W3C home > Mailing lists > Public > html-tidy@w3.org > January to March 2002

Re: Terminating..

From: Richard A. O'Keefe <ok@cs.otago.ac.nz>
Date: Tue, 19 Mar 2002 15:41:50 +1200 (NZST)
Message-Id: <200203190341.PAA301435@atlas.otago.ac.nz>
To: html-tidy@w3.org, reinis@apollo.lv
reinis@apollo.lv wrote:
	Using the latest tidy (on linux(RH), the one of 15 march) ...
	HTML Tidy for Linux/x86 (vers 1st March 2002; built on Mar 15 2002, at 23:33:08)
	Parsing "new.htm"
	line 4 column 1 - Warning: <table> lacks "summary" attribute
	line 12 column 20 - Error: <p> missing '>' for end of tag
	...

When we look at the text in question, we find
	<P class=MsoNormal><?xml:namespace prefix = o /><o:p>&nbsp;</o:p></P></TD>
                           ^^                         ^^

In SGML, processing instructions begin with '<?' and end with '>'.
In XML,  processing instructions begin with '<?' and end with '?>'.

If the input is supposed to be XML, then Tidy would necessarily have
swallowed the rest of the input at this point looking for the closing
'?>'.  Does Tidy need a flag to treat '/>' as if it were '?>' ?

I note that this is NOT how you declare XML namespaces.
In XML, you can't have a name space prefix without saying what URI
the prefix represents.  If the whole thing is a single DIV, you'd
have
    <DIV class=Section1 xmlns:o="http://www.msdn.com/foo/bar/ick.asp">
or whatever the URI was.
Received on Monday, 18 March 2002 23:10:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 April 2012 06:13:51 GMT