From: Mark Hubbart <discord@...>
Date: 2004-04-02T03:15:09+09:00
Subject: Re: HTML parsing by REXML


On Apr 1, 2004, at 9:14 AM, Dario Linsky wrote:

> Hi,
>
> On Fri, 02 Apr 2004 00:58:24 +0900, Paul Argentoff wrote:
>> Sorry for returning to sorta well-discussed (but not in a sense I 
>> need)
>> topic.
>> I can't parse xml files by rexml since some tags in html are open 
>> (such as
>> <link>, etc). Document.new errors with a message about such a tag.
> Do I understand your problem right, that REXML gives you an Exception
> because you did not close a tag? If so, a possible solution would be to
> use XHTML instead of normal HTML.

to expand on this:
html is not xml. xhtml *is* xml.
in html, you can have unclosed tags that look like this:
   <link rel="stylesheet" href="default.css">
but in xhtml, the tag has to close itself:
   <link rel="stylesheet" href="default.css"/>
xhtml does this to be compatible with xml, which it is based on.

If you have a bunch of regular html files that you need to parse, I 
would suggest running them through "HTML Tidy", which can convert them 
to well-formed xhtml for you. You should then be able to parse them 
with REXML. See http://tidy.sourgeforge.net/

--Mark