Cleaned up oauth code based on feedback. Created FlowThreeLegged and OAuthCredentials classes.
diff --git a/httplib2/__init__.py b/httplib2/__init__.py
new file mode 100644
index 0000000..61e9caa
--- /dev/null
+++ b/httplib2/__init__.py
@@ -0,0 +1,1208 @@
+from __future__ import generators
+"""
+httplib2
+
+A caching http interface that supports ETags and gzip
+to conserve bandwidth.
+
+Requires Python 2.3 or later
+
+Changelog:
+2007-08-18, Rick: Modified so it's able to use a socks proxy if needed.
+
+"""
+
+__author__ = "Joe Gregorio (joe@bitworking.org)"
+__copyright__ = "Copyright 2006, Joe Gregorio"
+__contributors__ = ["Thomas Broyer (t.broyer@ltgt.net)",
+ "James Antill",
+ "Xavier Verges Farrero",
+ "Jonathan Feinberg",
+ "Blair Zajac",
+ "Sam Ruby",
+ "Louis Nyffenegger"]
+__license__ = "MIT"
+__version__ = "$Rev$"
+
+import re
+import sys
+import email
+import email.Utils
+import email.Message
+import email.FeedParser
+import StringIO
+import gzip
+import zlib
+import httplib
+import urlparse
+import base64
+import os
+import copy
+import calendar
+import time
+import random
+import errno
+# remove depracated warning in python2.6
+try:
+ from hashlib import sha1 as _sha, md5 as _md5
+except ImportError:
+ import sha
+ import md5
+ _sha = sha.new
+ _md5 = md5.new
+import hmac
+from gettext import gettext as _
+import socket
+
+try:
+ from httplib2 import socks
+except ImportError:
+ socks = None
+
+# Build the appropriate socket wrapper for ssl
+try:
+ import ssl # python 2.6
+ _ssl_wrap_socket = ssl.wrap_socket
+except ImportError:
+ def _ssl_wrap_socket(sock, key_file, cert_file):
+ ssl_sock = socket.ssl(sock, key_file, cert_file)
+ return httplib.FakeSocket(sock, ssl_sock)
+
+
+if sys.version_info >= (2,3):
+ from iri2uri import iri2uri
+else:
+ def iri2uri(uri):
+ return uri
+
+def has_timeout(timeout): # python 2.6
+ if hasattr(socket, '_GLOBAL_DEFAULT_TIMEOUT'):
+ return (timeout is not None and timeout is not socket._GLOBAL_DEFAULT_TIMEOUT)
+ return (timeout is not None)
+
+__all__ = ['Http', 'Response', 'ProxyInfo', 'HttpLib2Error',
+ 'RedirectMissingLocation', 'RedirectLimit', 'FailedToDecompressContent',
+ 'UnimplementedDigestAuthOptionError', 'UnimplementedHmacDigestAuthOptionError',
+ 'debuglevel']
+
+
+# The httplib debug level, set to a non-zero value to get debug output
+debuglevel = 0
+
+
+# Python 2.3 support
+if sys.version_info < (2,4):
+ def sorted(seq):
+ seq.sort()
+ return seq
+
+# Python 2.3 support
+def HTTPResponse__getheaders(self):
+ """Return list of (header, value) tuples."""
+ if self.msg is None:
+ raise httplib.ResponseNotReady()
+ return self.msg.items()
+
+if not hasattr(httplib.HTTPResponse, 'getheaders'):
+ httplib.HTTPResponse.getheaders = HTTPResponse__getheaders
+
+# All exceptions raised here derive from HttpLib2Error
+class HttpLib2Error(Exception): pass
+
+# Some exceptions can be caught and optionally
+# be turned back into responses.
+class HttpLib2ErrorWithResponse(HttpLib2Error):
+ def __init__(self, desc, response, content):
+ self.response = response
+ self.content = content
+ HttpLib2Error.__init__(self, desc)
+
+class RedirectMissingLocation(HttpLib2ErrorWithResponse): pass
+class RedirectLimit(HttpLib2ErrorWithResponse): pass
+class FailedToDecompressContent(HttpLib2ErrorWithResponse): pass
+class UnimplementedDigestAuthOptionError(HttpLib2ErrorWithResponse): pass
+class UnimplementedHmacDigestAuthOptionError(HttpLib2ErrorWithResponse): pass
+
+class RelativeURIError(HttpLib2Error): pass
+class ServerNotFoundError(HttpLib2Error): pass
+
+# Open Items:
+# -----------
+# Proxy support
+
+# Are we removing the cached content too soon on PUT (only delete on 200 Maybe?)
+
+# Pluggable cache storage (supports storing the cache in
+# flat files by default. We need a plug-in architecture
+# that can support Berkeley DB and Squid)
+
+# == Known Issues ==
+# Does not handle a resource that uses conneg and Last-Modified but no ETag as a cache validator.
+# Does not handle Cache-Control: max-stale
+# Does not use Age: headers when calculating cache freshness.
+
+
+# The number of redirections to follow before giving up.
+# Note that only GET redirects are automatically followed.
+# Will also honor 301 requests by saving that info and never
+# requesting that URI again.
+DEFAULT_MAX_REDIRECTS = 5
+
+# Which headers are hop-by-hop headers by default
+HOP_BY_HOP = ['connection', 'keep-alive', 'proxy-authenticate', 'proxy-authorization', 'te', 'trailers', 'transfer-encoding', 'upgrade']
+
+def _get_end2end_headers(response):
+ hopbyhop = list(HOP_BY_HOP)
+ hopbyhop.extend([x.strip() for x in response.get('connection', '').split(',')])
+ return [header for header in response.keys() if header not in hopbyhop]
+
+URI = re.compile(r"^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?")
+
+def parse_uri(uri):
+ """Parses a URI using the regex given in Appendix B of RFC 3986.
+
+ (scheme, authority, path, query, fragment) = parse_uri(uri)
+ """
+ groups = URI.match(uri).groups()
+ return (groups[1], groups[3], groups[4], groups[6], groups[8])
+
+def urlnorm(uri):
+ (scheme, authority, path, query, fragment) = parse_uri(uri)
+ if not scheme or not authority:
+ raise RelativeURIError("Only absolute URIs are allowed. uri = %s" % uri)
+ authority = authority.lower()
+ scheme = scheme.lower()
+ if not path:
+ path = "/"
+ # Could do syntax based normalization of the URI before
+ # computing the digest. See Section 6.2.2 of Std 66.
+ request_uri = query and "?".join([path, query]) or path
+ scheme = scheme.lower()
+ defrag_uri = scheme + "://" + authority + request_uri
+ return scheme, authority, request_uri, defrag_uri
+
+
+# Cache filename construction (original borrowed from Venus http://intertwingly.net/code/venus/)
+re_url_scheme = re.compile(r'^\w+://')
+re_slash = re.compile(r'[?/:|]+')
+
+def safename(filename):
+ """Return a filename suitable for the cache.
+
+ Strips dangerous and common characters to create a filename we
+ can use to store the cache in.
+ """
+
+ try:
+ if re_url_scheme.match(filename):
+ if isinstance(filename,str):
+ filename = filename.decode('utf-8')
+ filename = filename.encode('idna')
+ else:
+ filename = filename.encode('idna')
+ except UnicodeError:
+ pass
+ if isinstance(filename,unicode):
+ filename=filename.encode('utf-8')
+ filemd5 = _md5(filename).hexdigest()
+ filename = re_url_scheme.sub("", filename)
+ filename = re_slash.sub(",", filename)
+
+ # limit length of filename
+ if len(filename)>200:
+ filename=filename[:200]
+ return ",".join((filename, filemd5))
+
+NORMALIZE_SPACE = re.compile(r'(?:\r\n)?[ \t]+')
+def _normalize_headers(headers):
+ return dict([ (key.lower(), NORMALIZE_SPACE.sub(value, ' ').strip()) for (key, value) in headers.iteritems()])
+
+def _parse_cache_control(headers):
+ retval = {}
+ if headers.has_key('cache-control'):
+ parts = headers['cache-control'].split(',')
+ parts_with_args = [tuple([x.strip().lower() for x in part.split("=", 1)]) for part in parts if -1 != part.find("=")]
+ parts_wo_args = [(name.strip().lower(), 1) for name in parts if -1 == name.find("=")]
+ retval = dict(parts_with_args + parts_wo_args)
+ return retval
+
+# Whether to use a strict mode to parse WWW-Authenticate headers
+# Might lead to bad results in case of ill-formed header value,
+# so disabled by default, falling back to relaxed parsing.
