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

IRC log for #mojo, 2014-08-20

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

All times shown according to UTC.

Time Nick Message
00:00 d4rkie joined #mojo
00:27 zivester joined #mojo
00:28 elik Is there a way to make a route to a static file? I'm trying to make / point to the index.html file in public/
00:28 purl Hmm.  No matches for that, elik.
00:36 disputin joined #mojo
00:39 disputin2 joined #mojo
00:44 fhelmber_ joined #mojo
01:00 janus joined #mojo
01:05 laouji joined #mojo
01:17 TitanOfOld joined #mojo
01:22 woz joined #mojo
01:29 sujithm joined #mojo
01:39 klapperl joined #mojo
01:51 br_hadrian joined #mojo
01:55 woz joined #mojo
02:42 adhoc joined #mojo
02:46 noganex joined #mojo
02:46 D4RK-PH0ENiX joined #mojo
02:52 GabrielVieira joined #mojo
02:56 jnbek elik: https://metacpan.org/pod/Mojolicious::Controller#render_static
03:02 rwp joined #mojo
03:04 woz joined #mojo
03:11 pfc joined #mojo
03:17 elik jnbek: thank you, that's what I ended up doing indeed: http://blogs.perl.org/users/joel_berger/2013/01/a-quick-static-file-webserver.html
03:40 zivester joined #mojo
03:43 disputin joined #mojo
04:06 woz joined #mojo
04:18 irq joined #mojo
04:39 jberger purl /testing/
04:39 purl jberger?
05:07 woz joined #mojo
05:37 preaction_ joined #mojo
05:38 Eke- joined #mojo
05:50 laouji joined #mojo
05:54 Eke|| joined #mojo
06:09 woz joined #mojo
06:10 batman nicomen_: that is indeed a non-trivial question :(
06:11 batman nicomen_: but add a TO_JSON method, and that method will be called to serialize the object.
06:11 batman No idea if that's even close to what you're asking though...
06:15 Vandal joined #mojo
06:34 basiliscos joined #mojo
06:41 KCL_ joined #mojo
06:49 batman marcus: have you ever written the server side part for M::P::Oauth2 ?
06:50 batman i want to enable all the apps we are writing @work to be authenticated by single sign on (different domains), but i guess i need some server side part to do that
06:51 batman marcus: could we also deprecate the on_xxx stuff in M::P::Oauth2? and run perltidy on it?
06:53 marcus batman: we could. I haven't written an oauth2 provider yet, but it should be pretty easy =]
06:53 batman cool.
06:53 * batman starts up perltidy :)
06:54 batman marcus: can i use convos' perltidyrc or should i use the one from mojo?
06:54 marcus batman: the diff is the line length?
06:54 batman i think so
06:55 marcus I'm fine with convos'
06:55 batman cool. do you want a pull request for the tidy part?
06:55 marcus yes pls
06:55 batman okidoki
06:59 batman marcus: https://github.com/marcusramberg/Mojolicious-Plugin-OAuth2/pull/32 #enjoy :)
07:00 cosimo joined #mojo
07:07 rawler joined #mojo
07:09 nicomen joined #mojo
07:11 woz joined #mojo
07:17 _eugen joined #mojo
07:17 Eke- joined #mojo
07:19 batman marcus++
07:24 preaction joined #mojo
07:26 preaction_ joined #mojo
07:35 woz joined #mojo
07:37 batman marcus: https://github.com/marcusramberg/Mojolicious-Plugin-OAuth2/pull/33
07:39 batman marcus: i guess i could just use t/blocking.t as a starting point for the server side part...
07:39 batman i'm thinking about the "fake_auth" resource
07:40 marcus Yeah, that seems reasonable.
07:40 batman sweet
07:41 marcus You'll need some more crypto tho, better read the spec ;)
07:41 batman yeah... probably :/
07:50 KCL joined #mojo
08:21 denis_boyun joined #mojo
08:26 hrupp joined #mojo
08:35 denis_boyun_ joined #mojo
08:50 TitanOfO` joined #mojo
08:51 dp_ joined #mojo
08:52 fhelmber_ joined #mojo
09:02 laouji joined #mojo
09:05 HtbaaPi perhaps a stupid question, but where is session data stored? Is it stored inside the cookie or somewhere in the tmp filesystem? Documentation says its stored in HMAC-SHA1 signed cookies, but does that also mean the session data is stored in there?
09:08 laouji joined #mojo
09:09 HtbaaPi ah, in Mojolicious::Session it indeed says all data is stored client side
09:10 HtbaaPi I actually need to store some file uploads in between screens (wizard form). Was thinking about storing this data inside the session, but I think I need another solution. Any best practices on this kind of stuff?
09:11 arthas joined #mojo
09:12 TitanOfO` HtbaaPi, you'll want to use hidden fields or a database.
09:12 laouji joined #mojo
09:13 TitanOfO` HtbaaPi, session data is stored client-side, but you want to be careful how much data you store as most, if not all, browsers restrict cookie size to 4KiB.
09:14 HtbaaPi TitanOfO`: yeah, I'm used to PHP sessions which stores all data server side. https://metacpan.org/pod/Mojolicious::Plugin::ServerSession seems like a flexible way to go around with
09:15 HtbaaPi but it hasn't seen any update in over 2 years and Mojolicious has evolved quite a bit since then, so I'm not sure if it'll work but best to try
09:22 laouji joined #mojo
09:23 batman HtbaaPi: you got plugins for connection the client side session with a server side store, but i think i would do what TitanOfO` suggested: use a hidden field which point to a temp file
09:23 batman i would create a temp file with a unique name (md5sum or something) to make sure not two wizards doesn't collide
09:24 batman the md5 sum could be client ID (username, ...) + file content or file name.
09:24 HtbaaPi batman: ah yes I see. I can store the file and keep a reference to the filepath. Unique name, yes. Only downside is having to clean up after the form has been fully submitted or the user kills the session
09:24 HtbaaPi but that can be overcome
09:25 HtbaaPi I can simply list the files inside the tmp directory and delete everything older than say 4 hours or so
09:25 batman but i think you need to clean up anyway..?
09:25 batman unless you got modern clients and can store the file in browser storage (unless the file is big)
09:26 TitanOfO` HTML5's web storage?
09:26 HtbaaPi have used the File API in Javascript before. Has some quirks and since it's a public form I can't depend on it
09:27 batman TitanOfO`: yes.
09:27 TitanOfO` http://www.html5rocks.com/en/features/storage
09:27 HtbaaPi but thanks for giving the suggestions. I'll keep a reference to a tmp file path
09:28 batman HtbaaPi: i agree. i would not depend on it either... makes things hard if you suddenly need support for big files (>5MB iirc)
09:28 TitanOfO` I forgot about that option, but there's a lot of people still on IE7.
09:28 HtbaaPi yeah I implemented it in the backoffice part of this webapplication (full Backbone JS app). I can't say it made me too happy :)
09:28 batman HtbaaPi: you could clean up the files either using a crontab (find + date + xargs + rm) if you're on linux or simply remove old files on each request (might be slow)
09:28 batman haha
09:29 batman each request = when you create a new "temp" file
09:29 HtbaaPi since the backbone.js app uses REST requests and all data is in JSON I had to implement file uploads in a rather nasty way
09:29 HtbaaPi got it
09:30 batman HtbaaPi: i think i would remove old files when you create new files, unless the platform has a lot of users
09:30 HtbaaPi using the File API I read the data of the selected file, then had to base64 encode it and send it to the server through HTTP. But then I would need another library for decoding the JSON without making Mojolicious eat up all memory :)
09:30 batman if you're programming non-blocking, you can offload the work to jberger's Mojo::IOLoop::ForkCall
09:31 HtbaaPi if you need to work with big JSON data use JSON:SL. I found it to be really fast
09:31 batman the work = traversing the directory and deleting files.
09:31 HtbaaPi got it :)
09:31 HtbaaPi I haven't invested much time in the non blocking aproach yet. Should do that some time
09:31 batman also --- if you have A LOT of visitors, you might want to make sub dirs (like temp/aa/bb/ccccddd) so to make it easier to traverse
09:32 batman or even temp/dir/$day-$hour/$file or something...
09:32 batman yeah, non-blocking kick ass. makes you run in circles around other web frameworks.
09:32 HtbaaPi yeah I think that would be best practice to start with right away
09:33 batman uhm... not sure if "run in circles" is a norwegian expression... meaning mojolicious kick other-other-framework's-ass :)
09:33 HtbaaPi being able to run non-blocking apps depends on the server as well right? e.g. CGI wouldn't be able to do it?
09:34 HtbaaPi ah well I got the gist of it :)
09:34 batman yes. you need to run a mojolicious based server.
09:34 HtbaaPi I always use hypnotoad
09:34 batman you could have nginx in front, doing reverse proxy to hypnotoad, daemon or prefork.
09:35 batman look for the servers here: http://mojolicio.us/perldoc#HIGHLIGHTS
09:36 batman i favor running hypnotoad (does not work on windows)
09:36 HtbaaPi I exclusively run my projects in a Vagrant based virtual machine (VirtualBox)
09:36 HtbaaPi so that stuff is no longer an issue with me
09:36 batman at @work i run it behind nginx. @home i only run hypnotoad directly on internet (since i'm just doing mojo apps)
09:36 HtbaaPi used to have issues with morbo restarting on Windows as well
09:36 batman ok
09:37 HtbaaPi I usually have either lighttpd or nginx in front of hypnotoad
09:38 batman wow! haven't heard about lighttpd in many years :)
09:38 batman i would only use it for legacy reasons.
09:38 batman i've heard it's pretty bad (can't remember if it's because of security or just in general)
09:39 HtbaaPi hmm well years ago there used to be a memory leak which has been fixed years ago, yet that argument still pops up :)
09:40 batman aha. ok. that might be it...
09:43 HtbaaPi time for me to continue work on my project. Thanks for the help
09:43 batman you're welcome :)
09:43 batman i would very much like if you came back and told us about the solution you selected and if it's a success :)
09:45 laouji joined #mojo
09:45 HtbaaPi well the one I'm going for is storing the file upload to a tmp file (unique name and in sub dirs), store that filename inside my session and when the final form has been submitted use that file. Afterwards, clean up. Probably use a cron to clean up old files
10:11 woz joined #mojo
10:12 laouji joined #mojo
10:20 dp_ joined #mojo
10:57 meshl joined #mojo
10:58 dexteruk joined #mojo
10:58 dexteruk_ joined #mojo
11:01 d4rkie joined #mojo
11:17 ryozi joined #mojo
11:19 laouji joined #mojo
11:21 Eke- joined #mojo
11:22 woz joined #mojo
11:23 Kripton joined #mojo
11:24 d4rkie joined #mojo
11:26 mattp_ joined #mojo
11:29 dp_ joined #mojo
11:31 laouji joined #mojo
11:38 laouji joined #mojo
11:53 Eke- joined #mojo
11:54 neilhwatson joined #mojo
11:55 GabrielVieira joined #mojo
12:06 woz joined #mojo
12:11 Kripton I am using postgres with json, and that was a terrible idea
12:12 Kripton I have lots of problems with Sql::Abstract, dbi, and the postgres json implementation
12:13 Kripton I need to make lots of workarounds, in too many places, the solution perl + postgres/json is not mature, and you should take care
12:14 Kripton now I need to  go back to "traditional" data model, converting the json columns to something else
12:18 Kripton I am saying that as warning, because the sollution was good on paper, and after some basic testing was good, but now, with more complexity, is a nightmare
12:24 odc good to know
12:34 laouji joined #mojo
12:35 sh4 joined #mojo
12:42 hesco joined #mojo
12:47 irq joined #mojo
12:55 sujithm joined #mojo
13:07 woz joined #mojo
13:17 gnephiak joined #mojo
13:25 disputin joined #mojo
13:38 Nei hi, I'm trying to debug my mojo reactor and it receives some "remove" calls like remove('5ff42155ceff37fb6a0d25670807e00f'), that are coming from the IOLoop, _stream, $stream->on(close => sub { $self && $self->_remove($id) });
13:39 Nei seems like it is "designed" to ask the reactor first if this is a timer?
13:42 sujithm_ joined #mojo
13:49 Akron joined #mojo
13:55 punter joined #mojo
13:55 punter Is anyone attending YAPC?
13:55 punter ::EU?
13:55 purl somebody said ::EU was very soon?
14:00 irq_ joined #mojo
14:08 woz joined #mojo
14:39 Nei sri, if a reactor gets reset externally is there any exising way to communicate that back to mojo (the ioloop etc) or can I somehow else gracefully handle that?
14:59 GabrielVieira joined #mojo
15:06 disputin joined #mojo
15:09 Dandre joined #mojo
15:10 woz joined #mojo
15:22 Nei stephan48: http://paste.scsys.co.uk/417096 this will work as long as you don't unload the "dataX" script
15:27 disputin joined #mojo
15:40 * sri is jealous of rails today
15:40 sri they can just announce that rails 5 will only work on ruby 2.2 and everybody is happy
15:42 lipizzan joined #mojo
15:43 meshl joined #mojo
15:44 good_news_everyon joined #mojo
15:44 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/3fkiLA
15:44 good_news_everyon mojo/master 3b20f39 Sebastian Riedel: fixed server descriptions
15:44 good_news_everyon left #mojo
15:46 marcus sri: Mojolicious 6 5.20 cutoff?
15:46 marcus :)
15:46 sri so wish we could just depend on 5.20 and use signatures for everything, to really differentiate us from everything else in perl... push forward more aggressively
15:46 marcus I wish that too
15:46 sri the docs would look sooo sweet
15:47 marcus What if we strongly encouraged 5.20, and documented that style everywhere, but had fallback for inferior perls?
15:47 btyler_ personally rather enjoy postderef as well, although I've heard there's a wide range of opinions on that front
15:47 marcus I still can't stomach it, tbh.
15:48 stephan48 maybe ask the guy who maintains mojo-legacy to maintain a mojo-newer-than-legacy-but-older-than-current branch?
15:48 sri marcus: then we would have to document "use experimental 'signature';" in every example
15:48 sri *+s
15:49 sri enabling it automatically in Mojo::Base on 5.20 is a little problematic, due to prototypes working differently... need :prototype(), which only works on 5.20 as well
15:50 basiliscos joined #mojo
15:50 btyler_ I wonder if there are any stats floating around on adoption rates for perl versions
15:51 marcus sri: Can't even do it with string eval?
15:51 disputin joined #mojo
15:51 btyler_ my sense is that a lot of the conversations about abandoning older perl versions for whatever reason get hit by one person saying "my company still uses 5.8, you can't do this to us!"
15:51 btyler_ and there are no numbers to suggest how many users would be impacted either way
15:51 marcus my reply to that person is 'you need to get your head out of your ass and stop using the system perl for apps'.
15:52 btyler_ well, right, of course
15:53 btyler_ I just have the sense that the conversation could be less emotionally grounded (the hold-outs vs the progressives, so to speak) if some numbers were around
15:53 btyler_ I'm not sure where you'd go to start gathering that kind of data
15:56 marcus just install something that dials home with the current perl version in the next mojo. I'm sure that won't cause any controversy.
16:09 sri marcus: wat do you mean with string eval?
16:10 sri anyway, we have 3 instances of prototype use in the code case i believe
16:11 sri ojo::n, Mojolicios::Lite::group, and Mojo::Util::steady:time
16:11 woz joined #mojo
16:12 ryozi_ joined #mojo
16:13 sri tbh. i don't think just changing the docs would work, too many bugs we would miss that way
16:13 marcus sri: never mind, just tested it doesn't work to do use feature inside any eval.
16:13 sri there will be callbacks that don't get the right number of arguments and the like
16:15 marcus Anyways, it seems a bit risky to change everything to use an experimental perl5 feature by default.
16:16 sri it's not going to disappear
16:16 marcus but it could change.
16:17 sri not the core feature set
16:17 sri maybe the radical stuff like return in defaults, but not basics
16:18 sri showing off modern perl like that could be very good for the perception of the language
16:19 sri i hate the fact that we even have to debate it like this :(
16:20 sri it's so perl/python
16:21 sri ruby and javascript communities just keep on pushing forward
16:29 fitnerd joined #mojo
16:51 disputin joined #mojo
16:53 Eke- joined #mojo
17:00 batman https://coderwall.com/p/1gm_lq # is this a joke..?
17:01 denis_boyun joined #mojo
17:01 batman sri: all the old apps would still work, even if the core use 5.20 features right?
17:01 batman but people would have the ability to write new cool code?
17:02 batman if i haven't misunderstood anything, i think i want this when we have perl 5.22, but i'm not going to make a big argument for _not_ changing now.
17:03 batman i just think like it when things get a bit more mature...
