pictuga
/
morss
1
1
Fork 0
Code Issues Pull Requests Actions Packages Releases Activity

readabilite: run the new cleaning code

master
pictuga 2018-10-25 01:08:25 +02:00
parent 58fe5243af
commit 7d005e9a65
1 changed files with 8 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
morss/readabilite.py
View File

@ -132,6 +132,12 @@ def write_score_all(root, grades):
node.attrib['score'] = str(int(grades.get(node, 0))) node.attrib['score'] = str(int(grades.get(node, 0)))
def clean_root(root):
for node in list(root):
clean_root(node)
clean_node(node)
def clean_node(node): def clean_node(node):
parent = node.getparent() parent = node.getparent()
@ -277,4 +283,6 @@ def get_article(data, url=None, encoding=None):
if url: if url:
best.make_links_absolute(url) best.make_links_absolute(url)
clean_root(best)
return lxml.etree.tostring(best, pretty_print=True) return lxml.etree.tostring(best, pretty_print=True)