Camelia, the Perl 6 bug

IRC log for #mojo, 2011-12-27

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

All times shown according to UTC.

Time Nick Message
00:10 mire__ joined #mojo
00:53 alnewkirk joined #mojo
01:20 d4rkie joined #mojo
02:08 mercutioviz joined #mojo
02:13 alnewkirk joined #mojo
02:29 alnewkirk joined #mojo
02:38 ccushing joined #mojo
03:39 ccushing does useragent have a shortcut for doing http auth that's not in the uri?
03:40 ccushing or should I just build the headers manually
03:41 Psyche^ joined #mojo
04:32 ccushing basically experiencing a problem where the username is an email address
04:33 ccushing but user@host.foo.tld:password@apihost.tld:port doesn't parse out right
05:21 tempire ccushing: you can build it using Mojo::URL
05:24 ccushing tempire I ended up doing an on( start => ... and added the headers before the request manually
05:24 ccushing it'd be really nice if there was an easier way to monkey with the $ua headers
05:25 ccushing and even better yet is said headers had a ->basic_auth style call that would encode things for me
05:26 tempire if you just want to mess with headers, that's easy enough.  $ua->get('foo' => {'My-Cool-Header', 'bar'});
05:27 tempire and use Mojo::ByteStream to do the encoding
05:27 ccushing yeah but it seems like a much sloppier way than being able to do $ua->headers->accept or $ua->headers->basic_auth( $username, $password);
05:28 ccushing or some such
05:28 ccushing I ended up using MIME::Base64 what's the advantage of ByteStream?
05:29 kaare joined #mojo
05:33 tempire You get ByteStream for free.  It's already loaded.
05:33 tempire doesn't matter in terms of results, though.
05:34 * ccushing wonders why Mojo has avoided api compatibility with LWP, MIME::Base64, etc
05:34 tempire http://search.cpan.org/dist/Mojolicious/lib/Mo​jolicious/Guides/FAQ.pod#Why_reinvent_wheels?
05:35 ccushing tempire I get that, but that says why you wrote your own... why not provide an api compatibility layer?
05:36 ccushing just sayin'
05:36 tempire what would be the point?
05:36 tempire if you like the other APIs, just use those APIs
05:37 ccushing I spent quite a bit of time refactoring things just to use the slightly different api calls because it makes more sense to use the builtins
05:38 ccushing doesn't matter much to me at this point
05:41 ccushing eh... bytestream seems too complicated for a single encoding
05:48 osprey left #mojo
05:56 xaka i think having basic auth in mojo core is a good idea and it has low cost for supporting
06:12 koban joined #mojo
07:04 sugar joined #mojo
07:13 Vandal joined #mojo
07:20 spleenjack joined #mojo
08:42 tholen42 joined #mojo
09:09 mire__ joined #mojo
09:27 osprey joined #mojo
10:43 judofyr joined #mojo
10:43 judofyr why doesn't $dom->at('h1')->parent work? :/
10:44 judofyr is it intentional, or am I doing something wrong?
10:46 Foxcool joined #mojo
10:51 judofyr hm, doing a "my $d = $tx->res->dom" helped
11:03 kaare joined #mojo
11:06 sri moin
11:23 kaare joined #mojo
11:27 kaare joined #mojo
11:29 kaare joined #mojo
11:30 kaare joined #mojo
11:37 d4rkie joined #mojo
12:02 judofyr joined #mojo
12:26 xoma joined #mojo
12:34 judofyr joined #mojo
12:40 kaare joined #mojo
12:48 d4rkie joined #mojo
12:57 kaare joined #mojo
13:31 jnap joined #mojo
14:05 abra joined #mojo
14:12 kaare joined #mojo
14:14 MojoGuest678 joined #mojo
14:14 MojoGuest678 From: http://networkedblogs.com/33Knp (30 hits)
14:36 asim joined #mojo
14:41 amoore joined #mojo
14:41 abra joined #mojo
14:45 inokenty joined #mojo
15:20 tempire searching for my name on duckduckgo brings up the search suggestion of "tempire" and "mojolicious"
15:20 tempire \o/
15:24 sri \o/
15:31 cstamas lol
16:16 mire__ joined #mojo
16:22 xaka joined #mojo
16:24 batman joined #mojo
16:46 noganex joined #mojo
17:12 dch joined #mojo
17:20 dch hi, I've started using perl again after a hiatus, after seeing mojo - it's awesome!
17:21 dch what's the recommended perl install for a mac?
17:21 sri glad you like it, use perlbrew
17:22 dch sri: thanks! perlbrew === homebrew for perl then.
