Test for completely broken HTML.

[bookmarks_db.git] / Robots / parse_html_beautifulsoup.py

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index fa5139f9e210be3dfb5d658f918689b329c0a220..d3778fbbfc55d529aa03f88c7d094c7bc22e07d3 100644 (file)
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -1,23 +1,13 @@
 """
    HTML Parser using BeautifulSoup
 
-   Written by Broytman. Copyright (C) 2007, 2008 PhiloSoft Design
+   Written by Broytman. Copyright (C) 2007-2010 PhiloSoft Design
 """
 
 import re
 from sgmllib import SGMLParser, SGMLParseError
-from HTMLParser import HTMLParser
 from BeautifulSoup import BeautifulSoup, CData
-
-
-class BSoupParser(HTMLParser):
-   def __init__(self, charset, meta, title, refresh, icon):
-      object.__init__(self)
-      self.charset = charset
-      self.meta_charset = meta
-      self.title = title
-      self.refresh = refresh
-      self.icon = icon
+from parse_html_util import HTMLParser
 
 
 # http://groups.google.com/group/beautifulsoup/browse_thread/thread/69093cb0d3a3cf63
@@ -87,16 +77,18 @@ def parse_html(filename, charset=None, log=None):
       # Lookup TITLE in the root
       title = root.title
 
-   if title is not None:
-      if title.string:
-         title = title.string.encode(_charset)
-      else:
-         parts = []
-         for part in title:
-            if not isinstance(part, basestring):
-               part = unicode(part)
-            parts.append(part.strip())
-         title = ''.join(parts).encode(_charset)
+   if title is None:
+      return None
+
+   if title.string:
+      title = title.string
+   else:
+      parts = []
+      for part in title:
+         if not isinstance(part, basestring):
+            part = unicode(part)
+         parts.append(part.strip())
+      title = ''.join(parts)
 
    meta = head.find(_find_contenttype, recursive=False)
    if meta:
@@ -105,7 +97,7 @@ def parse_html(filename, charset=None, log=None):
          if meta_content:
              __charset = meta_content.lower().split('charset=')[1].split(';')[0]
          else:
-             meta_charset = False
+             __charset = False
       except IndexError: # No charset in the META Content-Type
          meta_charset = False
       else:
@@ -113,6 +105,9 @@ def parse_html(filename, charset=None, log=None):
    else:
       meta_charset = False
 
+   if charset or meta_charset:
+      title = title.encode(charset or meta_charset)
+
    meta = head.find(_find_refresh, recursive=False)
    if meta:
       refresh = meta.get("content")
@@ -125,7 +120,7 @@ def parse_html(filename, charset=None, log=None):
    else:
       icon = None
 
-   return BSoupParser(_charset, meta_charset, title, refresh, icon)
+   return HTMLParser(_charset, meta_charset, title, refresh, icon)
 
 def _find_contenttype(Tag):
    return (Tag.name == "meta") and \