Camelia, the Perl 6 bug

IRC log for #mojo, 2012-01-12

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

All times shown according to UTC.

Time Nick Message
00:03 ashleydev perl6 pod has been implemented...
00:11 abra_ joined #mojo
00:55 abra__ joined #mojo
01:29 tempire I know I've asked this before, but I can't find the answer…sri: what's the main thing that's preventing a perl6 port…blocking something?
01:31 tempire oh right
01:31 tempire non-blocking io
01:31 purl it has been said that non-blocking io is nicer than multiple fork()s and crap. :)
01:37 gabriel joined #mojo
01:42 abra joined #mojo
01:50 xaka tempire: what's wrong with non-blocking io in perl6?
01:52 tempire I don't know about the language spec, but rakudo apparently does not support it
01:52 tempire currently
02:04 jnap joined #mojo
02:11 abra joined #mojo
02:21 vel joined #mojo
02:35 hshong joined #mojo
03:05 abra joined #mojo
03:08 abra_ joined #mojo
03:11 xaka joined #mojo
03:38 Psyche^ joined #mojo
04:12 mercutioviz joined #mojo
06:06 batman joined #mojo
06:07 batman joined #mojo
06:13 koban joined #mojo
06:41 sri ashleydev: but ther is no documentation to display
06:42 sri moin btw.
06:43 sri tempire: apparently non-blocking io is implemented in perl 6 now, but i'm not too keen on reading the source to figure out every single feature i need
06:43 tempire now the tweet makes sense
06:43 * tempire bewares a frustrated kraih
06:55 batman joined #mojo
06:56 ashleydev joined #mojo
07:14 baton8 joined #mojo
07:15 abra joined #mojo
07:18 spleenjack joined #mojo
07:21 Foxcool joined #mojo
07:56 Vandal joined #mojo
08:06 sugar joined #mojo
08:13 berov joined #mojo
09:19 GitHub45 joined #mojo
09:19 GitHub45 [mojo] kraih pushed 1 new commit to master: http://git.io/M8_fAg
09:19 GitHub45 [mojo/master] better delay examples - Sebastian Riedel
09:19 GitHub45 left #mojo
09:19 sri marcus: that might help with api design in the future
09:20 sri the only portable way to synchronize non-blocking stuff portably
09:22 sri umm
09:22 sri s/portable//
09:23 marcus salsa shark!
09:23 purl I think we're gonna need a bigger boat
09:25 * sri is getting hungry
09:25 cosmincx joined #mojo
09:28 ki0 joined #mojo
09:50 marcus sri: dino steak!
09:52 GitHub71 joined #mojo
09:52 GitHub71 [mojo] kraih pushed 1 new commit to master: http://git.io/U7s91w
09:52 GitHub71 [mojo/master] added delay recipe to cookbook - Sebastian Riedel
09:52 GitHub71 left #mojo
09:52 sri marcus: and that should sum it up for you
09:53 marcus sri: yeah, I already did a big project with judofyr's parallol, which basically does this
09:53 sri good :)
09:54 marcus I'm not sure I like how the callbacks tend to end up cluttering your controllers tho
09:54 abra joined #mojo
09:54 sri well, that's the price for going non-blocking
09:54 marcus I want to have my cake and eat it too!
09:54 sri give me coroutines in core and we can make it more pretty
09:55 marcus coreroutines
09:55 sri !
09:56 sri marcus: you could also flatten your callback chains if they get too deep
09:57 sri think we talked about that like half a year ago :)
09:57 marcus sri: well, we usually don't have more than 1-2 levels, parallol helps a lot with that.
09:58 sri like https://github.com/creationix/step
09:58 sri but yea, abstraction sure helps too
09:59 sri hmm, delay has so much documentation now, we should really consider making it stable
10:00 sri i really can't do it unless some tells me to though ;p
10:01 marcus sri: make it stable, please.
10:02 marcus goodbye to daddy!
