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

Re: Report page in validation service is not well-formed XML when the validation was successful

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Wed, 28 May 2008 15:54:52 +0300
Message-ID: <00fe01c8c0c2$06c855c0$0500000a@DOCENDO>
To: "Svensson, Lars" <l.svensson@d-nb.de>, <www-validator-css@w3.org>

Svensson, Lars wrote:

>> I have no idea of why the report contains the spurious line 33,
>> containing just an unmatched "</a>" tag.
> Yes. That is the obvious error I mean. I might be mistaken, but to me
> this makes the document not well-formed (and thus automatically
> invalid since well-formedness is a prerequisit...).

Right. Quite right. I didn't think of it, since I'm so used to seeing 
people confuse well-formedness with validity.

> Since the document is supposed to be XHTML, I guess
> the quotes must be mandatory anyway.

Right, so not quoting the href value causes somewhat different 
confusion. Actually the validator says

Line 34, Column 51: an attribute value specification must be an 
attribute value literal unless SHORTTAG YES is specified.

which is a somewhat indirect way of saying "you forgot the quotes, 
dude!" and reflects the fact that the Markup Validator is really an SGML 
validator hacked up a bit to sort-of work as an XML validator too.

Then confusion arises because in
<a href=http://www.w3.org/TR/CSS21/>
the first slash is taken as part of markup...

But regarding the CSS Validator, it is sufficient that the markup is 
fixed to use quotes (and the spurious "</a>" is removed).

Oh wait... there's still an error on lines 105 to 107:

        <ul id="lang_choice">


At least one <li> element must be present, or the <ul> element needs to 
be removed.

Jukka K. Korpela ("Yucca")
Received on Wednesday, 28 May 2008 12:55:46 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:01:02 UTC