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

IRC log for #mojo, 2015-05-11

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

All times shown according to UTC.

Time Nick Message
00:09 tardisx joined #mojo
00:11 tardisx joined #mojo
00:20 pete left #mojo
00:26 dmanto joined #mojo
01:01 sh4|2 joined #mojo
01:06 klapperl_ joined #mojo
01:38 mattastrophe joined #mojo
02:08 gryphon joined #mojo
02:09 noganex joined #mojo
02:38 dmanto joined #mojo
02:42 nf joined #mojo
02:45 dmanto left #mojo
02:54 mattastrophe joined #mojo
03:19 basic6_ joined #mojo
04:04 kaare joined #mojo
04:08 mattastrophe joined #mojo
04:44 absolut_todd :b1
04:55 preaction if i wanted to do an online game of some kind, what kind of database or synchronization thing should i use? is this what Redis is for?
04:59 Grinnz you could use redis for part of it. but there's a lot of relational data in online games
05:00 Grinnz i would go postgres
05:01 preaction i was planning on some X thing for the game state part, and then, yes, a relational database for the stuff that doesn't have to be hot
05:02 preaction it's a lobby-type game. but ensuring the game is synced between clients requires some master copy of the game's state
05:02 preaction can't use the mojo app, because horizontal scaling
05:03 Grinnz yeah, game state in redis is a good idea
05:03 Grinnz and if you wanted to use websockets, you could do pub/sub communication over either :P
05:04 Grinnz you could also just store game state in postgres though, as a jsonb
05:05 preaction right, it's all going to be websockets, but if one client connected to one hypnotoad worker updates a game, the clients connected to the other hypnotoad workers also need to know about it. so in memory isn't good enough
05:05 preaction so some central place that has pub/sub needs to exist
05:06 Grinnz postgres and redis can both do pub/sub
05:06 Grinnz is what i meant
05:06 preaction ahh
05:06 preaction i suppose i could use DBIx::Class and Mojo::Pg simultaneously... or perhaps i could make a DBIx::Class::Async...
05:06 Grinnz i was using pub/sub over redis for a game i was working on
05:07 Grinnz no postgres just because i wasn't planning on keeping a lot of relational data other than the cards, which aren't too difficult to store
05:08 Grinnz but i was thinking about making it able to run on postgres as well
05:10 Vertig0 joined #mojo
05:10 preaction yeah, let's start with Mojo::Pg and see where that goes. thanks, Grinnz
05:43 Vertig0 joined #mojo
05:49 bramirez joined #mojo
05:50 mattastrophe joined #mojo
05:59 dod joined #mojo
05:59 hshong joined #mojo
06:03 dod joined #mojo
06:56 McA joined #mojo
06:58 McA joined #mojo
07:20 trone joined #mojo
07:21 bjakubski joined #mojo
07:24 Renaud joined #mojo
07:47 odc joined #mojo
08:15 Vandal31341 joined #mojo
08:34 Dandre joined #mojo
09:03 misty_g3ar joined #mojo
09:04 coolo joined #mojo
09:56 nf joined #mojo
10:17 bramirez joined #mojo
10:33 sh4 joined #mojo
10:46 bramirez joined #mojo
11:10 OmarOthman joined #mojo
11:19 OmarOthman Hi All
11:19 OmarOthman I want to share a painful experience I had today with Mojolicious, so that you can open a ticket for it and fix it if you see relevant.
11:19 OmarOthman I had this regular route, let's say like this:
11:19 OmarOthman $r->get   (q{/monframe/(:user_id)/dashboards/my/(:item_id)}, { item_id => 0 })
11:19 OmarOthman ->to('monframe#dashboards_my_read');
11:19 OmarOthman So, the controller is class Monframe and the method is dashboards_my_read.
11:19 OmarOthman Someone changed the name to MonFrame instead of Monframe...
11:19 OmarOthman ... and now I kept getting 404!
11:20 OmarOthman And because there were numerous other changes at the same time, and because this application is a part of a much larger system that is also being developed, I had a gazillion things to think about at the same time.
11:20 OmarOthman Now I know that this is because of the camelization thing, which I literally have no idea why it might be even useful!
11:20 OmarOthman But anyway, my point is: Why should Mojolicious tell me that this resource in not found (404) when in reality what happened was finding the resource but not finding its handler?!
11:20 OmarOthman That's the "UX" bug I would like to report...
11:20 OmarOthman Thanks!
11:21 nicomen OmarOthman: could be that you guys are using a 404 page that says resource not found, and it just uses that one?
11:22 OmarOthman No no, that I got using CURL, from Mojolicious directly!
11:23 OmarOthman My point is, this is not a "resource not found" problem! The route actually matched!
11:25 tardisx The log is pretty specific about not finding the controller
11:25 tardisx [Mon May 11 20:55:05 2015] [debug] Controller "TestyWesty::Missing" does not exist.
11:25 OmarOthman Mojolicious: None of these routes could generate a response for your <code>GET</code> request for <code>/monframe/oothman/dashboards/my</code>, maybe you need to add a new one?
11:26 OmarOthman That's what I got from
11:26 tardisx at the bottom it shows the log that shows "... does not exist"
11:26 OmarOthman curl http://127.0.0.1:3000/monframe/oothman/dashboards/my
11:26 OmarOthman That's not what happened with me.
11:27 OmarOthman <title>Page not found (development mode)</title>
11:27 OmarOthman May that be related?
11:27 OmarOthman (I've just grepped the output by the way)
11:28 tardisx the logs should also be there in that output. HTML is not a terribly useful way to look at them unless you're in a browser. You should probably be tailing the log file on disk as part of your normal development
11:29 tardisx unless you like looking at ampersands and angle brackets
11:30 OmarOthman Sorry, where is that by default?
11:30 OmarOthman I usually tail STDERR and it is enough
11:30 nicomen should be in STDERR
11:30 OmarOthman No, nothing appeared there *at all*!
11:31 tardisx if you have a log directory, there will be a development.log file
11:31 OmarOthman I just got a 404 from the other web application that was sending this request!
11:32 OmarOthman find . | grep -i log
11:32 OmarOthman Nothing
11:32 OmarOthman . is the application folder
11:34 OmarOthman Anyway, I have a meeting. Sorry for withdrawing. As I've said, this is a bug report, feel free to take a look at and investigate. For me, I've learned the lesson, but that was really a waste of time; and I would not like someone else to run into it.
11:34 OmarOthman Thanks for your time!
11:56 romel OmarOthman: you should create 'log' directory inside your app's home manually in case you want redirect STDERR in a log file
12:00 romel and i personally think that 404 status code has correct meaning in terms of http in the case you're talking about
12:05 neilhwatson joined #mojo
12:05 bramirez joined #mojo
12:14 niczero joined #mojo
12:17 niczero Anyone been in the position where your routing table got so big it became worthwhile to avoid routing conditions?
12:17 niczero I currently have ~200 routes, expecting that to go above 1000
12:18 dod joined #mojo
12:18 niczero and almost every route requires the user to be authenticated and their role to be known
12:18 niczero On the one hand I'm thinking I might regret not having routing cache
12:19 niczero on the other hand, I'm thinking I could get into a silly inefficient muddle without routing conditions
12:20 niczero Benchmarks would answer it, but writing the alternative version is quite a bit of work
12:20 niczero hmm, I guess that answers the question
12:21 mattastrophe joined #mojo
12:24 ajr_ joined #mojo
12:26 romel niczero: i'm just curious about the case when you need so many routes
12:28 niczero It's a fairly complex app
12:29 niczero it's kind of a 'suite' of apps rolled into one
12:31 marcus maybe you should be using app embedding?
12:33 romel niczero: so are these routes being generated dynamically based on some data set? asking because i'm working on something alike
12:36 asarch joined #mojo
12:37 niczero marcus: I was just thinking the same
12:38 batman Is 1000 routes "many routes"..?
