"""
-import sys, os, string
+import sys, os
import time, urllib
from base64 import b64encode
from urlparse import urljoin
}
def __init__(self, errcode, newurl):
Exception.__init__(self, "(%s) to %s" % (self.reloc_dict[errcode], newurl))
+ self.url = newurl
class MyURLopener(urllib.URLopener):
urllib._urlopener = MyURLopener()
-server_version = "bookmarks_db (Python-urllib/%s)" % urllib.__version__
-urllib._urlopener.addheaders[0] = ('User-agent', server_version)
+# Fake headers to pretend this is a real browser
+_version = "Links (2.1; Linux 2.6 i686; 80x25)"
+urllib._urlopener.addheaders[0] = ('User-Agent', _version)
+_version = "bookmarks_db (Python %d.%d.%d; urllib/%s)" % (
+ sys.version_info[0], sys.version_info[1], sys.version_info[2], urllib.__version__)
+urllib._urlopener.addheader('X-User-Agent', _version)
+
+urllib._urlopener.addheader('Connection', 'close')
+urllib._urlopener.addheader('Content-Length', '0')
+urllib._urlopener.addheader('Accept', '*/*')
+urllib._urlopener.addheader('Accept-Language', 'ru,en')
+urllib._urlopener.addheader('Cache-Control', 'max-age=300')
+urllib._urlopener.addheader('Referer', 'http://www.yahoo.com/')
def get_error(msg):
- if type(msg) == type(""):
+ if isinstance(msg, str):
return msg
else:
s = []
for i in msg:
- s.append("'%s'" % string.join(string.split(str(i), "\n"), "\\n"))
- return "(%s)" % string.join(s)
+ s.append("'%s'" % str(i).replace('\n', "\\n"))
+ return "(%s)" % ' '.join(s)
urllib_ftpwrapper = urllib.ftpwrapper
def __init__(self, user, passwd, host, port, dirs):
urllib_ftpwrapper.__init__(self, user, passwd, host, port, dirs)
global ftpcache_key
- ftpcache_key = (user, host, port, string.join(dirs, '/'))
+ ftpcache_key = (user, host, port, tuple(dirs))
urllib.ftpwrapper = myftpwrapper
if self.log: self.log(" no charset in Content-Type header")
if content_type == "text/html":
parser = parse_html(fname, charset, self.log)
- title = parser.title.replace('\r', '').replace('\n', ' ').strip()
- bookmark.real_title = parser.unescape(title)
- if self.log: self.log(" final title : %s" % bookmark.real_title)
+ bookmark.real_title = parser.title
if parser.refresh:
refresh = parser.refresh
try:
except (IndexError, ValueError):
raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh))
else:
+ try:
+ timeout = int(refresh.split(';')[0])
+ except ValueError:
+ pass # float timeout
raise RedirectException("html", "%s (%s sec)" % (url, timeout))
# Get favicon.ico
if not icon:
icon = "/favicon.ico"
icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon)
- if self.log: self.log(" icon : %s" % icon)
+ if self.log: self.log(" looking for icon at: %s" % icon)
try:
- fname, headers = urllib.urlretrieve(icon)
+ for i in range(8):
+ try:
+ fname, headers = urllib.urlretrieve(icon)
+ except RedirectException, e:
+ icon = e.url
+ if self.log: self.log(" redirect to : %s" % icon)
+ else:
+ break
+ else:
+ raise IOError("Too many redirects")
except:
etype, emsg, tb = sys.exc_info()
if self.log: self.log(" no icon : %s %s" % (etype, emsg))
icon = icon_file.read()
icon_file.close()
bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+ if self.log: self.log(" got icon : %s" % content_type)
else:
- if self.log: self.log(" no icon : %s" % content_type)
+ if self.log: self.log(" no icon : bad content type '%s'" % content_type)
os.remove(fname)
except KeyError: