current_charset = default_encoding.replace("windows-", "cp")
DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+parsers = []
try:
- from parse_html_beautifulsoup import parse_html as _parse_html
+ from parse_html_beautifulsoup import parse_html
except ImportError:
- from parse_html_htmlparser import parse_html as _parse_html
+ pass
+else:
+ parsers.append(parse_html)
+
+from parse_html_htmlparser import parse_html
+parsers.append(parse_html)
import re
output = []
for part in entity_re.split(title):
if entity_re.match(part):
- part = unichr(int(part[2:-1])).encode(charset, "replace")
+ try:
+ part = unichr(int(part[2:-1])).encode(charset)
+ except UnicodeEncodeError:
+ pass # Leave the entity as is
output.append(part)
return ''.join(output)
except (ValueError, LookupError):
charset = None # ...try charset from HTML
- parser = _parse_html(filename, charset)
- title = parser.title
+ for p in parsers:
+ parser = p(filename, charset)
+ if parser:
+ break
+ else:
+ if log: log("Parser %s.%s failed, trying next one." % (p.__module__, p.__name__))
+ title = parser.title
if not parser.charset:
try:
unicode(title, "ascii")
except LookupError:
if log: log(" unknown charset: `%s' or `%s'" % (parser.charset, current_charset))
- parser.title = recode_entities(title, current_charset)
+ title = recode_entities(title, current_charset)
+ parts = [s.strip() for s in title.replace('\r', '').split('\n')]
+ title = ' '.join([s for s in parts if s])
+ if log: log(" final title : %s" % title)
+ parser.title = title
return parser