12:38 niczero Abstractly I could do with the power of conditions at the top level (check auth+authz then dispatch to sub-app)
12:38 batman It sounds weird to have that many, but still...
12:38 niczero and have caching within the router of each sub-app
12:39 niczero batman: it sounds like you've just answered your question :)
12:39 batman niczero: I would advice benchmarking it before making it into a problem
12:39 batman (I could be wrong)
12:39 niczero That's a challenge, like I said
12:40 batman niczero: Not really. My first comment was about speed. The second was about design.
12:40 niczero Your first implies it's not many; your second implies the opposite :P
12:41 niczero For me it seems quite a lot
12:41 niczero I guess if it gets to the point where I might fail UAT, I'll have to try one of the alternatives
12:42 batman Rule of relativity.
12:43 niczero batman: With your farm business, are there separate routers for the top level and each sub-app?
12:45 batman I don't get "farm business"
12:46 * niczero searches github for "Batman's farm business"
12:46 batman Something else: one route should point to one resource. So having 1000 routes should imply that you have 1000 different kind of resources.
12:46 batman It sound strange to me to have that many resources.
12:46 dod joined #mojo
12:47 batman Example of a resource could be "profile page". So even if you have 1000 users, you still only have one resource that return different result.
12:47 niczero https://github.com/jhthorsen/toadfarm
12:48 batman I don't think I could cope with that many resources within a single app.
12:48 batman Not saying it's wrong. Just saying it sounds like something went wrong in the design decision...
12:50 niczero it is somewhat ridiculous to judge a design decision without knowing any of the background
12:50 niczero batman: with toadfarm are there separate routers for the top level and the sub-apps?
12:51 batman niczero: Yes. One route per app in Toadfarm. I haven't implemented shared routes.
12:51 niczero that sounds useful
12:51 niczero I'm guessing I could create a pseudo app for the shared bits (user profile, etc)
12:51 batman niczero: That's why I'm not judging. I'm just implying, so you might reconsider.
12:52 batman But I have no idea what you've done, so it's impossible to judge.
12:52 * batman out.
12:52 zivester joined #mojo
12:53 niczero "it sounds like something went wrong in the design decision" is very judgemental for an impossible-to-judge situation
12:54 batman That's why I used the word "sounds". If I wanted to judge I would say "you're design is crazy"
12:54 batman (I didn't mean to judge)
12:54 niczero no no no, YOUR design is crazy
12:54 niczero :)
12:55 batman niczero: Yeah. Autocorrect--
12:55 gryphon joined #mojo
12:55 batman Thanks anyway.
12:55 niczero no, thank you for the info on toadfarm
12:55 niczero it's one of those things I haven't ever gotten around to spinning
12:56 batman Just use it. I've spent quite some time putting it together.
12:56 marcus Crazier than coconuts
12:56 batman Next is probably shared routes (if it's even possible)
12:56 batman marcus?
12:56 purl well, marcus is pretty sure lisp adoption is being hampered by being named after a speech defect. or http://picasaweb.google.com/olga.ramberg/Lisboa#5240395957801479314 or a moose farmer?
12:57 niczero Fairly recently the design changed so that each team gets their own sidebar
12:57 niczero at which point app-of-apps starts looking more attractive
12:57 batman niczero: Please open an issue if Toadfarm lacks functionality or if you spot any typos in the POD. I'm sure there's plenty.
12:58 niczero "I have spent three weeks analysing your software and I can report that you are missing two apostrophes"
12:59 batman niczero: I hope not rack sidebar is a unique resource. If so, I'm very confident the design is messed up. (Again: can't really judge, so sorry about the use of words)
12:59 batman Uhm... "Not rack" = "not each"
13:05 Kripton joined #mojo
13:12 niczero batman: Is there anywhere that says what toadfarm provides?
