Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Replace html5lib in news download subsytem

Browse Source
This commit is contained in:
Kovid Goyal 2017-07-09 08:57:16 +05:30
parent 46de368b4b
commit 636cb5e654
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/calibre/web/feeds/news.py
View File

@ -686,14 +686,14 @@ class BasicNewsRecipe(Recipe):
else: else:
_raw = _raw.decode(self.encoding, 'replace') _raw = _raw.decode(self.encoding, 'replace')
if as_tree: if as_tree:
import html5lib from html5parser import parse
from calibre.ebooks.chardet import strip_encoding_declarations, xml_to_unicode from calibre.ebooks.chardet import strip_encoding_declarations, xml_to_unicode
from calibre.utils.cleantext import clean_xml_chars from calibre.utils.cleantext import clean_xml_chars
if isinstance(_raw, unicode): if isinstance(_raw, unicode):
_raw = strip_encoding_declarations(_raw) _raw = strip_encoding_declarations(_raw)
else: else:
_raw = xml_to_unicode(_raw, strip_encoding_pats=True, resolve_entities=True)[0] _raw = xml_to_unicode(_raw, strip_encoding_pats=True, resolve_entities=True)[0]
return html5lib.parse(clean_xml_chars(_raw), treebuilder='lxml', namespaceHTMLElements=False) return parse(clean_xml_chars(_raw))
massage = list(BeautifulSoup.MARKUP_MASSAGE) massage = list(BeautifulSoup.MARKUP_MASSAGE)
enc = 'cp1252' if callable(self.encoding) or self.encoding is None else self.encoding enc = 'cp1252' if callable(self.encoding) or self.encoding is None else self.encoding