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

Re: No source code to validate?

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Sun, 3 Feb 2008 21:21:10 +0200
Message-ID: <02b501c86699$f05a0540$0300000a@DOCENDO>
To: <www-validator@w3.org>

Micke Ströberg wrote:

> When checking my website with your validator I get the impression
> that there is no source code to validate...

The W3C Validator indeed finds no source code and no other content 
either.

> The website is http://www.designen.se and the main page is
> http://www.designen.se/index.php.

The server www.designen.se has been misconfigured or the PHP code has 
been written wrongly.

Use e.g. http://www.rexswain.com/httpview.html to check what the server 
actually sends. The response body is 5 octets: the digit "0" followed by 
two CR LF pairs.

This may depend on the Accept-Language header. If I send a request with 
such a header, I get a normal-looking response (though with many markup 
errors). The reason why haven't noticed the problem using a browser is 
that browsers _usually_ send an Accept-Language header (usually with 
disinformation content, but I digress). You would see the problem on a 
browser, too, if you removed all the languages from the so-called 
language settings in the browser.

So probably the PHP code or the server tries some content negotiation 
on language and fails when the user agent sends no Accept-Language 
header (which is quite acceptable and in fact the only sensible setting 
in many contexts).

Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/ 
Received on Sunday, 3 February 2008 19:21:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:28 GMT