X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_parse_html.py;h=922c745346ecfa8f8f9c2988ef94f85345dc167a;hb=d454f1d6aa7f2430d502d847693515f69489c66c;hp=69d9035f5001759cb7689221775572b3e49b1373;hpb=9e2bf3eaee5a8ee84c61ee5eac9c55090d45f63f;p=bookmarks_db.git

diff --git a/parse_html/bkmk_parse_html.py b/parse_html/bkmk_parse_html.py
index 69d9035..922c745 100644
--- a/parse_html/bkmk_parse_html.py
+++ b/parse_html/bkmk_parse_html.py
@@ -14,21 +14,16 @@ __all__ = ['parse_html', 'parse_filename', 'universal_charset']
 import codecs
 import os
 import re
-from htmlentitydefs import name2codepoint
+try:
+    from html.entities import name2codepoint
+except ImportError:
+    from htmlentitydefs import name2codepoint
 
 from compat import unicode, unichr
 
 DEFAULT_CHARSET = "cp1251"  # Stupid default for Russian Cyrillic
 parsers = []
 
-try:
-    from . import bkmk_ph_beautifulsoup4
-except ImportError:
-    pass
-else:
-    bkmk_ph_beautifulsoup4.DEFAULT_CHARSET = DEFAULT_CHARSET
-    parsers.append(bkmk_ph_beautifulsoup4.parse_html)
-
 try:
     from . import bkmk_ph_beautifulsoup
 except ImportError:
@@ -38,33 +33,26 @@ else:
     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)
+    parsers.append(bkmk_ph_lxml.parse_html)
 
 universal_charset = "utf-8"
 entity_re = re.compile("(&\\w+;)")
@@ -98,7 +86,7 @@ BKMK_DEBUG_HTML_PARSERS = os.environ.get("BKMK_DEBUG_HTML_PARSERS")
 
 
 def parse_html(html_text, charset=None, log=None):
-    if not parsers:
+    if not html_text or not parsers:
         return None
 
     if charset:
@@ -165,7 +153,7 @@ def parse_html(html_text, charset=None, log=None):
     #            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:
             title.decode("ascii")
         except UnicodeDecodeError:
@@ -223,7 +211,7 @@ def parse_html(html_text, charset=None, log=None):
 
 
 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: