# Decoding Unicode URLs Properly

**URL:** <https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308>\
**Category:** Rails\
**Created:** [June 28, 2006, 1:23am UTC](https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308 "2006-06-28T01:23:42Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![patched](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/p/ba9def/32.png) [@patched](https://www.ruby-forum.com/u/patched)\
**Post date:** [June 28, 2006, 1:23am UTC](https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308/1 "2006-06-28T01:23:42Z")

</div>

Please help,

I have an application I switched over to UTF-8, but I am unable to  
decode URLs with unicode characters in them properly. For instance, if  
I have “item=Pur%E9ed” as a parameter in my url, when it’s decoded by  
rails, it returns “Pur?ed” and not “Puréed”.

Is there an easy way to fix my controller to accept these characters  
properly?

Or is there perhaps a plugin someone wrote that is less extreme then  
unicode\_hacks?

Thanks in advance,

Gregg P.  
Orlando Ruby U. Group  
[http://www.ORUG.org](http://www.ORUG.org)

---

<div class="post-metadata">

**Author:** ![patched](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/p/ba9def/32.png) [@patched](https://www.ruby-forum.com/u/patched)\
**Post date:** [June 28, 2006, 1:46am UTC](https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308/2 "2006-06-28T01:46:02Z")

</div>

On Wed Jun 28, 2006 at 01:23:42AM +0200, Gregg P. wrote:

> Or is there perhaps a plugin someone wrote that is less extreme then  
> unicode\_hacks?

i think you have to wait for Ruby to suport UTF-8, or resort to various  
hacks/patches/plugins. i made a local file browser in rails the other  
day and ran into this same issue…

---

<div class="post-metadata">

**Author:** ![patched](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/p/ba9def/32.png) [@patched](https://www.ruby-forum.com/u/patched)\
**Post date:** [June 28, 2006, 2:11am UTC](https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308/3 "2006-06-28T02:11:11Z")

</div>

But there has to be a way to simply decode HTML parameters in the URL.

The problem with doing this in a controller action, is that it’s already  
too late. Rails has already attempted to decode it, and a ? appears  
instead of the %E9.

I guess I’ll have to dive into ActionController rails code?

carmen wrote:

> On Wed Jun 28, 2006 at 01:23:42AM +0200, Gregg P. wrote:
> 
> > Or is there perhaps a plugin someone wrote that is less extreme then  
> > unicode\_hacks?
> 
> i think you have to wait for Ruby to suport UTF-8, or resort to various  
> hacks/patches/plugins. i made a local file browser in rails the other  
> day and ran into this same issue…

---

<div class="post-metadata">

**Author:** ![patched](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/p/ba9def/32.png) [@patched](https://www.ruby-forum.com/u/patched)\
**Post date:** [June 28, 2006, 4:06am UTC](https://www.ruby-forum.com/t/decoding-unicode-urls-properly/96308/4 "2006-06-28T04:06:10Z")

</div>

> carmen wrote:
> 
> > On Wed Jun 28, 2006 at 01:23:42AM +0200, Gregg P. wrote:
> > 
> > > Or is there perhaps a plugin someone wrote that is less  
> > > extreme then  
> > > unicode\_hacks?
> > 
> > i think you have to wait for Ruby to suport UTF-8, or resort to  
> > various  
> > hacks/patches/plugins. i made a local file browser in rails the other  
> > day and ran into this same issue…

On Jun 27, 2006, at 8:11 PM, Gregg P. wrote:

> But there has to be a way to simply decode HTML parameters in the URL.
> 
> The problem with doing this in a controller action, is that it’s  
> already  
> too late. Rails has already attempted to decode it, and a ? appears  
> instead of the %E9.
> 
> I guess I’ll have to dive into ActionController rails code?

But %E9 isn’t a valid UTF-8 character. To encode é (U+00E9), you’d  
need %C3%A9

> > $KCODE=‘u’  
> > =\> “u”
> 
> > CGI::unescape(“item=Pur%E9ed”)  
> > =\> “item=Pur?ed”
> 
> > parm = “item=Pur%C3%A9ed”  
> > =\> “item=Pur%C3%A9ed”
> 
> > CGI::unescape(parm)  
> > =\> “item=Puréed”
> 
> > CGI::escape(CGI::unescape(parm))  
> > =\> “item%3DPur%C3%A9ed”

Hmm… That works. Perhaps the problem is the user agent that encodes  
the parameters into the URL? Or in making the assumption that the  
original string is UTF-8.

-Rob