Camelia, the Perl 6 bug

IRC log for #mojo, 2011-08-02

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

All times shown according to UTC.

Time Nick Message
00:17 sri tempire: why not $self->param('foo')?
00:18 * tempire has no good reason
00:18 tempire shortcuts are for suckers
00:18 tempire there's my reason
00:18 * sri bonks tempire on the noggin
00:20 sri "To find your robot name, take the first 16 digits of your credit card & combine with the start/expiry dates and security code. What's yours?"
00:20 sri lol!
00:27 xaka left #mojo
00:30 Kulag left #mojo
00:36 Kulag joined #mojo
00:37 sri cosimo: no need to mention handler for every render call
00:38 sri we have handler auto detection
00:38 sri and html is the default format
00:42 SmokeMachine joined #mojo
00:43 sri http://mojolicio.us/perldoc/Mojo​licious/Renderer#default_handler
00:43 sri it's really just a fallback
00:44 Kulag left #mojo
00:52 Alias_ left #mojo
00:52 perlrocks Twitter: "I still do Mojolicious." (ja) --taiju http://twitter.com/taiju/status/98194441612951552
01:00 sri cosimo: the mojolicious renderer is all about allowing you to use multiple template systems parallel
01:01 sri since different template systems have different strengths
01:03 stephen left #mojo
01:13 preflex left #mojo
01:48 Alias_ joined #mojo
01:52 xaka joined #mojo
02:12 abra left #mojo
02:16 preflex joined #mojo
02:21 stephen joined #mojo
02:39 sri http://chocolatapp.com/ # ooooh, this looks nice
02:39 sri like what i would expect from textmate2
02:41 crab display multiple files at once!
02:47 sri love the minimalism
02:47 tempire anyone used skillshare.com?
02:49 jwang_ joined #mojo
02:57 jwang_ left #mojo
03:08 sri woah, it gets so much hate on hackernews
03:09 sri http://hackerne.ws/item?id=2835346
03:21 perlrocks Twitter: "morbo helo.pl #mojolicious mee like" --nyet http://twitter.com/nyet/status/98231807631306753
03:21 Alias_ You know what I hate about Mac projects?
03:21 Alias_ Their websites are always do good
03:21 Alias_ :(
03:21 * Alias_ sighs at Padre
03:21 Alias_ The main design principle of our website was "Content should be abstracted from style"
03:21 sri not like padre itself is pretty
03:22 Alias_ That become more important than EITHER the actual content or the actual style
03:22 Alias_ sri: Give me some time
03:22 sri i would be all over it if it was
03:22 Alias_ We've never had graphical gui design until last month
03:22 sri you might want to take a look at sublime text 2
03:23 sri they've taken cross platform editors to a whole new level
03:23 Alias_ http://padre.perlide.org/trac/wiki/Screenshots
03:23 Alias_ That's still fairly ugly, but you can see that at least with the Form Builder dialog we're closer to at least using fonts and iconography
03:23 sri Alias_: http://www.sublimetext.com/x
03:24 Alias_ And we can play with and change the layout every release with no real effort
03:24 sri it's mostly python
03:24 Alias_ Form Builder already looks quite different
03:24 Alias_ "Sublime Text X has graduated to become Sublime Text 2"
03:25 Alias_ I find that fairly ugly
03:25 Alias_ All style, little substance
03:25 sri it is completely themeable
03:25 Alias_ (Although I get the point)
03:25 sri https://github.com/buymeasoda/soda-theme/
03:26 Alias_ The scaled down version of the file on the right is interesting
03:26 Alias_ And I might steal their find and replace stuff at some point
03:26 Alias_ Anyways
03:26 Alias_ I reaffirm that Padre has been historically ugly because it had to be
03:27 Alias_ And hopefully with this new generation of dialog code that should start to be fixable
03:27 SmokeMachine left #mojo
03:28 Alias_ I like the option buttons and such
03:28 Alias_ I shall add this to my list of editors to steal good ideas from
03:29 sri if i spend all day working with a program, it better be drop dead gorgeous
03:29 Alias_ I wonder what gui toolkit they use
03:30 stephan48 left #mojo
03:30 Alias_ I mostly want mine to be fast and not distract me
03:30 sri custom toolkit, specifically avoiding lowest common denominator
03:30 Alias_ Commercial right?
