I saw a number of sites that forbid "Mozilla compatible".

[bookmarks_db.git] / Robots / parse_html_beautifulsoup.py

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index 8c51d0500b6e78d19475f1b25e6c29c250b3c8c9..3408def7162a50deec9fee0f4894fdff5e37c9ea 100644 (file)
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -20,34 +20,29 @@ class BSoupParser(HTMLParser):
 
 def parse_html(filename, charset=None):
    infile = open(filename, 'r')
-   root = BeautifulSoup(infile, fromEncoding=charset)
+   try:
+      root = BeautifulSoup(infile, fromEncoding=charset)
+   except TypeError:
+      return None
    infile.close()
 
    _charset = root.originalEncoding
    try:
       title = root.html.head.title.string.encode(_charset)
    except AttributeError:
-      title = ''
+      return None
 
-   try:
-      meta = root.html.head.find(_find_refresh, recursive=False)
-   except AttributeError:
-      refresh = None
+   meta = root.html.head.find(_find_refresh, recursive=False)
+   if meta:
+      refresh = meta.get("content")
    else:
-      if meta:
-         refresh = meta.get("content")
-      else:
-         refresh = None
+      refresh = None
 
-   try:
-      meta = root.html.head.find(_find_icon, recursive=False)
-   except AttributeError:
-      icon = None
+   meta = root.html.head.find(_find_icon, recursive=False)
+   if meta:
+      icon = meta.get("href")
    else:
-      if meta:
-         icon = meta.get("href")
-      else:
-         icon = None
+      icon = None
 
    return BSoupParser(_charset, _charset != charset, title, refresh, icon)