SyntaxError: invalid predicate using lxml iterfind-CodePudding

I am currently struggling with iterating through a XPath expression. I am trying to retrieve all the system-out nodes that contains a substring of "[[SOMETHING|". The issue is that I get the following syntax error that points to the tree.iterfind.

    for elem in tree.iterfind('.//system-out[contains(.,"[[SOMETHING|")]'):
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "src/lxml/etree.pyx", line 2288, in lxml.etree._ElementTree.iterfind
  File "src/lxml/etree.pyx", line 1588, in lxml.etree._Element.iterfind
  File "src/lxml/_elementpath.py", line 312, in lxml._elementpath.iterfind
  File "src/lxml/_elementpath.py", line 295, in lxml._elementpath._build_path_iterator
  File "src/lxml/_elementpath.py", line 237, in lxml._elementpath.prepare_predicate
SyntaxError: invalid predicate

tree = etree.parse(test_file)
for elem in tree.iterfind('.//system-out[contains(.,"[[SOMETHING|")]'):
     print("do something")

The above is my code. As far as I can see I don't have any syntax error. And I have also tried to test the xpath expression using a free formatter tool, and that seems to work. I just can't seem to see what is wrong. I have attempted to use the "findall" function provided by lxml but I receive the same error. I have also tried to compile the xpath expression using the etree.XPath function into an attribute, however I received an TypeError that says the following, which makes sense.

TypeError: 'lxml.etree.XPath' object is unsliceable

Is there something I am missing? Or is just an unsupported expression by the lxml package itself?

CodePudding user response：

In case SOMETHING instead of [[SOMETHING| still can be used and will be a unique enough I'd suggest instead of this .//system-out[contains(.,"[[SOMETHING|")] to use just this:

'.//system-out[contains(.,"SOMETHING")]'

So the entire code line will be

for elem in tree.iterfind('.//system-out[contains(.,"SOMETHING")]'):

CodePudding user response：

As Martin Honnen explained in the comments, the find methods (iterfind, find, findall) in ElementTree and lxml does not support the full XPath 1.0 syntax which explains the SyntaxError: invalid predicate error.

I used the lxml.etree.xpath() function instead, which does support the XPath 1.0 syntax. Being able to retrieve the text in the XML file I then used the result of the xpath() function to iterate over all of the occurrences by using a much simpler XPath expression that iterfind can understand.

occ = tree.xpath('.//system-out[contains(.,"[[SOMETHING|")]')[0].text
for elem in tree.iterfind(f'.//*[.="{occ}"]'):
     print("do something")