Camelia, the Perl 6 bug

IRC log for #mojo, 2012-06-15

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

All times shown according to UTC.

Time Nick Message
00:39 Tom joined #mojo
00:56 laouji joined #mojo
01:12 tempire the problems that folks have when starting is they don't read the guides in order
01:12 tempire for some reason, it's not apparent to everyone.
01:14 scott_ joined #mojo
01:46 asarch joined #mojo
01:50 Averna1 joined #mojo
01:56 cfedde the problem people have starting is that there are not enough clues decorating the docs to take users back to the golden path.
01:57 d4rkie joined #mojo
01:59 rem_lex| joined #mojo
02:02 cfedde for example the word "plugin" does not appear in the Mojo::Template file
02:03 cfedde s/file/doc/
03:05 ispy_ joined #mojo
03:16 russum joined #mojo
03:39 tetraodontidae joined #mojo
03:40 tetraodontidae joined #mojo
03:48 tetraodontidae joined #mojo
03:55 mikegrb m
03:56 mikegrb avenj: I always forget you are in here
03:56 avenj mikegrb: me too a lot of the time :x
03:57 Leandr joined #mojo
04:20 spaceships joined #mojo
04:33 russum Is there anyone who's using Mojolicious::Plugin::Mail and are the mail templates working for you? I keep getting "[error] no data in this part" when I try to load the content of the email from a template…
05:01 laouji joined #mojo
05:08 ovnimancer joined #mojo
05:10 russum I got it, I missed the part that the template needs to be in name.mail.ep format but used name.html.ep…
05:13 Leandr joined #mojo
05:30 tempire sooo
05:30 tempire I talked to the head rakudo guy
05:31 tempire during his lightning talk, he went over some serious performance improvements in the past year
05:31 Foxcool joined #mojo
05:32 tempire I reminded him that non-blocking is one of the things that's blocking a mojolicious port
05:32 tempire he said he's going to contact some people who have expressed interest in it, and see about getting that focused on.
05:32 sri cfedde: but why would the word plugin appear in Mojo::Template?
05:33 sri tempire: i think they have non-blocking sockets now
05:34 sri but there are more smaller features still missing for a full port, like DATA sections
05:34 tempire anything else?
05:35 sri i find something new every time i start ;p
05:37 tempire oh, I saw a dude with one of the raptor shirts from the kraih store
05:39 sri \o/
05:39 maettu joined #mojo
05:40 tempire About an hour ago, there was a guy who came up while I was showing mojoexample to someone, and said, "oooh, mojolicious!  that thing is awesome"
05:40 * tempire waits for the videos to be posted so he can watch his own talk
05:40 sri those are now collectibles, i lost the files for the t-shirts :o
05:48 tempire I'm disappointed that when I was answering questions afterwards, no one noticed that the nyan cat was pulsating in and out
05:52 sri oh my, there will be a .transformers tld
05:56 fhelmber_ joined #mojo
06:00 JerJer joined #mojo
06:46 marcus will it be owned by asus?
06:49 sri hasbro ;p
06:50 sri there is so much wrong with those new domains
06:50 sri amazon will get .author and .book... google .app
06:55 omega it is insane
07:23 laouji joined #mojo
07:37 Foxcool joined #mojo
07:45 Vandal joined #mojo
07:47 Mad_Dud- joined #mojo
07:47 cosmincx joined #mojo
08:00 andrefs joined #mojo
08:00 janliesendahl_de joined #mojo
08:08 diegok we should get the .shagadelic tld :-p
08:22 * sri bonks diegok on the noggin
08:22 * diegok ouch
08:26 cosmincx joined #mojo
08:31 * sri wonders if the Mojo::Template documentation should be removed
08:32 sri it seems to cause confusion every now and then
08:33 sri the description had already been shortened though
08:35 stephan48 maybe make a link to a template guide or something explaining EPL?
08:35 sri already in there
08:36 stephan48 so wheres the prob then?
08:36 sri it seems to still cause confusion
08:36 maettu left #mojo
08:37 stephan48 maybe ask these users what is confusing for them?
