1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
|
# -*- coding: utf-8 -*-
from __future__ import absolute_import
"""
oauthlib.common
~~~~~~~~~~~~~~
This module provides data structures and utilities common
to all implementations of OAuth.
"""
import random
import re
import string
import time
import urllib
import urlparse
UNICODE_ASCII_CHARACTER_SET = (string.ascii_letters.decode('ascii') +
string.digits.decode('ascii'))
always_safe = (u'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
u'abcdefghijklmnopqrstuvwxyz'
u'0123456789' u'_.-')
def quote(s, safe=u'/'):
encoded = s.encode("utf-8")
quoted = urllib.quote(encoded, safe)
return quoted.decode("utf-8")
def unquote(s):
encoded = s.encode("utf-8")
unquoted = urllib.unquote(encoded)
return unquoted.decode("utf-8")
def urlencode(params):
utf8_params = encode_params_utf8(params)
urlencoded = urllib.urlencode(utf8_params)
return urlencoded.decode("utf-8")
def encode_params_utf8(params):
"""Ensures that all parameters in a list of 2-element tuples are encoded to
bytestrings using UTF-8
"""
encoded = []
for k, v in params:
encoded.append((
k.encode('utf-8') if isinstance(k, unicode) else k,
v.encode('utf-8') if isinstance(v, unicode) else v))
return encoded
def decode_params_utf8(params):
"""Ensures that all parameters in a list of 2-element tuples are decoded to
unicode using UTF-8.
"""
decoded = []
for k, v in params:
decoded.append((
k.decode('utf-8') if isinstance(k, str) else k,
v.decode('utf-8') if isinstance(v, str) else v))
return decoded
urlencoded = set(always_safe) | set(u'=&;%+~')
def urldecode(query):
"""Decode a query string in x-www-form-urlencoded format into a sequence
of two-element tuples.
Unlike urlparse.parse_qsl(..., strict_parsing=True) urldecode will enforce
correct formatting of the query string by validation. If validation fails
a ValueError will be raised. urllib.parse_qsl will only raise errors if
any of name-value pairs omits the equals sign.
"""
# Check if query contains invalid characters
if query and not set(query) <= urlencoded:
raise ValueError('Invalid characters in query string.')
# Check for correctly hex encoded values using a regular expression
# All encoded values begin with % followed by two hex characters
# correct = %00, %A0, %0A, %FF
# invalid = %G0, %5H, %PO
invalid_hex = u'%[^0-9A-Fa-f]|%[0-9A-Fa-f][^0-9A-Fa-f]'
if len(re.findall(invalid_hex, query)):
raise ValueError('Invalid hex encoding in query string.')
query = query.decode('utf-8') if isinstance(query, str) else query
# We want to allow queries such as "c2" whereas urlparse.parse_qsl
# with the strict_parsing flag will not.
params = urlparse.parse_qsl(query, keep_blank_values=True)
# unicode all the things
return decode_params_utf8(params)
def extract_params(raw):
"""Extract parameters and return them as a list of 2-tuples.
Will successfully extract parameters from urlencoded query strings,
dicts, or lists of 2-tuples. Empty strings/dicts/lists will return an
empty list of parameters. Any other input will result in a return
value of None.
"""
if isinstance(raw, basestring):
try:
params = urldecode(raw)
except ValueError:
params = None
elif hasattr(raw, '__iter__'):
try:
dict(raw)
except ValueError:
params = None
except TypeError:
params = None
else:
params = list(raw.items() if isinstance(raw, dict) else raw)
params = decode_params_utf8(params)
else:
params = None
return params
def generate_nonce():
"""Generate pseudorandom nonce that is unlikely to repeat.
Per `section 3.3`_ of the OAuth 1 RFC 5849 spec.
Per `section 3.2.1`_ of the MAC Access Authentication spec.
A random 64-bit number is appended to the epoch timestamp for both
randomness and to decrease the likelihood of collisions.
.. _`section 3.2.1`: http://tools.ietf.org/html/draft-ietf-oauth-v2-http-mac-01#section-3.2.1
.. _`section 3.3`: http://tools.ietf.org/html/rfc5849#section-3.3
"""
return unicode(unicode(random.getrandbits(64)) + generate_timestamp())
def generate_timestamp():
"""Get seconds since epoch (UTC).
Per `section 3.3`_ of the OAuth 1 RFC 5849 spec.
Per `section 3.2.1`_ of the MAC Access Authentication spec.
.. _`section 3.2.1`: http://tools.ietf.org/html/draft-ietf-oauth-v2-http-mac-01#section-3.2.1
.. _`section 3.3`: http://tools.ietf.org/html/rfc5849#section-3.3
"""
return unicode(int(time.time()))
def generate_token(length=30, chars=UNICODE_ASCII_CHARACTER_SET):
"""Generates a non-guessable OAuth token
OAuth (1 and 2) does not specify the format of tokens except that they
should be strings of random characters. Tokens should not be guessable
and entropy when generating the random characters is important. Which is
why SystemRandom is used instead of the default random.choice method.
"""
rand = random.SystemRandom()
return u''.join(rand.choice(chars) for x in range(length))
def add_params_to_qs(query, params):
"""Extend a query with a list of two-tuples."""
queryparams = urlparse.parse_qsl(query, keep_blank_values=True)
queryparams.extend(params)
return urlencode(queryparams)
def add_params_to_uri(uri, params):
"""Add a list of two-tuples to the uri query components."""
sch, net, path, par, query, fra = urlparse.urlparse(uri)
query = add_params_to_qs(query, params)
return urlparse.urlunparse((sch, net, path, par, query, fra))
def safe_string_equals(a, b):
""" Near-constant time string comparison.
Used in order to avoid timing attacks on sensitive information such
as secret keys during request verification (`rootLabs`_).
.. _`rootLabs`: http://rdist.root.org/2010/01/07/timing-independent-array-comparison/
"""
if len(a) != len(b):
return False
result = 0
for x, y in zip(a, b):
result |= ord(x) ^ ord(y)
return result == 0
class Request(object):
"""A malleable representation of a signable HTTP request.
Body argument may contain any data, but parameters will only be decoded if
they are one of:
* urlencoded query string
* dict
* list of 2-tuples
Anything else will be treated as raw body data to be passed through
unmolested.
"""
def __init__(self, uri, http_method=u'GET', body=None, headers=None):
self.uri = uri
self.http_method = http_method
self.headers = headers or {}
self.body = body
self.decoded_body = extract_params(body)
self.oauth_params = []
@property
def uri_query(self):
return urlparse.urlparse(self.uri).query
@property
def uri_query_params(self):
return urlparse.parse_qsl(self.uri_query, keep_blank_values=True,
strict_parsing=True)
|