Re: Validator ignores malformed closing comment tag

David Dorward wrote:
> In HTML, <style> elements contain CDATA. It isn't an invalid comment, 
> it is just a part of the stylesheet. ...
> There is nothing invalid about this, but it is something that would be 
> useful for a general HTML lint to pick up.

Thank you, David, for explaining why ->> inside <style> isn't flagged by 
Validator.  That makes sense.  Still, although I always fix what 
Validator complains about, I don't use it to make Netscape versions 2 or 
3 work.  I also want common browsers to render pages correctly.

I found that Internet Explorer 6, 7, and 8 all produce blank pages 
because of ->> in the <style> CDATA.  I guess that's technically 
Microsoft's fault, so I understand that Validator shouldn't flag it.  
Still, a checkbox on Validator that says "also run HTML lint" -- to 
catch mistakes that aren't technically wrong, but which probably weren't 
intended -- could help non-expert HTML coders produce web pages that 
work better.  (In the meantime, I'll find an HTML lint to use along with 
Validator.)  Thanks, and sorry for the "noise" on the list.


Received on Tuesday, 22 September 2009 02:48:14 UTC