+# Set to true to turn on, usefull for testing servers.
+USE_WWW_AUTH_STRICT_PARSING = 0
+
+# In regex below:
+# [^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+ matches a "token" as defined by HTTP
+# "(?:[^\0-\x08\x0A-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?" matches a "quoted-string" as defined by HTTP, when LWS have already been replaced by a single space
+# Actually, as an auth-param value can be either a token or a quoted-string, they are combined in a single pattern which matches both:
+# \"?((?<=\")(?:[^\0-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?(?=\")|(?<!\")[^\0-\x08\x0A-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+(?!\"))\"?
+WWW_AUTH_STRICT = re.compile(r"^(?:\s*(?:,\s*)?([^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+)\s*=\s*\"?((?<=\")(?:[^\0-\x08\x0A-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?(?=\")|(?<!\")[^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+(?!\"))\"?)(.*)$")
+WWW_AUTH_RELAXED = re.compile(r"^(?:\s*(?:,\s*)?([^ \t\r\n=]+)\s*=\s*\"?((?<=\")(?:[^\\\"]|\\.)*?(?=\")|(?<!\")[^ \t\r\n,]+(?!\"))\"?)(.*)$")
+UNQUOTE_PAIRS = re.compile(r'\\(.)')
+def _parse_www_authenticate(headers, headername='www-authenticate'):
+ """Returns a dictionary of dictionaries, one dict
+ per auth_scheme."""
+ retval = {}
+ if headers.has_key(headername):
+ authenticate = headers[headername].strip()
+ www_auth = USE_WWW_AUTH_STRICT_PARSING and WWW_AUTH_STRICT or WWW_AUTH_RELAXED
+ while authenticate:
+ # Break off the scheme at the beginning of the line
+ if headername == 'authentication-info':
+ (auth_scheme, the_rest) = ('digest', authenticate)
+ else:
+ (auth_scheme, the_rest) = authenticate.split(" ", 1)
+ # Now loop over all the key value pairs that come after the scheme,
+ # being careful not to roll into the next scheme
+ match = www_auth.search(the_rest)
+ auth_params = {}
+ while match:
+ if match and len(match.groups()) == 3:
+ (key, value, the_rest) = match.groups()
+ auth_params[key.lower()] = UNQUOTE_PAIRS.sub(r'\1', value) # '\\'.join([x.replace('\\', '') for x in value.split('\\\\')])
+ match = www_auth.search(the_rest)
+ retval[auth_scheme.lower()] = auth_params
+ authenticate = the_rest.strip()
+ return retval
+
+
+def _entry_disposition(response_headers, request_headers):
+ """Determine freshness from the Date, Expires and Cache-Control headers.
+
+ We don't handle the following:
+
+ 1. Cache-Control: max-stale
+ 2. Age: headers are not used in the calculations.
+
+ Not that this algorithm is simpler than you might think
+ because we are operating as a private (non-shared) cache.
+ This lets us ignore 's-maxage'. We can also ignore
+ 'proxy-invalidate' since we aren't a proxy.
+ We will never return a stale document as
+ fresh as a design decision, and thus the non-implementation
+ of 'max-stale'. This also lets us safely ignore 'must-revalidate'
+ since we operate as if every server has sent 'must-revalidate'.
+ Since we are private we get to ignore both 'public' and
+ 'private' parameters. We also ignore 'no-transform' since
+ we don't do any transformations.
+ The 'no-store' parameter is handled at a higher level.
+ So the only Cache-Control parameters we look at are:
+
+ no-cache
+ only-if-cached
+ max-age
+ min-fresh
+ """
+
+ retval = "STALE"
+ cc = _parse_cache_control(request_headers)
+ cc_response = _parse_cache_control(response_headers)
+
+ if request_headers.has_key('pragma') and request_headers['pragma'].lower().find('no-cache') != -1:
+ retval = "TRANSPARENT"
+ if 'cache-control' not in request_headers:
+ request_headers['cache-control'] = 'no-cache'
+ elif cc.has_key('no-cache'):
+ retval = "TRANSPARENT"
+ elif cc_response.has_key('no-cache'):
+ retval = "STALE"
+ elif cc.has_key('only-if-cached'):
+ retval = "FRESH"
+ elif response_headers.has_key('date'):
+ date = calendar.timegm(email.Utils.parsedate_tz(response_headers['date']))
+ now = time.time()
+ current_age = max(0, now - date)
+ if cc_response.has_key('max-age'):
+ try:
+ freshness_lifetime = int(cc_response['max-age'])
+ except ValueError:
+ freshness_lifetime = 0
+ elif response_headers.has_key('expires'):
+ expires = email.Utils.parsedate_tz(response_headers['expires'])
+ if None == expires:
+ freshness_lifetime = 0
+ else:
+ freshness_lifetime = max(0, calendar.timegm(expires) - date)
+ else:
+ freshness_lifetime = 0
+ if cc.has_key('max-age'):
+ try:
+ freshness_lifetime = int(cc['max-age'])
+ except ValueError:
+ freshness_lifetime = 0
+ if cc.has_key('min-fresh'):
+ try:
+ min_fresh = int(cc['min-fresh'])
+ except ValueError:
+ min_fresh = 0
+ current_age += min_fresh
+ if freshness_lifetime > current_age:
+ retval = "FRESH"
+ return retval
+
+def _decompressContent(response, new_content):
+ content = new_content
+ try:
+ encoding = response.get('content-encoding', None)
+ if encoding in ['gzip', 'deflate']:
+ if encoding == 'gzip':
+ content = gzip.GzipFile(fileobj=StringIO.StringIO(new_content)).read()
+ if encoding == 'deflate':
+ content = zlib.decompress(content)
+ response['content-length'] = str(len(content))
+ # Record the historical presence of the encoding in a way the won't interfere.
+ response['-content-encoding'] = response['content-encoding']
+ del response['content-encoding']
+ except IOError:
+ content = ""
+ raise FailedToDecompressContent(_("Content purported to be compressed with %s but failed to decompress.") % response.get('content-encoding'), response, content)
+ return content
+
+def _updateCache(request_headers, response_headers, content, cache, cachekey):
+ if cachekey:
+ cc = _parse_cache_control(request_headers)
+ cc_response = _parse_cache_control(response_headers)
+ if cc.has_key('no-store') or cc_response.has_key('no-store'):
+ cache.delete(cachekey)
+ else:
+ info = email.Message.Message()
+ for key, value in response_headers.iteritems():
+ if key not in ['status','content-encoding','transfer-encoding']:
+ info[key] = value
+
+ # Add annotations to the cache to indicate what headers
+ # are variant for this request.
+ vary = response_headers.get('vary', None)
+ if vary:
+ vary_headers = vary.lower().replace(' ', '').split(',')
+ for header in vary_headers:
+ key = '-varied-%s' % header
+ try:
+ info[key] = request_headers[header]
+ except KeyError:
+ pass
+
+ status = response_headers.status
+ if status == 304:
+ status = 200
+
+ status_header = 'status: %d\r\n' % response_headers.status
+
+ header_str = info.as_string()
+
+ header_str = re.sub("\r(?!\n)|(?<!\r)\n", "\r\n", header_str)
+ text = "".join([status_header, header_str, content])
+
+ cache.set(cachekey, text)
+
+def _cnonce():
+ dig = _md5("%s:%s" % (time.ctime(), ["0123456789"[random.randrange(0, 9)] for i in range(20)])).hexdigest()
+ return dig[:16]
+
+def _wsse_username_token(cnonce, iso_now, password):
+ return base64.b64encode(_sha("%s%s%s" % (cnonce, iso_now, password)).digest()).strip()
+
+
+# For credentials we need two things, first
+# a pool of credential to try (not necesarily tied to BAsic, Digest, etc.)
+# Then we also need a list of URIs that have already demanded authentication
+# That list is tricky since sub-URIs can take the same auth, or the
+# auth scheme may change as you descend the tree.
+# So we also need each Auth instance to be able to tell us
+# how close to the 'top' it is.
+
+class Authentication(object):
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ (scheme, authority, path, query, fragment) = parse_uri(request_uri)
+ self.path = path
+ self.host = host
+ self.credentials = credentials
+ self.http = http
+
+ def depth(self, request_uri):
+ (scheme, authority, path, query, fragment) = parse_uri(request_uri)
+ return request_uri[len(self.path):].count("/")
+
+ def inscope(self, host, request_uri):
+ # XXX Should we normalize the request_uri?
+ (scheme, authority, path, query, fragment) = parse_uri(request_uri)
+ return (host == self.host) and path.startswith(self.path)
+
+ def request(self, method, request_uri, headers, content):
+ """Modify the request headers to add the appropriate
+ Authorization header. Over-rise this in sub-classes."""
+ pass
+
+ def response(self, response, content):
+ """Gives us a chance to update with new nonces
+ or such returned from the last authorized response.
+ Over-rise this in sub-classes if necessary.
+
+ Return TRUE is the request is to be retried, for
+ example Digest may return stale=true.
+ """
+ return False
+
+
+
+class BasicAuthentication(Authentication):
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
+
+ def request(self, method, request_uri, headers, content):
+ """Modify the request headers to add the appropriate
+ Authorization header."""
+ headers['authorization'] = 'Basic ' + base64.b64encode("%s:%s" % self.credentials).strip()
+
+
+class DigestAuthentication(Authentication):
+ """Only do qop='auth' and MD5, since that
+ is all Apache currently implements"""
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
+ challenge = _parse_www_authenticate(response, 'www-authenticate')
+ self.challenge = challenge['digest']
+ qop = self.challenge.get('qop', 'auth')
+ self.challenge['qop'] = ('auth' in [x.strip() for x in qop.split()]) and 'auth' or None
+ if self.challenge['qop'] is None:
+ raise UnimplementedDigestAuthOptionError( _("Unsupported value for qop: %s." % qop))
+ self.challenge['algorithm'] = self.challenge.get('algorithm', 'MD5').upper()
+ if self.challenge['algorithm'] != 'MD5':
+ raise UnimplementedDigestAuthOptionError( _("Unsupported value for algorithm: %s." % self.challenge['algorithm']))
+ self.A1 = "".join([self.credentials[0], ":", self.challenge['realm'], ":", self.credentials[1]])
+ self.challenge['nc'] = 1
+
+ def request(self, method, request_uri, headers, content, cnonce = None):
+ """Modify the request headers"""
+ H = lambda x: _md5(x).hexdigest()
+ KD = lambda s, d: H("%s:%s" % (s, d))
+ A2 = "".join([method, ":", request_uri])
+ self.challenge['cnonce'] = cnonce or _cnonce()
+ request_digest = '"%s"' % KD(H(self.A1), "%s:%s:%s:%s:%s" % (self.challenge['nonce'],
+ '%08x' % self.challenge['nc'],
+ self.challenge['cnonce'],
+ self.challenge['qop'], H(A2)
+ ))
+ headers['Authorization'] = 'Digest username="%s", realm="%s", nonce="%s", uri="%s", algorithm=%s, response=%s, qop=%s, nc=%08x, cnonce="%s"' % (
+ self.credentials[0],
+ self.challenge['realm'],
+ self.challenge['nonce'],
+ request_uri,
+ self.challenge['algorithm'],
+ request_digest,
+ self.challenge['qop'],
+ self.challenge['nc'],
+ self.challenge['cnonce'],
+ )
+ self.challenge['nc'] += 1
+
+ def response(self, response, content):
+ if not response.has_key('authentication-info'):
+ challenge = _parse_www_authenticate(response, 'www-authenticate').get('digest', {})
+ if 'true' == challenge.get('stale'):
+ self.challenge['nonce'] = challenge['nonce']
+ self.challenge['nc'] = 1
+ return True
+ else:
+ updated_challenge = _parse_www_authenticate(response, 'authentication-info').get('digest', {})
+
+ if updated_challenge.has_key('nextnonce'):
+ self.challenge['nonce'] = updated_challenge['nextnonce']
+ self.challenge['nc'] = 1
+ return False
+
+
+class HmacDigestAuthentication(Authentication):
+ """Adapted from Robert Sayre's code and DigestAuthentication above."""
+ __author__ = "Thomas Broyer (t.broyer@ltgt.net)"
+
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
+ challenge = _parse_www_authenticate(response, 'www-authenticate')
+ self.challenge = challenge['hmacdigest']
+ # TODO: self.challenge['domain']
+ self.challenge['reason'] = self.challenge.get('reason', 'unauthorized')
+ if self.challenge['reason'] not in ['unauthorized', 'integrity']:
+ self.challenge['reason'] = 'unauthorized'
+ self.challenge['salt'] = self.challenge.get('salt', '')
+ if not self.challenge.get('snonce'):
+ raise UnimplementedHmacDigestAuthOptionError( _("The challenge doesn't contain a server nonce, or this one is empty."))
+ self.challenge['algorithm'] = self.challenge.get('algorithm', 'HMAC-SHA-1')
+ if self.challenge['algorithm'] not in ['HMAC-SHA-1', 'HMAC-MD5']:
+ raise UnimplementedHmacDigestAuthOptionError( _("Unsupported value for algorithm: %s." % self.challenge['algorithm']))
+ self.challenge['pw-algorithm'] = self.challenge.get('pw-algorithm', 'SHA-1')
+ if self.challenge['pw-algorithm'] not in ['SHA-1', 'MD5']:
+ raise UnimplementedHmacDigestAuthOptionError( _("Unsupported value for pw-algorithm: %s." % self.challenge['pw-algorithm']))
+ if self.challenge['algorithm'] == 'HMAC-MD5':
+ self.hashmod = _md5
+ else:
+ self.hashmod = _sha
+ if self.challenge['pw-algorithm'] == 'MD5':
+ self.pwhashmod = _md5
+ else:
+ self.pwhashmod = _sha
+ self.key = "".join([self.credentials[0], ":",
+ self.pwhashmod.new("".join([self.credentials[1], self.challenge['salt']])).hexdigest().lower(),
+ ":", self.challenge['realm']
+ ])
+ self.key = self.pwhashmod.new(self.key).hexdigest().lower()
+
+ def request(self, method, request_uri, headers, content):
+ """Modify the request headers"""
+ keys = _get_end2end_headers(headers)
+ keylist = "".join(["%s " % k for k in keys])
+ headers_val = "".join([headers[k] for k in keys])
+ created = time.strftime('%Y-%m-%dT%H:%M:%SZ',time.gmtime())
+ cnonce = _cnonce()
+ request_digest = "%s:%s:%s:%s:%s" % (method, request_uri, cnonce, self.challenge['snonce'], headers_val)
+ request_digest = hmac.new(self.key, request_digest, self.hashmod).hexdigest().lower()
+ headers['Authorization'] = 'HMACDigest username="%s", realm="%s", snonce="%s", cnonce="%s", uri="%s", created="%s", response="%s", headers="%s"' % (
+ self.credentials[0],
+ self.challenge['realm'],
+ self.challenge['snonce'],
+ cnonce,
+ request_uri,
+ created,
+ request_digest,
+ keylist,
+ )
+
+ def response(self, response, content):
+ challenge = _parse_www_authenticate(response, 'www-authenticate').get('hmacdigest', {})
+ if challenge.get('reason') in ['integrity', 'stale']:
+ return True
+ return False
+
+
+class WsseAuthentication(Authentication):
+ """This is thinly tested and should not be relied upon.
+ At this time there isn't any third party server to test against.
+ Blogger and TypePad implemented this algorithm at one point
+ but Blogger has since switched to Basic over HTTPS and
+ TypePad has implemented it wrong, by never issuing a 401
+ challenge but instead requiring your client to telepathically know that
+ their endpoint is expecting WSSE profile="UsernameToken"."""
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
+
+ def request(self, method, request_uri, headers, content):
+ """Modify the request headers to add the appropriate
+ Authorization header."""
