__all__ = ['parse_html']
-import re
+import warnings
+
from bs4 import BeautifulSoup
+
from .bkmk_ph_util import HTMLParser
+from compat import string_type
+
+warnings.filterwarnings(
+ 'ignore', 'No parser was explicitly specified')
+warnings.filterwarnings(
+ 'ignore',
+ "It looks like you're parsing an XML document using an HTML parser.")
universal_charset = "utf-8"
DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+
def _parse_html(html_text, charset):
try:
- return BeautifulSoup(html_text, from_encoding=charset)
+ if isinstance(html_text, bytes):
+ return BeautifulSoup(html_text, from_encoding=charset)
+ else:
+ return BeautifulSoup(html_text)
except TypeError:
return None
+
def parse_html(html_text, charset=None, log=None):
+ if not html_text:
+ return None
root = _parse_html(html_text, charset)
if root is None:
return None
else:
parts = []
for part in title:
- if not isinstance(part, basestring):
- part = unicode(part)
- parts.append(part.strip())
+ #if not isinstance(part, string_type):
+ # part = part.decode()
+ if part.strip:
+ parts.append(part.strip())
+ else:
+ parts.append(' ') # Skip tags, they're usually `<br>`
title = ''.join(parts)
meta = head.find(_find_contenttype, recursive=False)
try:
meta_content = meta.get("content")
if meta_content:
- __charset = meta_content.lower().split('charset=')[1].split(';')[0]
+ __charset = meta_content.lower().split('charset=')[1].\
+ split(';')[0]
else:
__charset = False
except IndexError: # No charset in the META Content-Type
if meta_content:
meta_charset = _charset = meta_content.lower()
- if title and (_charset or meta_charset):
- try:
- title = title.encode(_charset or meta_charset)
- except LookupError:
- title = title.encode(universal_charset)
- _charset = universal_charset
+ #if title and (_charset or meta_charset):
+ # try:
+ # title = title.encode(_charset or meta_charset)
+ # except LookupError:
+ # title = title.encode(universal_charset)
+ # _charset = universal_charset
meta = head.find(_find_refresh, recursive=False)
if meta:
return None
return HTMLParser(_charset, meta_charset, title, refresh, icon)
+
def _find_contenttype(Tag):
return (Tag.name == "meta") and \
(Tag.get_attribute_list("http-equiv", '')[0].lower() == "content-type")
+
def _find_charset(Tag):
return (Tag.name == "meta") and Tag.get("charset", '')
+
def _find_refresh(Tag):
return (Tag.name == "meta") and \
(Tag.get_attribute_list("http-equiv", '')[0].lower() == "refresh")
+
def _find_icon(Tag):
- return (Tag.name == "link") and \
- (Tag.get_attribute_list("rel", '')[0].lower() in ('icon', 'shortcut icon'))
+ if Tag.name != "link":
+ return False
+ rel = ' '.join(Tag.get_attribute_list("rel", ''))
+ return rel in ('icon', 'shortcut icon')