import codecs
-from m_lib.defenc import default_encoding
-current_charset = default_encoding.replace("windows-", "cp")
-
universal_charset = "utf-8"
DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
parsers = []
try:
- from parse_html_beautifulsoup import parse_html
+ import parse_html_beautifulsoup
+ parse_html_beautifulsoup.DEFAULT_CHARSET = DEFAULT_CHARSET
except ImportError:
pass
else:
- parsers.append(parse_html)
+ parsers.append(parse_html_beautifulsoup.parse_html)
from parse_html_htmlparser import parse_html
parsers.append(parse_html)
def recode_entities(title, charset):
output = []
- for part in num_entity_re.split(title):
+ for part in entity_re.split(title):
+ if entity_re.match(part):
+ part = entitydefs.get(part[1:-1], part)
+ output.append(part)
+
+ output2 = []
+ for part in num_entity_re.split(''.join(output)):
if num_entity_re.match(part):
try:
part = unichr(int(part[2:-1])).encode(charset)
except UnicodeEncodeError:
pass # Leave the entity as is
- output.append(part)
-
- output2 = []
- for part in entity_re.split(''.join(output)):
- if entity_re.match(part):
- part = entitydefs.get(part[1:-1], part)
- if num_entity_re.match(part):
- try:
- part = unichr(int(part[2:-1])).encode(charset)
- except UnicodeEncodeError:
- pass # Leave the entity as is
output2.append(part)
return ''.join(output2)
charset = None # ...try charset from HTML
for p in parsers:
- parser = p(filename, charset)
+ charsets = [universal_charset, DEFAULT_CHARSET]
+ if charset not in charsets:
+ charsets.insert(0, charset)
+ parser = None
+ for c in charsets:
+ try:
+ parser = p(filename, c)
+ break
+ except UnicodeEncodeError:
+ pass
if parser:
break
else:
if parser.meta_charset:
if log: log(" META charset : %s" % parser.charset)
else:
- if log: log(" HTTP charset : %s" % parser.charset)
+ if log: log(" guessed charset: %s" % parser.charset)
if log: log(" current charset: %s" % universal_charset)
if log: log(" title : %s" % title)
- save_title = title
try:
converted_title = unicode(title, parser.charset).encode(universal_charset)
except UnicodeError:
- if parser.meta_charset and parser.charset.endswith("1252") and \
- not DEFAULT_CHARSET.endswith("1252") and (DEFAULT_CHARSET <> universal_charset):
- parser.charset = DEFAULT_CHARSET
- if log: log(" incorrect conversion from cp1252, converting from %s" % DEFAULT_CHARSET)
- converted_title = unicode(save_title, DEFAULT_CHARSET, "replace").encode(universal_charset, "replace")
- else:
- converted_title = unicode(title, parser.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)
except LookupError:
if log: log(" unknown charset: `%s' or `%s'" % (parser.charset, universal_charset))
if __name__ == '__main__':
import sys
+ from m_lib.defenc import default_encoding
+ current_charset = default_encoding.replace("windows-", "cp")
+
parser = parse_html(sys.argv[1], universal_charset)
print parser.charset
print parser.title