Camelia, the Perl 6 bug

IRC log for #mojo, 2010-10-21

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

All times shown according to UTC.

Time Nick Message
00:03 MojoGuest478 joined #mojo
00:03 MojoGuest478 left #mojo
01:17 perlrocks Twitter: "Mojolicious (PSGI) Starman deploy the app - Getting Started with Perl code samples http://icio.us/15exxh" (ja) --hayajo_bookmark http://twitter.com/hayajo_b​ookmark/status/27983257291
01:18 tardisx left #mojo
01:18 tardisx joined #mojo
01:20 perlrocks Twitter: "RT! Hayajo_bookmark: Mojolicious (PSGI) Starman deploy the app - Getting Started with Perl code samples http://icio.us/15exxh" (ja) --perlism http://twitter.com/perlism/status/27983484513
01:39 ltriant left #mojo
01:42 ltriant joined #mojo
02:35 janus left #mojo
02:49 Alias_ joined #mojo
03:00 janus joined #mojo
04:04 tl joined #mojo
04:06 tholen42 left #mojo
04:13 meraxes left #mojo
04:13 meraxes joined #mojo
04:13 yakudzo sri (: wow you are not loosing time ((: left as it is yesterday evening, woke up in the morning and see new patch (:
04:19 perlrocks Twitter: "Mojolicious 0.999930 / Mojolicious / allframeworks.ru - Blog about modern web development and tools .. http://bit.ly/ddUw3Y" (ru) --Foxcool_ru http://twitter.com/Foxcool_ru/status/27996281234
04:29 perlrocks Twitter: "And Mojolicous Mojolicious: Lite was wrong. generate app will be generated in a more Mojolicious it. Well no, but the roots will be the same." (ja) --shingoyoneda http://twitter.com/shingoyoneda/status/27996826189
05:24 ysyrota joined #mojo
05:55 kvorg left #mojo
06:15 MojoGuest341 joined #mojo
06:15 MojoGuest341 From: http://twitter.com/justinv​incent/status/12550383961 (1 hits)
06:16 MojoGuest662 joined #mojo
06:18 MojoGuest341 left #mojo
06:26 MojoGuest662 left #mojo
06:26 ltriant left #mojo
06:52 kvorg joined #mojo
07:59 koban joined #mojo
08:24 perlrocks Twitter: "Mojolicious:: Lite try." (ja) --polaroidoon http://twitter.com/polaroidoon/status/28007840250
08:56 und3f joined #mojo
09:09 sri moin
09:10 spleenjack joined #mojo
09:13 yakudzo sri, hello (:
09:14 yakudzo sri, with before and after I can create elements by giving path to DOM (:
09:14 yakudzo sri, http://paste.scsys.co.uk/53759
09:15 sri i know :)
09:15 yakudzo cool stuff (:
09:17 vti_ stephan48: please reclone it, i messed up the history :-[
09:21 fhelmber_ joined #mojo
09:23 MojoGuest12 joined #mojo
09:23 MojoGuest12 From: http://news.ycombinator.com/item?id=1277067 (12 hits)
09:23 MojoGuest12 left #mojo
09:43 memememomo_ is now known as memememomo
09:44 memememomo is now known as memememomo_
10:04 kvorg sri: do you think it would be perlish to add "E[foo~~'b.*r']" which would work with regular expressions? (perhaps bad syntax, but ~= is taken)
10:05 sri only if it's in the spec
10:05 kvorg also, it is confusing that the pod examples say "E[foo="bar"]", 'E[foo="bar"]' would be better, no?
10:06 kvorg no, it isnt't, but it is 'customary' to support regexps as perlish extensions when implementing XML specs
10:07 sri for now i'd rather stay spec compliant
10:07 sri and let the api settle
10:07 kvorg anyway, just an idea, if you want to follow the specs and avoid extensions, it is a design/policy decision. i only mention this because i suspect it would be easy to implement, but tricky to tack on from the outside.
10:07 kvorg of course, you are right then. and anyway, Mojo::DOM is making quite a lot of buzz ... and for a good reason.
