与xpath，Lxml等效的Findall

Question

I am extracting text with respect to tags and I need to get them in a list form wrt p tags. I have this xpath expression as:

 find =  etree.XPath("//w:p//.//*[local-name() = 'ins']//text()" ,namespaces={'w':"http://schemas.openxmlformats.org/wordprocessingml/2006/main"}) 

And i want to use it in a findall expression. I tried:

inserted_list_1=[]
for p in lxml_tree.findall('.//{' + w + '}p'):
    inserted_list_1.append([t.text for t in p.findall('.//{' + w + '}ins')])

but all this returns is a list full of None values whilst the former xpath works perfectly.
I think there's some intermediate path missing.

Answer 1

You cannot use that expression with findall() ; the findall() method deliberately keeps compatibility with the limited ElementTree API XPath support .

Use the xpath() method instead:

for p in lxml_tree.xpath('.//w:p', namespaces={'w': w}):

and just use namespace prefixes for much more readable queries.

If you just wanted to extract all contained text, you can use:

[t for t in p.xpath('../w:p//w:ins//text()',namespaces={'w': w})]