+ headers['Authorization'] = 'WSSE profile="UsernameToken"'
+ iso_now = time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime())
+ cnonce = _cnonce()
+ password_digest = _wsse_username_token(cnonce, iso_now, self.credentials[1])
+ headers['X-WSSE'] = 'UsernameToken Username="%s", PasswordDigest="%s", Nonce="%s", Created="%s"' % (
+ self.credentials[0],
+ password_digest,
+ cnonce,
+ iso_now)
+
+class GoogleLoginAuthentication(Authentication):
+ def __init__(self, credentials, host, request_uri, headers, response, content, http):
+ from urllib import urlencode
+ Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
+ challenge = _parse_www_authenticate(response, 'www-authenticate')
+ service = challenge['googlelogin'].get('service', 'xapi')
+ # Bloggger actually returns the service in the challenge
+ # For the rest we guess based on the URI
+ if service == 'xapi' and request_uri.find("calendar") > 0:
+ service = "cl"
+ # No point in guessing Base or Spreadsheet
+ #elif request_uri.find("spreadsheets") > 0:
+ # service = "wise"
+
+ auth = dict(Email=credentials[0], Passwd=credentials[1], service=service, source=headers['user-agent'])
+ resp, content = self.http.request("https://www.google.com/accounts/ClientLogin", method="POST", body=urlencode(auth), headers={'Content-Type': 'application/x-www-form-urlencoded'})
+ lines = content.split('\n')
+ d = dict([tuple(line.split("=", 1)) for line in lines if line])
+ if resp.status == 403:
+ self.Auth = ""
+ else:
+ self.Auth = d['Auth']
+
+ def request(self, method, request_uri, headers, content):
+ """Modify the request headers to add the appropriate
+ Authorization header."""
+ headers['authorization'] = 'GoogleLogin Auth=' + self.Auth
+
+
+AUTH_SCHEME_CLASSES = {
+ "basic": BasicAuthentication,
+ "wsse": WsseAuthentication,
+ "digest": DigestAuthentication,
+ "hmacdigest": HmacDigestAuthentication,
+ "googlelogin": GoogleLoginAuthentication
+}
+
+AUTH_SCHEME_ORDER = ["hmacdigest", "googlelogin", "digest", "wsse", "basic"]
+
+class FileCache(object):
+ """Uses a local directory as a store for cached files.
+ Not really safe to use if multiple threads or processes are going to
+ be running on the same cache.
+ """
+ def __init__(self, cache, safe=safename): # use safe=lambda x: md5.new(x).hexdigest() for the old behavior
+ self.cache = cache
+ self.safe = safe
+ if not os.path.exists(cache):
+ os.makedirs(self.cache)
+
+ def get(self, key):
+ retval = None
+ cacheFullPath = os.path.join(self.cache, self.safe(key))
+ try:
+ f = file(cacheFullPath, "rb")
+ retval = f.read()
+ f.close()
+ except IOError:
+ pass
+ return retval
+
+ def set(self, key, value):
+ cacheFullPath = os.path.join(self.cache, self.safe(key))
+ f = file(cacheFullPath, "wb")
+ f.write(value)
+ f.close()
+
+ def delete(self, key):
+ cacheFullPath = os.path.join(self.cache, self.safe(key))
+ if os.path.exists(cacheFullPath):
+ os.remove(cacheFullPath)
+
+class Credentials(object):
+ def __init__(self):
+ self.credentials = []
+
+ def add(self, name, password, domain=""):
+ self.credentials.append((domain.lower(), name, password))
+
+ def clear(self):
+ self.credentials = []
+
+ def iter(self, domain):
+ for (cdomain, name, password) in self.credentials:
+ if cdomain == "" or domain == cdomain:
+ yield (name, password)
+
+class KeyCerts(Credentials):
+ """Identical to Credentials except that
+ name/password are mapped to key/cert."""
+ pass
+
+
+class ProxyInfo(object):
+ """Collect information required to use a proxy."""
+ def __init__(self, proxy_type, proxy_host, proxy_port, proxy_rdns=None, proxy_user=None, proxy_pass=None):
+ """The parameter proxy_type must be set to one of socks.PROXY_TYPE_XXX
+ constants. For example:
+
+p = ProxyInfo(proxy_type=socks.PROXY_TYPE_HTTP, proxy_host='localhost', proxy_port=8000)
+ """
+ self.proxy_type, self.proxy_host, self.proxy_port, self.proxy_rdns, self.proxy_user, self.proxy_pass = proxy_type, proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass
+
+ def astuple(self):
+ return (self.proxy_type, self.proxy_host, self.proxy_port, self.proxy_rdns,
+ self.proxy_user, self.proxy_pass)
+
+ def isgood(self):
+ return (self.proxy_host != None) and (self.proxy_port != None)
+
+
+class HTTPConnectionWithTimeout(httplib.HTTPConnection):
+ """HTTPConnection subclass that supports timeouts"""
+
+ def __init__(self, host, port=None, strict=None, timeout=None, proxy_info=None):
+ httplib.HTTPConnection.__init__(self, host, port, strict)
+ self.timeout = timeout
+ self.proxy_info = proxy_info
+
+ def connect(self):
+ """Connect to the host and port specified in __init__."""
+ # Mostly verbatim from httplib.py.
+ msg = "getaddrinfo returns an empty list"
+ for res in socket.getaddrinfo(self.host, self.port, 0,
+ socket.SOCK_STREAM):
+ af, socktype, proto, canonname, sa = res
+ try:
+ if self.proxy_info and self.proxy_info.isgood():
+ self.sock = socks.socksocket(af, socktype, proto)
+ self.sock.setproxy(*self.proxy_info.astuple())
+ else:
+ self.sock = socket.socket(af, socktype, proto)
+ self.sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
+ # Different from httplib: support timeouts.
+ if has_timeout(self.timeout):
+ self.sock.settimeout(self.timeout)
+ # End of difference from httplib.
+ if self.debuglevel > 0:
+ print "connect: (%s, %s)" % (self.host, self.port)
+
+ self.sock.connect(sa)
+ except socket.error, msg:
+ if self.debuglevel > 0:
+ print 'connect fail:', (self.host, self.port)
+ if self.sock:
+ self.sock.close()
+ self.sock = None
+ continue
+ break
+ if not self.sock:
+ raise socket.error, msg
+
+class HTTPSConnectionWithTimeout(httplib.HTTPSConnection):
+ "This class allows communication via SSL."
+
+ def __init__(self, host, port=None, key_file=None, cert_file=None,
+ strict=None, timeout=None, proxy_info=None):
+ httplib.HTTPSConnection.__init__(self, host, port=port, key_file=key_file,
+ cert_file=cert_file, strict=strict)
+ self.timeout = timeout
+ self.proxy_info = proxy_info
+
+ def connect(self):
+ "Connect to a host on a given (SSL) port."
+
+ if self.proxy_info and self.proxy_info.isgood():
+ sock = socks.socksocket(socket.AF_INET, socket.SOCK_STREAM)
+ sock.setproxy(*self.proxy_info.astuple())
+ else:
+ sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+ sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
+
+ if has_timeout(self.timeout):
+ sock.settimeout(self.timeout)
+ sock.connect((self.host, self.port))
+ self.sock =_ssl_wrap_socket(sock, self.key_file, self.cert_file)
+
+
+
+class Http(object):
+ """An HTTP client that handles:
+- all methods
+- caching
+- ETags
+- compression,
+- HTTPS
+- Basic
+- Digest
+- WSSE
+
+and more.
+ """
+ def __init__(self, cache=None, timeout=None, proxy_info=None):
+ """The value of proxy_info is a ProxyInfo instance.
+
+If 'cache' is a string then it is used as a directory name
+for a disk cache. Otherwise it must be an object that supports
+the same interface as FileCache."""
+ self.proxy_info = proxy_info
+ # Map domain name to an httplib connection
+ self.connections = {}
+ # The location of the cache, for now a directory
+ # where cached responses are held.
+ if cache and isinstance(cache, str):
+ self.cache = FileCache(cache)
+ else:
+ self.cache = cache
+
+ # Name/password
+ self.credentials = Credentials()
+
+ # Key/cert
+ self.certificates = KeyCerts()
+
+ # authorization objects
+ self.authorizations = []
+
+ # If set to False then no redirects are followed, even safe ones.
+ self.follow_redirects = True
+
+ # Which HTTP methods do we apply optimistic concurrency to, i.e.
+ # which methods get an "if-match:" etag header added to them.