10:10 sri actually it's not too hard to do from the outside
10:11 sri find('foo > bar[baz]')->each(sub {print $_->text if $_->attrs->{baz} =~ /.../ });
10:11 sri you can always just check for presence and iterate
10:42 marcus sri: thought about the new mba?
10:43 sri mba?
10:44 marcus macbook air?
10:44 sri looks very nice
10:44 marcus agree
10:46 vti_ i want it too
10:47 marcus I really can't justify it tho
10:47 marcus I just bought this 15" pro
10:49 vti_ finally usb ports are separated
11:11 tardisx left #mojo
11:11 tardisx joined #mojo
11:13 tardisx left #mojo
11:13 tardisx joined #mojo
11:13 spleenjack left #mojo
11:16 dotan joined #mojo
11:19 perlrocks left #mojo
11:20 perlrocks joined #mojo
11:20 tardisx left #mojo
11:21 tardisx joined #mojo
11:21 y0j joined #mojo
11:22 forwardever joined #mojo
11:22 Eugene joined #mojo
11:23 spleenjack joined #mojo
11:24 ms joined #mojo
11:24 Eugene hi. SOmetimes I receive warning when using Mojo::Client:
11:24 Eugene Use of uninitialized value in concatenation (.) or string at C:/usr/site/lib/Mojo/Cookie/Request.pm line 63
11:24 Eugene it's here:
11:24 Eugene # Render
11:24 Eugene my $cookie = $self->name . '=' . $self->value;
11:25 Eugene I dumped the object and the value was undef
11:25 Eugene is it possible to add a check?
11:25 Eugene for value
11:26 sri everything is possible
11:26 sri but what does the spec say about empty names/values?
11:26 Eugene good question, need to check
11:29 Eugene not sure it's the one: http://www.faqs.org/rfcs/rfc2965.html
11:29 Eugene av-pair     =     attr ["=" value]              ; optional value
11:29 Eugene under 3.1  Syntax:  General
11:33 Eugene probably I was wrong and it's here: http://www.w3.org/Protocols/rfc2109/rfc2109
11:33 Eugene 4.2.2  Set-Cookie Syntax
11:33 Eugene although I can't find there anything regarding non-empty value
11:40 spleenjack left #mojo
11:43 spleenjack joined #mojo
11:44 forwardever what is the correct command to create a mojolicious app, mojo generate or mojolicious generate or both ?
11:45 GitHub120 joined #mojo
11:45 GitHub120 mojo: master Sebastian Riedel * 9fd4823 (5 files in 4 dirs): fixed optional value support in Mojo::Cookie - http://bit.ly/buCBLA
11:45 GitHub120 left #mojo
11:45 sri forwardever: mojolicious command is gone
11:45 sri Eugene: should be fixed
11:48 Eugene great, thank you very much
11:50 sri btw. we now have over 4100 tests :)
11:53 marcus yay
11:53 marcus how many unique tests?
11:53 sri unique?
11:53 marcus not looped?
11:54 sri about 4100 ;p
11:54 marcus nice
12:02 forwardever sri: still looking for a way to determine the app name, so if a user is told to go to is app folder, e.g. /my_app/, I would than check the lib folder for .pm files with a startup method and read the class name, how reliable could that be?
12:03 sri why not do it like the routes command?
12:03 forwardever will check that
12:03 sri think i told you two times now
12:04 forwardever yea, this was the second time, maybe when I checked first, there was something that prevented me from applying this, will check again now
12:05 marcus forwardever: just tell him he's a lying liar. That will piss him off plenty.
12:05 sri naah, got tasty brownies here, nothing could piss me off
12:06 marcus mellow seb
12:06 GitHub77 joined #mojo
12:06 GitHub77 mojo: master Sebastian Riedel * c272d75 (2 files in 1 dirs): made tests more windows friendly - http://bit.ly/cQ8llF
12:06 GitHub77 left #mojo
12:08 und3f left #mojo
12:08 sri *munch*
12:11 sri if anyones interested in some major karma, turn http://github.com/und3f/MojoX-D​NS/blob/master/lib/MojoX/DNS.pm into a proper A + AAAA resolver
12:12 forwardever don't get it, you only way the routes command works is: perl script/my_app routes, but this already requires to pass the app name ?
