def fetch(self, code):
resp = self._http_fetch(code)
if resp.status == 200:
return self._fetch_200(code)
elif resp.status == 301:
location = resp.getheader("Location")
if not location:
raise exceptions.CodeBlockedException(
"No Location header after HTTP status 301")
tiny = resp.getheader("X-tiny")
if tiny and tiny[:3] == "aff":
return self._preview(code)
return location
elif resp.status == 302:
raise exceptions.CodeBlockedException()
elif resp.status == 404:
raise exceptions.NoRedirectException()
elif resp.status == 500:
# Some "errorhelp" URLs result in HTTP status 500, which goes away when trying a different server
self._conn.close()
raise exceptions.ServiceException("HTTP status 500")
else:
raise exceptions.ServiceException("Unknown HTTP status %i" %
resp.status)
return resp.status
def fetch(self, code):
resp = self._http_fetch(code)
if resp.status == 301:
location = resp.getheader("Location")
if not location:
raise exceptions.ServiceException(
"No Location header after HTTP status 301")
if resp.reason == "Moved": # Normal bit.ly redirect
return location
elif resp.reason == "Moved Permanently":
# Weird "bundles" redirect, forces connection close despite
# sending Keep-Alive header
self._conn.close()
raise exceptions.CodeBlockedException()
else:
raise exceptions.ServiceException(
"Unknown HTTP reason %s after HTTP status 301" %
resp.reason)
elif resp.status == 302:
location = resp.getheader("Location")
if not location:
raise exceptions.ServiceException(
"No Location header after HTTP status 302")
return self._parse_warning_url(code, location)
elif resp.status == 403:
raise exceptions.BlockedException()
elif resp.status == 404:
raise exceptions.NoRedirectException()
elif resp.status == 410:
raise exceptions.CodeBlockedException()
else:
raise exceptions.ServiceException("Unknown HTTP status %i" %
resp.status)
def _fetch_blocked(self, code):
resp = self._http_fetch(code, "GET")
data = resp.read()
if resp.status != 200:
raise exceptions.ServiceException(
"HTTP status changed from 200 to %i on second request" %
resp.status)
if not data:
raise exceptions.CodeBlockedException(
"Empty response on status 200")
if self.RATE_LIMIT_STRING in data:
raise exceptions.BlockedException()
position = data.find(self.BLOCKED_STRING_START)
if position == -1:
raise exceptions.ServiceException(
"Unexpected response on status 200")
data = data[position + len(self.BLOCKED_STRING_START):]
position = data.find(self.BLOCKED_STRING_END)
if position == -1:
raise exceptions.ServiceException(
"Unexpected response on status 200")
url = data[:position].decode("utf-8")
return HTMLParser.HTMLParser().unescape(url).encode("utf-8")
def fetch(self, code):
if code == "500":
raise exceptions.CodeBlockedException()
url = super(Trimnew, self).fetch(code)
if url == "http://tr.im/404":
raise exceptions.NoRedirectException()
return url
def fetch(self, code):
location = super(Snipurl, self).fetch(code)
try:
if location.decode("ascii") == "/site/getprivate?snip=" + code:
raise exceptions.CodeBlockedException("Private key required")
except UnicodeDecodeError:
pass
return location
def _parse_blocked(self, code, data):
match = re.search(
"<p>For reference and to help those fighting spam the original destination of this URL is given below \(we strongly recommend you don't visit it since it may damage your PC\): -<br />(.*)</p><h2>is\.gd</h2><p>is\.gd is a free service used to shorten long URLs\.",
data)
if not match:
raise exceptions.ServiceException(
"Could not find target URL in 'Link Disabled' page")
url = match.group(1).decode("utf-8")
url = HTMLParser.HTMLParser().unescape(url).encode("utf-8")
if url == "":
raise exceptions.CodeBlockedException("Empty URL on preview")
return url
def _parse_json(self, data):
try:
data = json.loads(data)
except ValueError:
raise exceptions.ServiceException("Could not decode response")
if not "kind" in data or data["kind"] != "urlshortener#url":
raise exceptions.ServiceException("No/bad type given")
if not "status" in data:
raise exceptions.ServiceException("No status given")
if not "longUrl" in data:
raise exceptions.CodeBlockedException("Status: %s" %
data["status"])
return data["longUrl"]
def fetch(self, code):
resp = self._http_fetch(code)
if resp.status == 200:
return self._fetch_blocked(code)
elif resp.status == 301:
location = resp.getheader("Location")
if not location:
raise exceptions.CodeBlockedException(
"No Location header after HTTP status 301")
return location
elif resp.status == 404:
raise exceptions.NoRedirectException()
else:
raise exceptions.ServiceException("Unknown HTTP status %i" %
resp.status)
def fetch(self, code):
resp = self._http_head(code)
if resp.status in self.http_status_redirect:
location = resp.getheader("Location")
if not location:
raise exceptions.ServiceException(
"No Location header after HTTP status 301")
return location
elif resp.status in self.http_status_no_redirect:
raise exceptions.NoRedirectException()
elif resp.status in self.http_status_code_blocked:
raise exceptions.CodeBlockedException()
elif resp.status in self.http_status_blocked:
raise exceptions.BlockedException()
else:
return self.unexpected_http_status(code, resp)
def unexpected_http_status(self, code, resp):
if resp.status != 200:
return super(Isgd, self).unexpected_http_status(code, resp)
resp, data = self._http_get(code)
if resp.status != 200:
raise exceptions.ServiceException(
"HTTP status changed from 200 to %i on second request" %
resp.status)
if not data:
raise exceptions.CodeBlockedException(
"Empty response on status 200")
if "<div id=\"main\"><p>Rate limit exceeded - please wait 1 minute before accessing more shortened URLs</p></div>" in data:
raise exceptions.BlockedException()
if "<div id=\"disabled\"><h2>Link Disabled</h2>" in data:
return self._parse_blocked(code, data)
if "<p>The full original link is shown below. <b>Click the link</b> if you'd like to proceed to the destination shown:" in data:
return self._parse_preview(code, data)