03:30 sri ye
03:31 Alias_ Righto then
03:31 Alias_ I try to resist comparing us to commercial code
03:31 sri i still wonder why you didn't go the komodo route and reused mozilla xul
03:31 Alias_ If I could work on Padre all day it would be SO much better already
03:32 Alias_ sri: Because I (and in this case, it's my opinion the project adopted) think that the most successful development tools long term are written in the language they are used to program
03:32 Alias_ Emacs, Eclipse,
03:33 Alias_ If you have to learn Javascript to do the GUI design for a Perl editor, it closes off the number of people that can work on it
03:33 Alias_ Makes perfect sense for a Browser to be partly written in web technologies or lookalikes, because that's the skillset of their primary userbase
03:33 sri not like wx is that much more encouraging
03:34 Alias_ Having wxFormBuilder improves things greatly
03:34 Alias_ Let me show you the total code for the new Preferences dialog
03:34 Alias_ http://cpansearch.perl.org/src/PLAVEN/Padr​e-0.86/lib/Padre/Wx/Dialog/Preferences.pm
03:34 Alias_ That's all of it
03:35 stephan48 joined #mojo
03:36 Alias_ That's not so far away making the Wx API surface as tractable in Perl as the DOM API is in JavaScript
03:36 sri dunno, i just couldn't develop ugly guis
03:36 Alias_ Nothing to stop you using Wx to develop prettier ones
03:36 Alias_ Just more work
03:36 Alias_ It's a matter of priorities
03:36 sri i've never seen a pretty wx app
03:37 Alias_ I want a working MANIFEST-aware folder recursive search and replace before I want it pretty
03:37 Alias_ BTW, themeability is coming
03:37 Alias_ Right now there's a really crude 1.0 implementation for Scintilla editors only
03:38 sri i couldn't care less about such advanced editing features
03:38 Alias_ But I'm chipping away at a 2.0 replacement that at least extends colour/font/etc control out to other gui elements
03:38 Alias_ sri: What about not having to select a file name when you save
03:38 sri if i was designing an editor i would make it pretty and only provide the most basic tools
03:39 sri then let plugins take care of the rest
03:39 Alias_ Who's going to write the plugins?
03:39 sri everybody
03:39 purl well, everybody is depraved
03:39 Alias_ Notepad++ is a simple editor with plugins
03:39 sri see textmate bundles
03:40 Alias_ People don't do your work for you until you're already successful
03:40 sri thsese days there are addons for different sidebars and lion fullscreen
03:40 Alias_ Anyways, this discussion is moot :)
03:41 Alias_ Chocolate has some fairly neato dialog designs
03:41 sri it covers the basics very well
03:43 sri i don't think you need too many features to reach critical mass, in fact less features and being scriptable in perl could be an advantage
03:43 sri your killer feature is being scriptable with perl for sure
03:43 Alias_ Yup
03:43 Alias_ That's one of the main reasons to stick with Wx in the end
03:43 Alias_ The other being installability from cpan
03:44 Alias_ The strategic value of "cpan -i Padre" is pretty lartge
03:44 Alias_ large
03:44 Alias_ It both simplifies packaging downstream, and it's a blatant statement "We are Perl People"
03:45 perlrocks Twitter: "Mojolicious:: Lite in anyway re-inventing" (ja) --narazaka http://twitter.com/narazak​a/status/98238080460406784
03:47 sri padre might already be too complex to ever become pretty
03:47 Alias_ Nah, I've been careful
03:47 sri for an editor it requires a certain level of minimalism
03:47 Alias_ There's been craptons of refactoring this year
03:47 Alias_ And then there's the feature_ settings
03:47 Alias_ Which we don't expose yet
03:48 Alias_ feature_* settings let you disable entire features you don't use
03:48 Alias_ Removes everything to do with them, including the code bloat, menu entries, key bindings, the lot
03:49 sri my first question would be, why have an icon bar?
03:49 Alias_ Personally, I turn off code folding, bookmarks, cursor memory, sessions, the debugger
03:49 Alias_ You can turn that off too
03:49 sri yea, but why have it at all?
03:49 sri nobody really uses that crap
03:49 Alias_ I use some of it
03:50 Alias_ We'll know that soon
03:50 Alias_ There's a "popularity contest" plugin that comes in the default install
03:50 Alias_ Which will count the number of times you use each menu entry and which toolbar buttons you click, and a few other things
03:50 perlrocks Twitter: "Mojolicious-Plugin-PlackMiddleware 0.21 by JAMADAM - http://metacpan.org/release/JAMADAM/Mo​jolicious-Plugin-PlackMiddleware-0.21/" --cpan_new http://twitter.com/cpan_ne​w/status/98239249127714816
03:50 Alias_ The reports those counts to us
03:50 Alias_ Then
03:51 Alias_ So we'll be able to tell which features people (or at least, the ones that are willing to tell us) actually use
03:52 Alias_ BTW, it's off by default on Mac :)
03:52 sri that's good :)
03:53 Alias_ Thank shlomi
03:53 Alias_ He files a bug report saying no Mac applications have toolbars
03:53 Alias_ Which isn't true, it turns out, but seems to be true for many simple normal things
03:53 Alias_ So we made it default off
03:54 Alias_ And nobody has cared enough to force it back
03:55 Alias_ You've just given me a great idea though...
