X-Git-Url: https://git.phdru.name/?p=bookmarks_db.git;a=blobdiff_plain;f=parse_html%2Fbkmk_parse_html.py;h=af9395b0d8d735689288a02071a0e1aee80814de;hp=7bc4640d31ccad2a902335c8a2aad978b02bb09c;hb=7fc9a9ac1bfa749aa30e3ae1d730ac4f266db950;hpb=c88cb7a75e7caf1d67466cfa107981d95115fa0c

diff --git a/parse_html/bkmk_parse_html.py b/parse_html/bkmk_parse_html.py
index 7bc4640..af9395b 100644
--- a/parse_html/bkmk_parse_html.py
+++ b/parse_html/bkmk_parse_html.py
@@ -18,6 +18,14 @@ 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:
@@ -136,6 +144,24 @@ def parse_html(html_text, charset=None, log=None):
         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'
+
     converted_title = title = parser.title
     if title and (not parser.charset):
         try: