Camelia, the Perl 6 bug

IRC log for #mojo, 2012-03-01

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

All times shown according to UTC.

Time Nick Message
00:15 stephen joined #mojo
00:37 jnap joined #mojo
00:48 mattp joined #mojo
00:56 arpadszasz joined #mojo
01:08 scott joined #mojo
01:48 poundyz joined #mojo
01:49 jnap joined #mojo
02:13 janus joined #mojo
02:29 arpadszasz joined #mojo
05:12 jmmills joined #mojo
06:02 Foxcool joined #mojo
06:27 hshong joined #mojo
06:41 spleenjack joined #mojo
06:44 abra joined #mojo
06:47 marcus good localtime, #mojo
07:08 black joined #mojo
07:08 AmeliePoulain joined #mojo
07:46 crab goodness gracious, #marcus.
07:47 jamesw joined #mojo
08:08 jmmills joined #mojo
08:08 lukep joined #mojo
08:12 batman joined #mojo
09:00 random joined #mojo
09:04 random joined #mojo
09:08 berov joined #mojo
09:12 random joined #mojo
09:14 cosmincx joined #mojo
09:17 random left #mojo
09:21 Alias__ joined #mojo
09:31 heytrav joined #mojo
09:36 marcus crabbie, old chap!
09:37 bo joined #mojo
09:37 bo_ joined #mojo
09:38 bo joined #mojo
10:08 pfig joined #mojo
10:30 judofyr joined #mojo
10:30 judofyr sri: how do you find the SPDY?
10:30 judofyr err
10:30 cosimo joined #mojo
10:30 judofyr the SPDY spec*
10:37 marcus NOBODY EXPECTS THE GOOGLE SPDY
10:40 pfig_ joined #mojo
10:49 judofyr ALL GLORY TO THE GOOGLE SPY
11:17 daxim joined #mojo
11:32 Foxcool joined #mojo
12:01 black joined #mojo
12:05 d4rkie joined #mojo
12:37 black joined #mojo
13:02 GabrielVieira joined #mojo
13:05 sri judofyr: it's ok
13:14 d4rkie joined #mojo
13:20 baton8_ joined #mojo
13:27 cosmincx joined #mojo
13:40 sri judofyr: i just hate the handshake and ssl requirement
13:41 sri those two could easily prevent mainstream adoption
13:43 sri oh, new css4 selector spec too, selector subject didn't change once again
13:47 batman joined #mojo
13:51 kokey joined #mojo
13:51 batman joined #mojo
13:51 kokey yay, found it
13:51 kokey quick question...
13:51 kokey how do I render multiple ep templates?
13:51 kokey just discovered multiple render lines doesn't seem to be the way
13:55 bo joined #mojo
13:55 judofyr kokey: render_partial?
13:59 bo Hello. I am quite new to Mojolicious. I try to find out the clients IP address. I found the info somewhere in the datastructure of the object ($self). I guess I should get it with something like $self->{tx}->{remote_address} but it is not working. Am I trying a completely wrong approach?
13:59 jnap joined #mojo
14:01 bo oh.. got it!
14:01 bo $self->{tx}{remote_address} is working
14:07 inokenty joined #mojo
14:17 judofyr bo: please don't use internal attributes
14:17 judofyr bo: there's a reason it's blessed
14:18 judofyr bo: $self->tx->remote_address should owkr
14:18 judofyr bo: tx is documented here: https://metacpan.org/module​/Mojolicious::Controller#tx
14:18 judofyr and here's the base class to $self->tx: https://metacpan.org/module/Mojo::Transaction
14:20 bo judofyr, thank you. -_-
14:20 judofyr bo: why -_-? :-(
14:21 bo judofyr, I'm just not to used to web development anymore and it's quite some work to get into it again :-)
14:22 judofyr bo: ah, anything you feel is missing from the Mojolicious documentation?
14:23 bo judofyr, since I'm not a native english speaker I find it quite hard to read myself into topics. So the problem is defenitely me, not the docs! After I made some progress it will also be easier for me to understand the documentation you know
14:24 bo judofyr, thanks a lot for encouraging me to stick to the documentation :-)
14:24 judofyr bo: ah, English knowledge is (unfortunately) something we'll have to take for granted in the docs (:
14:26 bo my english is quite fine... but I found that reading technical things in a foreign language adds quite some complexity.
14:26 bo maybe it's just me
14:26 judofyr ah
14:26 judofyr reference documentation is probably even more difficult
14:26 judofyr bo: what about the guides?
14:28 bo I have not read them all. I'm just trying my first steps. Haha. Maaaannn... yes yes I will read them aaaaaaall before asking again :-P
14:28 judofyr bo: hehe. some people might be annoyed when you ask before you read; I really don't mind.
14:31 bo judofyr, do I have to check all ->param('bla') stuff and kind with the defined() function before using? I'm getting errors that make me think I should!
14:32 plu joined #mojo
14:33 bo like if(defined $self->param('bla') and $self->param('bla') eq 'blue') instead of just if($self->param('bla'))
14:33 judofyr bo: that depends on how you want to handle empty params
14:34 judofyr if the URL is "?q=", then $self->param('q') will return an empty string
14:34 judofyr (and therefore it will be false)
14:34 bo ok, so my problem is probably somewhere else
14:35 judofyr what's the error?
14:35 purl the error is probably KABOOM!
14:35 judofyr purl++
14:36 bo judofyr, I mean it doesn't matter where the variable is coming from right? it can also come from a button like "<input name="BLA" type="submit" value="BLA" />"
14:36 bo I just call it the same way... $self->param('BLA') in this case
14:37 judofyr bo: yes. it also doesn't matter if it's a action="post" or action="get"
14:38 bo ok, then I'm at least trying it the rigeht way :-) This is a good moment to take a break and eat some chocolate :-)
14:38 judofyr yum yum
14:38 bo judofyr, thx for the kind help
14:38 judofyr you're welcome
14:43 judofyr :)
14:47 sri MOAR KABOOM
15:08 kaare joined #mojo
15:19 random joined #mojo
15:29 d4rkie joined #mojo
15:36 cosmincx joined #mojo
15:41 abra joined #mojo
15:43 d4rkie joined #mojo
15:45 mattastrophe joined #mojo
16:05 GitHub59 joined #mojo
16:05 GitHub59 [mojo] kraih pushed 1 new commit to master: http://git.io/Ftf5bQ
16:05 GitHub59 [mojo/master] made a few lines with rather dense code more readable - Sebastian Riedel
16:05 GitHub59 left #mojo
16:08 judofyr sri: any plans for SPDY support in Mojo, or are you going to leave it to nginx/apache?
16:10 sri judofyr: https://github.com/kraih/mojo/issues/293
16:10 judofyr NPN support?
16:11 sri yes
16:11 judofyr what's that?
16:11 purl hmmm... that is .nl
16:12 daxim purl, forget that
16:12 purl daxim: I forgot that
16:12 sri next protocol negotiation
16:12 judofyr ah
16:12 sri a TLS extension that allows the next protocol to be negotiated during the TLS handshake
16:13 sri server says "i have http/2.0 and spdy/1.0"... client says "cool story bro, i'll take spdy/1.0"
16:13 sri only recently got into openssl
16:13 judofyr ah
16:14 judofyr I see
16:14 sri it's rather annoying
16:14 sri yet another way to upgrade connections
16:15 judofyr sri: what do you think about server push? automatic extraction from HTML? another method on Controller?
16:15 sri i've not planned that far yet
16:16 sri in the end we might never actually support spdy, but some dumbed down version they adopt as http 2.0
16:16 sri imo the one important feature is multiplexing
16:16 sri i couldn't care less about everything else
16:17 sri header compression is ok, but it's not a game changer like multiplexing
16:19 sri i'd rather pipe large files over a websocket with deflate extension on a multiplexed http 2.0 stream
16:25 * crab decides to depend on /js/jquery.js
16:27 sri judofyr: regarding push, i think we will end up with letting nginx handle it with a new header "X-Push: /foo.png, /bar.js, /baz.css" or so
16:27 * sri sets crab on fire
16:28 crab !
16:29 crab am i not supposed to use it, or what?
16:29 sri nope
16:39 xaka joined #mojo
16:40 sri crab: well, you can depend on it, but you should be aware that it is not covered by the deprecation policy
16:42 crab after you go to sleep, i'll commit a test that checks for it ;-)
16:42 sri it's also undocumented
16:43 sri and you need a vote to make it a feature ;p
16:44 sri but i think it's unrealistic to cover bundled javascript we don't actually maintain with our deprecation policy
16:45 sri what if they make a non-backcompat change, are we suddenly cut off from bugfixes for a few months?
16:45 sri would we have to start backporting javascript fixes?
16:48 sri it will never pass a vote, way too much trouble :)
16:52 crab sri: it's not too much trouble to depend on it (for me), since i can always toss a jquery.js into public/js if i need to.
16:56 sri https://github.com/kraih/mojo/issues/289 # is this discussion over or not?
17:03 crab what is TO_JSON?
17:03 purl TO_JSON is always static, TO_SQL depends on db
17:04 crab my $tj = $ref->can('TO_JSON'); if ($tj) { $obj = $ref->$tj() }?
17:06 sri think so
17:06 * sri hates the idea of encouraging all upper case method names
17:06 crab why does it have to be uppercase?
17:07 sri because to_json is already often in use or so
17:07 crab sri: i don't think anything needs to be done. i have dealt with the same problem, and it was easy enough to have a method that does the work.
17:07 sri i actually have no idea who made it up
17:08 crab in my case, it's called ->display_hash
17:08 sri why not ->to_hash? :)
17:08 crab sri: because there are other ->db_hash etc. methods too
17:08 sri i see
17:09 crab all of which return a hash of some sort based on the object's current values
17:09 sri think i would still use ->to_db_hash and ->to_display_hash
17:10 sri like ->to_psgi_app in mojolicious
17:10 crab makes sense too
17:13 sri ah, JSON.pm actually supports TO_JSON
17:14 sri "The TO_JSON method may safely call die if it wants. If TO_JSON returns other blessed objects, those will be handled in the same way. TO_JSON must take care of not causing an endless recursion cycle (== crash) in this case. The name of TO_JSON was chosen because other methods called by the Perl core (== not by the user of the object) are usually in upper case letters and to avoid collisions with the to_json function or method."
17:14 sri not exactly pretty
17:23 xaka joined #mojo
17:34 batman joined #mojo
18:30 marcus sri: prettier than having to deserialize an object graph yourself.
18:31 * marcus is using arne's old 17". This thing is a beast!
18:32 marcus wonder if they will make them without dvd players, they will be pretty weird.
18:32 sri marcus: ugly is ugly
18:33 marcus sri: I didn't pick perl for it's looks. :p
18:35 * marcus is still in a happy daze about not having to call self. all over the place in his obj-c classes
18:37 tempire ugh. obj-c.
18:37 marcus tempire: with arc and the new compiler features, it gets downright pretty.
18:38 tempire arc?
18:38 purl i think arc is Lisp-like or not as big as it should be. Probably getting consumed by alloc blocks from the L2ARC.
18:38 marcus xcode can even autocomplete stuff like [foo.bar.baz ba...]
18:38 tempire auto ref counting, ok
18:38 marcus yepp
18:38 tempire do you still have to declare accessors twice?
18:39 tempire accessors to visual items, I think it was
18:39 tempire to be fair, it's not strictly obj-c that I didn't like.  it was cocoa
18:39 tempire but these days I'm kind of disgusted by all things not functional.
18:39 tempire chomp makes me feel icky
18:39 marcus yeah, you have to @synthizise them.
18:40 marcus but you don't have to declare the ivars anymore
18:40 tempire well that's good
18:40 marcus tempire: cocoa has gotten a lot more functional as well, with blocks support all over the place.
18:41 * marcus has actually had a flashback to visual basic this week, with dragging gui elements to the code to autogenerate accessors for the storyboard items :)
18:41 marcus and actions
18:44 marcus shit, the perlito guy really keeps a good pace - https://github.com/fglock/Perlito/commits/master
18:44 marcus maybe it'll be usable for actual stuff sometime soon.
18:46 GitHub48 joined #mojo
18:46 GitHub48 [mojo] kraih pushed 1 new commit to master: http://git.io/8qPniA
18:46 GitHub48 [mojo/master] deprecated custom listen values in Mojo::Server::Daemon->listen in favor of URLs, this change also affects Morbo and Hypnotoad - Sebastian Riedel
18:46 GitHub48 left #mojo
18:46 sri this should be a lot more dwim
18:47 marcus sri: btw, I was wondering, isn't the request timeout in place for psgi requests?
18:47 sri wat?
18:47 purl Watt?  Watt?  I can't hear you, there's this buzzing noise where my brain should be.
18:48 marcus I was looking, and it seemed like it's implemented in Server.pm?
18:48 sri all timeouts are server features
18:49 marcus sri: right. so if you deploy a request with psgi, it won't time out ever if it doesn't render? Just block the psgi server?
18:49 marcus unless that server times out the request, ofcourse.
18:49 sri depends entirely on the psgi server
18:49 sri we just do what the psgi spec says, that's it
18:51 marcus I see FCGI::Engine::PSGI just uses Plack::Util::run_app
18:52 marcus so I guess it has no protection against blocking children.
18:56 tempire the cert/key/ca as query params seems weird to me
18:56 sri tempire: how so?
18:56 tempire although the old syntax is weird too, objectively
18:57 sri i've just hacked on the mongodb uri parsing code and notcied the ugly custom hack we used for listen values :)
18:57 sri mongodb uses query params too
18:57 sri the good thing is, query params are extendable
18:58 tempire woah.  perlito is...does this work?
18:58 sri we could in the future just support more values
18:58 marcus tempire: I was able to make a minimal Mojo::Lite app with it the last time I tried.
18:58 sri i'm thinking spdy=1 and the like
18:58 batman is it possible to do something like "% $foo and form_for "foo", begin" ?
18:59 batman looks like begin/end doesn't behave nicely with conditions
18:59 batman which i understand :)
18:59 sri batman: rendering guide has examples that will show you exactly what begin/end compiles to
19:02 batman i know that part.
19:02 batman just wondering if there was a way to accomplish what i wanted...
19:03 batman oh. i think i know now..
19:03 sri tempire: you would go back to the previous listen value?
19:04 tempire no, the new one clearly makes more sense.
19:04 sri k
19:05 sri btw. maybe a release soonish? i'd like to see the new iowatcher tests on cpantesters ;)
19:10 * sri is also still waiting for asset plugins... now thatwe have support for multiple public and templates directories
19:10 * sri remembers marcus promising to package bootstrap 2.0
19:23 GitHub148 joined #mojo
19:23 GitHub148 [mojo] kraih pushed 1 new commit to master: http://git.io/D93Ppw
19:23 GitHub148 [mojo/master] mention that deprecation only affects HTTPS - Sebastian Riedel
19:23 GitHub148 left #mojo
19:32 marcus sri: that is still vaguely somewhere on my todo.
19:32 * sri cracks the whip
19:36 gshank joined #mojo
19:38 gshank_ joined #mojo
19:42 sri CAN HAZ RELEASE?
19:43 * sri shrugs and goes to get some pizza
19:47 marcus joined #mojo
19:57 mire joined #mojo
20:05 marcus joined #mojo
20:13 vervain Is there a way to 'turn off' ep evaluation for a section of a template?  I want to use an underscore template but they use the same tag identifiers... and I"d rather not change ep syntax for my whole ap.
20:28 vervain Seems I can escape all my tags with <%% but it would be nice to do this 'blockwise'.