13:14 niczero ah, I think I was expecting too much of it.  It says it's just for starting and stopping apps
13:17 punter joined #mojo
13:19 Ptolemarch joined #mojo
13:19 bramirez joined #mojo
13:28 nicholas_ joined #mojo
13:33 dvinciguerra joined #mojo
13:50 dmanto joined #mojo
13:51 jb360 joined #mojo
13:54 batman niczero: Yeah. It has several manuals.
13:55 batman It's much more that starting and stopping daemons.
13:55 batman Where did you read that it was the only thing..?
13:58 jabberwok some poor woman in Morocco just died as a result of trying to stop daemons... oh, wait, that was "demons"
14:00 batman niczero: The main purpose of Toadfarm is to route requests to the correct application. Pretty much like a virtual host definition in Apache or Nginx
14:00 Vertig0 joined #mojo
14:09 AirDisa joined #mojo
14:10 dod joined #mojo
14:23 coolo joined #mojo
14:37 mattastrophe joined #mojo
14:43 Ptolemarch joined #mojo
14:52 bramirez joined #mojo
15:04 bramirez joined #mojo
15:09 bramirez joined #mojo
15:10 dmanto hi, couple days ago I submitted a helper to dinamically add steps to the ioloop that I think could be interesting
15:11 dmanto jberger was going to take a look
15:15 dmanto anyway, want to know if there is other ways to do the same, or if it is interesting at all
15:15 dmanto https://github.com/dmanto/mojo-delay-add-steps
15:22 mattastrophe joined #mojo
15:23 ajr_ joined #mojo
15:26 niczero batman: You should probably say that in Toadfarm.pm pod
15:26 niczero at the moment the reader is left pondering why they'd want to use it
15:26 niczero "Toadfarm is a module for configuring and starting your Mojolicious applications"
15:35 bramirez joined #mojo
15:36 jabberwok true enough, but perhaps worth saying:  Applications in the sense of, something you could mount with Mojolicious::Plugin::Mount.
15:55 jberger dmanto: personally I don't think it works for the workflow that most people use
15:55 mattastrophe joined #mojo
15:55 jberger and I also think that you might have some easier time of keeping the steps and re-adding via the "remaining" attribute rather than closing over them
15:57 jberger anyway, all that said, no one is going to stop you from putting something on CPAN, just please use a descriptive name for the plugin
15:58 jberger Mojolicious::Plugin::RestartableDelay or something like that
16:11 GabrielVieira joined #mojo
16:11 batman niczero, jabberwok: i don't see how it doesn't say that in the POD, so please specify where the text should be moved from and to where.
16:11 batman or just to where.
16:24 Ptolemarch joined #mojo
16:27 AirDisa joined #mojo
16:32 mattastrophe joined #mojo
16:33 dmanto jberger:thanks for your comments
16:34 dmanto i'll start taking a look on the "remaining" attribute as you suggested
16:34 dmanto sound like the right way to do it
16:35 jberger you are on an interesting track, I just think your current version has a bunch of extra architecture
16:37 dmanto jberger ah i have another question to you
16:37 dmanto about the "reactors" to use
16:38 dmanto i had the worst performance with MOJO_REACTOR="Mojo::Reactor::EV"
16:38 dmanto is that normal or is something in my system
16:39 dmanto is ubuntu 14
16:40 dmanto and i thought EV should be better than Epoll, that should be better than Poll
16:42 Grinnz_ dmanto: EV is just a system library which uses different backends; on linux it usually defaults to Epoll, i've had slightly better perofrmance than the built-in Poll usually
16:42 dmanto ok
16:42 Grinnz_ dmanto: but Poll can certainly be better performance than Epoll under certain circumstances (all else equal)
16:42 dmanto yes i mean adding also LIBEV_FLAGS=4
16:43 dmanto ah ok interesting
16:43 Grinnz_ dmanto: epoll is optimized for large numbers of handles, but as a result has slight overhead in adding/removing watchers for each handle
16:45 dmanto ah ok didn't know that
16:45 ToApolytoXaos joined #mojo
16:46 dmanto you have a link or something i could read?