03:55 * Alias_ ponders
03:55 sri native toolbars have a whole different style
03:55 Alias_ Ditto windows now tbh
03:55 sri those cross platform ones are a total no go
03:55 Alias_ It's lowest common
03:55 Alias_ New windows stuff uses ribbons
03:57 Alias_ I'm mainly trying to avoid change at the moment
03:57 Alias_ I'll start experimenting once the popcon numbers start accumulating
04:06 sri http://www.cpantesters.org/cpan/report​/45e05890-ba13-11e0-aefb-871f66328602
04:06 sri :S
04:06 * sri runs to the meme generator
04:07 Alias_ sri: Dependency on Task::Weaken needed?
04:07 Alias_ Not that you can actually do anything about that really
04:08 Alias_ But might want to steal the code it uses to "detect" weaken
04:09 sri i think the question really is why would there exist a perl without weaken?
04:10 Alias_ Because that's the way Perl compiles by default
04:10 Alias_ Or used to
04:10 Alias_ So if you did the build without turning it on, you got that problem
04:11 Alias_ MOST places the downstream distribution builders turned it on
04:11 Alias_ But not RedHat
04:11 Alias_ So on RedHat... erm... 4 and a few other places with default-build corporate builds (Oracle's bundled Perl maybe) you don't get weaken
04:12 sri i've never built a perl without weaken
04:12 sri and i mostly stick with the defaults
04:12 Alias_ Neither have I, but others have, and I've encountered them
04:12 Alias_ That default off was fixed in Perl 5.10.something?
04:12 Alias_ So mostly you encounter it in 5.8.8 era machines
04:12 sri that would explain it
04:13 sri oh, this is 5.15
04:13 sri prolly just broken
04:13 Alias_ yup
04:13 Alias_ Or he's doing some kind of weird flag-search break thing
04:13 Alias_ Testers with rare settings switched to see which modules behave
04:14 Alias_ I keep meaning to set up a CPAN Testers box with Perl installed into a path with spaces in it
04:14 Alias_ Or unicode directories
04:14 Alias_ etc
04:15 Alias_ oh
04:15 Alias_ He might have the "no compiler" setup
04:15 Alias_ And it's installed a newer version of Scalar::Util which overwrites the default one
04:16 Alias_ And the non-XS one has no weaken
04:17 sri http://images.memegenerator.n​et/instances/500x/9183359.jpg
04:18 Alias_ You can set an explicit dep in the most recent versions
04:20 Alias_ hrm
04:20 Alias_ ok, even more changes in dev versions
04:20 Alias_ never mind
04:20 * Alias_ tries to concentrate on work
04:20 Alias_ later
04:20 Alias_ left #mojo
04:22 stephen left #mojo
04:26 sri http://blog.chromium.org/2011/08/new​-websocket-protocol-secure-and.html
04:26 sri :)
04:45 crab sri: did you implement binary support already?
04:46 sri how and why?
04:46 crab how and why what?
04:47 sri binary support
04:47 crab no idea, you mentioned it the other day and the chromium announcement referred to it, so i remembered.
04:53 sri not sure how to do it yet
04:54 sri we prolly also need frame size control
05:13 garu sri: in html, is it legal to have something like "<table> <tr> <tr>"?
05:13 garu I'm asking because I just saw it, and Mojo::DOM is jumping over the second tr, since it thinks it's a child of the first one
05:14 garu (while the web browser correctly fixes this by adding a </tr> just before the new one
05:14 garu )
05:15 garu I know it's an impossible fight against crappy my-nephew-did-it-for-5-bucks websites, but they do exist and sometimes we just gotta parse them =(
05:15 sri garu: don't know, you'll have to check the specs
05:16 GitHub84 joined #mojo
05:16 GitHub84 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/96​1b119835b5d3fc38b3fd9e600111006b216e63
05:16 GitHub84 [mojo/master] improved 64bit support in Mojo::Transaction::WebSocket - Sebastian Riedel
05:16 GitHub84 left #mojo
05:16 garu sri: if I check and it's not possible, do you think we can add this "fix" to Mojo::DOM?
05:17 sri if the spec says so, sure
05:17 garu yay
05:17 * garu goes to read some w3c docs :)
05:19 sri garu: i think you're wrong though
05:19 sri as in, Mojo::DOM already handles that case
05:20 abra joined #mojo
05:21 garu sri: can I show you the test case I'm looking at?
