W3C home > Mailing lists > Public > www-validator@w3.org > June 2008

Re: How to validate pages of a website with national characters in domain name

From: Frank Ellermann <nobody@xyzzy.claranet.de>
Date: Thu, 19 Jun 2008 01:48:54 +0200
To: www-validator@w3.org
Message-ID: <g3c6pl$dep$1@ger.gmane.org>

Raija Ylonen wrote:

> www.ylöjärventaideyhdistys.fi
> When I try to validate it, I get error messages.

Works for me.  IOW when I clicked on your string my
MUA *guessed* that this is a http IRI to be handled
by a Web browser, based on the "www".  The MUA also
knew that your message uses Latin-1 (iso-8859-1) as
its charset, noted in the Content-Type header field:

<http://article.gmane.org/gmane.org.w3c.validator/10999/raw>

My OS then informed the Web browser on duty that an
IRI <http://www.ylöjärventaideyhdistys.fi> requires
its attention.  The browser on duty was Firefox 2,
and FF2 can transform IRIs with the domain in any
supported non-ASCII charset into equivalent URIs:

<URL:http://www.xn--yljrventaideyhdistys-dzb80b.fi/>

Actually I'm not sure at which point the IRI was
transformed into an URI, but I know that FF2 can
do this for RFC 3987 <i-host> names in an IRI.

What went wrong for you is that you tried to copy
and paste the IRI into the "Address" field of the
<http://validator.w3.org> Web form.

When I try it, the Latin-1 to UTF-8 translation is
handled by clipboard + FF2, but after that neither
FF2 on my side nor the validator on the other side
take care of transforming the IRI into an URL.  

Of course FF2 cannot know that the field is about
IRIs, only the validator knows this.  I think the
Web form should say "URL" instead of "Address" as
long as it does not support IRIs.

 Frank
Received on Wednesday, 18 June 2008 23:47:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:29 GMT