17:23 sri not quite, it allows you to run many perls next to each other
17:24 sri like rvm and virtualenv
17:43 sri for installing modules also get cpanm
17:43 sri (inside perlbrew)
17:44 kaare joined #mojo
17:56 dch sri: thanks, I already have cpanm, tried using it to install App::perlbrew but even with --force I have a few failures
17:58 sri http://perlbrew.pl/
17:58 sri actually don't remember how i started my os x installation, think i just downloaded the single file script version :)
18:00 dch sri: thanks for the pointers, I need to head off and will see if I can get a clean state to start from first.
18:02 dch sri: first bunch of errors are around trying to install stuff into /usr/share/* -> permissions. I want to keep all in /usr/local so I can bail easily if reqd.
18:02 dch thanks again
18:03 sri right, you shouldn't install perlbrew that way, it can be done without touching anything outside your home directory
18:03 dch more homework reqd first :-)
18:03 sri worth it though :)
18:04 sri <3 perlbrew
18:09 osprey left #mojo
18:10 tempire I like that perlbrew has an arbitrary ninja
18:17 sri just don't mouseover it
18:20 ccushing does Test::Mojo actually start an instance of the daemon?
18:28 mire__ joined #mojo
18:36 tempire ccushing: yes
18:37 tempire that is, if you're talking about ->test_server
18:43 sri tempire: didn't you say you wanted to skip experimental features?
18:44 sri collections are mostly experimental but shown in e5
18:46 baton8 joined #mojo
18:49 sri not much of a choice now but to remove experimental status
18:51 GitHub115 joined #mojo
18:51 GitHub115 [mojo] kraih pushed 1 new commit to master: http://git.io/7KQbTA
18:51 GitHub115 [mojo/master] removed experimental status from Mojo::Collection - Sebastian Riedel
18:51 GitHub115 left #mojo
18:58 ccushing tempire well I'm just curious if it starts up an actual instance without that... I think psgi stuff it's possible to run tests without actually starting a daemon
18:58 ccushing bbl
19:02 sri hmm, guess morbo and the mount plugin could become stable too
19:02 sri they've proven themselves by now
19:13 GitHub45 joined #mojo
19:13 GitHub45 [mojo] kraih pushed 1 new commit to master: http://git.io/U8NBLg
19:13 GitHub45 [mojo/master] removed experimental status from a few more classes and methods - Sebastian Riedel
19:13 GitHub45 left #mojo
19:13 sri gonna keep morbo experimental for now, since the class api is still unclear
19:14 tempire hmm
19:14 sri it's the same problem we have with the test daemon
19:14 tempire my only defense is that you didn't notice until now :)
19:14 tempire either
19:14 sri there's no good way to expose all daemon settings without a lot of repetition
19:15 sri tempire: :)
19:16 tempire we really need an experimental feature detector
19:19 sri would be a bit silly if they printed warnings like deprecated features
19:19 tempire I'm thinking of a separate tool.
19:19 sri on the other hand it would force people to react
19:20 tempire could be an optional switch.  it's kind of hard to know whether you're using an experimental feature at the moment; easy to miss it.
19:20 tempire I didn't even think of checking to see whether collections were experimental
19:21 tempire it's not a problem for me, 'cause I would consider it my own fault if they were removed
19:21 tempire obviously some other folks don't feel the same way
19:21 sri how do you warn about a class that is experimental if it is loaded always anyway?
19:22 tempire I was thinking a scanning tool for an app
19:22 sri some experimental methods are also used for internals… some *a lot*
19:22 tempire like test_server
19:22 sri don't think that would be easy
19:23 tempire maybe there should be a different status than experimental, then
19:23 tempire something additional
19:23 purl something additional is at play that I don't see
19:23 tempire experimental: we're not sure if this is a good idea
19:23 tempire another term: this is cool, but it's in flux as to exactly how it will work
19:24 sri people already have a hard time understanding what experimental means
19:24 sri and it is really simple
19:24 tempire it could be because it's really several concepts merged
19:25 tempire actually, you're right
19:25 tempire they just want  to use whatever they see and damn the consequences
19:25 * sri nods
19:26 tempire this is why a tool is a good idea.  it gives people transparency as to what's going on, at any time
19:26 sri "just don't use anything marked as experimental, if you're interested in that functionality give us feedback"
19:26 tempire myapp report_experimental
19:27 * tempire looks for the best pod parser/searcher in core
19:27 sri you want something from cpan
19:27 sri seriously
19:27 purl is that for real?
19:28 sri your next problem is differentiating between mojo core and other modules
19:28 sri then there are experimental method arguments, hooks and events
19:28 tempire does it matter?  search Mojolicious:: and Mojo:: for an index of experimental methods/classes
19:29 tempire there are experimental method arguments?
19:29 sri yes there are
19:29 tempire hum
19:29 tempire example?
