X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_parse_html.py;h=2e7df1a2b90fde750d0d47aafcf073a11c60c3a5;hb=8a33ee42d7f8edff076734eba8c1718da8127b22;hp=6afb7f9df735f6a487b6b7537fc23f2c3daa71c2;hpb=cb9c36b39ed72cd1fa272130d2bcf162a89c3013;p=bookmarks_db.git

diff --git a/parse_html/bkmk_parse_html.py b/parse_html/bkmk_parse_html.py
index 6afb7f9..2e7df1a 100644
--- a/parse_html/bkmk_parse_html.py
+++ b/parse_html/bkmk_parse_html.py
@@ -13,6 +13,13 @@ __all__ = ['parse_html', 'parse_filename', 'universal_charset']
 
 import codecs
 import os
+import re
+from htmlentitydefs import name2codepoint
+
+from compat import unicode, unichr
+
+DEFAULT_CHARSET = "cp1251"  # Stupid default for Russian Cyrillic
+parsers = []
 
 try:
     from . import bkmk_ph_beautifulsoup4
@@ -59,15 +66,8 @@ else:
 # else:
 #     parsers.append(bkmk_ph_etreetidy.parse_html)
 
-import re
-from htmlentitydefs import name2codepoint
-
 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]+;)")
 
 
@@ -167,7 +167,7 @@ def parse_html(html_text, charset=None, log=None):
     converted_title = title = parser.title
     if title and (not parser.charset):
         try:
-            unicode(title, "ascii")
+            title.decode("ascii")
         except UnicodeDecodeError:
             parser.charset = DEFAULT_CHARSET
 
@@ -186,7 +186,7 @@ def parse_html(html_text, charset=None, log=None):
             if log: log("   title          : %s" % title)
             if parser.charset != universal_charset:
                 try:
-                    converted_title = unicode(title, parser.charset).\
+                    converted_title = title.decode(parser.charset).\
                         encode(universal_charset)
                 except UnicodeError:
                     if log:
@@ -194,7 +194,7 @@ def parse_html(html_text, charset=None, log=None):
                             "converting from %s"
                             % (parser.charset, DEFAULT_CHARSET))
                     converted_title = \
-                        unicode(title, DEFAULT_CHARSET, "replace").\
+                        title.decode(DEFAULT_CHARSET, "replace").\
                         encode(universal_charset, "replace")
                     parser.charset = DEFAULT_CHARSET
             if log and (converted_title != title):