kraih pushed 1 new commit to master: http://git.io/1lF6Ag
21:31 GitHub39 [mojo/master] simplify Makefile.PL - Sebastian Riedel
21:31 GitHub39 left #mojo
21:33 fibo joined #mojo
21:33 sri no point keeping that list around when a perl 6 port is still so far away
21:33 * sri wonders if it makes the installation a little faster
21:35 batman joined #mojo
21:43 vervain tempire: You've infected my computer with ruby ;-)
21:58 vervain tempire: ping
22:01 vervain installation of Mojolicious-Command-deploy-heroku-0.01 bombs... looks like you need to require Devel::Dwarn.  I've put an issue on github.
22:04 vervain is './script/myapp deploy heroku --create' supposed to do anything more than provide a usage: message?
22:06 vervain my one and only git alias: gleen='git clean -fd; git reset --hard HEAD' :-)
22:19 marty joined #mojo
22:20 Alias joined #mojo
22:21 tempire forgot to remove Devel::Dwarn
22:21 tempire vervain: yes, it's supposed to give you a lot of output
22:21 tempire can you screenshot or screenr the command and output?
22:22 tempire marcus: devel::dwarn is a part of Data-Dumper-Concise
22:23 * tempire pushes to cpan again
22:25 abra_ joined #mojo
22:33 tempire cpanm Data::Dumper::Concise or cpanm https://github.com/tempire/net-herok​u/raw/master/Net-Heroku-0.04.tar.gz
22:42 vervain Just reset perlbrew... trying again.
22:46 vervain tempire: In your 'Workflow'... what does 3. Login mean?
22:47 tempire click the quickstart link
22:47 tempire it lays out the 3 heroku steps
22:49 vervain I suggest that you should put 'heroku login' next to your 3. Login step.
22:49 tempire I'm probably going to remove those 3 steps from the readme complete.
22:49 tempire *completely
22:50 tempire well
22:50 tempire maybe not
22:51 vervain This is what I get if I try deploy --create prior to heroku login. http://pastie.org/3475674
22:52 vervain This is what I get if I try deploy --create after heroku login: http://pastie.org/3475677
22:54 tempire vervain: paste the output of: MOJO_USERAGENT_DEBUG=1 script/my_app deploy heroku --create
22:56 sri crab: heh, just noticed 2.53 was 1mb while all normal releases are 500kb :) http://api.metacpan.org/so​urce/AMS/Mojolicious-2.53/
22:57 sri YO DAWG, I HEAR YOU LIKE MOJOLICIOUS... :D
22:59 tempire Time for an addition to .gitignore
22:59 tempire Mojolicious-?.??
22:59 tempire Oh I guess that wouldn't do anything for make dist
23:00 * tempire neverminds
23:00 sri you mean MANIFEST.SKIP
23:02 vervain MOJO_USERAGENT_DEBUG=1 script/my_app deploy heroku --create : http://pastie.org/3475735
23:03 mire joined #mojo
23:03 tempire !?
23:03 alnewkirk joined #mojo
23:04 tempire ls ~/.heroku/
23:07 sri http://paulasmuth.com/blog/gith​ub_please_fix_the_watch_button/ # at least i'm not the only one noticing it
23:08 vervain tempire: I think I was logged out again... after confirming my heroku login: http://pastie.org/3475777
23:09 tempire output from ls ~/.heroku ?
23:09 vervain credentials
23:09 purl credentials are irrelevant in an argument in which you can absolutely verify the argument.
23:10 tempire mind pasting your credentials file?  or pm me.
23:11 vervain I do mind yes ;-)
23:12 tempire first line email, second line api key?
23:12 vervain Looks that way
23:13 vervain various heroku commands seem to be working.
23:14 tempire try it with -a <api-key>
23:14 vervain Try what?
23:14 purl just a little bit harder !
23:15 tempire script/my_app deploy heroku --create -a <api-key>
23:16 vervain Same
23:16 sri tempire: btw. still no problems at all with filevault ;)
23:16 tempire strange.
23:16 purl But true.
23:17 tempire vervain: does "heroku keys" list any ssh keys?
23:17 vervain Yep
23:19 vervain During heroku login it asked me if I wanted to use one of my public keys... so I chose an appropriate key... it's now listed in 'heroku keys'
23:35 tempire ok, try this new build: https://github.com/tempire/mojolicious-​command-deploy-heroku/raw/master/Mojoli​cious-Command-deploy-heroku-0.01.tar.gz
23:39 GitHub109 joined #mojo
23:39 GitHub109 [mojo] kraih pushed 1 new commit to master: http://git.io/pyiUsA
23:39 GitHub109 [mojo/master] tweaked a few cookbook recipes - Sebastian Riedel
23:39 GitHub109 left #mojo
23:55 tempire how do I stop a IOLoop->recurring?
23:55 sri drop
23:56 tempire events are neat
23:57 sri i can confirm this
23:58 tempire and shnikies, I can't believe I've never used eventsource before.

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