10:03 foomator joined #mojo
10:04 marcus joined #mojo
10:04 spleenjack joined #mojo
10:04 perlite joined #mojo
10:05 abra_ joined #mojo
10:07 tempire joined #mojo
10:08 perlite joined #mojo
10:08 pl0p joined #mojo
10:09 d4rkie joined #mojo
10:09 jwang joined #mojo
10:25 junkpiano joined #mojo
10:29 charsbar_ joined #mojo
10:37 tty234 joined #mojo
11:01 xaka joined #mojo
11:26 Alias joined #mojo
12:29 ki0 joined #mojo
12:29 koban hi ppl
12:29 marcus hi koban
12:29 sri ho ho ho
12:29 purl and a bottle of lube!
12:29 marcus ohoh
12:30 koban how can i disable httponly flag for cookies?
12:31 sri it's not enabled by default
12:34 koban tnx )
12:36 koban sri: https://github.com/kraih/mojo/blob/ma​ster/lib/Mojolicious/Sessions.pm#L70
12:37 sri why would you want to disable it for the session cookie?
12:37 koban because flash seem not to set session cookie if htttonly flag set
12:38 koban /httt/http/
12:41 marcus koban: I thought flash just had broken session handling, period.
12:41 marcus at least it doesn't work for flash uploader.
12:42 sri koban: do you have any links on that subject?
12:43 sri i can't find anything with google
12:45 marcus "The Flash Player has a few issues with cookies (cookies aren't sent at all in many browsers). "
12:51 koban i can't find any proper link by now, i just try our flash uploader - without httponly it work
12:52 marcus koban: I think you will find it's browser dependent, and that you should make some fallback mechanism to handle it
12:52 marcus like - http://swfupload.org/forum/generaldiscussion/158
12:52 koban how about have it enabled by default, but give ability to disable it when needed?
12:53 marcus http://stackoverflow.com/questions/6785736/​should-a-session-cookie-always-be-httponly
12:56 GitHub100 joined #mojo
12:56 GitHub100 [mojo] kraih pushed 1 new commit to master: http://git.io/BbDVpw
12:56 GitHub100 [mojo/master] removed experimental status from Mojo::IOLoop::Delay and removed defer method from Mojo::IOLoop - Sebastian Riedel
12:56 GitHub100 left #mojo
12:57 koban marcus, thx for the link
12:58 marcus koban: fwiw, that is how I fixed it when I was using swfuploader with catalyst.
13:06 cosmincx joined #mojo
13:19 d4rkie joined #mojo
13:20 metaperl joined #mojo
13:26 D4RK-PH0ENiX joined #mojo
13:31 Vandal joined #mojo
13:35 jnap joined #mojo
13:38 diegok Hi, is there something in core like http://search.cpan.org/~jhthorsen/Mojolicious-Plug​in-UrlWith-0.04/lib/Mojolicious/Plugin/UrlWith.pm ?
13:40 diegok ^ if not, shouldn't be?
13:40 marcus we wrote our own, because batman's didn't quite work for us - https://gist.github.com/1600581
13:42 metaperl joined #mojo
13:59 sri diegok: semantics seem pretty messy
14:00 sri and it doesn't even use core functionality
14:00 sri url_for->path('append/this')->query([merge => 'this']);
14:01 diegok sri: I don't say we should slurp that code, just something solving a common case of create a url from current adding/changing query params. Mostly pagers I think :)
14:02 diegok oh!, query merge looks it is :)
14:02 sri what i wrote above works already
14:02 sri query() can replace, merge and append
14:02 sri http://mojolicio.us/perldoc/Mojo/URL#query
14:03 marcus url_for won't bring along existing query params tho?
14:05 diegok marcus: oh!, that's invalidate the merge solution :(
14:05 sri that's true
14:05 sri $self->req->url->clone->path(​'append/this')->query([merge => 'this']);
14:06 marcus starting to get a little ugly ;)
14:06 diegok so, there is a plugin, marcus has his own helper... I would use plugin or copy the marcus one... are those evidence of the need somethin like this in core? :)
14:06 sri well, url_with semantics are ugly too
14:06 sri diegok: make it not ugly
14:07 diegok cruel!
14:07 diegok :)
14:07 marcus we use this all over the place, like [% h.url_with('artifact.gallery') %]
14:08 marcus or [%h.url_with(from_year => undef, to_year => undef)%]
14:09 marcus HURL WITH!
14:09 netfeed looks like rails :)
14:09 marcus netfeed: TT
14:09 sri that's not even in line with ->query semantics :S
14:11 amoore joined #mojo
14:11 sri i'm not against the idea, but i want something pretty that fits in
14:12 diegok marcus: so, you use url_with to go to different path with same query params ?
