XSLT: Parse string as XML Node Set (concret: transform HTML-String to a node-set)?
Question
I am facing the problem that I have in XML a node with a string, representing HTML. I need to cut this string, but, of course, this can result in an invalide HTML-Markup (eg if I cut the string always after 30 characters I can easily lose closing elements like </ul> , etc.). What to do? It seems to be very difficult, because I don't find real help via google.
My idea so far: with "analyze-string" and regex select nodes and contents and write them es XML-element-nodes. But I get big problems to handle all cases, especially the nested nodes.
Does someone has some ideas?
FYI: My notepad:
- Regex: Catch first tag
- Read tagname of first node
- Put tagname in regex and search whole tag, select also the rest of the string (to go on with it later)
- Check content: More tags? yes: -> Step 1, no: -> Step 5
- Write tag as node-element
- Take rest of the string -> Step 1
Here is the XML-doc:
<?xml version="1.0" encoding="UTF-8"?>
<html>
<data>
<![CDATA[
<h2>header</h2><p>A little article. <b>Here</b> it's already done!</p>
]]>
</data>
</html>
What I want to do:
In I have a string (html) and just want to output a special amount of characters (eg the first 25). When I do this just on the string, I get this result:
"<h2>header</h2><p>A little article"
In a next step I put this string in an HTML-output, but on this point I get invalid markup because the <p> -Tag is not closed.
So my first approach: Parsing this string to get XML-Nodes for each tag, and then go over each node, write an xml-element (to make sure the final tag will be valid) and copy as many characters until reached the limit, in this example would be 25 characters.
2 answers
solution1
0 2012-01-03 09:50:18
If you have an XML node which represents HTML, then this should have been entity encoded, ie open and close braces converted to < and > - this means you can cut it wherever you like and still have a valid XML document.
solution2
0 2012-01-03 11:10:52
As you mention analyze-string you seem to use XSLT 2.0. That way you have two options, with Saxon 9 there is an extension function http://www.saxonica.com/documentation/extensions/functions/parse.xml (and even one http://www.saxonica.com/documentation/extensions/functions/parse-html.xml in case you want to parse HTML), and then there is David Carlisle's pure XSLT 2.0 implementation of an HTML parser http://code.google.com/p/web-xslt/source/browse/trunk/htmlparse you can import in your stylesheet and then use on the contents of your data element.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.