X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html_beautifulsoup.py;h=db291f9fd46df8607709333dbaa304eaba00007e;hb=73f38dccb4a11e857ae3497a61089a4ea89cd654;hp=93a4d6fd9a2ba27ba0c1de2fdcdb3fe9971461c0;hpb=896c8750d5b4e7a833156cc87a1f2f5847729904;p=bookmarks_db.git

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index 93a4d6f..db291f9 100644
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -39,7 +39,7 @@ class BadDeclParser(BeautifulSoup):
              except SGMLParseError:
                  # Could not parse the DOCTYPE declaration
                  # Try to just skip the actual declaration
-                 match = re.search(r'<!DOCTYPE([^>]*?)>', self.rawdata[i:], re.MULTILINE)
+                 match = re.search(r'<!DOCTYPE([^>]*?)>', self.rawdata[i:], re.MULTILINE|re.IGNORECASE)
                  if match:
                      toHandle = self.rawdata[i:match.end()]
                  else:
@@ -60,42 +60,44 @@ def _parse_html(filename, charset):
 
 def parse_html(filename, charset=None, log=None):
    root = _parse_html(filename, charset)
+   if root is None:
+      return None
 
    _charset = root.originalEncoding
    if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"): # Replace default
       _charset = DEFAULT_CHARSET
       root = _parse_html(filename, _charset)
+      if root is None:
+         return None
 
-   try:
-      html = root.html
-   except AttributeError:
-      if log: log("No HTML in root")
-      html = root
-
+   html = root.html
    if html is None:
       html = root
 
-   try:
-      head = html.head
-   except AttributeError:
-      if log: log("No HEAD in HTML")
-      head = html
-
+   head = html.head
    if head is None:
       head = html # Some sites put TITLE in HTML without HEAD
 
-   try:
-      title = head.title.string.encode(_charset)
-   except AttributeError:
-      title = '' # HEAD but no TITLE
-
-   if (not title) and (head is not html):
+   title = head.title
+   if (title is None) and (html is not head):
       # Some sites put TITLE in HTML outside of HEAD
+      title = html.title
 
-      try:
-         title = html.title.string.encode(_charset)
-      except AttributeError:
-         title = '' # no TITLE in HTML too
+   if title is None:
+      # Lookup TITLE in the root
+      title = root.title
+
+   if title is None:
+      title = ''
+   elif 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)
 
    meta = head.find(_find_contenttype, recursive=False)
    if meta: