From: Oleg Broytman <phd@phdru.name>
Date: Sun, 24 Feb 2008 21:37:24 +0000 (+0000)
Subject: Used name2codepoint directly; recode it.
X-Git-Tag: v4.5.3~192
X-Git-Url: https://git.phdru.name/?a=commitdiff_plain;h=e7035a2a59f6a2cb55aa7ca852e2cff3b66b59d9;p=bookmarks_db.git

Used name2codepoint directly; recode it.


git-svn-id: file:///home/phd/archive/SVN/bookmarks_db/trunk@184 fdd5c36f-1aea-0310-aeeb-c58d7e2b6c23
---

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 57f6498..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]+;)")
@@ -34,19 +34,20 @@ def recode_entities(title, charset):
    for part in entity_re.split(title):
       if part not in ("&amp;", "&lt;", "&gt;", "&quote;", "&nbsp;") and \
             entity_re.match(part):
-         part = entitydefs.get(part[1:-1], 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):