import codecs
import os
-
+import re
try:
- from . import bkmk_ph_beautifulsoup4
+ from html.entities import name2codepoint
except ImportError:
- pass
-else:
- bkmk_ph_beautifulsoup4.DEFAULT_CHARSET = DEFAULT_CHARSET
- parsers.append(bkmk_ph_beautifulsoup4.parse_html)
+ from htmlentitydefs import name2codepoint
+
+from compat import unicode, unichr
+
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+parsers = []
try:
from . import bkmk_ph_beautifulsoup
parsers.append(bkmk_ph_beautifulsoup.parse_html)
try:
- from . import bkmk_ph_html5
+ from . import bkmk_ph_beautifulsoup4
except ImportError:
pass
else:
- parsers.append(bkmk_ph_html5.parse_html)
+ bkmk_ph_beautifulsoup4.DEFAULT_CHARSET = DEFAULT_CHARSET
+ parsers.append(bkmk_ph_beautifulsoup4.parse_html)
try:
- from . import bkmk_ph_lxml
+ from . import bkmk_ph_htmlparser
except ImportError:
pass
else:
- parsers.append(bkmk_ph_lxml.parse_html)
+ parsers.append(bkmk_ph_htmlparser.parse_html)
try:
- from . import bkmk_ph_htmlparser
+ from . import bkmk_ph_lxml
except ImportError:
pass
else:
- parsers.append(bkmk_ph_htmlparser.parse_html)
-
-# ElementTidy often segfaults
-# try:
-# from . import bkmk_ph_etreetidy
-# except ImportError:
-# pass
-# else:
-# parsers.append(bkmk_ph_etreetidy.parse_html)
-
-import re
-from htmlentitydefs import name2codepoint
+ parsers.append(bkmk_ph_lxml.parse_html)
universal_charset = "utf-8"
-DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
-
-parsers = []
-
-entity_re = re.compile("(&\w+;)")
+entity_re = re.compile("(&\\w+;)")
num_entity_re = re.compile("(&#[0-9]+;)")
entity_re.match(part):
_part = name2codepoint.get(part[1:-1], None)
if _part is not None:
- part = unichr(_part).encode(charset)
+ part = unichr(_part)
output.append(part)
title = ''.join(output)
for part in num_entity_re.split(title):
if num_entity_re.match(part):
try:
- part = unichr(int(part[2:-1])).encode(charset)
+ part = unichr(int(part[2:-1]))
except UnicodeEncodeError:
pass # Leave the entity as is
output.append(part)
def parse_html(html_text, charset=None, log=None):
- if not parsers:
+ if not html_text or not parsers:
return None
if charset:
p, parser = _parsers[0]
if log: log(" Using %s" % p.__module__)
- title = parser.title
- if isinstance(title, unicode):
- if parser.charset:
- parser.title = title.encode(parser.charset)
- else:
- try:
- parser.title = title.encode('ascii')
- except UnicodeEncodeError:
- try:
- parser.title = title.encode(DEFAULT_CHARSET)
- except UnicodeEncodeError:
- parser.title = title.encode(universal_charset)
- parser.charset = universal_charset
- else:
- parser.charset = DEFAULT_CHARSET
- else:
- parser.charset = 'ascii'
+ #title = parser.title
+ #if isinstance(title, unicode):
+ # if parser.charset:
+ # parser.title = title.encode(parser.charset)
+ # else:
+ # try:
+ # parser.title = title.encode('ascii')
+ # except UnicodeEncodeError:
+ # try:
+ # parser.title = title.encode(DEFAULT_CHARSET)
+ # except UnicodeEncodeError:
+ # parser.title = title.encode(universal_charset)
+ # parser.charset = universal_charset
+ # else:
+ # parser.charset = DEFAULT_CHARSET
+ # else:
+ # parser.charset = 'ascii'
converted_title = title = parser.title
- if title and (not parser.charset):
+ if title and isinstance(title, bytes) and (not parser.charset):
try:
- unicode(title, "ascii")
+ title.decode("ascii")
except UnicodeDecodeError:
parser.charset = DEFAULT_CHARSET
if log: log(" guessed charset: %s" % parser.charset)
# if log: log(" current charset: %s" % universal_charset)
if log: log(" title : %s" % title)
- if parser.charset != universal_charset:
- try:
- 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")
- parser.charset = DEFAULT_CHARSET
- if log and (converted_title != title): log(" converted title: %s" % converted_title)
+ #if parser.charset != universal_charset:
+ # try:
+ # converted_title = title.decode(parser.charset).\
+ # encode(universal_charset)
+ # except UnicodeError:
+ # if log:
+ # log(" incorrect conversion from %s,"
+ # "converting from %s"
+ # % (parser.charset, DEFAULT_CHARSET))
+ # converted_title = \
+ # title.decode(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'" % 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
- if isinstance(icon, unicode):
- try:
- parser.icon = icon.encode('ascii')
- except UnicodeEncodeError:
- if parser.charset:
- parser.icon = icon.encode(parser.charset)
+ #icon = parser.icon
+ #if isinstance(icon, unicode):
+ # try:
+ # parser.icon = icon.encode('ascii')
+ # except UnicodeEncodeError:
+ # if parser.charset:
+ # parser.icon = icon.encode(parser.charset)
return parser
def parse_filename(filename, charset=None, log=None):
- fp = open(filename, 'r')
+ fp = open(filename, 'rt', encoding=charset)
try:
parser = parse_html(fp.read(), charset=charset, log=log)
finally: