I'm trying to identify an element that has certain text but I only want to identify the element if the desired text occurs a specific number of times.
For example, imagine we have the following two HTML snippets on the same page:
Snippet 1:
<span id="price">
$36.46
<span>
($0.38 / Count)
</span>
</span>
Snippet 2:
<span id="price">$38.38</span>
I could identify both elements using the XPath: .//span[contains(text(),'$')] However, I only want to identify the element if it (or any descendant of span element) contain at least two instances of the character: $
In above example, it would only identify the first snippet because the second snippet only contains one instance of $, not two.
What is the correct XPath syntax to use?
You can use the XPath //span[count(.//text()[contains(., "$")]) >= 2]
This is a moderately complicated XPath, so to explain it some by expanding outwards:
.//text()[contains(., "$")]
Select all text elements descending from the current node whose self contains "$".
count(.//text()[contains(., "$")])
Count the number of text elements descending from the current node whose self contains "$".
//span[count(.//text()[contains(., "$")]) >= 2]
Select all span elements with two or more text descendants whose self contains "$"
As a caveat, this only works if the dollar sign is in two different text elements. If you want to include the span in this example:
<span>
$$
<span>
foo
</span>
</span>
...then you'll need a different approach:
//span[string-length(.) - string-length(translate(., "$", "")) >= 2]
This predicate compares the string length of the span to the string length of the same span with all "$" characters removed.