12:13 sri the routes command works flawless
12:13 sri try mojo routes
12:13 sri it detects that there is nothing
12:14 forwardever it get Application has no routes
12:14 sri exactly
12:14 forwardever ?
12:15 forwardever app has example#welcome route
12:15 sri look, you get the current application instance, thats all you need
12:15 stephan48 vti_: please pull git://github.com/vti/bootylicious.git
12:16 vti_ stephan48: please reclone
12:16 stephan48 vti_: mh?
12:16 vti_ i screwed up the history :(
12:16 stephan48 ye i done
12:16 vti_ stephan48: git clone agan
12:16 vti_ and?
12:16 stephan48 after fucking 2h with github
12:16 vti_ sorry
12:17 sri forwardever: you hook into routes and get all the information you need
12:17 stephan48 ye i used wrong acc and then failed to find out why it wont accept the pub key :)
12:17 stephan48 http://github.com/vti/bootylici​ous/blob/master/bootylicious#L7
12:17 sri thats better than any application class detection you could ver come up with
12:17 stephan48 here you say the dir for the plugin is: contrib/mojolicious-plugin-bot_protection/lib
12:17 vti_ stephan48: fixed
12:17 sri you automatically support all special cases, because you hook directly into the workin app
12:18 stephan48 but when you access it in browser: http://github.com/vti/bootylicious/blob/master​/contrib/mojolicious-plugin-bot_protection/lib it gives 404 :)
12:18 sri forwardever: app->routes->namespace *hint hint*
12:18 vti_ stephan48: just fixed it! :)
12:19 stephan48 ye
12:19 forwardever sri: need some time to understand will get back in couple of minutes (or hours :)
12:19 vti_ stephan48: i've written a simple admin page where you can add and edit articles from the web...
12:19 stephan48 coooooool
12:19 vti_ it's ugly, but works
12:20 marcus like your mum
12:20 Eugene left #mojo
12:40 forwardever sri: not sure whether we mean the same thing: will just describe again: in my case, there is no app running, i just want to create a scaffold via a generate resource command, which needs the app name to get the directory structure right, so I tried my $app = Mojo::Server->new->app; $app->routes->namespace which produces: Can't locate object method "routes" via package "Mojo::HelloWorld" ... I'm
12:40 forwardever lost
12:41 * sri gives up
12:42 forwardever hmm, too many new things I have to understand...
12:49 sri ok one last hint, script/myapp generate resource is the only way to make it work
12:50 sri mojo generate resource is not possible
12:56 forwardever thats like passing the app name to mojo generate resource, don't like the api, will instead look for a script directory, open the file in it, and than read $ENV{MOJO_APP}
12:57 sri why?
12:58 forwardever i want: mojo generate resource resource_name, right now I have: mojo generate resource app_name resource_name, which is too long, as there will be more options to be added
12:58 sri you can't do "mojo daemon" to start your app either
12:59 sri oh well, no point repeating myself :S
13:01 rhaen mmmmmooooin!
13:01 sri moin
13:03 stephan48 forwardever: you have your dammed script right?
13:03 stephan48 so try substiture the damn cmd line with "perl appfile generate resource resource_name"
13:05 forwardever rhaen: there is broken link in the wiki, someone made a change on the start page :)
13:05 rhaen really?
13:05 rhaen let me have a look :)
13:06 sri forwardever: lets rephrase, the mojo command should only be used if there is no application
13:06 forwardever the red one
13:06 rhaen yup, I see
13:06 stephan48 sri: thanks sri ;)
13:06 forwardever sri: I think it should be easy to get the app name if I'm in the app directory
13:07 stephan48 no
13:07 sri no!
13:07 sri :)
13:07 stephan48 i have here a dir with about 10 apps in it
13:07 sri ^^^
13:07 stephan48 forwardever: if you want to try to write a mindselection interface go and do it
13:07 stephan48 but i dont think it would always find the correct app i want to issue a cmd on
13:08 sri the default app skeleton is just a suggestion
13:08 GabrielVieira left #mojo
13:08 stephan48 wenn you have a non lite app it might work but still its ugly
13:08 sri app class might even be inlined somewhere
13:09 sri the only way to properly get at that information is the real application
13:10 sri detection is wrong wrong wrong
13:10 sri it never works properly
13:10 * sri is out of brownies :(
13:13 * sri goes to teh bakery to get more
13:15 rhaen oh my god!