14:12 marcus diegok: yeah, for both use cases, to change query params on this url, and to change the path and keep the query params
14:13 sri maybe open an issue for it so you can collect a list of use cases and clean up semantics?
14:13 marcus and even to change the path and modify the query param.
14:13 diegok marcus: yeah, I'm seeing that :)
14:13 diegok ^ crazy :p
14:13 diegok but looks compact :)
14:13 marcus it's very convenient.
14:14 diegok sri: I'll open it.
14:18 sri wouldn't something like <%= current_url->path('append/this')->query([merge => 'this']) %> work?
14:19 sri or even url->
14:19 diegok sri: I think the merge thing should be implicit on the helper.
14:22 diegok https://github.com/kraih/mojo/issues/262
14:23 diegok I somehow like the marcus one. Will start using it to test drive it :)
14:25 diegok But not sure you should trust my opinion. I have proposed/supported the begin "keyword" and now I hate it :(
14:26 metaperl joined #mojo
14:29 sri diegok: why do you hate it?
14:30 diegok sri: because a lot of times it's larger than the text it enclose
14:30 diegok like ...begin %>Home<% end %>
14:30 sri well, it's not like there are many other options for keywords that make sense
14:31 marcus { }
14:31 * sri pokes out his eyes
14:31 diegok yeh, it make sense, I know, but too verbose.
14:32 diegok I didn't liked the { } at that time... but now I also dislike the begin keyword :(
14:32 * diegok dring another mate
14:33 sri be glad it's not …sub { %>Home<% } %> ;p
14:33 diegok :D
14:33 diegok yeah, the worst part of the { } is the closing tag.
14:34 diegok I like <% end %>
14:34 diegok but not begin :)
14:34 sri you want …do %>Home<% end %> ? :)
14:34 sri damn you perl for using it yourself!
14:35 diegok I want nothing on the opening probably :)
14:35 diegok but understand it's not possible
14:37 sri yea
14:37 netfeed is there a way to make a json object out of a param other then inclusing Mojo::JSON?
14:37 marcus %{ Home ]%
14:37 marcus }% even
14:38 sri marcus: actually, for helpers with small content i'd just make it a normal argument
14:38 sri like link_to can do
14:39 sri or, you could actually turn ...begin %>Home<% end %> into …sub { 'Home' }
14:40 sri if the helper didn't support plain arguments
14:46 sri the problem of url_with is that once you add url_for('named') support it becomes chaos
14:46 diegok https://github.com/kraih/mojo/i​ssues/262#issuecomment-3464163
14:47 sri diegok: yea, total chaos
14:47 diegok sri: yes, named vs path is a bit messy
14:47 sri diegok: you're forgetting a lot there too
14:48 sri like ->url_for('named', foo => 'bar')
14:48 diegok yes :(
14:48 sri placeholder values that get merged back into the route
14:48 diegok yes, I'm following you
14:50 diegok first argument would probably be an arrayref to mean named, so ^ would be (['named', foo => 'bar'])
14:50 diegok so, there are only two optional args. First tell about path, second about query params to merge.
14:51 sri how does that look in template engines like TT?
14:52 sri and you only want to support param merging?
14:52 sri no append or replace?
14:54 marcus tt automatically turns ('foo',bar=>'baz') into ('foo,{bar=>'baz')
14:55 marcus ditto for (bar=>'baz',foo')
14:55 diegok sri: merge won't append/replace?
14:56 sri diegok: those are all very different things
14:56 * diegok going back to docs
14:56 sri i think the examples cover it
14:57 sri marcus: what about the array hash mix?
14:57 sri $self->url_with([ 'named', foo => 'bar' ], { merge => 'this' } );
14:57 sri that was the last proposal
14:58 marcus sri: I *think* that should work untouched
14:59 sri well, i'm gonna leave the discussion to you for a bit now, will look again later
14:59 diegok Understand it is possible to append same param many times... I was only covering the case where something like hash merge happens on the query args...
15:03 sri for core we have to consider all use cases
15:04 sri a list of examples for all possible use cases would be great, like the examples in Mojo::URL->query, then we have the test cases right away too
15:05 GabrielVieira joined #mojo
15:09 diegok sri: so, append and replace should be in?.. I think that's not the work of url_with, well, not with replace... can be with append...
