X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=680ad5573a741bb99402449a77d969202e49abde;hb=e7035a2a59f6a2cb55aa7ca852e2cff3b66b59d9;hp=0d7b20b5ce82b88d596b1cb25aac90084a997735;hpb=3a4e6b782c229014a6cc49f1e68681696c1ed1f3;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 0d7b20b..680ad55 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -24,7 +24,7 @@ parsers.append(parse_html)
 
 
 import re
-from htmlentitydefs import entitydefs
+from htmlentitydefs import name2codepoint
 
 entity_re = re.compile("(&\w+;)")
 num_entity_re = re.compile("(&#[0-9]+;)")
@@ -32,20 +32,22 @@ num_entity_re = re.compile("(&#[0-9]+;)")
 def recode_entities(title, charset):
    output = []
    for part in entity_re.split(title):
-      if entity_re.match(part):
-         part = entitydefs.get(part[1:-1], part)
+      if part not in ("&", "<", ">", "&quote;", " ") and \
+            entity_re.match(part):
+         part = unichr(name2codepoint.get(part[1:-1], part)).encode(charset)
       output.append(part)
+   title = ''.join(output)
 
-   output2 = []
-   for part in num_entity_re.split(''.join(output)):
+   output = []
+   for part in num_entity_re.split(title):
       if num_entity_re.match(part):
          try:
             part = unichr(int(part[2:-1])).encode(charset)
          except UnicodeEncodeError:
             pass # Leave the entity as is
-      output2.append(part)
+      output.append(part)
 
-   return ''.join(output2)
+   return ''.join(output)
 
 
 def parse_html(filename, charset=None, log=None):