13:15 rhaen forwardever: thanks for the hint!
13:15 rhaen fixed
13:16 forwardever stephan48: would like to understand your point, you wrote: "i have here a dir with about 10 apps", so in the case one of these apps is a none lite app, one could go into that directory and assume that controllers are located at lib/app_name
13:17 dotan forwardever: you know that all the directories can be overridden (templates, static, etc), right?  app->templates('my_stuff').
13:17 sri forwardever: 10 scripts in the script directory
13:17 dotan You need the actual app if you want to put everything in the right place
13:17 perlrocks Twitter: "I read the Perl source to study the beginner to intermediate learners What might be a CPAN module. Kano Hana Plagger or Mojo or otherwise, but also to listen." (ja) --koba04 http://twitter.com/koba04/status/28025568504
13:17 perlrocks Twitter: "RT @ koba04: Perl to read a good source to study the beginner to intermediate learners What might be a CPAN module. Kano Hana Plagger or Mojo or otherwise, but also to listen." (ja) --AE_PerlIsm http://twitter.com/AE_PerlIsm/status/28025569293
13:17 sri 10 app classes in the lib dir
13:18 sri we do app embedding, i expect people to mix stuff up over time
13:19 sri there could even be multiple apps running inside the same process
13:20 perlrocks Twitter: "RT! Koba04: Perl to read a good source to study the beginner to intermediate learners What might be a CPAN module. Kano Hana Plagger or Mojo or otherwise, but also to listen." (ja) --perlism http://twitter.com/perlism/status/28025779797
13:20 sri (to sum it up)
13:20 sri everything is dynamic, for us an app is just an object at runtime
13:20 sri *BUT*
13:21 sri this objects route instance holds all namespace information
13:21 sri to load controllers dynamically
13:21 sri so thats the only point where you get proper information
13:23 sri as soon as someone does app->routes->namespace('Broken') whatever detection you have in place will go horribly wrong
13:23 sri ok, now off for more brownies :D
13:41 sri *munch*
13:41 forwardever dotan: I know that template paths and controller paths might be overridden, but as far as i understand, once I get access to the app, I can read that namespace stuff to get things done right, so the problem is just to find a way to get access to the app, which right now seems just to be possible by using script/myapp generate
13:42 sri you still don't understand
13:42 forwardever :(
13:42 sri there is absolutely no way the mojo command can get to the right app
13:42 sri script/myapp generate resource
13:43 sri thats the only way to get it right
13:45 vel joined #mojo
13:45 forwardever so there is absolutely no naming convention, not even that there should be a script in the script folder that contains a file with information on the app name...
13:45 sri thats correct
13:46 sri there might be multiple scripts with multiple applications in the script folder
13:48 und3f[A] is now known as und3f
13:48 kvorg left #mojo
13:51 forwardever why should a user put scripts for multiple apps in a directory that is designed for one app
13:52 und3f yo
13:53 sri forwardever: what makes you think it is designed for one app?
13:53 * sri waves
13:54 forwardever generate app creates a directory structure for one app (with multiple controllers of course) ?
13:54 koban left #mojo
13:55 sri you could just as well have multiple apps in there
13:57 forwardever hmm, that is true
14:01 dotan forwardever: Mojolicious' lax conventions are part of what makes in appealing, IMO - it makes it super flexible and adaptable to any existing legacy code layout or personal preference.
14:02 forwardever dotan: the good thing is that it also gives me the freedom to create a plugin that uses a lot of naming conventions to get things done quickly :)
14:03 stephan48 how should i implement xmlrpc in a plugin? i want to make it possible to add methods from other plugins/modules...
14:03 dotan forwardever: That said, there's no reason your plugin (for generating resource CRUD skeleton?) not to use it's own conventions  - your toy, your rules...
14:03 stephan48 when i only have a fixed set of methods its rather simple....
14:03 stephan48 but idk if i want it to be dynamic
14:06 forwardever dotan: yea, some kind of a controller, template and form generator, but with no specific ORM in mind, still working on it and will update incomplete version on git later today
14:10 sri stephan48: complicated topic
14:10 stephan48 ye
14:11 sri stephan48: think you'll have to find out for yourself the hard way
14:11 sri trial and error :)
14:11 stephan48 k
14:12 stephan48 is there a way to specifiy own keywords like get or post?