20:47 lukep joined #mojo
20:59 baton8 joined #mojo
21:04 * sri wonders where the core devs hide
21:07 fhelmbe__ joined #mojo
21:16 GitHub23 joined #mojo
21:16 GitHub23 [mojo] kraih pushed 1 new commit to master: http://git.io/sFNRyA
21:16 GitHub23 [mojo/master] list available parameters for listen URLs - Sebastian Riedel
21:16 GitHub23 left #mojo
21:20 sri oh
21:20 * sri knows what tempire is doing :)
21:20 sri tempire++
21:22 marcus I'm setting up this 17" laptop
21:23 marcus Installing /Users/marcus/perl5/perlbrew/build/perl-5.14.2 failed. See /Users/marcus/perl5/perlbrew/build.log to see why. :(
21:23 tempire failed tests?
21:23 * sri wonders if they'll make a 17" Air
21:25 marcus HiRes.t and posix.t
21:25 marcus sri: I wonder that too.
21:25 * sri doesn't want to think about the price of a 17" Air with retina display
21:25 marcus this non-air is really too heavy, but an air model would be awesome.
21:28 tempire I think all of them will be airs soon
21:30 sri it's a bit ridiculous that no other company has figured out yet how to make quality laptops
21:31 sri even those lenovos are still huge black bricks
21:31 tempire what about the hp that copied the apple commercial?
21:31 * tempire thinks it was hip
21:31 sri haha
21:31 tempire er
21:31 tempire *hp
21:32 sri it has a beats by dr dre logo, it must be hip!
21:32 marcus hp envy
21:32 marcus best name for an mbp ripoff ever.
21:32 sri that commercial was hilarious
21:41 fhelmbe__ joined #mojo
21:45 marcus hmm, still no vico 1.4
21:45 * marcus is losing faith
21:45 marcus http://blog.vicoapp.com/2012/01​/1-3-x-bug-fix-release#comments
21:49 omega vico is the flavour of the week editor?
21:49 omega marcus: has iterm2 + tmux stabelized?
21:50 marcus omega: vico is almost a good vim/textmate blend
21:50 marcus omega: I noticed it has progressed, but I didn't try it yet.
21:50 marcus I probably need to upgrade my tmux
21:53 omega latest iterm2 tmux is from 3rd of Feb
21:54 marcus http://www.youtube.com/watch?v=_sUeGC-8dyk phear
21:55 marcus shit, I forgot how slow a machine can be without ssd.
21:55 marcus this is sa
21:55 marcus sad
21:56 sri skynet \o/
21:56 marcus omega: You have run an unsupported version of tmux. Please install a version that is compatible with this build of iTerm2.
21:58 omega marcus: yeah, they need to match
21:58 marcus omega: how childish
21:59 tempire I have created the most useful vim shortcut in the entire world
22:00 marcus omega: I got the latest from http://code.google.com/p/iterm2/downlo​ads/detail?name=tmux-for-iTerm2-201202​03-precompiled.tar.gz&amp;can=2&amp;q= and it is still saying that.
22:00 tempire https://gist.github.com/1953527
22:02 marcus tempire: it opens perldoc?
22:02 tempire the source of the module your cursor is over
22:02 geck sri: my father in law got a Samsung Series 5 and the build quality seemed pretty high
22:04 marcus omega: figured it out, had to kill the running tmux -C
22:08 omega aha
22:12 marcus omega: seems more robust now
22:15 marcus I take that back :/
22:16 omega ohh?
22:21 marcus omega: fucked up layout on cmd-d, put all my windows at the top of the screen
22:21 marcus (on one occation(
22:22 omega huh for me it opened a new window
22:22 omega and in the "old" it showed the menu
22:23 omega is that normal?
22:33 marcus yeah
22:34 marcus and all the windows in the new window will automatically be in your tmux
22:36 sri so, rails is over?
22:37 d4rkie joined #mojo
22:38 tempire it's no longer releasing records.  but it's making a killing producing for other artists.
22:39 marcus hate the hype, love the hyper.
23:12 d4rkie joined #mojo
23:16 sri https://github.com/kraih/mojo/issues/295 # one thing that needs to be thought through before stabilizing websockets
23:24 crab hi
23:24 purl salut, crab.
23:24 crab sri: release-ish?
23:25 crab hmm. 2.56. we should release 5.12 next.
23:25 sri good morning india
23:28 sri and good night europe :)

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