19:29 sri json pointer, dom text trimming
19:30 tempire yeah, that's a potentially huge project
19:31 sri events and hooks aren't any easier
19:32 sri imo the only way to make it work is core modifications that come with a hefty maintenance and performance cost
19:32 tempire I'm assuming PPI is smart enough to determine arguments.  once you have a list of events, you can just check every odd argument to ->on
19:33 tempire don't know enough about perl parsing to say anything more at the moment
19:34 tempire I could always retcon the collections part of e5
19:35 tempire I like the collections, though.
19:37 tempire dernit.  why does google always want to include the html video chapters text file in their subtitles.
19:44 sri unless someone has concerns collections are now stable
19:54 tempire mojolicious logo: myriad web pro, or myriad pro?
19:55 sri myriad pro
20:30 mire__ joined #mojo
20:35 sri https://github.com/kraih/mojo/pull/257 # more comments please
20:37 judofyr joined #mojo
20:55 tempire http://www.codinghorror.com/blog/2008/0​8/protecting-your-cookies-httponly.html
20:56 sri tempire: my last comment sensible?
20:56 sri (added a pointer at the irc channel)
20:56 tempire it was sensible, though I don't know that it was necessary.
20:57 tempire hopefully he's not too sensitive; don't want to shoot down those willing to put the effort forth, only point them in the right direction.
20:57 mercutioviz joined #mojo
20:57 sri same
20:58 sri but those commits really suffer from lack of discussion
20:58 tempire true enough
20:58 tempire "If you like, you can drop by #mojo to discuss the issue in real time, we're open to new opinions"
20:59 tempire that's what I would have said, and let the rest be implicit
20:59 tempire I don't know, people were harsh with me when I first started in IRC, and I got the message and adjusted
20:59 sri ok, updated
20:59 tempire not everyone reacts that way; it's hard to say what the best approach is
21:00 sri well, you need a thick skin to survive in the open source jungle
21:00 tempire rawr!
21:00 sri RAWR!
21:00 * tempire hides
21:01 tempire on a side note, I *really* hate smiley faces
21:01 tempire and little emoticons
21:01 sri :D
21:01 tempire but it seems they're necessary
21:01 bvierra joined #mojo
21:01 judofyr joined #mojo
21:01 tempire now they've grown on me
21:01 bvierra Hey all
21:01 tempire I even like lolcats
21:01 * sri waves
21:01 * tempire wonders where he went wrong
21:02 sri tempire: btw. http://www.igvita.com/2011/12/1​9/dont-push-your-pull-requests/
21:02 bvierra so I was a Catalyst user (well developed apps using Catalyst) and it seems that the new perl rage is Mojolicious, so here I am :)
21:02 sri i liked that
21:03 * sri hands bvierra his hipster badge
21:03 * bvierra tries to eat it like a cookie
21:03 sri it's a chocolate badge
21:03 * tempire is the new rage: http://www.youtube.com/watch?v=7mZZd4gQrrg
21:03 * bvierra thought it tasted pretty good for a badge
21:04 bvierra have a new project, basically a web caching server that collects stats with a pretty front end
21:04 bvierra have it working fine now, just no framework... basically a POC to make sure it would work as I needed it
21:05 bvierra but figured I would jump in here as I started getting it all going, somewhere to ask the experts when I screw up ;)
21:05 tempire good article
21:05 sri tempire: didn't we discuss and accept httponly for sessions recently?
21:06 tempire wasn't me
21:06 tempire don't remember seeing that conversation
21:06 * sri only vaguely remembers
21:06 sri could have been with crab
21:07 sri http://irclog.perlgeek.de/​mojo/2011-10-01#i_4507446
21:07 sri looks like we forgot
21:13 stephan48 httponly for sessions?
21:13 stephan48 wont that break ajax?
21:14 * stephan48 hides again in the pit
21:15 stephan48 oh ok nvm
21:15 stephan48 i annoyed you with it already!
21:16 sri tempire: has group {} proven itself yet?
21:16 tempire I was pretty convinced when it was introduced
21:17 tempire I haven't heard anything about it, but before it was introduced, questions about under used to happen all the time.
21:17 tempire since then, crickets
21:19 sri indeed
21:20 GitHub126 joined #mojo
21:20 GitHub126 [mojo] kraih pushed 1 new commit to master: http://git.io/1ZvxYA
21:20 GitHub126 [mojo/master] improved session cookies to always enable the HttpOnly flag and removed experimental status from group feature of Mojolicious::Lite - Sebastian Riedel
21:20 GitHub126 left #mojo
21:25 sri https://github.com/kraih/mojo/pull/257 # also updated
21:36 perlite_ joined #mojo
21:46 bvierra ok really random question... the colorscheme for the code sections of mojolicio.us/perldoc/... is what?
21:46 tempire slightly customized version of textmate's twilight
21:48 bvierra got a link for the vim equivelant?
21:49 xaka joined #mojo
21:49 tempire here's something close: https://gist.github.com/1525272
21:49 tempire used in the mojocasts
21:49 bvierra nice :)
21:52 sugar joined #mojo
22:17 judofyr sri: got a good name for a Capybara-clone in Perl?
22:18 sri GuineaPig
22:18 purl i heard GuineaPig was a symlink to hamster
22:18 sri Hamster
22:19 sri Chinchilla
22:19 judofyr hehe
22:19 judofyr Test::Mojo::Hamster
22:19 sri Test::Mojo::GuineaPig
22:20 judofyr sri: also: Mojo::DOM::CSS doesn't like XML-namespaced elements :/
22:21 sri namespaces suck in css
22:21 judofyr sri: we'll see if I can open-source Test::Mojo::GuineaPig. right now it's just living as a half-way implemented version in t/lib.
22:22 sri you have to manually check namespaces for elements
22:23 sri works rather reliable though
22:24 sri i'm sure there are examples in the tests
22:54 xxtjaxx http://paste.debian.net/150339/ << Somethings definitely wrong with the default RewriteCond's
22:58 xxtjaxx #+
22:58 sri https://github.com/kraih/mojo/pull/257 # needs comments again
22:59 sri i'm not sure about sessions that die with the browser window
23:00 sri i do know it has been requested a few times now
23:01 MojoGuest142 joined #mojo
23:01 MojoGuest142 From: http://www.google.it/url?sa=t&amp;rct=j&amp;​q=html5%20irc%20client&amp;source=web&amp;cd​=4&amp;ved=0CDoQFjAD&amp;url=http%3A%2F%2Fde​v.xantus.org%2F&amp;ei=Hk76TvHyBM2p-gaH6ZWgA​Q&amp;usg=AFQjCNGw6nZv0wO42uzUnHhWcD0vcKNULg (1 hits)
23:03 MojoGuest562 joined #mojo
23:03 MojoGuest562 From: http://networkedblogs.com/33Knp (31 hits)
23:03 alnewkirk joined #mojo
23:03 MojoGuest562 left #mojo
23:04 xxtjaxx sri: It still worries/suprises me that it prependds /app.pl/ to the css file ... even without rewrites.
23:15 jnap joined #mojo
23:23 mire__ joined #mojo
23:27 sri diegok: didn't you want session cookie without expiration?
23:27 sri *+s
23:27 sri tempire/marcus: thoughts?
23:27 purl Moonlight shines through the dark night / clouds move overhead, casting shadows / dancing in the firelight
23:28 tempire I don't care one way or the other.  100000 is the same as 0.  Didn't marcus want 0?
23:28 sri 0 is session dies with browser window
23:28 tempire er, I mean expiration
23:28 tempire not sure why I said 0
23:30 sri personally i'm not to keen on it, but i know it has been requested a few times
23:30 tempire does it hurt?
23:30 sri more rope to shoot yourself in the foot
23:31 sri i'll take a look at the code and see how much change it actually requires
23:31 sri the patch is a little verbose
23:37 Quai hmf, any idea why this would happen? https://gist.github.com/1525470 Im running 2.26 (I know, I know, Im stuck on perl 5.10 atm)
23:40 Quai the code is deployed under Hypnotoad atm
23:40 tempire the latest mojolicious version runs on 5.10 just fine
23:46 Quai well, the Makefile.PL wants 5.10.1
23:46 Quai (I think I edited my way around that last time, so maybe I should complain about strange errors like that ;P)
23:47 tempire that's the minimum version
23:49 Quai I know, and Im stuck on 5.10.0 (lenny...) right now
23:51 tempire the logging seems like more of a coding error than anything else.
23:51 tempire I presume AIS is a separate namespace
23:52 tempire oh, it works once and then disdisappears?
23:52 Quai It might be a coding error, I just find it strange that the namespace will disappear from one request to the next
23:52 Quai yes
23:52 tempire running in morbo?
23:52 Quai no, hypnotoad
23:52 Quai under apache
23:53 tempire lite or full?
23:53 Quai full
23:54 rsimoes1 joined #mojo
23:54 tempire same thing happen in direct hypnotoad, without apache?
23:55 sri hmm, 0 sessions can be done with 2 additional lines of code
23:56 tempire sri: http://mojocasts.com/
23:56 tempire thoughts?
23:56 purl Moonlight shines through the dark night / clouds move overhead, casting shadows / dancing in the firelight
23:57 sri tempire: like it
23:57 tempire menus should make learning mojolicious even more transparent
23:58 tempire I also made the url keep the hash thingy

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