Rails - Jan 2006 - textarea problem with accentued chars

Hello, I have a problem with accentued characters return from a textarea

I have 3 simple files to show example ( below ).

index.rhml : If I put "?nial" inside the textarea, then submit to 
:action => post

post.rhml : wrote

?nial

195

"\303\251nial"


//

params[:comment][:message]    return  => ?nial

params[:comment][:message][1]    return => 195   ! ( 195 is not "?"
,
http://www.asciitable.com/ )

params[:comment][:message].inspect    return => "\303\251nial"

and in ddb I store "??nial"


Someone can help me please ?


//

test_controller.rb :
------------------

class TestController < ActionController::Base

def index
end

def post
    @var1 = params[:comment][:message]
    @var2 = params[:comment][:message][0]
    @var3 = params[:comment][:message].inspect

    hash = { :name => params[:comment][:message] }
    obj = Color.new( hash )
    obj.save
end

end

index.rhtml :
------------

<%= start_form_tag :action => ''post'' %>
  <p><label
for="comment_message">Message</label><br/>
  <%= text_area "comment" , "message" %></p>
  <%= submit_tag "Post" %>
<%= end_form_tag %>

post.rhtml :
-----------

<p><%= @var1 %></p>
<p><%= @var2 %></p>
<p><%= @var3 %></p>

On 17 Jan ''06, at 12:04 PM, oo00oo wrote:
> Hello, I have a problem with accentued characters return from a  
> textarea
> index.rhml : If I put "?nial" inside the textarea, then submit  
> to :action => post
> post.rhml : wrote
> ?nial
> 195
> "\303\251nial"
Looks like a classic encoding mismatch: the browser is returning  
UTF-8 and you''re expecting something different (probably WinLatin1,  
aka CP-1252). Make sure your HTML has a <meta> tag specifying the  
intended encoding.
> params[:comment][:message][1]    return => 195   ! ( 195 is not  
> "?" , http://www.asciitable.com/ )
195 may not be "?" in the particular encoding you''re used to,
but
that doesn''t mean it isn''t in some other encoding. That  
"asciitable.com" site is rather na?ve ... "ascii" refers
only to the
low 7 bits (0..127), and there is absolutely no agreed-on "extended  
ascii" standard. There are dozens and dozens of different encodings  
for (128..255), and there are encodings that encode some characters  
using two or more bytes. UTF-8 is slowly becoming the lingua franca,  
though.

This brings up the question in my mind of how to deal with UTF-8 in  
Ruby. Last  checked, there was no native support for different  
character encodings or Unicode text. Has that been improved since  
1.6? Or are there optional libraries we can use to work with Unicode?

--Jens
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails/attachments/20060118/ff31ea4a/attachment.html

Jens Alfke wrote:
> This brings up the question in my mind of how to deal with UTF-8 in 
> Ruby. Last  checked, there was no native support for different character 
> encodings or Unicode text. Has that been improved since 1.6? Or are 
> there optional libraries we can use to work with Unicode?
The canned response is KCODE=''u''; require
''jcode''; but that''s got a
whole bunch of holes in it.  By strange coincidence, though, _why posted 
this a couple of days ago:

http://redhanded.hobix.com/inspect/unicodeLibForRuby18.html

-- 
Alex

2006/1/18, Alex Young <alex@blackkettle.org>:
> Jens Alfke wrote:
> > This brings up the question in my mind of how to deal with UTF-8 in
> > Ruby. Last  checked, there was no native support for different
character
> > encodings or Unicode text. Has that been improved since 1.6? Or are
> > there optional libraries we can use to work with Unicode?
>
> The canned response is KCODE=''u''; require
''jcode''; but that''s got a
> whole bunch of holes in it.  By strange coincidence, though, _why posted
> this a couple of days ago:
>
> http://redhanded.hobix.com/inspect/unicodeLibForRuby18.html
>
Optional library: http://rubyforge.org/projects/icu4r - still in development :)

I have the same problem now, did u "win" it?

Thanks,
Mikhail