+ self.optimistic_concurrency_methods = ["PUT"]
+
+ # If 'follow_redirects' is True, and this is set to True then
+ # all redirecs are followed, including unsafe ones.
+ self.follow_all_redirects = False
+
+ self.ignore_etag = False
+
+ self.force_exception_to_status_code = False
+
+ self.timeout = timeout
+
+ def _auth_from_challenge(self, host, request_uri, headers, response, content):
+ """A generator that creates Authorization objects
+ that can be applied to requests.
+ """
+ challenges = _parse_www_authenticate(response, 'www-authenticate')
+ for cred in self.credentials.iter(host):
+ for scheme in AUTH_SCHEME_ORDER:
+ if challenges.has_key(scheme):
+ yield AUTH_SCHEME_CLASSES[scheme](cred, host, request_uri, headers, response, content, self)
+
+ def add_credentials(self, name, password, domain=""):
+ """Add a name and password that will be used
+ any time a request requires authentication."""
+ self.credentials.add(name, password, domain)
+
+ def add_certificate(self, key, cert, domain):
+ """Add a key and cert that will be used
+ any time a request requires authentication."""
+ self.certificates.add(key, cert, domain)
+
+ def clear_credentials(self):
+ """Remove all the names and passwords
+ that are used for authentication"""
+ self.credentials.clear()
+ self.authorizations = []
+
+ def _conn_request(self, conn, request_uri, method, body, headers):
+ for i in range(2):
+ try:
+ conn.request(method, request_uri, body, headers)
+ except socket.gaierror:
+ conn.close()
+ raise ServerNotFoundError("Unable to find the server at %s" % conn.host)
+ except socket.error, e:
+ if e.errno == errno.ECONNREFUSED: # Connection refused
+ raise
+ except httplib.HTTPException:
+ # Just because the server closed the connection doesn't apparently mean
+ # that the server didn't send a response.
+ pass
+ try:
+ response = conn.getresponse()
+ except (socket.error, httplib.HTTPException):
+ if i == 0:
+ conn.close()
+ conn.connect()
+ continue
+ else:
+ raise
+ else:
+ content = ""
+ if method == "HEAD":
+ response.close()
+ else:
+ content = response.read()
+ response = Response(response)
+ if method != "HEAD":
+ content = _decompressContent(response, content)
+ break
+ return (response, content)
+
+
+ def _request(self, conn, host, absolute_uri, request_uri, method, body, headers, redirections, cachekey):
+ """Do the actual request using the connection object
+ and also follow one level of redirects if necessary"""
+
+ auths = [(auth.depth(request_uri), auth) for auth in self.authorizations if auth.inscope(host, request_uri)]
+ auth = auths and sorted(auths)[0][1] or None
+ if auth:
+ auth.request(method, request_uri, headers, body)
+
+ (response, content) = self._conn_request(conn, request_uri, method, body, headers)
+
+ if auth:
+ if auth.response(response, body):
+ auth.request(method, request_uri, headers, body)
+ (response, content) = self._conn_request(conn, request_uri, method, body, headers )
+ response._stale_digest = 1
+
+ if response.status == 401:
+ for authorization in self._auth_from_challenge(host, request_uri, headers, response, content):
+ authorization.request(method, request_uri, headers, body)
+ (response, content) = self._conn_request(conn, request_uri, method, body, headers, )
+ if response.status != 401:
+ self.authorizations.append(authorization)
+ authorization.response(response, body)
+ break
+
+ if (self.follow_all_redirects or (method in ["GET", "HEAD"]) or response.status == 303):
+ if self.follow_redirects and response.status in [300, 301, 302, 303, 307]:
+ # Pick out the location header and basically start from the beginning
+ # remembering first to strip the ETag header and decrement our 'depth'
+ if redirections:
+ if not response.has_key('location') and response.status != 300:
+ raise RedirectMissingLocation( _("Redirected but the response is missing a Location: header."), response, content)
+ # Fix-up relative redirects (which violate an RFC 2616 MUST)
+ if response.has_key('location'):
+ location = response['location']
+ (scheme, authority, path, query, fragment) = parse_uri(location)
+ if authority == None:
+ response['location'] = urlparse.urljoin(absolute_uri, location)
+ if response.status == 301 and method in ["GET", "HEAD"]:
+ response['-x-permanent-redirect-url'] = response['location']
+ if not response.has_key('content-location'):
+ response['content-location'] = absolute_uri
+ _updateCache(headers, response, content, self.cache, cachekey)
+ if headers.has_key('if-none-match'):
+ del headers['if-none-match']
+ if headers.has_key('if-modified-since'):
+ del headers['if-modified-since']
+ if response.has_key('location'):
+ location = response['location']
+ old_response = copy.deepcopy(response)
+ if not old_response.has_key('content-location'):
+ old_response['content-location'] = absolute_uri
+ redirect_method = ((response.status == 303) and (method not in ["GET", "HEAD"])) and "GET" or method
+ (response, content) = self.request(location, redirect_method, body=body, headers = headers, redirections = redirections - 1)
+ response.previous = old_response
+ else:
+ raise RedirectLimit( _("Redirected more times than rediection_limit allows."), response, content)
+ elif response.status in [200, 203] and method == "GET":
+ # Don't cache 206's since we aren't going to handle byte range requests
+ if not response.has_key('content-location'):
+ response['content-location'] = absolute_uri
+ _updateCache(headers, response, content, self.cache, cachekey)
+
+ return (response, content)
+
+ def _normalize_headers(self, headers):
+ return _normalize_headers(headers)
+
+# Need to catch and rebrand some exceptions
+# Then need to optionally turn all exceptions into status codes
+# including all socket.* and httplib.* exceptions.
+
+
+ def request(self, uri, method="GET", body=None, headers=None, redirections=DEFAULT_MAX_REDIRECTS, connection_type=None):
+ """ Performs a single HTTP request.
+The 'uri' is the URI of the HTTP resource and can begin
+with either 'http' or 'https'. The value of 'uri' must be an absolute URI.
+
+The 'method' is the HTTP method to perform, such as GET, POST, DELETE, etc.
+There is no restriction on the methods allowed.
+
+The 'body' is the entity body to be sent with the request. It is a string
+object.
+
+Any extra headers that are to be sent with the request should be provided in the
+'headers' dictionary.
+
+The maximum number of redirect to follow before raising an
+exception is 'redirections. The default is 5.
+
+The return value is a tuple of (response, content), the first
+being and instance of the 'Response' class, the second being
+a string that contains the response entity body.
+ """
+ try:
+ if headers is None:
+ headers = {}
+ else:
+ headers = self._normalize_headers(headers)
+
+ if not headers.has_key('user-agent'):
+ headers['user-agent'] = "Python-httplib2/%s" % __version__
+
+ uri = iri2uri(uri)
+
+ (scheme, authority, request_uri, defrag_uri) = urlnorm(uri)
+ domain_port = authority.split(":")[0:2]
+ if len(domain_port) == 2 and domain_port[1] == '443' and scheme == 'http':
+ scheme = 'https'
+ authority = domain_port[0]
+
+ conn_key = scheme+":"+authority
+ if conn_key in self.connections:
+ conn = self.connections[conn_key]
+ else:
+ if not connection_type:
+ connection_type = (scheme == 'https') and HTTPSConnectionWithTimeout or HTTPConnectionWithTimeout
+ certs = list(self.certificates.iter(authority))
+ if scheme == 'https' and certs:
+ conn = self.connections[conn_key] = connection_type(authority, key_file=certs[0][0],
+ cert_file=certs[0][1], timeout=self.timeout, proxy_info=self.proxy_info)
+ else:
+ conn = self.connections[conn_key] = connection_type(authority, timeout=self.timeout, proxy_info=self.proxy_info)
+ conn.set_debuglevel(debuglevel)
+
+ if method in ["GET", "HEAD"] and 'range' not in headers and 'accept-encoding' not in headers:
+ headers['accept-encoding'] = 'gzip, deflate'
+
+ info = email.Message.Message()
+ cached_value = None
+ if self.cache:
+ cachekey = defrag_uri
+ cached_value = self.cache.get(cachekey)
+ if cached_value:
+ # info = email.message_from_string(cached_value)
+ #
+ # Need to replace the line above with the kludge below
+ # to fix the non-existent bug not fixed in this
+ # bug report: http://mail.python.org/pipermail/python-bugs-list/2005-September/030289.html
+ try:
+ info, content = cached_value.split('\r\n\r\n', 1)
+ feedparser = email.FeedParser.FeedParser()
+ feedparser.feed(info)
+ info = feedparser.close()
+ feedparser._parse = None
+ except IndexError:
+ self.cache.delete(cachekey)
+ cachekey = None
+ cached_value = None
+ else:
+ cachekey = None
+
+ if method in self.optimistic_concurrency_methods and self.cache and info.has_key('etag') and not self.ignore_etag and 'if-match' not in headers:
+ # http://www.w3.org/1999/04/Editing/
+ headers['if-match'] = info['etag']
+
+ if method not in ["GET", "HEAD"] and self.cache and cachekey:
+ # RFC 2616 Section 13.10
+ self.cache.delete(cachekey)
+
+ # Check the vary header in the cache to see if this request
+ # matches what varies in the cache.
+ if method in ['GET', 'HEAD'] and 'vary' in info:
+ vary = info['vary']
+ vary_headers = vary.lower().replace(' ', '').split(',')
+ for header in vary_headers:
+ key = '-varied-%s' % header
+ value = info[key]
+ if headers.get(header, None) != value:
+ cached_value = None
+ break
+
+ if cached_value and method in ["GET", "HEAD"] and self.cache and 'range' not in headers:
+ if info.has_key('-x-permanent-redirect-url'):
+ # Should cached permanent redirects be counted in our redirection count? For now, yes.
+ (response, new_content) = self.request(info['-x-permanent-redirect-url'], "GET", headers = headers, redirections = redirections - 1)
+ response.previous = Response(info)
+ response.previous.fromcache = True
+ else:
+ # Determine our course of action:
+ # Is the cached entry fresh or stale?
+ # Has the client requested a non-cached response?
+ #
+ # There seems to be three possible answers:
+ # 1. [FRESH] Return the cache entry w/o doing a GET
+ # 2. [STALE] Do the GET (but add in cache validators if available)
+ # 3. [TRANSPARENT] Do a GET w/o any cache validators (Cache-Control: no-cache) on the request
+ entry_disposition = _entry_disposition(info, headers)
+
+ if entry_disposition == "FRESH":
+ if not cached_value:
+ info['status'] = '504'
+ content = ""
+ response = Response(info)
+ if cached_value:
+ response.fromcache = True
+ return (response, content)
+
+ if entry_disposition == "STALE":
+ if info.has_key('etag') and not self.ignore_etag and not 'if-none-match' in headers:
+ headers['if-none-match'] = info['etag']
+ if info.has_key('last-modified') and not 'last-modified' in headers:
+ headers['if-modified-since'] = info['last-modified']
+ elif entry_disposition == "TRANSPARENT":
+ pass
+
+ (response, new_content) = self._request(conn, authority, uri, request_uri, method, body, headers, redirections, cachekey)
+
+ if response.status == 304 and method == "GET":
+ # Rewrite the cache entry with the new end-to-end headers
+ # Take all headers that are in response
+ # and overwrite their values in info.
+ # unless they are hop-by-hop, or are listed in the connection header.
+
+ for key in _get_end2end_headers(response):
+ info[key] = response[key]
+ merged_response = Response(info)
+ if hasattr(response, "_stale_digest"):
+ merged_response._stale_digest = response._stale_digest
+ _updateCache(headers, merged_response, content, self.cache, cachekey)
+ response = merged_response
+ response.status = 200
+ response.fromcache = True
+
+ elif response.status == 200:
+ content = new_content
+ else:
+ self.cache.delete(cachekey)
+ content = new_content
+ else:
+ cc = _parse_cache_control(headers)
+ if cc.has_key('only-if-cached'):
+ info['status'] = '504'
+ response = Response(info)
+ content = ""
+ else:
+ (response, content) = self._request(conn, authority, uri, request_uri, method, body, headers, redirections, cachekey)
+ except Exception, e:
+ if self.force_exception_to_status_code:
+ if isinstance(e, HttpLib2ErrorWithResponse):
+ response = e.response
+ content = e.content
+ response.status = 500
+ response.reason = str(e)
+ elif isinstance(e, socket.timeout):
+ content = "Request Timeout"
+ response = Response( {
+ "content-type": "text/plain",
+ "status": "408",
+ "content-length": len(content)
+ })
+ response.reason = "Request Timeout"
+ else:
+ content = str(e)
+ response = Response( {
+ "content-type": "text/plain",
+ "status": "400",
+ "content-length": len(content)
+ })
+ response.reason = "Bad Request"
+ else:
+ raise
+
+
+ return (response, content)
+
+
+
+class Response(dict):
+ """An object more like email.Message than httplib.HTTPResponse."""
+
+ """Is this response from our local cache"""
+ fromcache = False
+
+ """HTTP protocol version used by server. 10 for HTTP/1.0, 11 for HTTP/1.1. """
+ version = 11
+
+ "Status code returned by server. "
+ status = 200
+
+ """Reason phrase returned by server."""
+ reason = "Ok"
+
+ previous = None
+
+ def __init__(self, info):
+ # info is either an email.Message or
+ # an httplib.HTTPResponse object.
+ if isinstance(info, httplib.HTTPResponse):
+ for key, value in info.getheaders():
+ self[key.lower()] = value
+ self.status = info.status
+ self['status'] = str(self.status)
+ self.reason = info.reason
+ self.version = info.version
+ elif isinstance(info, email.Message.Message):
+ for key, value in info.items():
+ self[key] = value
+ self.status = int(self['status'])
+ else:
+ for key, value in info.iteritems():
+ self[key] = value
+ self.status = int(self.get('status', self.status))
+
+
+ def __getattr__(self, name):
+ if name == 'dict':
+ return self
+ else:
+ raise AttributeError, name
diff --git a/httplib2/iri2uri.py b/httplib2/iri2uri.py
new file mode 100644
index 0000000..70667ed
--- /dev/null
+++ b/httplib2/iri2uri.py
@@ -0,0 +1,110 @@
+"""
+iri2uri
+
+Converts an IRI to a URI.
+
+"""
+__author__ = "Joe Gregorio (joe@bitworking.org)"
+__copyright__ = "Copyright 2006, Joe Gregorio"
+__contributors__ = []
+__version__ = "1.0.0"
+__license__ = "MIT"
+__history__ = """
+"""
+
+import urlparse
+
+
+# Convert an IRI to a URI following the rules in RFC 3987
+#
+# The characters we need to enocde and escape are defined in the spec:
+#
+# iprivate = %xE000-F8FF / %xF0000-FFFFD / %x100000-10FFFD
+# ucschar = %xA0-D7FF / %xF900-FDCF / %xFDF0-FFEF
+# / %x10000-1FFFD / %x20000-2FFFD / %x30000-3FFFD
+# / %x40000-4FFFD / %x50000-5FFFD / %x60000-6FFFD
+# / %x70000-7FFFD / %x80000-8FFFD / %x90000-9FFFD
+# / %xA0000-AFFFD / %xB0000-BFFFD / %xC0000-CFFFD
+# / %xD0000-DFFFD / %xE1000-EFFFD
+
+escape_range = [
+ (0xA0, 0xD7FF ),
+ (0xE000, 0xF8FF ),
+ (0xF900, 0xFDCF ),
+ (0xFDF0, 0xFFEF),
+ (0x10000, 0x1FFFD ),
+ (0x20000, 0x2FFFD ),
+ (0x30000, 0x3FFFD),
+ (0x40000, 0x4FFFD ),
+ (0x50000, 0x5FFFD ),
+ (0x60000, 0x6FFFD),
+ (0x70000, 0x7FFFD ),
+ (0x80000, 0x8FFFD ),
+ (0x90000, 0x9FFFD),
+ (0xA0000, 0xAFFFD ),
+ (0xB0000, 0xBFFFD ),
+ (0xC0000, 0xCFFFD),
+ (0xD0000, 0xDFFFD ),
+ (0xE1000, 0xEFFFD),
+ (0xF0000, 0xFFFFD ),
+ (0x100000, 0x10FFFD)
+]
+
+def encode(c):
+ retval = c
+ i = ord(c)
+ for low, high in escape_range:
+ if i < low:
+ break
+ if i >= low and i <= high:
+ retval = "".join(["%%%2X" % ord(o) for o in c.encode('utf-8')])
+ break
+ return retval
+
+
+def iri2uri(uri):
+ """Convert an IRI to a URI. Note that IRIs must be
+ passed in a unicode strings. That is, do not utf-8 encode
+ the IRI before passing it into the function."""
