XPath SyntaxError: invalid predicate with text()

Question

What am I doing wrong?

When using what I believe to be a valid XPath expression in Python, I get a "SyntaxError: invalid predicate" result. My code is:

import xml.etree.ElementTree as ET
data = '''<root>
            <child>
                <p>aaa</p>
                <p>bbb</p>
            </child>
        </root>'''

root = ET.fromstring(data)

p_element = root.findall(".//p[text()='aaa']")

print(p_element[0].text)

Running this code produces an invalid predicate error, but I believe it is correct. I’m using Python version 3.10.0, and lxml version 4.9.2, and running this on Windows.

It seems to have something to do with the "text()" part of the XPath; with that removed, it does not error.

Asked By: anon

||

Source

Answer 1

.//p[text()='aaa'] is a perfectly valid XPath expression.

The problem is that ElementTree’s implementation of XPath is incomplete.

For your XML, testing the string value, .//p[.='aaa'], would be a work-around if you’re stuck with ElementTree. Better would be to only use an XPath library such as lxml that better follows the standard.

XPath SyntaxError: invalid predicate with text()

Question:

Answers:

See also