Re: Parser Error? from Lloyd Wood on 1999-06-14 (www-validator@w3.org from June 1999)

From: Lloyd Wood <L.Wood@surrey.ac.uk>
Date: Mon, 14 Jun 1999 21:05:48 +0100 (BST)
To: Jukka.Korpela@hut.fi
cc: www-validator@w3.org
Message-ID: <Pine.SOL.4.02.9906142058350.14483-100000@petra.ee.surrey.ac.uk>

On Thu, 10 Jun 1999 Jukka.Korpela@hut.fi wrote:

> On Thu, 10 Jun 1999, Lloyd Wood wrote:
> 
> > Yet a URL with form state includes & between attribute values, gets
> > checked and gets complained about by the validator, which is pretty
> > much where we came in.
> 
> That's because in CDATA, entity references are recognized.
> When your HTML document contains an HREF attribute (or _any_
> attribute, or plain text) containing, say, &foo, then an HTML
> parser recognizes it as entity reference.
> 
> So, for example, an &amp; _denotes_ the & character. When you
> have HREF="http://host.dom/foo.pl?a=1&amp;b=2", then the URL
> itself is http://host.dom/foo?a=1&b=2 but it must be written
> slightly differently in HTML. An HTML parser or a validator
> still does not parse the URL according to any specific syntax.
> It does the same basic thing with &stuff as it does elsewhere.

In theory. Practice can be a different thing.

http://www.mg.co.za/mg/m&e/archive.htm
 
Okay, yes, that's an ampersand in the URL and directory name, and
they're running Apache 1.3.4 (unix).
 
What I'd dearly like to know is why:
 
http://www.mg.co.za/mg/m&amp;e/archive.htm
 
doesn't work. Yes, it's probably due to Netscape 4.5 not parsing &amp;
correctly...
 
Good cartoon, btw.

L.

this is worse that &amp; not being recognised in page titles by browsers.

<L.Wood@surrey.ac.uk>PGP<http://www.ee.surrey.ac.uk/Personal/L.Wood/>

Received on Monday, 14 June 1999 16:05:53 UTC