08:38 sri i did, but couldn't get a clear answer from anyone yet
08:39 sri teh astrangest thing is that even after knowing about the rendering guide some seem to return to Mojo::Template and get confused again
08:39 sri *the strangest
08:40 * sri shrugs
08:40 sri i'm running out of ideas
08:41 sri even links leading to Mojo::Template have been reduced to the absolute minimum now
08:41 stephan48 i dont know, i think Mojo::Template should contain a meaningfull documentation by itself
08:41 stephan48 but howto help the newbies in understanding? idk
08:42 GitHub65 joined #mojo
08:42 GitHub65 [mojo] kraih pushed 1 new commit to master: http://git.io/i_fnTA
08:42 GitHub65 [mojo/master] document alternatives to render_* methods - Sebastian Riedel
08:42 GitHub65 left #mojo
08:42 sri stephan48: what is meaningful?
08:42 sri current version you mean?
08:44 stephan48 i would say yes
08:44 stephan48 it explains all important features M::T is capable of
08:44 stephan48 and for details f.e. a sort of tutorial/guide i would make a link at the beginning to the guides
08:45 sri perhaps a small reorganization would help
08:45 stephan48 maybe
08:46 sri putting the in depth description under a different header and linking to the rendering guide earlier
08:48 diegok sri: I thing the "See Mojolicious::Guides::Rendering for more." should be moved to the first line of the description and refrased to something more like "If you are looking for how to use views in mojolicious please see ..."
08:49 diegok hm, yes, or creating a header so it goes into the toc.
08:50 diegok but I guess experienced perl programmers tend to go directly to synopsis and description first :p
08:53 GitHub124 joined #mojo
08:53 GitHub124 [mojo] kraih pushed 1 new commit to master: http://git.io/8s8z8Q
08:53 GitHub124 [mojo/master] reorganized Mojo::Template documentation - Sebastian Riedel
08:53 GitHub124 left #mojo
08:53 sri that should be better
08:56 sri http://mojolicio.us/perldoc​/Mojo/Template#DESCRIPTION
08:56 stephan48 jup
09:01 pau4o joined #mojo
09:04 batman joined #mojo
09:06 diegok well, now it's just wait and see if this make it better...
09:06 * sri waits
09:06 batman hello
09:07 batman sri: did you see my comment about io() and watch() ? is there a way to combine the two, so the order does not matter?
09:07 sri batman: nope
09:07 sri didn't see it
09:08 sri and don't understand
09:09 sri if you don't have an active i/o watcher how would you change which i/o events to watch for?
09:09 batman i did $reactor->watch($fh, 1, 0)->io($fh, CODE); which then made the CODE receive both read and write events
09:09 sri yea, that makes no sense
09:10 sri if you don't put a watcher on that handle there is nothing to change
09:10 batman ok... what about allowing watch to take a callback? ->watch($fh, 1, 0, CODE); instead of using io() at all?
09:11 batman it's no biggy, i'm just curious...
09:11 sri it's not pretty, so i'm rather opposed
09:11 batman that's fine. thanks for the feedback :)
09:12 sri i've been thinking about the reactor api a lot and couldn't come up with something more pretty that didn't involve guard objects
09:12 batman no... it's a bit "funny", but at least it's 100x more readable than using POE :)
09:13 batman but maybe ->watch() should die, unless io() was called up front?
09:13 sri reactor is also a bit special, since it's meant to be subclassed, so too much sugar api could be counter productive
09:14 batman sound reasonable
09:14 batman *sounds
09:15 batman https://speakerdeck.com/u/danwrong/p/pu​tting-the-web-back-in-web-applications <-- what a boring error message :P
09:15 sri you're welcome to play with the api and design something simpler/more pretty, everything is possible in the 3.0 release ;)
09:17 batman i will think about it, and if i come up with something i will show you a diff on github...
09:17 sri in the meantime i'll add an example
09:18 D4RK-PH0ENiX joined #mojo
09:18 batman no die "need to call io() first" ?
09:20 sri i don't like adding code ;p
09:20 sri it's not really an end user api
09:21 batman :D
09:27 GitHub131 joined #mojo
09:27 GitHub131 [mojo] kraih pushed 1 new commit to master: http://git.io/qxgOAg
09:27 GitHub131 [mojo/master] better Mojo::Reactor examples - Sebastian Riedel
09:27 GitHub131 left #mojo
09:28 batman nais
09:29 sri sometimes i wish i went with a full event loop and guard objects -.-
09:32 sri but of course then we would just have different tradeoffs
09:32 lukep joined #mojo
09:34 batman i'm glad you went with this
09:45 d4rkie joined #mojo
09:48 andrefs joined #mojo
09:58 mire joined #mojo
10:04 daxim joined #mojo
10:11 d0as8 joined #mojo
10:15 sri tempire, marcus, crab: https://github.com/kraih/mojo/issues/338 # now it's not a theoretical issue anymore
10:15 sri there is an alternative to the I18N plugin and someone to maintain it
10:17 sri tempire, marcus, crab: i expect your votes
10:17 sri and everybody who cares, please comment too
11:07 * marcus votes Cthulu for president.
11:27 janliesendahl_de joined #mojo
12:30 ovnimancer joined #mojo
12:33 tetraodontidae joined #mojo
12:34 janliesendahl_de joined #mojo
12:36 janliesendahl_de joined #mojo
12:37 tetraodontidae joined #mojo
12:38 Debolaz[Lo] joined #mojo
12:38 janliesendahl_de joined #mojo
12:40 Debolaz[Lo] I'm pondering writing a Reactor class to run Mojo on IO::Async. Should this be fairly straight-forward?
12:46 Debolaz[Lo] The motivation being so I can run the Mojo useragent inside another program I have that uses IO::Async.
12:48 sri or you just fix IO::Async::Loop::EV
12:50 sri we theoretically already support all event loops, but IO::Async::Loop::EV and POE::Loop::EV are still broken
12:50 Debolaz[Lo] sri: The issue there is EV. I'm not really comfortable using it as long as it's maintained by Marc.
12:50 Debolaz[Lo] sri: And it has an icky dependency on pthreads which complicates things a bit.
12:51 sri it's not like there are alternatives atm
12:52 andrefs joined #mojo
12:52 Debolaz[Lo] sri: How in particular is IO::Async::Loop::EV broken btw?
12:52 sri I/O events don't work at all
12:52 sri LeoNerd never finished it
12:54 Debolaz[Lo] Anyway, the main issue for me wrt EV has been to even get it to compile.
12:54 AmeliePoulain joined #mojo
12:54 Debolaz[Lo] Doesn't matter if support for it is broken in some module, if it itself doesn't work.
12:55 sri if there was a sensible alternative i would switch in a heartbeat
12:55 sri but for now it's our best bet at supporting as many event loops as possible with one backend
13:03 Debolaz[Lo] sri: I'm not sure I follow that line of reasoning, EV is one specific event loop. It has some ability to merge into other loops, but most people don't actually use that approach, they use an abstract event loop wrapper such as AnyEvent or IO::Async to accomplish it, and leave EV completely out of the loop.
13:04 sri EV is simple and tiny, it is perfect for a shared backend when running IO::Async, AnyEvent and POE at the same time
13:05 sri they all can be consumers of a single shared EV backend
13:05 Debolaz[Lo] sri: But those aren't event loops though, like I said. It may have made more sense to use AnyEvent rather than EV.
13:05 sri now i don't follow
13:06 sri AnyEvent would not have made sense at all
13:07 sri marc actively prevents people from using AnyEvent as a backend
13:08 BinGOs He doesn't seem to have noticed POE::Loop::AnyEvent yet.
13:08 Debolaz[Lo] sri: AnyEvent would make perfect sense, but if Marc gets violent over people using it, then that might be a good reason to avoid it.
13:09 sri BinGOs: hehe, only a matter of time i guess
13:09 BinGOs it has been months now.
13:09 Debolaz[Lo] sri: With AnyEvent, Mojo would use any backend already loaded into the system.
13:10 Psyche^ joined #mojo
13:10 BinGOs The IO::Async AnyEvent loop only lasted weeks.
13:10 sri Debolaz[Lo]: that's not a good thing
13:11 sri anyway, discussion is moot, AnyEvent is dead as a platform due to politics
13:12 * Debolaz[Lo] also needs to figure out a clever way of doing XMLRPC. Probably going to use templates for generating XML requests, and Mojo::DOM for parsing responses.
13:12 Debolaz[Lo] Since XML::LibXML is out of the question.
13:22 robinsmidsrod joined #mojo
13:47 gryphon joined #mojo
13:50 tempire I don't have much experience with l18n, but if that api is vouched for by several people that do, and sharifulin wants to maintain that module...
13:50 tempire that's the ideal situation.
13:50 tempire and regarding the Mojo::Template confusion...
13:52 sri marcus, crab: +1 from me and tempire then, what's your vote?
13:52 sri and no, Cthulhu does not count
13:53 tempire I'm thinking maybe a standard message at the top of all api-doc pod, saying something like "This is API documentation.  Please read the Guides first".
13:53 sri we have links at the bottom
13:53 tempire the bottom
13:54 marcus sri: my vote doesn't matter anymore then, does it?
13:54 tempire marcus: you always matter to me
13:54 * tempire swoons
13:54 sri marcus: to me too!
13:54 * marcus rolls eyes.
13:54 marcus +1 # for the swooning
13:57 iansealy joined #mojo
13:58 sri \o/
14:11 tempire http://search.cpan.org/~drolsky/Cou​rriel-0.29/lib/Courriel/Builder.pm
14:12 tempire seems sensible
14:14 tempire also, useful list for rakudo support: http://perl6.org/compilers/features
14:18 sri https://github.com/kraih/mojo/issues/338 # ok, i've offered him the Mojolicious::Plugin::I18N namespace
14:23 pau4o joined #mojo
14:39 inokenty joined #mojo
14:42 noganex joined #mojo
14:47 spaceships joined #mojo
15:08 sri and he accepted \o/
15:16 zivester joined #mojo
15:20 GitHub138 joined #mojo
15:20 GitHub138 [mojo] kraih pushed 1 new commit to master: http://git.io/pFBTBw
15:20 GitHub138 [mojo/master] removed Mojolicious::Plugin::I18N so it can be maintained as a separate distribution (closes #338) - Sebastian Riedel
15:20 GitHub138 left #mojo
15:28 sri and that was that :)
15:36 brambles joined #mojo
15:36 Leandr joined #mojo
15:39 sri anyone know what would happen if he uploads a separate Mojolicious::Plugin::I18N distribution now?
15:42 brambles_ joined #mojo
15:42 tetraodontidae joined #mojo
15:45 d4rkie joined #mojo
15:51 * marty contemplates switching from his Kubuntu workstation to a new macbook pro with retina display.
15:58 tetraodontidae joined #mojo
15:58 CromeDome joined #mojo
16:03 sri i've sent a mail to the list about the I18N plugin
16:18 * sri is jealous of marty
16:26 * marty just ordered his new macbook pro retina via the apple store.   o/    (also got one of those 27" thuderbold displays)  weeeee
16:26 sri :o
16:29 marty It's primary use will be to run final cut pro.  but I'
16:29 marty I'm going to see how I like it for development.  If I do like it, I might switch. but I kinda like my Kubuntu.   :)
16:34 tempire Next year I need to give a yapc talk about social skills, and present them as equations.
16:34 tempire final cut!
16:36 russum joined #mojo
16:36 sri "subject a: sri, just do the oppsite of what he does"
16:37 sri oh, new releases of IO::Socket::IP and IO::Socket::SSL are out \o/
16:46 cosmincx joined #mojo
16:47 GitHub54 joined #mojo
16:47 GitHub54 [mojo] kraih pushed 1 new commit to master: http://git.io/38Tl_A
16:47 GitHub54 [mojo/master] switched back from IO::Socket::INET6 to IO::Socket::IP for IPv6 support - Sebastian Riedel
16:47 GitHub54 left #mojo
16:48 sri very smooth
16:48 sri we don't use the fallback feature of IO::Socket::IP yet though
16:49 sri once i find a public host on the web that actually requires it we will though ;p
16:50 CromeDome joined #mojo
16:50 sri the high version requirement will prolly make some people unhappy, but if we want to support ipv6 properly there is no way around that
16:52 marty Sounds reasonable.  You cannot expect to support new(er) features without high(er) versions.
16:53 sri and i think ipv6 is a big enough deal
16:53 * marty nods
17:02 pau4o left #mojo
17:07 tempire I don't see any problem with high version requirements on optional modules.
17:08 tempire hmm
17:08 sri we've done it in the past and it results in some complaints, because linux distros are always way behind
17:09 sri then again, those might be the same folks that use the backport fork :)
17:09 tempire return $self->redirect_to('perlbrew.pl');
17:10 sri that's not an absolute url
17:10 * tempire kicks sri's url absolutely
17:11 * sri kicks his url absolutely too
17:13 * sri would really like a host that resolves to ipv6 and ipv4 but only listens for ipv4
17:13 sri for science
17:13 marty I remember when I suggested perlbrew mitigated the need to maintain support for legacy versions and a lot of people disagreed.  I think mostly sysadmins who maintain a lot of machines and need to use the system perl.
17:14 sri they always use centos
17:21 tempire so someone needs to build and maintain an rpm for centos
17:23 tempire anyone in here have experience with building rpms?
17:24 tempire though I understand some sysadmins will still want to use the redhat-guaranteed distro
17:24 mire joined #mojo
17:26 asarch joined #mojo
17:40 Foxcool joined #mojo
18:23 sri *crickets*
18:23 * sri hugs perlbrew
18:28 spaceships joined #mojo
18:29 tetraodontidae joined #mojo
18:33 CromeDome joined #mojo
18:37 sh4 joined #mojo
18:41 GitHub141 joined #mojo
18:41 GitHub141 [mojo] kraih pushed 1 new commit to master: http://git.io/knthpQ
18:41 GitHub141 [mojo/master] mention version requirement of optional dependencies - Sebastian Riedel
18:41 GitHub141 left #mojo
18:46 GitHub27 joined #mojo
18:46 GitHub27 [mojo] kraih pushed 1 new commit to master: http://git.io/xQ1uXA
18:46 GitHub27 [mojo/master] mention version requirements everywhere - Sebastian Riedel
18:46 GitHub27 left #mojo
18:52 jfuller joined #mojo
18:54 jfuller I have run into a situation where I need to make a non-blocking call to a credit card processor. I am not exactly where to even begin. Basically, I want to make sure it doesn't block other requests while waiting, but I want the initiating request to wait until the result comes back.
18:57 GitHub135 joined #mojo
18:57 GitHub135 [mojo] kraih pushed 1 new commit to master: http://git.io/Dliq9w
18:57 GitHub135 [mojo/master] explain which Perl features are automatically enabled in templates and config files - Sebastian Riedel
18:57 GitHub135 left #mojo
18:58 tempire jfuller: just make a useragent request call and do something with the response.
18:58 tempire and be wary of the timeouts involved
19:00 jfuller tempire: This is a legacy system, basically I am inserting a record, then checking it once a second looking for the processed value. There is another service that monitors that database and actually handles the communication with the processor.
19:02 jfuller So, I guess what I'd like to know is how to not block while that loop is running
19:06 sri http://mojolicio.us/perldoc/Mojoliciou​s/Guides/Cookbook#Backend_web_services
19:07 andrefs joined #mojo
19:10 jfuller sri: So, you're saying I should setup another site that handles processing credit cards?
19:11 sri nope, i assumed it was http
19:12 jfuller sri: Basically, what I am doing is waiting for the value in a record in the db to change.
19:13 sri sounds messy
19:13 * sri hides
19:13 jfuller sri: lol, yeah. Its from a legacy system
19:32 tempire http://act.yapcna.org/2012/talk/146
19:32 tempire seems like this talk will be interesting, about to start
19:32 tempire "Why MVC is the wrong way to think about Web development" <- bold claim
19:33 tempire marcus ^ hateos references
19:37 sri "...choosing tools and frameworks that do not fully take advantage of all that HTTP has to offer..." <- i bet his framework doesn't either -.-
19:40 sri in fact, i bet on a simple webmachine clone
19:42 tempire well, he's basically saying mvc doesn't match perfectly to every aspect of web dev.  which is quite true.
19:49 sh4|2 joined #mojo
19:51 sri if anything MVC is on a comeback :)
19:51 sri now that the big ones have proven that heavy client side apps mostly suck
19:57 tempire have they?  it seems to me that the proper structure to keep it simple just hasn't been found
20:12 hipyhop joined #mojo
20:50 hipyhop joined #mojo
21:19 tempire oooooh
21:19 tempire look how pretty https://github.com/bkeepers/gaskit
21:48 tempire so it looks like travis has a value past cpantesters
22:21 russum left #mojo
23:56 tempire trying out github auto pages
23:56 tempire http://tempire.github.com/mojo​licious-command-deploy-heroku
23:56 tempire seems a bit too generic

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