W3C home > Mailing lists > Public > www-validator@w3.org > November 2013

Re: [VE] [65] False validation

From: David Dorward <david@dorward.me.uk>
Date: Sun, 10 Nov 2013 21:25:30 +0000
To: Marcel <info@deltanet-production.de>
Cc: www-validator@w3.org
Message-ID: <30C34DF2-8108-4D1B-B7A1-E17BB8084A65@dorward.me.uk>
On 8 Nov 2013, at 6:20, Marcel wrote:

> Validating http://forum.ygopro.de/devpro-forum.html
> Error [65]: "document type does not allow element X here; missing one 
> of Y start-tag"
>
> The given error is a false return because it says, that this error 
> will be on Line 411 but it's in line 3160.

The document doesn't have that many lines.

> The validator also doesn't understand, that this part is in a 
> nonscript tag and the nonscript tag is in a ul/li tag and these tags 
> are in a div tag.

It does understand that. The rules for where noscript elements are 
allowed and what elements they are allowed to contain are defined in the 
DTD that the validator compares the document to.

> A validator must check the html code not line by line

It has to do it in some order, and from top to bottom is how all HTML 
parsers work.

> , he must choose between the head and the body element and then he 
> must scan for all
> tags in there.

Aside from the choosing between the head and the body (since the 
validator will work through the document in order), that is what it 
does.

> After this process, he must check for some scipt tags (javascript) and 
> must follow these script.

No. Running JavaScript is out of scope for the validator. If you want to 
validate the resulting DOM after running JS, then you need to generate 
that DOM yourself, serialise it to HTML, then feed that HTML into the 
validator.

> At last your validator has to check if the page was coded self or if 
> the page was generatet throw an CMS like Typo 3 or the Woltlab Burning 
> Board.

An error is an error. It doesn't matter if the error was produced by the 
author directly or by some software that they choose to use, it should 
still be fixed.

> So please correct your validator and note that most CMS are working 
> with a languagesystems wich automaticly escape the "&" over the CDATA 
> block
> in the XML file.

If the CMS was fixing the & characters (instead of expecting the browser 
to do so) then the validator would receive the fixed version from the 
CMS and would not report an error.

-- 
David Dorward
http://dorward.co.uk/
Received on Sunday, 10 November 2013 21:26:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:18:09 UTC