import re
-from htmlentitydefs import entitydefs
+from htmlentitydefs import name2codepoint
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):
- if entity_re.match(part):
- part = entitydefs.get(part[1:-1], part)
+ if part not in ("&", "<", ">", ""e;", " ") and \
+ entity_re.match(part):
+ part = unichr(name2codepoint.get(part[1:-1], part)).encode(charset)
output.append(part)
+ title = ''.join(output)
- output2 = []
- for part in num_entity_re.split(''.join(output)):
+ output = []
+ for part in num_entity_re.split(title):
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)
+ output.append(part)
- return ''.join(output2)
+ return ''.join(output)
def parse_html(filename, charset=None, log=None):
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
- print parser.title.decode(universal_charset).encode(current_charset, "replace")
- print parser.refresh
- print parser.icon
+ parser = parse_html(sys.argv[1], universal_charset,
+ log=lambda s: sys.stdout.write(s + '\n'))
+ print " refresh:", parser.refresh
+ print " icon :", parser.icon