W3C home > Mailing lists > Public > www-validator@w3.org > April 2001

Re: Incorrect "catch" of valid data

From: Tim Bagot <tsb-w3-validator-0003@earth.li>
Date: Sun, 22 Apr 2001 00:21:40 +0000 (UTC)
To: <www-validator@w3.org>
Message-ID: <Pine.LNX.4.33.0104212342370.13905-100000@windle.hack>
At 2001-04-21T22:44-0000, Tangina Barrows wrote:-

> On Sat, 21 Apr 2001, Michael Bowen wrote:
> > When I validate documents (transitional DTD), if there are any errors
> > in a document, then the < &#153; > symbols are flagged as invalid.
> I thought it could be because on errors on the <meta tag referring to
> charset, but i've noticed that i validate, for example:
> it signals me that there's a missing </p> tag and an invalid sgml
> char...
> Could be part of the xml specification... or not?

The meta element is irrelevant because no matter what the encoding,
character references always refer to ISO 10646. (And the encoding ought to
be specified in the XML declaration.) #153 (#x99) is not the trademark
symbol - it is a C1 control (unnamed AFAICT); the trademark symbol is
&#x2122; or &#8482; or &trade;.

It does seem rather odd that the error is reported only when other errors
are also present; for this I have no explanation. #153 is included in
XML's Char production[1], but being a control, it is not a "legal graphic
character", so it should probably not be allowed.

[1] <http://www.w3.org/TR/REC-xml#NT-Char>

Tim Bagot
Received on Saturday, 21 April 2001 20:27:16 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:17:29 UTC