05:21 abra left #mojo
05:21 sri sure
05:21 sri (if it's small)
05:23 crab wow. i downloaded sublime text 2 for linux, and it displays a gtk warning for every single character i type.
05:24 * crab runs screaming back to xterm+vim
05:24 sri heh
05:28 garu sri: http://paste.scsys.co.uk/128108
05:28 garu maybe I'm missing something
05:28 garu that crappy html already melted most of my brain cells
05:28 sri eeep
05:29 garu I know
05:29 garu :)
05:29 sri can you reduce it to a minimal test case?
05:29 garu I can try
05:29 garu (just showing you a live sample first :)
05:30 sri chances are it's too broken
05:30 sri since we already handle <tr><tr> rather well
05:31 sri which would mean you have to tweak the actual html healing in Mojo::DOM::HTML
05:33 sri crab: we need to rethink the whole websocket api
05:35 sri or maybe not :)
05:36 sri actually on_message should work fine for chars and bytes
05:36 sri for sending we could use send_bytes in addition to send_message
05:40 garu sri: ok, looks like I found what's confusing the dom
05:41 garu it's not the <tr><tr> like you said
05:41 garu it's <tr><td><div></td>
05:41 garu sri: can I interest you in a failing test? :)
05:42 sri that might be a very hard case
05:43 sri since div is a block element
05:44 sri don't really have a strategy for fixing that
05:44 Cside left #mojo
05:45 plu you can feel the browser's developers pain :)
05:45 garu sri: not my fault people hire their 5 year old nephews to write html. Also, not my fault Mojo::DOM is so bloody awesome I want to use it everywhere =D
05:45 garu sri++
05:51 crab garu: my daughter is learning html in school. i asked her yesterday "what happened in the computer lab today?" and she said "we learned how to use the marquee element"
05:51 crab imagine my horror.
05:51 sri :O
05:51 garu oO
05:52 crab then i discovered that it's quite hard to explain the virtues of separating content from presentation to a "little child" (as she refers to herself).
05:53 sri garu: you're wrong again though
05:53 garu sri: I can't begin to pretend I understand Mojo::DOM::HTML's source
05:53 sri garu: that case gets fixed too
05:54 garu it's odd, because if I add the </div>, I get all 5 elements instead of just 2
05:54 marcus morning
05:54 garu but I'll reduce it as much as I can to help nail this
05:54 crab there should be a marcus element.
05:54 garu (it's in my best interest, after all :)
05:54 sri </marcus>
05:54 crab or marqus.
05:55 marcus thanks for the warm welcome :p
05:55 sri garu: chances are it's not such a hard case now
05:55 garu crab: I thought that was what the marquee element was all about ;)
05:55 * sri hugs marcus
05:56 sri what does the marcus element actually do?
05:56 marcus it makes stuff awesome
05:58 sri <marcus>
05:58 sri lets see!
05:59 sri garu: seriously, no matter what i throw at it… the dom repairs it :S
06:00 garu I'm almost finished now, 2 minutes :)
06:01 marcus see!
06:02 garu http://paste.scsys.co.uk/128109
06:02 garu sri: ^^
06:03 sri is that all necessary? :o
06:04 * garu further shortens the demo
06:05 marcus garu: at this point it could just as well be a unit test :)
06:05 sri garu: no need, i have the case isolated
06:05 garu yay
06:06 sri it could be hard though
06:06 sri <table><tr><td><div></td><td></td></tr></table>
06:07 Foxcool joined #mojo
06:10 sri and it is a hard case
06:10 garu sri: it's to_xml'ing to <div />
06:11 garu is that even possible?
06:11 sri we don't verify that
06:12 sri so far we only know about block elements and options end tags
06:12 sri s/options/optional/
06:14 sri <table><tr><td><div><td /></div></td></tr></table>
06:14 sri that's what it is repairing it to here
06:14 sri which is not the worst outcome
06:15 garu if we did verify that, would the issue go away?
06:15 sri i wouldn't know how
06:20 koban joined #mojo
06:20 garu does this help at all? http://paste.scsys.co.uk/128110
06:21 sri not at all
06:21 garu :(
06:21 sri i know exactly what breaks the case
06:21 sri but the problem is very deep in the healing strategy
06:25 sromanov joined #mojo
06:35 marcus http://blog.chromium.org/2011/08/new​-websocket-protocol-secure-and.html
06:35 jwang left #mojo
06:46 perlrocks Twitter: "/ Mojolicious reference to perform HTTP requests in parallel - http://t.co/OtmA9Fw Introduction to Perl with sample code # perl # mojolicious" (ja) --perlcodesample http://twitter.com/perlcodesa​mple/status/98283498187653120
06:46 AmeliePoulain joined #mojo
06:47 perlrocks Twitter: "Execute parallel HTTP requests / Mojolicious Reference http://t.co/OtmA9Fw # perl # mojolicious" (ja) --perlcodesample http://twitter.com/perlcodesa​mple/status/98283655830568961
07:06 cosmincx joined #mojo
07:11 xaka left #mojo
07:17 cosimo sri: yeah, thanks (renderer, defaults, ...)