14:13 sri look at how lite does it
14:13 sri but there is not for plugins
14:13 stephan48 k
14:14 sri you could make a new class that automatically imports lite functions too
14:14 stephan48 ye
14:14 forwardever rhaen: link works again, however, link contains the earlier corrected typo again (so link and file path would have to be corrected :)
14:23 ms left #mojo
15:18 sri http://peepcode.com/products/raphaeljs # oh, interesting new screencast
15:21 kvorg joined #mojo
15:28 perlrocks Twitter: "Mojolicious Web Framework - Join the revolution! Http://bit.ly/carI57" (ja) --wPerlnews http://twitter.com/wPerlnews/status/28036710349
15:28 und3f sri, got timeout and IPV6 queries
15:30 und3f sri, and nameservers from /etc/resolv :)
15:30 sri und3f++
15:31 sri my dns knowledge is rather bad, how important is recursion?
15:32 sri will there be a lot of dns servers that don't do it automatically?
15:32 und3f I can't answer for sure... Seems like root dns must have recursion closed
15:33 sri for now we could get away with a fallback to blocking dns lookup i guess
15:34 sri but maybe you're already working on it? :)
15:34 und3f now i am resting :)
15:34 sri :o
15:34 und3f yeah, fallback is nice
15:35 sri we basically get the fallback for free if we return the unresolved name
15:35 und3f Need to read spec, possible recursion is not so hard
15:36 sri don't think it's hard
15:36 und3f i just tried to make recursion query with nslookup to get spec in my wireshark
15:36 und3f But didn't found
15:36 und3f how
15:37 sri you need to leave out the server side recursion bit
15:37 forwardever left #mojo
15:37 sri 3rd or 4th flag i think
15:39 sri it is RD - Recursion Desired
15:40 sri 5th flag
15:41 forwardever joined #mojo
15:43 sri asking a root server for a domain from another root server should work as a test case
15:43 sri they don't do recursion themself i think
15:49 und3f sri, yes, they don't
15:49 und3f just tries
15:49 und3f *tried
15:49 sri dig @128.63.2.53 nasa.gov
15:49 und3f a.root-server.net
15:49 sri that should be a good scenario for a test
15:50 sri oh you got it already :)
15:50 und3f you forget? I am fast :)
15:50 sri -.-
15:50 und3f sri, should i return string like "Recursion desired?"
15:50 sri ?
15:51 sri just have an attribute max_recursion or so
15:51 sri if it is 0 then it is disabled
15:51 und3f oh, you want me to do recusrion
15:51 sri of course :D
15:52 sri you are faster
15:52 und3f right...
15:52 sri ;p
15:54 und3f is now known as und3f[A]
15:54 und3f[A] is now known as und3f
15:54 und3f Seems like we don't need recursion
15:55 und3f I just setup root server as my dns in resolv
15:55 und3f And i was unable to use any site ;)
15:55 sri Oo
15:55 und3f seems like gethostbyaddr doesn't do recursive queries
15:56 und3f In wireshark i saw just queries to root server
15:56 sri i would expect that to be platform specific
15:56 sri still very weird
15:58 sri same for gethostbyname?
15:59 sri maybe it's just reverse lookup
16:00 und3f didn't tried reverse
16:00 und3f sri, but that seems right
16:01 und3f sri, users doesn't know hierarchy of dns
16:01 und3f or know?
16:01 sri not sure what you mean
16:01 und3f i need to read more docs..
16:02 sri if you ask the .mil root server for a .gov address it will give you the address for the .gov dns server
16:03 sri or actually a bunch of them
16:03 sri dig @128.63.2.53 nasa.gov
16:04 und3f any way, seems like every user has DNS with enable recursion
16:05 sri oh interesting
16:05 sri nasa.gov doesn't actually resolve to a host
16:05 sri just a zone record
16:06 sri and www.nasa.gov resolves to a cname
16:06 sri www.nasa.gov.speedera.net
16:06 sri do you support cnames?
16:07 und3f sri, they just works )
16:08 sri because we expect the server to do recursion?
16:08 sri right
16:08 perlrocks Twitter: "#perl What is better? #catalyst or #mojo?" --_sattellite http://twitter.com/_sattellite/status/28040123214
16:08 und3f yeah.
16:09 sri !tweet @_sattellite #mojolicious of course, viva la revolucion!
16:09 perlrocks sri: http://twitter.com/perlrocks/status/28040218354
16:09 perlrocks Twitter: "@_sattellite #mojolicious of course, viva la revolucion! --sri" --perlrocks http://twitter.com/perlrocks/status/28040218354
16:11 und3f sri, just try module - you can use other names in tests
16:11 und3f to test it right
16:13 sri oh
16:13 sri do you allow a and aaaa records to be queried together?
16:14 und3f no :|
16:14 sri :(
16:14 und3f Or what do you meen?
16:14 sri thats the normal use case
16:14 und3f sri, semi time?
16:14 und3f sri, or in one packet ?
16:14 sri can it be done in one packet?
16:14 und3f Yes...
16:15 sri then that would be desirable
16:15 und3f But than we got lot of problems, like truncated packets, etc...
16:15 sri well, we always want both
16:16 und3f sri, you can do $dns->resolve('host, 'A', $cb); $dns->resolve('host', 'AAAA', $cb); $loop->start
16:16 sri well...asking for ipv4 first and then ipv6 would work too i guess
16:16 und3f that should work
16:16 und3f sri, i got asking 'A', then 'AAAA' in tests http://github.com/und3f/MojoX-​DNS/blob/master/t/MojoX-DNS.t
16:17 und3f i'll make parallel queries in test
16:18 perlrocks Twitter: "@ Perlrocks thx:)" (de) --_sattellite http://twitter.com/_sattellite/status/28040951301
16:19 sri und3f: you might be dropping descriptors
16:19 sri umm leaking
16:19 und3f sri, test improved
16:19 sri because you don't drop connections
16:19 und3f sri, i do
16:20 und3f oh, right
16:20 und3f in success case i don't
16:20 sri do you reuse the connection?
16:21 sri otherwise you leak on success
16:21 und3f sri, fixed
16:22 ysyrota left #mojo
16:27 sri und3f: how much faster is it than gethostbyname?
16:28 und3f why it suppose to be?
16:29 sri just out of interest
16:29 und3f hm... could try
16:29 sri of course not blocking is the main goal, but parallel resolving should be noticeable
16:39 und3f sri, it is slower :|
16:39 und3f sri, http://pastebin.ca/1969148
16:40 sri cache?
16:40 und3f sri, in wireshark i see all queries to dns
16:41 sri interesting
16:41 sri what if you go up to 1000 queries?
16:41 sri or at least 100
16:42 sri maybe you should be reusing one socket
16:43 sri and just match request ids
16:44 und3f sri, gethostbyname uses new sockets
16:44 und3f but i found something interesting in wireshark
16:44 und3f servers react differ on my queries
16:46 sri we of course have some more perl overhead, for almost sequential requests i would expect gethostbyname to be faster
16:47 und3f sri, i think so too
16:47 sri but that should even out with more paralllel requests
16:47 sri like 100
16:55 und3f sri, found a problem :)
16:55 und3f in 4 serial  questions we got:
16:56 und3f gethostbyname 0.006612
16:56 und3f Mojo 0.008024
16:56 und3f If do parallel queries with Mojo:
16:56 und3f gethostbyname 0.008747
16:56 und3f Mojo 0.004373
16:56 und3f sri, so it rocks :)
16:56 sri \o/
16:59 sri would it be hard to support resolving txt records too?
17:00 und3f sri, i guess that not
17:01 sri oh well, time for me to clean up the ioloop for the async resolver :)
17:02 sri gonna need new connection ids
17:02 und3f also PTR resolving is nice thing :)
17:03 sri decoding more record types is like adding a oneliner right?