15:10 sri diegok: those are questions that need to be asked and answered in the discussion ;)
15:10 sri we need more opinions about that
15:10 diegok sri: sure, I'm just asking and giving my point of view ;)
15:10 diegok sure!
15:10 mattastrophe joined #mojo
15:12 GabrielVieira Hi there, I'm trying to deploy Mojolicious::Lite with Apache... first I tried Apache2  + Mojolicious::Lite + fcgi, but with no success... now I'm trying with hypnotoad, but it always shows the '/' route result... the code and httod.conf http://pastebin.com/z0zxZ5vJ
15:17 GabrielVieira I just found out that I'm in the need of doing http://localhost/mojo/[anything_here]/login/ to get the /login route
15:18 metaperl joined #mojo
15:20 gshank joined #mojo
15:21 noganex_ joined #mojo
15:24 sri diegok: my main problem atm is that query semantics are different from Mojo::URL->query, where merge is [...]
15:26 diegok sri: https://github.com/kraih/mojo/i​ssues/262#issuecomment-3464800
15:26 sri yes
15:26 sri you have { page => 3 } instead of [ page => 3 ]
15:27 diegok exactly, that is what I've commented at the end of that comment...
15:27 diegok I think we need more eyes/brains on this :)
15:27 sri oh
15:27 sri yea
15:28 diegok but it's a start :)
15:28 GabrielVieira I got it working! just removed the end slash from the proxy url
15:28 * sri pokes marcus, tempire and crab
15:28 * sri cracks the whip
15:29 sri GabrielVieira: configuring apache is an art :)
15:29 diegok counting elements to see it is is even/odd when there is only one argument seems a bit weird...
15:29 sri diegok: we do the counting trick all over the place, that's no big deal
15:29 diegok and it's the most common usage...
15:30 sri i'm only concerned with complexity of variations
15:30 diegok sri: well, it should be complex to cover all this is covering :p
15:30 sri i don't want it to look like its own little programming language ;p
15:31 diegok well, it is much more of what I started asking for :)
15:31 GabrielVieira sri: :) I followed the wiki's example in github, now tried the example in the cookbook and it's perfect
15:31 GabrielVieira thanks
15:32 diegok I was expecting only the query args merging, no path change or anything like that till marcus example :)
15:32 marcus Blame judofyr
15:34 diegok my only concern about this last proposal I've made is the most common use case using []... but it can be fixed using nothing, right?. We only need to asume _with will never replace.
15:34 diegok marcus: those ruby boys.... :)
15:35 GabrielVieira sri: is hypnotoad a good solution for ecommerce websites?
15:35 GabrielVieira I mean.. it can handle?
15:35 GabrielVieira or should I use apache + fcgi or apache + plack + proxy?
15:35 marcus GabrielVieira: hypnotoad is most excellent.
15:36 marcus and it can handle...
15:36 GabrielVieira nice
15:36 GabrielVieira thanks!
15:36 GabrielVieira =)
15:39 judofyr joined #mojo
15:39 judofyr hey #mojo
15:39 judofyr have you missed me?
15:44 marcus judofyr: yes
15:44 judofyr yay
15:45 diegok judofyr: hi! :)
15:49 metaperl|2 joined #mojo
15:54 mercutioviz joined #mojo
15:57 inokenty joined #mojo
15:59 crab i am converting to Mojolicious(::Lite) a simple cgi program that reads albums.txt and photographs.txt and responds to requests for e.g. .../, .../album, or .../album/photo
16:00 crab i can easily read the two index files into memory on startup and use that data. but is there a more graceful way to handle it when those files change, without just forcing the issue by restarting hypnotoad?
16:00 crab (and, it goes without saying, also without reading the files on every request)
16:07 alnewkirk joined #mojo
16:07 sri mtime check in a timer?
16:10 sri no shared memory then though
16:17 marcus osx has a daemon for this afaik
16:31 cosmincx joined #mojo
16:36 grim_fandango joined #mojo
16:40 sri diegok: what does url_with('named', {foo => 'bar'}) exactly do again?
16:41 sri i mean, what's the point of having it support routes at all?
16:41 diegok nope, that just add 'named' to the path
16:42 sri then url_with(['named'])
16:42 sri why support that?
16:42 diegok right
16:42 sri it seems pointless
16:42 sri considering url_for('named')->query({lalala...})
16:42 diegok sri: well, if we didn't suport named, why do we support path changes?
16:43 sri diegok: i'm asking *you* :D
16:43 sri i have no idea why
16:43 diegok :)
16:43 diegok hm... because of judofyr :p
16:44 sri for all i care it could be url_with($something_query_also_supports)
16:44 diegok sri: that will simplify a lot and is ok with me.
16:44 judofyr I'm innocent!
16:45 sri neat, lets sacrifice him
16:45 diegok judofyr: that's what marcus've said :)
16:45 * marcus is ready to back his claim with git blame
16:46 sri sooo
16:46 judofyr sri: our use-case: we have different urls for different types of search: /search/articles, /search/objects, but they take the same filters/query-params. so the tabs are written like <a href="[% h.url_with('other_thing') %]">
16:46 diegok sri: I wast just supporting all that path/route stuff because I've liked the example given by marcus on the gist :)
16:46 sri url_with(…) eq $self->req->url->clone->query(…)?
16:46 judofyr gah, that didn't make much sense, or what?
16:47 * marcus whistles
16:47 diegok I do like the complexity added by marcus & judofyr
16:48 diegok and I don't like url_with to be abble to full "replace" query
16:48 judofyr but yeah, we have certain links which needs to preserve the query params
16:49 sri judofyr: you may want to join the github discussion
16:49 diegok judofyr: yes, I don't think it's the most common use, but I can see it's useful :)
16:50 diegok sri: actually, I was about to add code to let url_with->(page => 1) to perform as url_with->([page => 1])
16:51 diegok ^ closing the possibility of "replace" with it :)
16:51 judofyr sri: does ->query({foo => undef}) delete the param?
16:52 judofyr I wonders why I write this https://gist.github.com/1600581 with a loop
16:52 sri judofyr: don't think so
16:52 judofyr ah
16:53 judofyr well, being able to remove params is kinda important for our use of url_with
16:53 sri [foo => undef] however might
16:53 judofyr but isn't [] replace?
16:53 judofyr gah
16:53 judofyr I hate that {} and [] behaves differently
16:53 * judofyr reads the doc
16:54 sri i'd like for {} and [] switch places
16:54 sri but it's too late
16:54 judofyr what's the difference between merge and append? merge overwrites while append turns it into an array?
16:55 diegok sri: yea, that is something I had in my mind al along this conversation :)
16:55 diegok judofyr: yes
16:55 judofyr my url_with clearly use merge/overwrite with a hashref :/
16:57 diegok judofyr: I also like that, but as sri had pointed, it must keep the same semantic as Mojo::URL->query to be in core
16:59 sri different proposal
17:00 sri url_with->query([foo => 'bar']); and url_with('named', some => 'placeholder')->query([foo => 'bar']);
17:01 sri arguments are the same as url_for, but it inherits query params
17:02 sri oh
17:02 judofyr url_with doesn't have to cover all cases, but there should at least be a consistent way to solve them all
17:02 sri that would be one
17:02 judofyr yeah, I like that
17:03 judofyr sri: empty url_with is the same as $req->url?
17:03 sri yes
17:03 judofyr or is empty url_for already the same as $req->url?
17:03 sri nope
17:03 judofyr ah
17:04 xaka joined #mojo
17:04 sri empty url_with would have better performance than empty url_for
17:04 judofyr so that's the only difference then. that and inheriting of query params.
17:04 * sri nods
17:04 judofyr seems good to me
17:04 judofyr or, how would you remove params?
17:05 sri url_with->path('append/this')
17:05 sri path also covered
17:05 * judofyr tries ->query([old => undef])
17:05 sri that works, just tried it
17:06 judofyr well, it creates ?old=
17:06 judofyr so it doesn't strictly remove it
17:06 sri well, that's minor imo
17:06 judofyr only empty it
17:06 sri you could consider that a bug and make undef special
17:06 sri '' for empty, undef for delete
17:06 marcus +1 on that suggestion
17:07 marcus let's hope there's no tests? :)
17:07 judofyr yeah, that would be nice
17:07 sri marcus: if we classify it as a bug tests don't matter ;p
17:08 marcus sri: there's no tests for it either, afaict.
17:08 diegok +1 on that too
17:08 judofyr although it's not that important. most of the time you treat undef and '' equally (when they come from <input>)
17:08 sri marcus: what about the url_with proposal?
17:08 marcus sri: I like it
17:09 * sri looks for tests
17:09 judofyr url_with => sub { my $self = shift; @_ ? $self->req->url->clone : $self->req->url_for(@_)->qu​ery($self->req->url->query) }
17:10 judofyr $self->url_for *
17:10 diegok I don't dislike the url_with poposal, but what I don't like is that the most common case of it is too vebose :(
17:10 marcus hmm
17:11 diegok url_with->query([page => 1]) being for me the most common use case.
17:11 sri diegok: slightly verbose or complex enough to be its own programming language ;p
17:12 judofyr it's hard to make it consistent and short though :/
17:12 sri "url with query" sounds good though
17:12 diegok mine is optimized for this simple use case :)
17:12 diegok actually it works for url_with(page => 1)
17:12 sri diegok: but url_for support breaks your proposal
17:13 sri this one covers everything, and benefits from all future additions to Mojo::URL
17:13 diegok sri: yes, that's what I like
17:14 sri still would like to switch {} and [] though :S
17:14 sri but i don't see a deprecation path
17:14 diegok probably with this new proposal url_with is not the best name...
17:14 sri name seems perfect
17:15 * diegok hide for a while and listen...
17:16 sri marcus: doesn't break tests btw
17:17 marcus sri: that's what I said :)
17:17 marcus sri: should I add that feature ? Since it's not controversial.
17:17 marcus and a test..
17:17 purl a test is not there :-/
17:17 marcus parameters.t ?
17:17 sri already on it :S
17:18 sri marcus: i did it with one additional line, can you do it without adding one? :)
17:19 marcus sri: no.
17:19 * marcus goes back to fuckipedia
17:20 tempire neat.  githire tells you your ranking.
17:21 tempire wonder how it's done
17:22 sri tempire: thoughts on url_with?
17:23 tempire reading backlog..
17:23 sri argh, of course i messed up
17:23 * sri should have let marcus do it
17:29 marcus :D
17:29 elb0w You ever just get tasked with something you completely dont agree with
17:29 elb0w and just want to shoot yourself in the face
17:30 sri TEST Y U NO PASS?!
17:30 elb0w and thats really the only thing that describes the feeling properly, "shoot myself in the face"
17:30 marcus elb0w: no, I want to shoot the guy who tasked it.
17:30 marcus usually.
17:30 purl somebody said usually was the keyword
17:30 * sri just wants to shoot
17:30 elb0w sri what do you do that you can work on mojo all day
17:30 elb0w im jealous
17:31 marcus sri is a luxus prostitute.
17:31 elb0w nice
17:31 sri BUSTED!
17:31 elb0w no clue what a luxus is
17:31 marcus luxury
17:31 elb0w oh
17:35 sri consulting is a form of prostitution i suppose
17:37 elb0w ah
17:38 tempire I dig url_with
17:38 tempire seems like the discussion has already been covered
17:39 sri every now and then i do consider getting a real job again though, now that remote working is going mainstream
17:40 tempire I switch back and forth.  I learn a lot about management and politics whenever I get a real job.
17:40 tempire and ways to manage ridiculously large legacy code bases
17:41 sri yea, cleaning up legacy code sucks big time
17:41 tempire if you play the game right, you can often get around it
17:42 tempire just depends on the company, I guess
17:42 tempire I don't mind legacy code as long as there is permission to fix it
17:43 tempire problem is, I spend too much time trying to make everything pretty
17:43 sri kinda would like to try a startup
17:44 tempire you should start one
17:45 sri i'm not the business type
17:48 sri germany is also not such a great country for startups
17:48 purl okay, sri.
17:53 kaare_ joined #mojo
17:59 memowe https://github.com/memowe/mojolicious-p​lugin-relativeurlfor/blob/master/lib/Mo​jolicious/Plugin/RelativeUrlFor.pm#L41
17:59 memowe does that make sense?