16:46 Grinnz_ dmanto: in a lot of cases, the benchmark itself is what causes discrepancies, there are lots of little overheads
16:47 Grinnz_ dmanto: well i'm not sure, i learned this mostly by reading the poll and epoll syscall docs ;)
16:49 AirDisa joined #mojo
17:03 Vertig0 joined #mojo
17:10 dmanto Grinnz_:it makes sense, i probably shouldn't expect big differences between Poll and Epoll,
17:11 disputin joined #mojo
17:11 dmanto and second I may messed up with the EV FLAGS and ended up with a default reactor like SELECT that actually gave the bad performance
17:12 Grinnz_ dmanto: you can check the EV backend being used like this: https://metacpan.org/source/DBOOK/POE-Loop-EV-0.12/t/00_info.t
17:17 ajr_ joined #mojo
17:25 Ptolemarch joined #mojo
17:29 dmanto aha
17:30 mgrimes joined #mojo
17:30 dmanto Grinnz_: i have some missconfiguration here
17:30 dmanto look at this oneliner (based on what you commented)
17:30 dmanto perl -MEV -E 'say EV::backend()'
17:31 dmanto it prints 2, which is Poll
17:31 dmanto and if I
17:31 dmanto LIBEV_FLAGS=1 perl -MEV -E 'say EV::backend()'
17:31 dmanto prints 1
17:32 dmanto but with LIBEV_FLAGS=4 i got an error :(
17:33 dmanto EV: cannot initialise libev backend. bad $ENV{LIBEV_FLAGS}? at /usr/l...
17:33 dmanto so i'm not getting epoll
17:33 dmanto shouldn't hypnotoad complain about this?
17:34 Grinnz dmanto, no, hypnotoad has no idea what backend you're using
17:34 Grinnz dmanto, how did you install EV?
17:35 dmanto ehmmm
17:35 dmanto I would say probably with atp-get
17:35 dmanto don't remember
17:35 purl Remember what?
17:36 Grinnz perhaps that version did not install epoll support
17:36 mattastrophe joined #mojo
17:36 dmanto ahh good tip
17:36 Grinnz i would consider a local::lib to install EV in with cpanm, or perl-build a separate perl
17:36 dmanto well i will review that installation
17:37 Grinnz what kernel are you using?
17:41 dmanto Linux daniel-Latitude-E4200 3.13.0-52-generic #86-Ubuntu SMP Mon May 4 04:32:59 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
17:42 dmanto thats the uname -a
17:42 Grinnz yeah, that should certainly support epoll
17:42 ajr_ joined #mojo
17:42 berov joined #mojo
17:43 dmanto don't worry about that, i will revise installation, etc and let you know after
17:44 dmanto still don't like that hypnotoad doesn't complain about that :(
17:44 dmanto i can enter
17:44 dmanto LIBEV_FLAGS=64 hypnotoad myapp
17:45 dmanto clearly 64 is a mistake
17:45 dmanto but the thing is that it silently doesn't work
17:46 Grinnz try with hypnotoad -f, it may warn when it fails to load EV
17:47 Grinnz the thing is, the only thing is it checking for is if the EV reactor loads; EV may not be installed at all, in which case it doesn't need to warn about it
17:48 dmanto yes in foreground i see it complains
17:49 dmanto ahhh i saw this "worker xxx stopped too early, shooting down" before...