17:05 und3f sri, right, but also you need to encode query, for example for query PTR 2001:470:b825::1 you need to query string '1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.​0.0.5.2.8.b.0.7.4.0.1.0.0.2.ip6.arpa'
17:05 sri oh right
17:05 sri ptr was special, forgot
17:06 und3f but in most cases in gonna be oneliners
17:06 sri txt should be like unpack (C/a*)* or so
17:07 und3f sri, all answer in same format, but for ip address we need to decode binary ip to string
17:07 sri ah
17:15 dotan left #mojo
17:38 su-bzero_ Resolve da w0r1d!!!
17:38 su-bzero_ All IPs!
17:39 su-bzero_ IPV6 too.
17:46 fhelmber_ left #mojo
17:46 GabrielVieira joined #mojo
17:57 alnewkirk joined #mojo
18:02 rhaenerc joined #mojo
18:02 rhaenerc hm.
18:06 spleenjack left #mojo
18:17 rhaenerc it feels strange to use emacs for chat purposes, but hej - it works
18:20 rhaenerc bbl
18:20 rhaenerc left #mojo
18:21 sri wow
18:21 su-bzero_ Indeed.
18:21 su-bzero_ It works perfectly. ;)
18:21 sri there will be free ec2 instances
18:22 sri thats going to be interesting
18:23 sri http://aws.amazon.com/free/
18:29 sri free for a year will still scare some people away
18:30 sri think they could get more people hooked if it had more limitations but without a time limit
18:31 sri thats possibly where google will attack
18:55 fish_ sri: a pure perl dns resolver for mojo? nice! =)
18:55 sri there is already gethostbyname() ;p
18:56 sri the key here is that it's non blocking
18:56 sri ok well... we can allow retrieving other record types
18:56 und3f sri, that will rock )
18:57 sri indeed
19:02 polvo left #mojo
19:04 y0j left #mojo
19:10 kvorg left #mojo
19:12 kvorg joined #mojo
19:19 forwardever left #mojo
19:20 su-bzero joined #mojo
19:21 spleenjack joined #mojo
19:25 su-bzero_ left #mojo
19:57 achromic joined #mojo
20:08 GitHub151 joined #mojo
20:08 GitHub151 mojo: master Sebastian Riedel * 596a322 (1 files in 1 dirs): cleanup - http://bit.ly/a0SP3T
20:08 GitHub151 left #mojo
21:00 ltriant joined #mojo
21:32 vel left #mojo
21:38 forwardever joined #mojo
22:45 forwardever sri: already sleeping?
22:47 und3f is now known as und3f[A]
22:54 sri forwardever: almost
22:55 forwardever i have improved my plugin and need a tester, it takes only 2 minutes, promised :)
22:57 forwardever added template scaffolding with tag helpers and data is saved in a class var, so one can get a working example out of the box
22:57 forwardever http://github.com/forwardever/mojo​licious-plugin-resourceful_routes
22:58 forwardever you would have to follow instructions in the README
23:01 ashleydev left #mojo
23:08 sri forwardever: interesting
23:09 forwardever does it work?
23:09 sri forwardever: btw. you can get the absolute template path from the app too
23:09 sri reading code
23:09 sri $app->renderer->root
23:09 sri oh, you already use it
23:10 sri that error message confused me
23:10 spleenjack left #mojo
23:10 forwardever but not sure what to do if user has changed controller root path
23:10 sri the path is absolute
23:11 sri controller root path?
23:11 forwardever yes
23:12 sri you have to transform the namespace to a path
23:12 sri namespace is often Foo::Bar
23:12 sri MyApp::Controller
23:26 sri it needs a model layer imo
23:26 sri right now the resources seem a bit lost
23:29 forwardever maybe mojolicious has to define some kind of a standard (api), so that orm plugins and scaffolding plugins can work well together
23:30 sri for the core thats very unlikely to happen
23:30 sri but there can always emerge a standard via plugins
23:30 sri some kind of crud framework on top of mojolicious
23:31 sri well, you're already half way there :)
23:32 forwardever okay, will thing about it tomorrow, no more energy left for thinking today :)
23:34 forwardever but CRUD framework might belong into the core, as you need critical mass to convince users to adopt it
23:36 forwardever will have to sleep now, thanks for your help :) nn#
23:40 ashleydev joined #mojo
23:41 sri nn
23:44 forwardever left #mojo
23:46 sri kvorg: whats the difference between check_box and check_box_x?

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