application/xhtml+xml is HTML, too.

[bookmarks_db.git] / Robots / parse_html_beautifulsoup.py

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index 25719ca9d643ad86e2c2844c572b938d1bf6f792..93a4d6fd9a2ba27ba0c1de2fdcdb3fe9971461c0 100644 (file)
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -49,39 +49,51 @@ 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)
+
+   _charset = root.originalEncoding
+   if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"): # Replace default
+      _charset = DEFAULT_CHARSET
+      root = _parse_html(filename, _charset)
+
    try:
-      head = root.html.head
+      html = root.html
    except AttributeError:
-      if log: log("No HTML in root or no HEAD in HTML")
-      return None
+      if log: log("No HTML in root")
+      html = root
 
-   if head is None:
-      head = root.html # Some sites put TITLE in HTML without HEAD
+   if html is None:
+      html = root
 
-   _charset = root.originalEncoding
-   if _charset == "windows-1252": # Replace default
-      _charset = DEFAULT_CHARSET
+   try:
+      head = html.head
+   except AttributeError:
+      if log: log("No HEAD in HTML")
+      head = html
+
+   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 root.html):
+   if (not title) and (head is not html):
       # Some sites put TITLE in HTML outside of HEAD
 
       try:
-         title = root.html.title.string.encode(_charset)
+         title = html.title.string.encode(_charset)
       except AttributeError:
          title = '' # no TITLE in HTML too