07:21 dekimsey left #mojo
07:23 dekimsey joined #mojo
07:29 sri garu: i have a solution
07:29 sri it's a little radical though
07:35 sri i'm a little scared
07:36 sri this could result in may new broken cases
07:37 garu I like it already :P
07:39 spleenjack joined #mojo
07:43 marcus ohoh
07:51 sri tag closing could be too aggressive now
07:51 GitHub18 joined #mojo
07:51 GitHub18 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/ad​0b3b3fcaacffe39fea34b126cd927e3f02ec78
07:51 GitHub18 [mojo/master] improved HTML healing capabilities of Mojo::DOM::HTML - Sebastian Riedel
07:51 GitHub18 left #mojo
07:52 sri we'll see
07:52 * sri pokes garu
07:52 sri at some point i might get fed up and switch to a slower state machine parser ;p
07:54 garu hehehe
07:54 garu sorry, 5am here
07:55 garu I get a little slow :P
07:55 * garu tries
07:58 cosimo left #mojo
08:04 garu sri++
08:04 garu but.
08:04 garu it may have broken stuff
08:04 garu I mean
08:05 garu that little test I pasted? it's still broken, but now it's broken differently
08:05 * garu pastes
08:08 garu sri: hmm... is it expected to translate <td></td> into <td /> ?
08:08 sri yes
08:08 sri it doesn't know better
08:09 garu cool, then it works beautifully :D
08:09 sri and i think that might be valid in xhtml5
08:09 sri if it's not we can fix that very easily
08:09 sri like in 2 minutes
08:10 sri but someone has to check the spec first *cough*
08:10 garu http://paste.scsys.co.uk/128113
08:10 garu spec? what spec?
08:10 * garu hides
08:10 * sri cracks the whip
08:11 nickl joined #mojo
08:12 garu sri: all 4 tests on that paste work now btw :)
08:12 sri \o/
08:12 garu sri++
08:15 garu oO
08:15 garu html5 is still a working draft?!
08:15 jamesw forever
08:15 garu heh
08:15 jamesw no really
08:16 sri rly
08:16 garu oO
08:16 sri html5 is a moving spec
08:17 sri there are no plans to ever *finish* it to my knowledge
08:17 garu wow, most of the topics start with "This section is non-normative"
08:18 garu talk about "anythong goes"
08:18 sri :)
08:19 garu crab: "The use of presentational elements leads to poorer accessibility"
08:20 garu you can tell your daughter the people that invented html5 said so :)
08:26 garu wow, there's an element named "ruby"
08:31 garu how lovely. The draft skips from 2.6 directly to 4.3
08:31 garu and, of course, element semantics are defined in 3.2 :)
08:32 garu ok, just the draft is empty
08:32 garu meaning... they're changing that bit??
08:40 tabbi joined #mojo
08:40 tabbi left #mojo
08:41 vasundhar joined #mojo
08:41 garu ... <div /> ... => This document was successfully checked as XHTML 1.0 Strict!
08:44 garu *but*
08:44 garu not in html5
08:45 garu sri: ...<div />... (validating againts <!DOCTYPE html>) => "Self-closing syntax (/>) used on a non-void HTML element. Ignoring the slash and treating as a start tag."
08:46 perlrocks Twitter: "Further study group meeting mojolicious INTRODUCTION" (ja) --mountainK http://twitter.com/mountain​K/status/98313695200940033
08:48 sromanov left #mojo
08:51 garu sri: void elements in html5 => area, base, br, col, command, embed, hr, img, input, keygen, link, meta, param, source, track, wbr
08:52 garu 8.1.2 of http://www.w3.org/TR/html5/author/spec.html
08:52 garu (finally)
08:55 garu wow, there are some funky rules about attributes
09:03 garu sri: ok, says there divs are "normal elements" and, of those, only html, head, body, li, dt, dd, p, rt, rp, optgroup, option, colgroup, thead, tbody, tfoot, tr, td, and th (phew) can have start/end tags ommited - and even so, only under very special and specific circumstances
09:03 garu damn, 6am
09:04 garu bedtime for me! cya guys
09:04 garu sri: thanks again for the help :)
09:04 garu &
09:04 sri garu: Xhtml5
09:04 sri nn ;p
09:06 arthas joined #mojo
09:49 vasundhar left #mojo
09:52 fmerges joined #mojo
10:01 xaka joined #mojo
10:02 sri http://i.imgur.com/0xVis.jpg
10:02 sri haha
10:04 sri love those clouds
10:10 xaka left #mojo
10:10 Foxcool left #mojo
10:10 marcus <3
10:13 Foxcool joined #mojo
10:14 Foxcool left #mojo
10:15 Foxcool joined #mojo
10:31 Foxcool left #mojo
10:37 GitHub182 joined #mojo
10:37 GitHub182 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/8a​9e0cee38904821ed6f3a94f6f30225323e4de2
10:37 GitHub182 [mojo/master] improved HTML rendering in Mojo::DOM::HTML - Sebastian Riedel
10:37 GitHub182 left #mojo
10:46 Foxcool joined #mojo
10:50 sri garu: ;)
10:52 vasundhar joined #mojo
10:55 mhi tempire: Thanks! (re: controller action)
10:56 GabrielVieira left #mojo
11:03 Foxcool left #mojo
11:07 Foxcool joined #mojo
11:13 Kulag joined #mojo
11:30 perlrocks Twitter: "/ Mojolicious reference to perform HTTP requests in parallel - http://bit.ly/nikQ5X Introduction to Perl with sample code" (ja) --wPGnews http://twitter.com/wPGnews​/status/98354897577181184
11:34 sri ohoh, that example looks flawed
11:36 crab how so?
11:36 sri http://mojolicio.us/perldoc/​Mojo/IOLoop#max_connections
11:39 sri quite sure he misunderstood that
11:40 sri not to mention that called on the class it just breaks
12:01 sri i hate having wrong examples out there, but no idea how to contact him :/
12:18 cosmincx1 joined #mojo
12:19 crab sri: er, just tweet to @perlcodesample?
12:20 cosmincx left #mojo
12:21 sri crab: the explanation doesn't really fit into 14 chars
12:21 sri *+0
12:21 sri considering the langauge barrier
12:21 sri eeep
12:21 mhi Upload a paste and tweet the URL. :)
12:21 sri sri no engrish? unpossible!
12:25 crab sri: so send mail to kimoto.yuki@gmail.com
12:27 sromanov joined #mojo
12:32 sri crab: afraid my attention span is too short :/
12:32 sri now i'm thinking about a new ioloop feature
12:33 vasundhar left #mojo
12:35 sri http://pastie.org/2308808
12:35 sri how about something like this to get rid of the stupid counters
12:36 sri not quite a condvar
12:38 sri still quite a bit too verbose i guess
12:42 crab beign/end doesn't seem too verbose to me.
12:42 sri it's not elegant enough
12:44 sri have a better idea though :)
12:50 sri hmm, actually i don't have a better idea :S
12:52 sri and it doesn't really help with our main use cases
13:02 GitHub37 joined #mojo
13:02 GitHub37 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/c6​167b4f9902dac97ec97b1e5ae45a7674e6272a
13:02 GitHub37 [mojo/master] fixed missing quote - Sebastian Riedel
13:02 GitHub37 left #mojo
13:14 abra joined #mojo
13:35 AmeliePoulain left #mojo
13:35 AmeliePoulain joined #mojo
13:54 sri http://pastie.org/2309132
13:54 sri how about this?
13:54 purl this is true
13:54 sri the callback is optional
13:54 AmeliePoulain left #mojo
13:54 sri ->start too
13:55 sri that way you can "trigger" a callback even inside a mojolicious app after multiple async operations
13:55 sri like a render call
13:56 sri ->start is a bit syntactic sugar for stand alone clients
13:57 sri it registers an event that stops the ioloop when end has been called the last time
13:58 sri much more like a condvar
14:01 perlrocks Twitter: "By the way: For those who find it cool # Sinatra, but would prefer instead of # # Perl Ruby, gibts # Mojolicious http://mojolicio.us/" (de) --nkoehring http://twitter.com/nkoehrin​g/status/98392989566631937
14:13 xaka joined #mojo
14:23 crab trigger seems a strange name for that
14:23 sri crab: got a better idea?
14:23 crab i'm thinking
14:24 sri seems appropriate considering it remote controls the event loop
14:27 crab like a bomb trigger?
14:28 Foxcool left #mojo
14:28 sri ye
14:29 crab but that just goes boom when you press it, it doesn't wait a balanced of presses/releases. and "trigger" in cs terms is usually a verb, so it's doubly confusing.
14:29 sri heh
14:29 sri come up with a better name ;p
14:30 crab seriously. i'm thinking about what to suggest.