+ if isinstance(uri ,unicode):
+ (scheme, authority, path, query, fragment) = urlparse.urlsplit(uri)
+ authority = authority.encode('idna')
+ # For each character in 'ucschar' or 'iprivate'
+ # 1. encode as utf-8
+ # 2. then %-encode each octet of that utf-8
+ uri = urlparse.urlunsplit((scheme, authority, path, query, fragment))
+ uri = "".join([encode(c) for c in uri])
+ return uri
+
+if __name__ == "__main__":
+ import unittest
+
+ class Test(unittest.TestCase):
+
+ def test_uris(self):
+ """Test that URIs are invariant under the transformation."""
+ invariant = [
+ u"ftp://ftp.is.co.za/rfc/rfc1808.txt",
+ u"http://www.ietf.org/rfc/rfc2396.txt",
+ u"ldap://[2001:db8::7]/c=GB?objectClass?one",
+ u"mailto:John.Doe@example.com",
+ u"news:comp.infosystems.www.servers.unix",
+ u"tel:+1-816-555-1212",
+ u"telnet://192.0.2.16:80/",
+ u"urn:oasis:names:specification:docbook:dtd:xml:4.1.2" ]
+ for uri in invariant:
+ self.assertEqual(uri, iri2uri(uri))
+
+ def test_iri(self):
+ """ Test that the right type of escaping is done for each part of the URI."""
+ self.assertEqual("http://xn--o3h.com/%E2%98%84", iri2uri(u"http://\N{COMET}.com/\N{COMET}"))
+ self.assertEqual("http://bitworking.org/?fred=%E2%98%84", iri2uri(u"http://bitworking.org/?fred=\N{COMET}"))
+ self.assertEqual("http://bitworking.org/#%E2%98%84", iri2uri(u"http://bitworking.org/#\N{COMET}"))
+ self.assertEqual("#%E2%98%84", iri2uri(u"#\N{COMET}"))
+ self.assertEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}"))
+ self.assertEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}")))
+ self.assertNotEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}".encode('utf-8')))
+
+ unittest.main()
+
+
diff --git a/httplib2/socks.py b/httplib2/socks.py
new file mode 100644
index 0000000..6f4f020
--- /dev/null
+++ b/httplib2/socks.py
@@ -0,0 +1,439 @@
+"""SocksiPy - Python SOCKS module.
+Version 1.00
+
+Copyright 2006 Dan-Haim. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+3. Neither the name of Dan Haim nor the names of his contributors may be used
+ to endorse or promote products derived from this software without specific
+ prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY DAN HAIM "AS IS" AND ANY EXPRESS OR IMPLIED
+WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
+EVENT SHALL DAN HAIM OR HIS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA
+OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
+OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+This module provides a standard socket-like interface for Python
+for tunneling connections through SOCKS proxies.
+
+"""
+
+"""
+
+Minor modifications made by Christopher Gilbert (http://motomastyle.com/)
+for use in PyLoris (http://pyloris.sourceforge.net/)
+
+Minor modifications made by Mario Vilas (http://breakingcode.wordpress.com/)
+mainly to merge bug fixes found in Sourceforge
+
+"""
+
+import socket
+import struct
+import sys
+
+if not hasattr(socket, 'socket'):
+ raise ImportError("Running on App Engine?")
+
+PROXY_TYPE_SOCKS4 = 1
+PROXY_TYPE_SOCKS5 = 2
+PROXY_TYPE_HTTP = 3
+
+_defaultproxy = None
+
+# Small hack for Python 2.x
+if sys.version_info[0] <= 2:
+ def bytes(obj, enc=None):
+ return obj
+
+class ProxyError(Exception):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class GeneralProxyError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks5AuthError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks5Error(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks4Error(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class HTTPError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+_generalerrors = ("success",
+ "invalid data",
+ "not connected",
+ "not available",
+ "bad proxy type",
+ "bad input")
+
+_socks5errors = ("succeeded",
+ "general SOCKS server failure",
+ "connection not allowed by ruleset",
+ "Network unreachable",
+ "Host unreachable",
+ "Connection refused",
+ "TTL expired",
+ "Command not supported",
+ "Address type not supported",
+ "Unknown error")
+
+_socks5autherrors = ("succeeded",
+ "authentication is required",
+ "all offered authentication methods were rejected",
+ "unknown username or invalid password",
+ "unknown error")
+
+_socks4errors = ("request granted",
+ "request rejected or failed",
+ ("request rejected because SOCKS server cannot connect to "
+ "identd on the client"),
+ ("request rejected because the client program and identd"
+ " report different user-ids"),
+ "unknown error")
+
+
+def setdefaultproxy(proxytype=None, addr=None, port=None, rdns=True,
+ username=None, password=None):
+ """setdefaultproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
+ Sets a default proxy which all further socksocket objects will use,
+ unless explicitly changed.
+ """
+ global _defaultproxy
+ _defaultproxy = (proxytype, addr, port, rdns, username, password)
+
+
+class socksocket(socket.socket):
+ """socksocket([family[, type[, proto]]]) -> socket object
+
+ Open a SOCKS enabled socket. The parameters are the same as
+ those of the standard socket init. In order for SOCKS to work,
+ you must specify family=AF_INET, type=SOCK_STREAM and proto=0.
+ """
+
+ def __init__(self, family=socket.AF_INET, type=socket.SOCK_STREAM,
+ proto=0, _sock=None):
+ socket.socket.__init__(self, family, type, proto, _sock)
+ if _defaultproxy != None:
+ self.__proxy = _defaultproxy
+ else:
+ self.__proxy = (None, None, None, None, None, None)
+ self.__proxysockname = None
+ self.__proxypeername = None
+
+ def __decode(self, bytes):
+ if getattr(bytes, 'decode', False):
+ try:
+ bytes = bytes.decode()
+ except Exception:
+ pass
+ return bytes
+
+ def __encode(self, bytes):
+ if getattr(bytes, 'encode', False):
+ try:
+ bytes = bytes.encode()
+ except Exception:
+ pass
+ return bytes
+
+ def __recvall(self, count):
+ """__recvall(count) -> data
+ Receive EXACTLY the number of bytes requested from the socket.
+ Blocks until the required number of bytes have been received.
+ """
+ data = bytes("")
+ while len(data) < count:
+ d = self.recv(count - len(data))
+ if not d:
+ raise GeneralProxyError(
+ (0, "connection closed unexpectedly"))
+ data = data + self.__decode(d)
+ return data
+
+ def sendall(self, bytes):
+ socket.socket.sendall(self, self.__encode(bytes))
+
+ def setproxy(self, proxytype=None, addr=None, port=None, rdns=True,
+ username=None, password=None):
+ """setproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
+ Sets the proxy to be used.
+ proxytype - The type of the proxy to be used. Three types
+ are supported: PROXY_TYPE_SOCKS4 (including socks4a),
+ PROXY_TYPE_SOCKS5 and PROXY_TYPE_HTTP
+ addr - The address of the server (IP or DNS).
+ port - The port of the server. Defaults to 1080 for SOCKS
+ servers and 8080 for HTTP proxy servers.
+ rdns - Should DNS queries be preformed on the remote side
+ (rather than the local side). The default is True.
+ Note: This has no effect with SOCKS4 servers.
+ username - Username to authenticate with to the server.
+ The default is no authentication.
+ password - Password to authenticate with to the server.
+ Only relevant when username is also provided.
+ """
+ self.__proxy = (proxytype, addr, port, rdns, username, password)
+
+ def __negotiatesocks5(self, destaddr, destport):
+ """__negotiatesocks5(self,destaddr,destport)
+ Negotiates a connection through a SOCKS5 server.
+ """
+ # First we'll send the authentication packages we support.
