Re: Bugs in XHTML validator

On Thursday 15 July 2010, Matej Lednár wrote:

> The XML declaration problem still continue (Validator creating comment
> and add XML declaration with UTF-8).

As Benjamin noted, this is not a bug, it is completely intentional.

The way the validator currently works, it needs to do it as well as override 
charset info in meta elements when passing the document to certain validation 
backends (libxml and at the moment) in order to avoid false 
positive errors from them.  We could choose to not show the actual source 
passed to these backends and just echo back what was given, but that would 
result in confusion if our charset overriding function was the reason for some 
errors they report - there would be no way for the end user to see what was 
actually validated.

Received on Thursday, 15 July 2010 15:58:05 UTC