XPath选择属性在Chrome上失败

Question

Here's a small test I ran with expect.js :

describe('xpath', function () {
    it('finds attribute nodes', function () {
        $(document.body).append('<string fooBar="bar"><data id="xyz"></data></string>');
        var xyz = document.getElementById('xyz');
        expect(xyz).not.to.be(null);

        var nodes = document.evaluate('../@fooBar', xyz, null, XPathResult.ANY_TYPE, null);
        var element = nodes.iterateNext();
        expect(element.nodeValue).to.equal('bar');
    });
});

I'm testing it on Chrome and Firefox with Testacular, and while it passes on Firefox, it fails on Chrome with an error saying that element is null . The really crazy part to me is that if I change fooBar to just foo , it passes on both systems. Chrome also breaks if I use Foo instead of foo . In other words, XPath querying on Chrome seems to break when the sought-for attribute has a capital letter in it. Is this normal behavior? Am I missing something? Or is this a bug in Chrome?

UPDATE 1

I added in Google's Wicked Good XPath, and now Internet Explorer works, too, but Chrome still fails.

UPDATE 2

The XPath query works on Chrome if the actual attribute has caps, but the XPath value is all lowercase.

Answer 1

Are you using XHTML? If so, the XHTML specification dictates that all element and attribute names must be in lower case .

You may find that Chrome is only geared up for evaluating XPaths on that basis, whereas other browsers are more compliant with XML (where any case is valid).