+ if (self.__proxy[4] != None) and (self.__proxy[5] != None):
+ # The username/password details were supplied to the
+ # setproxy method so we support the USERNAME/PASSWORD
+ # authentication (in addition to the standard none).
+ self.sendall("\x05\x02\x00\x02")
+ else:
+ # No username/password were entered, therefore we
+ # only support connections with no authentication.
+ self.sendall("\x05\x01\x00")
+ # We'll receive the server's response to determine which
+ # method was selected
+ chosenauth = self.__recvall(2)
+ if chosenauth[0] != "\x05":
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ # Check the chosen authentication method
+ if chosenauth[1] == "\x00":
+ # No authentication is required
+ pass
+ elif chosenauth[1] == "\x02":
+ # Okay, we need to perform a basic username/password
+ # authentication.
+ self.sendall("\x01" + chr(len(self.__proxy[4])) + self.__proxy[4] +
+ chr(len(self.__proxy[5])) + self.__proxy[5])
+ authstat = self.__recvall(2)
+ if authstat[0] != "\x01":
+ # Bad response
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ if authstat[1] != "\x00":
+ # Authentication failed
+ self.close()
+ raise Socks5AuthError((3, _socks5autherrors[3]))
+ # Authentication succeeded
+ else:
+ # Reaching here is always bad
+ self.close()
+ if chosenauth[1] == "\xFF":
+ raise Socks5AuthError((2, _socks5autherrors[2]))
+ else:
+ raise GeneralProxyError((1, _generalerrors[1]))
+ # Now we can request the actual connection
+ req = "\x05\x01\x00"
+ # If the given destination address is an IP address, we'll
+ # use the IPv4 address request even if remote resolving was specified.
+ try:
+ ipaddr = socket.inet_aton(destaddr)
+ req = req + "\x01" + ipaddr
+ except socket.error:
+ # Well it's not an IP number, so it's probably a DNS name.
+ if self.__proxy[3] == True:
+ # Resolve remotely
+ ipaddr = None
+ req = req + "\x03" + chr(len(destaddr)) + destaddr
+ else:
+ # Resolve locally
+ ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
+ req = req + "\x01" + ipaddr
+ req = req + self.__decode(struct.pack(">H", destport))
+ self.sendall(req)
+ # Get the response
+ resp = self.__recvall(4)
+ if resp[0] != "\x05":
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ elif resp[1] != "\x00":
+ # Connection failed
+ self.close()
+ if ord(resp[1]) <= 8:
+ raise Socks5Error((ord(resp[1]), _socks5errors[ord(resp[1])]))
+ else:
+ raise Socks5Error((9, _socks5errors[9]))
+ # Get the bound address/port
+ elif resp[3] == "\x01":
+ boundaddr = self.__recvall(4)
+ elif resp[3] == "\x03":
+ resp = resp + self.recv(1)
+ boundaddr = self.__recvall(ord(resp[4]))
+ else:
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ boundport = struct.unpack(">H", bytes(self.__recvall(2), 'utf8'))[0]
+ self.__proxysockname = boundaddr, boundport
+ if ipaddr != None:
+ self.__proxypeername = (socket.inet_ntoa(ipaddr), destport)
+ else:
+ self.__proxypeername = (destaddr, destport)
+
+ def getproxysockname(self):
+ """getsockname() -> address info
+ Returns the bound IP address and port number at the proxy.
+ """
+ return self.__proxysockname
+
+ def getproxypeername(self):
+ """getproxypeername() -> address info
+ Returns the IP and port number of the proxy.
+ """
+ return socket.socket.getpeername(self)
+
+ def getpeername(self):
+ """getpeername() -> address info
+ Returns the IP address and port number of the destination
+ machine (note: getproxypeername returns the proxy)
+ """
+ return self.__proxypeername
+
+ def __negotiatesocks4(self, destaddr, destport):
+ """__negotiatesocks4(self,destaddr,destport)
+ Negotiates a connection through a SOCKS4 server.
+ """
+ # Check if the destination address provided is an IP address
+ rmtrslv = False
+ try:
+ ipaddr = socket.inet_aton(destaddr)
+ except socket.error:
+ # It's a DNS name. Check where it should be resolved.
+ if self.__proxy[3] == True:
+ ipaddr = "\x00\x00\x00\x01"
+ rmtrslv = True
+ else:
+ ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
+ # Construct the request packet
+ req = "\x04\x01" + self.__decode(struct.pack(">H", destport)) + ipaddr
+ # The username parameter is considered userid for SOCKS4
+ if self.__proxy[4] != None:
+ req = req + self.__proxy[4]
+ req = req + "\x00"
+ # DNS name if remote resolving is required
+ # NOTE: This is actually an extension to the SOCKS4 protocol
+ # called SOCKS4A and may not be supported in all cases.
+ if rmtrslv==True:
+ req = req + destaddr + "\x00"
+ self.sendall(req)
+ # Get the response from the server
+ resp = self.__recvall(8)
+ if resp[0] != "\x00":
+ # Bad data
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ if resp[1] != "\x5A":
+ # Server returned an error
+ self.close()
+ if ord(resp[1]) in (91,92,93):
+ self.close()
+ raise Socks4Error((ord(resp[1]), _socks4errors[ord(resp[1])-90]))
+ else:
+ raise Socks4Error((94,_socks4errors[4]))
+ # Get the bound address/port
+ self.__proxysockname = (socket.inet_ntoa(resp[4:]),struct.unpack(">H",bytes(resp[2:4],'utf8'))[0])
+ if rmtrslv != None:
+ self.__proxypeername = (socket.inet_ntoa(ipaddr),destport)
+ else:
+ self.__proxypeername = (destaddr, destport)
+
+ def __negotiatehttp(self, destaddr, destport):
+ """__negotiatehttp(self,destaddr,destport)
+ Negotiates a connection through an HTTP server.
+ """
+ # If we need to resolve locally, we do this now
+ if self.__proxy[3] == False:
+ addr = socket.gethostbyname(destaddr)
+ else:
+ addr = destaddr
+ self.sendall(("CONNECT %s:%s HTTP/1.1\r\n"
+ "Host: %s\r\n\r\n") % (addr, destport, destaddr))
+ # We read the response until we get the string "\r\n\r\n"
+ resp = self.recv(1)
+ while resp.find("\r\n\r\n") == -1:
+ resp = resp + self.recv(1)
+ # We just need the first line to check if the connection
+ # was successful
+ statusline = resp.splitlines()[0].split(" ", 2)
+ if statusline[0] not in ("HTTP/1.0", "HTTP/1.1"):
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ try:
+ statuscode = int(statusline[1])
+ except ValueError:
+ self.close()
+ raise GeneralProxyError((1, _generalerrors[1]))
+ if statuscode != 200:
+ self.close()
+ raise HTTPError((statuscode, statusline[2]))
+ self.__proxysockname = ("0.0.0.0", 0)
+ self.__proxypeername = (addr, destport)
+
+ def connect(self, destpair):
+ """connect(self,despair)
+ Connects to the specified destination through a proxy.
+ destpar - A tuple of the IP/DNS address and the port number.
+ (identical to socket's connect).
+ To select the proxy server use setproxy().
+ """
+ # Do a minimal input check first
+ # TODO(durin42): seriously? type checking? do we care?
+ if ((not isinstance(destpair, (list, tuple))) or len(destpair) < 2
+ or not isinstance(destpair[0], str) or not isinstance(destpair[1], int)):
+ raise GeneralProxyError((5, _generalerrors[5]))
+ if self.__proxy[0] == PROXY_TYPE_SOCKS5:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 1080
+ socket.socket.connect(self,(self.__proxy[1], portnum))
+ self.__negotiatesocks5(destpair[0], destpair[1])
+ elif self.__proxy[0] == PROXY_TYPE_SOCKS4:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 1080
+ socket.socket.connect(self, (self.__proxy[1], portnum))
+ self.__negotiatesocks4(destpair[0], destpair[1])
+ elif self.__proxy[0] == PROXY_TYPE_HTTP:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 8080
+ socket.socket.connect(self, (self.__proxy[1], portnum))
+ self.__negotiatehttp(destpair[0], destpair[1])
+ elif self.__proxy[0] == None:
+ socket.socket.connect(self, (destpair[0], destpair[1]))
+ else:
+ raise GeneralProxyError((4, _generalerrors[4]))