How to parse author name and book title from scraped HTML using XPath?

Question

The HTML you see below is text I have scraped from a remote site, as-is, into a local variable.

Now I need to parse the authorName and bookTitle from the HTML tags into their own variables, given the following consistent format of the scraped text:

<p>
  William Faulkner - 'Light In August'
  <br/>
  William Faulkner - 'Sanctuary'
  <br/>
  William Faulkner - 'The Sound and the Fury'
</p>

Is it possible to do this in XPath?

Answer 1

Yes. And easy, too:

//p/text()

Will give you three separate text nodes:

"
  William Faulkner - 'Light In August'
  ",
"
  William Faulkner - 'Sanctuary'
  ",
"
  William Faulkner - 'The Sound and the Fury'
"

Remember that preceding and trailing whitespace (including any line breaks) is always part of the text node. Trim the result.

I take it that you do not need help with splitting the resulting strings into author and title.

Answer 2

You can get the $N-th author name with this XPath expression:

substring-before(normalize-space(p/text()[$N]), ' -')

You can get the $N-th title with this XPath expression:

substring-after(normalize-space(p/text()[$N]), ' - ')

You can get first the count of all text nodes with:

count(p/text())

then you can issue the first two XPath expressions, substituting $N with the numbers in the interval

[1,count(p/text())]

Answer 3

In XPath 1.0 you can select the text node childs of p:

/p/text()

You can also get the string before (author) and after (title) - character for each text node

substring-before(/p/text()[1],'-')

Result:

  William Faulkner 

substring-after(/p/text()[1],'-')

Result:

 'Light In August'       

In XPath 2.0:

/p/text()/substring-before(.,'-')

Result in a sequence of 3 items:

William Faulkner William Faulkner William Faulkner 

And

/p/text()/substring-after(.,'-')

Result also in a sequence of 3 items:

'Light In August' 'Sanctuary' 'The Sound and the Fury'