尝试使用来自自定义Firefox扩展/附加组件的xpath和javascript来引用页面中的文本失败

Question

I have been having issues with this for several days and have been searching all over StackOverflow and Google for the answer. This is code within a Firefox extension which I am referencing a label/text on a webpage. I am trying to get the text from the middle of <b></b> tags. The test is deep within a document so I figured I might be able to use xpath to reference it. I have been attempting to use the document.evaluate function to reference this with no luck. Here is the code snippet that I have most recently tried:

var result = document.evaluate( '//*[@id="page"]/table[3]/tbody/tr/td[3]/table[1]/tbody/tr[9]/td[2]/b', document, null, XPathResult.STRING_TYPE, null );
alert( 'Value: ' + result.stringvalue );

I am new to DOM and xpath so please let me know if this is at least on the right track. I have tried other methods as well with little success. The HTML code I'm trying to pull this out of looks like the following:

.
. Rest of the source
.
<b>This is title number 8955592</b>
.
.

I got the xpath from the chrome inspect element feature. I have also tried using firebug to see if anything in there would help as well.

Xpath:

//*[@id="page"]/table[3]/tbody/tr/td[3]/table[1]/tbody/tr[9]/td[2]/b

CSSPath:

#page > table:nth-child(3) > tbody > tr > td:nth-child(3) > table:nth-child(3) > tbody > tr:nth-child(9) > td:nth-child(2) > b

My hope is that I have a simple formatting error in my xpath string which I am just starting the string wrong. This same issue forced me to stop another project I was working on so I decided that I would try and actually address it this time. I must re-enforce that this is part of a firefox extension I am working on. I also looked into nodes as a way to do this but that got complicated pretty fast. Any help is very much appreciated.

Answer 1

First, you have

result.stringvalue

which should be

result.stringValue

(capital V). However, this should throw an error. I assume you have checked the console for errors?

Without seeing the page you're trying to extract data from, it's hard to say what's going wrong. But you could narrow it down as follows:

var result = document.evaluate(
  '//*[@id="page"]', // /table[3]/tbody/tr/td[3]/table[1]/tbody/tr[9]/td[2]/b',
  document, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null );

What I've done is just take the first part of the XPath expression and comment out the rest for now. I also asked for a node result type. Then see how many nodes are returned:

console.log(result.snapshotLength); // or alert() if you prefer

If the //*[@id="page"] expression yields 1 or more results, then extend the expression and try again:

 var result = document.evaluate(
  '//*[@id="page"]/table[3]', // /tbody/tr/td[3]/table[1]/tbody/tr[9]/td[2]/b',
  document, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null );

and again check the result.snapshotLength. When you get a zero for snapshotLength, that's the point where your XPath is failing. And that should help you figure out why. But if you can't figure out why, show us those results and we'll see if we can help.

Answer 2

Thank you for your help everyone. I finally figured this out, I apologize for the late response. I did misspell stringValue and however I didn't end up using this property. I was able to move forward in this and figure it out given the recommendations provided. I was getting a undefined result initially. When I took off the .stringValue from the end the result was a XPathResult object. I later determined that you can do a .singleNodeValue on a XPathResult object and get the actual HTMLElement object you are trying to work with. Here is the end result I eventually used to resolve this:

document.evaluate( '//*[@id="page"]/table[3]/tbody/tr/td[3]/table[1]/tbody/tr[9]/td[2]/b', document, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null ).singleNodeValue.innerHTML;

Which returns the contents of the <b>text</b> from the page. Even though this may look like a tedious way of getting this result, figuring this out basically taught me how these objects work which I didn't understand when I first asked this question. Using a systematic way to track if I am on the right path (as recommended) was extremely helpful. Again, thank you very much for your assistance.