18:07 elb0w !points
18:07 elb0w er
18:08 elb0w man irssi sucks at notifications
18:08 elb0w they should have something that tells you when you get pms at least
18:12 marcus pms?
18:12 purl pms is, like, that time of the month when i am cranky. or Pre-Marital Sex. or Please Massage Skrewtape or Ponies, Monkeys, Skrewtape (oh my) or pre-menstrual syndrome
18:16 inokenty joined #mojo
18:17 cosmincx joined #mojo
18:18 tempire colloquy notifies me of nearby pre-marital sex.
18:20 elb0w sex?
18:20 purl Sex is the question, "YES!" is the answer. or http://xrl.us/bgrq5
18:20 elb0w purl who is the fairest of them all?
18:20 purl well, the fairest of them all is Tom, of course!
18:21 elb0w lame
18:25 GitHub50 joined #mojo
18:25 GitHub50 [mojo] kraih pushed 1 new commit to master: http://git.io/IF-G4Q
18:25 GitHub50 [mojo/master] added experimental url_with method to Mojolicious::Controller - Sebastian Riedel
18:25 GitHub50 left #mojo
18:26 sri that should cover everything
18:35 sri marcus/diegok: speak up now if you don't like it
18:35 mike_ joined #mojo
18:36 batman joined #mojo
18:36 sri really wish we could switch query({…}) and query([…])
18:36 batman joined #mojo
18:36 sri but i think i mentioned that already :/
18:37 diegok sri: only a few times :p
18:39 diegok Well, other than my cry about not being optimized for the most common use case, i like :)
18:39 sri diegok: well, you can add a oneline helper now to make it even shorter :)
18:40 diegok sure!, but need to find a new name :)
18:40 sri helper url_for_diego -> sub { shift->url_with->query(@_) }
18:41 diegok yup!, but I'll have some serious issues to pass code review :p
18:43 sri marcus: sorry for stealing your task btw. i'll try to wait a little longer in the future :)
18:47 * sri wonders how githire ranking works
18:49 diegok link_to this_with( page => $pager->next_page ) looks like a better name for my helper :)
18:49 diegok ^ my one line helper thanks to the new experimental feature :D
18:50 netfeed joined #mojo
18:53 mire joined #mojo
18:55 memowe hehehe
18:56 sri but seriously… {…} should be switched with [...]
18:56 sri not sure i mentioned it before
19:07 metaperl|2 joined #mojo
19:16 netfeed was there any better way to to create a json object from a param? nothing similar to $ua->get()->res->json ? Must use Mojo::JSON directly?
19:48 sri hmm
19:48 sri a very ugly deprecation path would be to replace Mojo::URL->query with Mojo::URL->q or so
19:48 sri and later change it back
19:50 sri query -> q (3 months) -> query (3 months)
20:06 metaperl|2 joined #mojo
20:07 netfeed i guess not :/
20:13 sri netfeed: if something doesn't exist make a plugin
20:16 netfeed i'm not accusing or anything, i just wondered if there was an as nice way for a param as it is for UA content, i'm just fine with using Mojo::JSON directly
20:20 sri just responding to the frowny face
20:20 sri in fact...
20:20 purl in fact... are all squash New World?
20:20 sri EVERYBODY RELEASE A PLUGIN, NOW!
20:28 gshank_ joined #mojo
20:42 metaperl joined #mojo
21:08 netfeed what's the best way to do a 404? i've found documentation for naming the template, would it suffice to just do a render(status => 404)?
21:08 netfeed naming the template, but not how to force the calling of it*
21:12 tempire netfeed: http://mojolicious.org/perldoc/Mojo​licious/Controller#render_not_found
21:13 netfeed thank you
21:20 memowe sri: done. #plugin https://github.com/memowe/mojo​licious-plugin-relativeurlfor
21:20 sri memowe++
21:20 memowe Not sure if it makes sense.
22:28 GitHub76 joined #mojo
22:28 GitHub76 [mojo] kraih pushed 1 new commit to master: http://git.io/Mx-erQ
22:28 GitHub76 [mojo/master] fixed url_with to be compatible with url_for - Sebastian Riedel
22:28 GitHub76 left #mojo
22:49 xaka epic fail, dots are disallowed in mongodb for databases/collections/keys :(
23:17 memowe Good night! :)
23:44 metaperl joined #mojo

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