17:50 Grinnz this is why i usually set $SIG{__WARN__} to use the app logger
17:50 Grinnz though it is probably too late to set it in app startup
17:51 Ptolemarch joined #mojo
17:52 dmanto yes i guess it's too late
17:57 mst o/ bang bang, he shot me down o/
17:58 dmanto tried it, doesn't work
18:17 Ptolemarch joined #mojo
18:21 disputin joined #mojo
18:25 Vertig0 joined #mojo
18:29 Ptolemarch joined #mojo
18:30 dod joined #mojo
18:31 punter joined #mojo
18:31 basic6 joined #mojo
18:34 misty_g3ar joined #mojo
18:37 trone joined #mojo
18:41 Ptolemarch joined #mojo
19:12 amon joined #mojo
19:30 misty_g3ar joined #mojo
19:32 ajr_ joined #mojo
19:50 berov1 joined #mojo
20:12 AirDisa joined #mojo
20:13 ajr_ joined #mojo
20:31 Grinnz_ i kind of want to rename my bot framework to maverick, do you think that's too close to marvin :P
20:32 Grinnz_ it would keep the m theme at least
20:42 jberger just don't use my bot's name!
20:43 Grinnz_ lol
20:43 Grinnz_ what was that again? *takes out notepad*
20:57 sri jberger: why is #795 not fixed yet?
20:57 jberger sri!
20:58 sri o/
20:58 jberger there is only your vote, I guess I haven't been drumming up core members to vote though :/
20:58 jberger marcus, crab, batman, tempire: https://github.com/kraih/mojo/issues/795
20:58 sri marcus, tempire, batman, crab: why the hell did you not vote?
20:59 sri our docs look shitty right now, and jberger couldn't do a thing because you ignored the issue
20:59 jberger I have a fix which theory liked enough to adopt in his own code
20:59 jberger or we could revert Oleg's patch and not strip leading whitespace from any pod
21:01 jberger Grinnz_: you were looking at this with me too when we found it, so if you have an opinion I would be interested in that too
21:01 sri #796 can't get my vote because of bad style
21:01 Grinnz_ eh, i don't know what stripping the whitespace would be useful for, so my vote is "anything that fixes the ascii art"
21:02 sri it's about copypasta
21:02 sri you don't notice it when it works, but when it doesn't.... people keep popping up all day complaining that examples don't work
21:03 sri and then it turns out they left __DATA__ indented and the like
21:03 Grinnz_ ahh
21:04 Grinnz_ what did it do before oleg's patch?
21:04 sri Mojo::Util::unindent
21:04 sri there's a helper and the doc browser
21:04 sri oleg made the helper unindent too
21:05 sri which i personally don't care much about
21:05 Grinnz_ but the helper is what's used for mojolicio.us?
21:05 sri so, i'd just undo the patch if the helper requires ugly code
21:05 sri no
21:07 jberger can we prevent perltidy from mangling it?
21:07 Grinnz_ ah! unindent is already doing what jberger is trying to do
21:09 Grinnz_ or was, before the patch
21:16 Grinnz_ jberger: https://metacpan.org/pod/distribution/Perl-Tidy/bin/perltidy#Skipping-Selected-Sections-of-Code
21:16 sri Grinnz_: no
21:18 Grinnz_ then it has to be 4 lines?
21:25 sri 3
21:26 jberger I'm happy to change the PR to something less horrid as map EXPR, but which one? map BLOCK with the extra lines?
21:27 sri i'm on it now
21:28 * sri wanted to say it's not worth 3 lines... but the helper makes testing it a lot easier
21:29 disputin joined #mojo
21:34 good_news_everyon joined #mojo
21:34 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/vU0bc
21:34 good_news_everyon mojo/master 29ce807 Sebastian Riedel: fix indentation of ASCII art in documentation browser (closes #796)
21:34 good_news_everyon left #mojo
21:35 good_news_everyon joined #mojo
21:35 good_news_everyon [mojo] kraih pushed 3 new commits to master: http://git.io/vU0bX
21:35 good_news_everyon mojo/master 2f15ae0 Zoffix Znet: Sort methods alphabetically in the source code
21:35 good_news_everyon mojo/master d1a36ee Zoffix Znet: Sort methods alphabetically in the documentation
21:35 good_news_everyon mojo/master c8654b4 Sebastian Riedel: Merge pull request #794 from zoffixznet/method-sort...
21:35 good_news_everyon left #mojo
21:36 borkur joined #mojo
21:39 jberger sri++ nice hack
21:40 sri jberger++ # it's your hack
21:40 sri not that i like it...
21:40 mishanti1 So it is only for templates that ->stash( foo => 'bar' ); will cause $foo to be set? Using $foo in a layout not possible, and I should there use stash('foo') instead of $foo?
21:40 jberger I mean using a function reference
21:40 sri ah, doing that all the time actually
21:41 jberger looks much less ugly than my variants
21:41 jberger mishanti1: the stash is available in templates
21:41 jberger and layouts
21:42 jberger oh, your question is about the variables, I think so! try it
21:45 mishanti1 jberger: Thanks. I keep getting that 'Global symbol "$foo" requires explicit package name'-message. I'll see if I can track down why that happens.
21:45 mishanti1 Probably something I'm doing wrong.
21:46 borkur Installing Mojolicious::Plugin::ParamsAuth fails
21:46 borkur build.log says Can't locate object method "generate_port" via package "Mojo::IOLoop" at t/auth.t line 7.
21:46 borkur any idea?
21:46 purl it has been said that any idea is welcomed.
21:48 trone joined #mojo
21:52 mishanti1 Hmm. So it happens compile-time at least, not when the layout is used.
21:54 Grinnz_ borkur: https://metacpan.org/source/SRI/Mojolicious-6.10/Changes#L637
21:54 jberger sri: thoughts on Mojo::ByteStream::TO_JSON ?
21:54 Grinnz_ https://metacpan.org/pod/Mojo::IOLoop::Server#generate_port
21:54 Grinnz_ sounds like that module has not been updated in a long time
21:55 Grinnz_ jberger: doesn't stringify handle that?
21:55 jberger borkur: yeah, that method has been gone for a long time
21:56 jberger Grinnz_: JSON::XS isn't doing it without having extra options set
21:56 jberger which I can do, I just was curious
21:56 Grinnz_ right, that wouldnt
21:56 Grinnz_ i don't think it has an option to stringify objects
21:56 jberger oh also, sri, while you are around, is "inline" a reserved stash key?
21:56 jberger it is not in the list
21:57 borkur Thanks.
21:57 jberger sri: http://mojolicio.us/perldoc/Mojolicious/Controller#stash
22:07 trone joined #mojo
22:07 sri jberger: hmm, arguable
22:08 sri it depends on the handler
22:08 jberger what would happen if I set template => 'something', inline => 'something else'?
22:10 sri up to the handler
22:11 mattastrophe joined #mojo
22:12 sri but sinde the default handler supports it... i guess it counts as a reserved value
22:12 sri s/d/c/
22:12 trone joined #mojo
22:13 good_news_everyon joined #mojo
22:13 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/vUETV
22:13 good_news_everyon mojo/master 77da896 Sebastian Riedel: fix bug where inline was not considered a reserved stash value
22:13 good_news_everyon left #mojo
22:13 jberger maybe mention it in the same paragraph, not necessarily as reserved?
22:13 jberger yeah, that's better
22:13 sri better be consistent
22:13 jberger and who knows, maybe other handlers want inline support too
22:13 sri yet another special case doesn't help anyone imo
22:14 jberger yeah
22:14 sri so, http://ipv6.google.com doesn't work anymore?
22:14 jberger IPV6 IS TEH DEADZ
22:14 sri or is my isp screwing around with the routing?
22:15 bobkare joined #mojo
22:15 jberger I don't think I have ipv6 here though, so I can't effectively test
22:15 sri hahaha... and somehow i don't trust this one -.- http://downforeveryoneorjustme.com/ipv6.google.com
22:16 sri schroedingers ipv6
22:16 * jberger entangles . and :
22:18 go|dfish joined #mojo
22:18 Grinnz_ i would check, but i disabled my ipv6 tunnel since i can't turn off ipv6 in Mojo::IOLoop::Client :P
22:58 punter joined #mojo
23:30 AirDisa joined #mojo
23:55 dvinciguerra joined #mojo

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