urllib._urlopener = MyURLopener()
-# Some sites allow only Mozilla-compatible browsers; way to stop robots?
-server_version = "Mozilla/3.0 (compatible; Python-urllib/%s)" % urllib.__version__
+server_version = "bookmarks_db (Python-urllib/%s)" % urllib.__version__
urllib._urlopener.addheaders[0] = ('User-agent', server_version)
from parse_html import parse_html
class robot_simple(Robot):
- def check_url(self, bookmark, url_type, url_rest):
+ def check_url(self, bookmark):
if not self.tempfname:
self.tempfname = bookmark.tempfname
try:
try:
self.start = int(time.time())
+ bookmark.icon = None
+
+ url_type, url_rest = urllib.splittype(bookmark.href)
url_host, url_path = urllib.splithost(url_rest)
url_path, url_tag = urllib.splittag(url_path)
icon = parser.icon
if not icon:
icon = "/favicon.ico"
- icon = urljoin("%s://%s" % (url_type, url_host), icon)
+ icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon)
if self.log: self.log(" icon : %s" % icon)
try:
fname, headers = urllib.urlretrieve(icon)
emsg = None
tb = None
else:
- icon_file = open(fname, "rb")
- icon = icon_file.read()
- icon_file.close()
- os.remove(fname)
content_type = headers["Content-Type"]
- bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+ if content_type.startswith("image/"):
+ icon_file = open(fname, "rb")
+ icon = icon_file.read()
+ icon_file.close()
+ bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+ else:
+ if self.log: self.log(" no icon : %s" % content_type)
+ os.remove(fname)
except KeyError:
pass