14:30 sri it basically synchronizes events
14:31 sri and controls a loop from the outside
14:32 sri trigger also because it will trigger a boom when it's done
14:33 sri there is usually at least one callback at the end
14:33 sri my example actually uses a bomb timer :D
14:35 koban left #mojo
14:36 sri http://pastie.org/2309347
14:36 sri ;)
14:37 crab that would be much better
14:37 crab in that case, it's clear that trigger isn't a verb. it's still a bit unfamiliar, but that's ok i think.
14:57 cosmincx1 left #mojo
14:59 minimalist left #mojo
15:01 kaare joined #mojo
15:05 ph1g joined #mojo
15:15 minimalist joined #mojo
15:16 GitHub155 joined #mojo
15:16 GitHub155 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/11​117830dc974126fcb4e941017e5883a307dfaa
15:16 GitHub155 [mojo/master] added experimental module Mojo::IOLoop::Trigger - Sebastian Riedel
15:16 GitHub155 left #mojo
15:17 cosimo joined #mojo
15:17 cosimo ehm. I don't get the default helpers with the TT renderer. Is that correct?
15:18 sri all helpers can be called on the controller instance
15:18 sri so as long as that is available you can call them as methods
15:18 cosimo not from the template though
15:18 sri from everywhere
15:19 sri i can only guess, but it's prolly [% c.url_for %]  or so
15:19 cosimo alright, then something's not working right. thanks, i'll double check
15:28 GitHub33 joined #mojo
15:28 GitHub33 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/00​b0e39d7bc72474c2d84d05b361942bd8e4f1cb
15:28 GitHub33 [mojo/master] documentation tweaks - Sebastian Riedel
15:28 GitHub33 left #mojo
15:33 perlrocks Twitter: "Parallel requests should be a lot more fun with the #mojolicious user agent now. :) http://t.co/J82Btsq #perl" --kraih http://twitter.com/kraih/status/98416066736304128
15:35 cosimo sri: [% c.flash('something') %] works
15:36 cosimo i was missing the "c."
15:37 crab osimo
15:39 xaka left #mojo
15:39 mattastrophe joined #mojo
15:40 ph1g_ joined #mojo
15:42 GitHub166 joined #mojo
15:42 GitHub166 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/07​7dcdab6ad6ae66ed2ec7549b9319f1f011125f
15:42 GitHub166 [mojo/master] made tests more portable - Sebastian Riedel
15:42 GitHub166 left #mojo
15:45 ph1g left #mojo
15:47 ph1g_ left #mojo
15:49 jwang joined #mojo
15:53 xaka joined #mojo
15:54 GitHub170 joined #mojo
15:54 GitHub170 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/e2​411b5b49e425369a580d6fdf1b3da4e0359207
15:54 GitHub170 [mojo/master] Merge pull request #191 from Akron/doc-bug - Sebastian Riedel
15:54 GitHub170 left #mojo
15:55 cosimo left #mojo
15:55 ph1g joined #mojo
16:04 sh9 joined #mojo
16:04 nickl left #mojo
16:06 tempire Has anyone ever interacted with John Napiorkowski?
16:07 sri i might have… a long long time ago
16:09 marcus he's been on #catalyst afaik
16:11 ispy_ joined #mojo
16:13 mattastrophe left #mojo
16:17 tempire I don't think I could handle #catalyst anymore
16:17 tempire #mojo is so friendly.
16:17 tempire I've grown soft.
16:23 * sri punches tempire on the nose
16:28 sh7 joined #mojo
16:31 sh9 left #mojo
16:37 * tempire bleeds glitter
16:40 amoore joined #mojo
16:45 metaperl joined #mojo
16:54 alnewkirk lol
17:03 spleenjack left #mojo
17:04 sromanov left #mojo
17:25 stephen joined #mojo
17:34 fmerges left #mojo
17:39 tabbi joined #mojo
17:39 tabbi left #mojo
17:42 sromanov joined #mojo
17:43 mattastrophe joined #mojo
17:57 GitHub121 joined #mojo
17:57 GitHub121 [mojo] kraih pushed 1 new commit to master: https://github.com/kraih/mojo/commit/41​c1ef83ff59ac5a7be213243193d0d26023b32d
17:57 GitHub121 [mojo/master] fixed memory leak in Mojo::IOLoop::Client - Sebastian Riedel
17:57 GitHub121 left #mojo
18:05 sri hmm
18:05 sri it would be fun to rename Mojo::IOLoop::Trigger to Mojo::IOLoop::Unicorn
18:14 tempire is there an obscure or relatively functional reference in there?
18:14 tempire because that statement seems true for every module on cpan
18:28 amoore left #mojo
18:30 elb0w Didnt you used to be able to access all of cpan from mojolicious.us ?
18:30 elb0w mojolicio.us
18:36 sri tempire: naah, was just thinking we should have a unicorn module
18:36 sri elb0w: no
18:36 tempire I definitely support it.
18:36 tempire Whether Mojo::IOLoop::Trigger is the right one...
18:37 tempire The unicorn module should do something useful and relevant to its name
18:37 tempire or maybe unicorn could be a related project
18:38 tempire like mojo::dom could be unicorn
18:38 tempire Which would make the statement "Mojolicious: with Unicorns" true
18:39 SmokeMachine joined #mojo
18:39 tempire though Mojo::DOM is usefully descriptive, so maybe something else
18:43 elb0w http://elbowrage.com/europe/IMG_1032.jpg wearing a raptor like a boss
18:44 elb0w cant wait for winter
18:44 elb0w gonna get the hoody :D
18:48 jwang_ joined #mojo
18:50 elb0w pfft stable
18:50 elb0w whoops
18:50 sri :D
18:52 elb0w walked down the stairs at santorini. Lets just say if you are ever in need of Donkey sh*t I know the place
18:53 tempire Why is the Perl guy the farthest away from the girls?
18:53 elb0w im the perl guy
18:53 elb0w we just walked down 600 stairs
18:54 tempire The raptor must be too much for them.
18:54 tempire Girls are weak
18:54 tempire rawr!
18:56 marty left #mojo
18:57 elb0w I bring perl to new heights, http://elbowrage.com/europe/2.jpg
18:57 elb0w haha
18:58 tempire top of the food chain
18:58 elb0w literally
18:58 marty joined #mojo
19:02 abra left #mojo
19:10 mspo mm santorini
19:11 mspo elb0w: did you make it out to that big black outcropping?
19:11 mspo elb0w: http://mspo.com/pics/greec​e2006/images/DSC_0084.jpg
19:16 elb0w no where is that?
19:16 elb0w whats there?
19:16 purl there is
19:18 sri treasure?
19:18 purl well, treasure is in the "other other box"
19:20 elb0w purl: whats in the box?
19:20 purl Pain.
19:20 elb0w purl: whats in the other box?
19:20 purl i haven't a clue, elb0w
19:20 elb0w purl: whats in the other other box?
19:20 purl bugger all, i dunno, elb0w
19:20 elb0w HAH
19:23 elb0w purl: who is your daddy?
19:23 purl Lenzo's my daddy!
19:24 elb0w purl: and what does he do?
19:24 purl elb0w: i don't know
19:24 elb0w purl: where is your father?
19:24 purl well, my father is oznoid, son of the fifth house of Betazed or me or me
19:26 crab i wonder how lenzo is these days.
19:29 ph1g left #mojo
19:39 mspo elb0w: nothing
19:39 mspo elb0w: it's just a cool place to walk
19:39 mspo elb0w: on santorini
19:40 henq joined #mojo
19:43 elb0w yeah im greek and I always forget the spelling lol
19:43 henq left #mojo
19:47 stephan48 how do i check if i have a dom model of a response?
19:49 ph1g joined #mojo
20:01 ph1g left #mojo
20:06 vasundhar joined #mojo
20:06 SmokeMachine left #mojo
20:08 Fried joined #mojo
20:08 SmokeMachine joined #mojo
20:10 amoore joined #mojo
20:12 ispy_ left #mojo
20:14 elb0w I think W4R is a depressed bot
20:14 elb0w whoops
20:14 elb0w irssi noob ftl
20:21 Fried left #mojo
20:24 jwang_ left #mojo
20:24 jwang left #mojo
20:24 amoore left #mojo
20:24 sh7 left #mojo
20:26 dekimsey left #mojo
20:35 Fried joined #mojo
20:44 xaka is Trigger for internal using or it has external benefits? from docs i can't understand the main purpose of it :)
20:48 SmokeMachine left #mojo
20:49 jwang joined #mojo
20:50 SmokeMachine joined #mojo
20:53 SmokeMachine_ joined #mojo
20:57 SmokeMachine left #mojo
20:57 SmokeMachine_ is now known as SmokeMachine
21:07 arthas left #mojo
21:07 arthas joined #mojo
21:09 vasundhar left #mojo
21:13 xantus[] sri: in accept the remote address info is sometimes undef
21:13 xantus[] its very odd
21:19 arthas left #mojo
21:22 xantus[] I'll make a test case and figure it out
21:36 kaare left #mojo
22:12 SmokeMachine left #mojo
22:27 metaperl left #mojo
22:29 mattastrophe left #mojo
23:06 metaperl joined #mojo
23:56 sromanov left #mojo

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