Some parts of core's extract_content function are not understood. #287

jake-yi · 2023-01-17T03:19:44Z

jake-yi
Jan 17, 2023

Hello.

There's a part that I don't understand.

That is the 538th line of "https://github.com/adbar/trafilatura/blob/master/trafilatura/core.py"

ptest = subtree.xpath(//p/text())

In this part, it is marked as //p//text() so It's looking for the root's p-tag rather than the subtree's p-tag.

Isn't ./p//text() the right approach?

Thank you.

adbar · 2023-01-19T17:04:43Z

adbar
Jan 19, 2023
Maintainer

Hi @jake-yi, I believe it doesn't make any difference because the subtree is used as root and not the tree but I may be wrong, I'm going to make tests.

0 replies

adbar · 2023-01-19T17:08:23Z

adbar
Jan 19, 2023
Maintainer

>>> from lxml import etree
>>> tree = etree.fromstring('<html><body><div id="1"><p>ABC</p></div><div id="2"><p>DEF</p></div></body></html>')
>>> subtree = tree.xpath("//div")[0]
>>> subtree.xpath("//p/text()")
['ABC', 'DEF']

OK, my bad, thanks for your feedback!

0 replies

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Some parts of core's extract_content function are not understood. #287

{{title}}

Replies: 2 comments

{{title}}

{{title}}

Select a reply

Some parts of core's extract_content function are not understood. #287

jake-yi Jan 17, 2023

Replies: 2 comments

adbar Jan 19, 2023 Maintainer

adbar Jan 19, 2023 Maintainer

jake-yi
Jan 17, 2023

adbar
Jan 19, 2023
Maintainer

adbar
Jan 19, 2023
Maintainer