def htmlentityreplace_errors(exc):
if isinstance(exc, (UnicodeEncodeError, UnicodeTranslateError)):
res = []
codepoints = []
skip = False
for i, c in enumerate(exc.object[exc.start:exc.end]):
if skip:
skip = False
continue
index = i + exc.start
if utils.isSurrogatePair(exc.object[index:min([exc.end, index+2])]):
codepoint = utils.surrogatePairToCodepoint(exc.object[index:index+2])
skip = True
else:
codepoint = ord(c)
codepoints.append(codepoint)
for cp in codepoints:
e = encode_entity_map.get(cp)
if e:
res.append("&")
res.append(e)
if not e.endswith(";"):
res.append(";")
else:
res.append("&#x%s;"%(hex(cp)[2:]))
return (u"".join(res), exc.end)
else:
return xmlcharrefreplace_errors(exc)
unicode_encode_errors = "strict"
else:
unicode_encode_errors = "htmlentityreplace"
from lib.html5lib.constants import entities
encode_entity_map = {}
is_ucs4 = len(u"\U0010FFFF") == 1
for k, v in entities.items():
#skip multi-character entities
if ((is_ucs4 and len(v) > 1) or
(not is_ucs4 and len(v) > 2)):
continue
if v != "&":
if len(v) == 2:
v = utils.surrogatePairToCodepoint(v)
else:
try:
v = ord(v)
except:
print v
raise
if not v in encode_entity_map or k.islower():
# prefer < over < and similarly for &, >, etc.
encode_entity_map[v] = k
def htmlentityreplace_errors(exc):
if isinstance(exc, (UnicodeEncodeError, UnicodeTranslateError)):
res = []
codepoints = []
skip = False