17:13 woz joined #mojo
17:19 sri i suppose the install docs could have to versions... "$ cpanm https://github.com/jamadam/mojo-legacy" for anything older than Perl 5.20
17:19 sri *two versions
17:19 sri and "$ cpanm Mojolicious" for 5.20
17:20 sri or even an official branch forked from master that gets backports whenever someone feels like backporting
17:21 Kundun joined #mojo
17:28 GabrielVieira joined #mojo
17:31 sh4|2 joined #mojo
17:32 mattp_ joined #mojo
17:42 disputin joined #mojo
17:58 dp_ joined #mojo
18:11 woz joined #mojo
18:12 compadre joined #mojo
18:12 compadre Hey guys, anybody know how to open up the hypnotoad server to the outside world and not just have it on localhost?
18:16 risugg joined #mojo
18:16 woz joined #mojo
18:17 genio compadre: have you looked at the deployment section of the cookbook?  http://mojolicio.us/perldoc/Mojolicious/Guides/Cookbook#DEPLOYMENT
18:18 compadre will do "genio"
18:18 compadre sorry not sure how to direct at people specifically
18:18 compadre but thank you for the link
18:18 genio compadre: like this :)
18:19 risugg joined #mojo
18:19 compadre I did not know that the term was deployment, but now I do, this will likely help me a lot in all my future adventures on various developmental tasks
18:19 compadre genio: much thanks
18:41 KCL_ joined #mojo
18:55 disputin joined #mojo
19:13 sri recently i've had to deal a lot with rfc 3339 datetimes... and i'm starting to wonder if maybe Mojo:Date should just be able to parse those
19:14 sri i remember Akron made a proposal once... but that was way too powerful
19:15 genio yes to the standard date and time and the combined datetime.  I'm not so sure about the other tidbits
19:15 genio 2014-W34 is about as useless as it gets to me
19:17 sri atm. i'm thinking only combined datetime
19:17 sri i've not come across the others
19:22 genio wait, I might be confusing RFC 3339 and ISO 8601 in my mind
19:25 sri 2014-w34 is not rfc 3339 i believe
19:26 sri iso 8601 is also ranges and shit
19:26 purl okay, sri.
19:26 sri heh :)
19:27 sri stuff like 20140820 is also valid iso 8601
19:28 sri https://tools.ietf.org/html/rfc3339
19:31 sri atom is one of the most prominent users of rfc 3339 http://www.ietf.org/rfc/rfc4287.txt
19:38 punter joined #mojo
19:38 basiliscos joined #mojo
19:39 punter joined #mojo
19:56 meshl joined #mojo
20:15 sri https://gist.github.com/anonymous/3a56bc0d839d86b1d9af # this is what i had in mind
20:19 sri also a quick reminder, the datetime type uses that format too :) http://mojolicio.us/perldoc/Mojolicious/Plugin/TagHelpers#datetime_field
20:22 sri if anyone knows rfcs that use only the date or time part, please link!
20:22 sri or actually, any rfcs referring to rfc 3339 would be nice
20:22 sri personally i've encountered the format in json and atom
20:23 sri but more use cases would be nice
20:23 sri i'm on the edge about it currently... since it costs quite a few lines ;p
20:24 marcus https://github.com/Dobiasd/programming-language-subreddits-and-their-choice-of-words/blob/master/README.md
20:25 sri don't you derail my discussion with myself!
20:26 sri perl people are happier than python people, so that settles it
20:28 Kripton I am doing this route to enable the files on the public folder
20:28 Kripton $r->any('/' => sub { shift->render_static('index.html'); });
20:28 Kripton this is the "best" way ?
20:29 preaction_ joined #mojo
20:29 sri Akron: if you're reading the logs, i'd like to hear your opinion :)
20:30 stephan48 sri+++++ # thats super awesome!
20:32 sri even the github api is using the format https://api.github.com/search/repositories?q=tetris+language:assembly&sort=stars&order=desc
20:32 sri stephan48: what's your use case that you'd be so happy about it?
20:35 sri haha, conincidentally chansen just wrote a blog post about all the different iso 8601 formats http://blogs.perl.org/users/chansen/2014/08/lets-talk-about-timemoment-and-round-trip-of-strings.html
20:35 sri someone should replace Time::Piece with Time::Moment in core
20:36 sri holy format batman https://metacpan.org/pod/distribution/Time-Moment/lib/Time/Moment.pod#from_string
20:38 stephan48 i encountered this rfc3339 format in some apps already(incoming datetime format), each time i was searched for a way to parse it nicely into something i could easily reuse... i always ended up with DateTime and its corresponding formater, which meant quite a few extra modules to install.. just for converting a string into a epoch. having the support for it in mojo core would make it easier for me. no strong arguments just pure convinience
20:39 stephan48 especially since i tend to forget how the format is called each time ;)
20:40 sri right, just convenience for me as well
20:41 stephan48 2 times convenience in the ring against quite a few new lines
20:42 stephan48 is there an performance hit with the new RFC3339 support?
20:44 sri nope, the two formats below it are pretty much never used
20:44 sri they are just there because the htp rfc requires it
20:44 sri *http
20:44 stephan48 so essentially there are no arguments against it?
20:45 sri just a) more code, b) no "direct" use in core
20:45 sri although, i do see many use cases now
20:46 sri so many popular rest apis use the format
20:46 stephan48 wouldn't it be good to be able to parse the value of datetime fields? :)
20:47 bodgix joined #mojo
20:47 stephan48 my vote would be to include it... its not like something really depends on it which would make it impossible to remove it later if needed
20:47 sri haha, even the metacpan api
20:47 sri mojo get https://api.metacpan.org/v0/author/SRI /updated
20:47 hesco joined #mojo
20:47 stephan48 hue? they have no offset?
20:48 sri many don't
20:48 stephan48 are they then automatically UTC based?
20:52 sri think so
20:52 stephan48 if i understand the rfc correctly it is so
20:52 disputin joined #mojo
21:09 disputin joined #mojo
21:14 sri maybe generating the 3339 format might make sense too https://gist.github.com/anonymous/5da075d02f5f55e04428
21:17 disputin joined #mojo
21:39 sri marcus, tempire, jberger, batman, crab: consider yourself poked
21:41 sri i tend towards +1 now, since the format is used EVERYWHERE on the web
21:49 woz joined #mojo
21:49 good_news_everyon joined #mojo
21:49 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/iZR4UQ
21:49 good_news_everyon mojo/master 6c249f1 Sebastian Riedel: added basic support for RFC 3339
21:49 good_news_everyon left #mojo
21:50 disputin joined #mojo
21:50 bobkare joined #mojo
21:52 sri haha, almost everywhere... the twitter api of course uses something else
21:52 sri "Sat Aug 25 17:26:51 +0000 2012"
21:55 disputin joined #mojo
21:59 sri github, facebook and google use rfc 3339
21:59 sri and metacpan
21:59 purl hmmm... metacpan is one of the best stuff
21:59 * sri hugs purl
21:59 * purl smiles
22:00 sri i suspect at some point we will get requests for partial formats (perhaps from Akron), but i don't think we'll go there without lots of use cases
22:01 sri partial as in "2014", "2014-08", "2014-08-20"...
22:01 sri with missing parts defaulting to 01 or 00, depending on type
22:03 sri sorry Akron!
22:03 sri (don't punch me at the next mojoconf...)
22:04 disputin joined #mojo
22:14 good_news_everyon joined #mojo
22:14 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/ikZ5sw
22:14 good_news_everyon mojo/master fcd0ee3 Sebastian Riedel: epoch is more commonly used
22:14 good_news_everyon left #mojo
22:17 davido_ joined #mojo
22:24 tempire +1
22:24 purl 1
22:24 tempire I've had to pull in DateTime several times for just that purpose
22:31 good_news_everyon joined #mojo
22:31 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/1mg0Ow
22:31 good_news_everyon mojo/master f88586e Sebastian Riedel: no need to have that many examples
22:31 good_news_everyon left #mojo
22:34 woz joined #mojo
22:45 meshl joined #mojo
22:49 sri hehe, we can't actually support the minimal "2014" format ;p
22:49 sri since that clashes with epoch
22:52 sri i do wonder if ->to_datetime should append a Z or not
22:53 sri github does, metacpan does not
22:53 sri the parse.com api does
22:55 sri google seems to mix it up
22:55 sri facebook does not
22:56 sri akamai does
23:05 woz joined #mojo
23:05 woz joined #mojo
23:34 neilhwatson joined #mojo
23:44 d4rkie joined #mojo
23:45 d4rkie_ joined #mojo
23:48 disputin joined #mojo

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