import codecs
-
-universal_charset = "utf-8"
-DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
-
-parsers = []
+import os
try:
from . import bkmk_ph_beautifulsoup4
import re
from htmlentitydefs import name2codepoint
+universal_charset = "utf-8"
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+
+parsers = []
+
entity_re = re.compile("(&\w+;)")
num_entity_re = re.compile("(&#[0-9]+;)")
+
def recode_entities(title, charset):
output = []
for part in entity_re.split(title):
return ''.join(output)
-import os
BKMK_DEBUG_HTML_PARSERS = os.environ.get("BKMK_DEBUG_HTML_PARSERS")
+
def parse_html(html_text, charset=None, log=None):
if not parsers:
return None
if log: log(" title : %s" % title)
if parser.charset != universal_charset:
try:
- converted_title = unicode(title, parser.charset).encode(universal_charset)
+ converted_title = unicode(title, parser.charset).\
+ encode(universal_charset)
except UnicodeError:
- if log: log(" incorrect conversion from %s, converting from %s" % (parser.charset, DEFAULT_CHARSET))
- converted_title = unicode(title, DEFAULT_CHARSET, "replace").encode(universal_charset, "replace")
+ if log:
+ log(" incorrect conversion from %s,"
+ "converting from %s"
+ % (parser.charset, DEFAULT_CHARSET))
+ converted_title = \
+ unicode(title, DEFAULT_CHARSET, "replace").\
+ encode(universal_charset, "replace")
parser.charset = DEFAULT_CHARSET
- if log and (converted_title != title): log(" converted title: %s" % converted_title)
+ if log and (converted_title != title):
+ log(" converted title: %s" % converted_title)
except LookupError:
if log: log(" unknown charset: '%s'" % parser.charset)
else:
final_title = recode_entities(converted_title, universal_charset)
parts = [s.strip() for s in final_title.replace('\r', '').split('\n')]
final_title = ' '.join([s for s in parts if s])
- if log and (final_title != converted_title): log(" final title : %s" % final_title)
+ if log and (final_title != converted_title):
+ log(" final title : %s" % final_title)
parser.title = final_title
icon = parser.icon
parser.icon = icon.encode(parser.charset)
return parser
+
def parse_filename(filename, charset=None, log=None):
fp = open(filename, 'r')
try: