Ignore case for DOCTYPE.

[bookmarks_db.git] / Robots / parse_html_beautifulsoup.py

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index 3c52ce402f026876fdc2be8b7dc15613e2eeebcb..ca8ee2e1a0ecc2bdd4f78524e04638a664ec508c 100644 (file)
--- 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.I)
                  if match:
                      toHandle = self.rawdata[i:match.end()]
                  else:
@@ -49,16 +49,27 @@ class BadDeclParser(BeautifulSoup):
          return j
 
 
-def parse_html(filename, charset=None, log=None):
+def _parse_html(filename, charset):
    infile = open(filename, 'r')
    try:
-      root = BadDeclParser(infile, fromEncoding=charset)
+      return BadDeclParser(infile, fromEncoding=charset)
    except TypeError:
-      if log: log("TypeError")
       return None
    finally:
       infile.close()
 
+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:
@@ -77,10 +88,6 @@ def parse_html(filename, charset=None, log=None):
    if head is None:
       head = html # Some sites put TITLE in HTML without HEAD
 
-   _charset = root.originalEncoding
-   if _charset == "windows-1252": # Replace default
-      _charset = DEFAULT_CHARSET
-
    try:
       title = head.title.string.encode(_charset)
    except AttributeError: