Not evaluating expressions in IE.

[GoogleCodeExporter]

Sorry if this doesn't belong here, but maybe you can help. I have some code 
that works on Chrome and Firefox but not on IE (any version as far as I can 
tell). Basically, the paths work just fine but the expressions do not. I am not 
sure what other information I can provide other than the attached example.

Open a web page, press F12 to open the debugger, go to the script tab, choose 
the "Console" tab on the right, and copy-paste the attached code. 
"successNodes" has two elements and "success" points to the first; "failNodes" 
has no elements and "fail" is null.

Original issue reported on code.google.com by John.Pau...@gmail.com on 1 Mar 2013 at 6:00

Attachments:

• IE-test.txt

[GoogleCodeExporter]

Could you provide the full html you used for the test? It seems that you are 
using document.evaluate on IE. I assume you've included wicked-good-xpath? 
Also, the library didn't fully support XML documents, but we can take a look at 
it.

Original comment by zhoumoto...@gmail.com on 2 Mar 2013 at 2:10

[GoogleCodeExporter]

Here are all of the files in a ZIP. I commented out the parts that would 
actually retrieve the XML and just hardcoded it. It should fail on page load. I 
didn't comment out the stuff around the button, but if you press it, it will 
just redirect you somewhere.

Because it works just fine in FF and Chrome, I suspect this is something wonky 
that IE is doing. Thank you for taking the time to look into it.

Original comment by John.Pau...@gmail.com on 4 Mar 2013 at 3:37

Attachments:

• Trialist.zip

[GoogleCodeExporter]

Bump.

Sorry to bother. We will begin testing with our deployment in the next month, 
but IE is still unusable. This is, of course, our problem, but this is just 
such a wonderful tool that I would like to stick with it if possible.

Thank you.

Original comment by John.Pau...@gmail.com on 4 Apr 2013 at 7:13

[GoogleCodeExporter]

I'll take a look this week.

Original comment by gden...@google.com on 8 Apr 2013 at 1:37

[GoogleCodeExporter]

John, just to clarify, the error you are seeing on IE is "An unknown error 
occurred. The setup survey is missing." Correct?

Original comment by gden...@google.com on 8 Apr 2013 at 6:35

• Changed state: Accepted

[GoogleCodeExporter]

Hi John,

When you're evaluating XPaths on XML documents, you don't need this library, as 
every browser (including IE) has native XPath support for XML documents. The 
library is intended to provided cross-browser XPath support for HTML documents. 
In your code, you current have this for IE,
  document.evaluate(expression, node, null, XPathResult.ANY_TYPE, null);
since "node" in this case is a node in an XML document, you should just use 
IE's native selectNodes method like this:
  node.selectNodes(expression);

To be really correct, you probably want to set the SelectionLanguage to XPath 
first. Documentation here:
http://msdn.microsoft.com/en-us/library/windows/desktop/ms754523(v=vs.85).aspx

Be aware that it returns a IXMLDOMNodeList, which doesn't have an iterateNext 
method, but has equivalents:
http://msdn.microsoft.com/en-us/library/windows/desktop/ms767664(v=vs.85).aspx

As long as you are only evaluating XPaths on XML documents, you shouldn't need 
this library at all.

Original comment by gden...@google.com on 8 Apr 2013 at 7:58

• Changed state: WontFix

[GoogleCodeExporter]

I see. I felt like this should be the case, but I was unable to find this. I 
apologize if I wasted your time.

Original comment by John.Pau...@gmail.com on 9 Apr 2013 at 3:51

[GoogleCodeExporter]

In case anyone else comes across this issue, I have solved it without WGXP per 
the above recommendation. Below is the code I used, which seems to be working 
across all versions of IE, FF (15.0.1), and Chrome (26.0.1410.65).

Thanks again for your help and sorry again that this was my lack of knowledge 
more than anything else.

[code]        
        /**
         * Converts an XML string into a DOM object.
         */
        function convertXmlStringToDomObject(xml) {
            var xmlDoc = null;

            if(window.DOMParser)
            {
                var parser = new DOMParser();
                xmlDoc = parser.parseFromString(xml, "text/xml");
            }
            else // Internet Explorer
            {
                xmlDoc = new ActiveXObject("Microsoft.XMLDOM");
                xmlDoc.async = false;
                xmlDoc.loadXML(xml); 
            }

            return xmlDoc;
        }

        /**
         * Runs the document.evaluate() method based on the current browser.
         */
        function evaluateXpath(root, node, expression) {
            var result;
            if(window.DOMParser)
            {
                result =
                    root
                        .evaluate(
                            expression,
                            node,
                            null, 
                            XPathResult.ANY_TYPE,
                            null);
            }
            else // Internet Explorer
            {
                result = node.selectNodes(expression);
            }
            return result;
        }

        /**
         * Returns the next element from a list of elements.
         */
        function nextElement(node) {
            if(window.DOMParser) {
                return node.iterateNext();
            }
            else {
                return node.nextNode();
            }
        }

        /**
         * Returns the text for the node.
         */
        function elementValue(node) {
            if(window.DOMParser) {
                return node.textContent;
            }
            else {
                return node.text;
            }
        }
[/code]

Original comment by John.Pau...@gmail.com on 24 Apr 2013 at 5:04