« back to all changes in this revision
Viewing changes to commit-bot/_http.py
- browse files at revision 86
- compare with another revision
- download diff
- download tarball
- view history from revision 86
- Committer: tom.prince at ualberta
- Date: 2013-05-21 17:27:11 UTC
- Revision ID: tom.prince@ualberta.net-20130521172711-o9vifzpokcyf2i55
Remove commit-bot and diffresource.
The code and history for them now reside in seperate repositories at
https://github.com/twisted-infra.
The code and history for them now reside in seperate repositories at
https://github.com/twisted-infra.
- files removed:
- commit-bot
- diffresource
added
removed
commit-bot/_http.py
1
2
import urlparse
3
import md5, sha
4
5
from twisted.web import client, http
6
from twisted.internet import reactor
7
8
9
class Token(str):
10
__slots__=[]
11
tokens = {}
12
def __new__(self, char):
13
token = Token.tokens.get(char)
14
if token is None:
15
Token.tokens[char] = token = str.__new__(self, char)
16
return token
17
18
def __repr__(self):
19
return "Token(%s)" % str.__repr__(self)
20
21
22
http_tokens = " \t\"()<>@,;:\\/[]?={}"
23
http_ctls = "\x00\x01\x02\x03\x04\x05\x06\x07\x08\t\n\x0b\x0c\r\x0e\x0f\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f\x7f"
24
25
26
def tokenize(header, foldCase=True):
27
"""Tokenize a string according to normal HTTP header parsing rules.
28
29
In particular:
30
- Whitespace is irrelevant and eaten next to special separator tokens.
31
Its existance (but not amount) is important between character strings.
32
- Quoted string support including embedded backslashes.
33
- Case is insignificant (and thus lowercased), except in quoted strings.
34
(unless foldCase=False)
35
- Multiple headers are concatenated with ','
36
37
NOTE: not all headers can be parsed with this function.
38
39
Takes a raw header value (list of strings), and
40
Returns a generator of strings and Token class instances.
41
"""
42
tokens=http_tokens
43
ctls=http_ctls
44
45
string = ",".join(header)
46
list = []
47
start = 0
48
cur = 0
49
quoted = False
50
qpair = False
51
inSpaces = -1
52
qstring = None
53
54
for x in string:
55
if quoted:
56
if qpair:
57
qpair = False
58
qstring = qstring+string[start:cur-1]+x
59
start = cur+1
60
elif x == '\\':
61
qpair = True
62
elif x == '"':
63
quoted = False
64
yield qstring+string[start:cur]
65
qstring=None
66
start = cur+1
67
elif x in tokens:
68
if start != cur:
69
if foldCase:
70
yield string[start:cur].lower()
71
else:
72
yield string[start:cur]
73
74
start = cur+1
75
if x == '"':
76
quoted = True
77
qstring = ""
78
inSpaces = False
79
elif x in " \t":
80
if inSpaces is False:
81
inSpaces = True
82
else:
83
inSpaces = -1
84
yield Token(x)
85
elif x in ctls:
86
raise ValueError("Invalid control character: %d in header" % ord(x))
87
else:
88
if inSpaces is True:
89
yield Token(' ')
90
inSpaces = False
91
92
inSpaces = False
93
cur = cur+1
94
95
if qpair:
96
raise ValueError, "Missing character after '\\'"
97
if quoted:
98
raise ValueError, "Missing end quote"
99
100
if start != cur:
101
if foldCase:
102
yield string[start:cur].lower()
103
else:
104
yield string[start:cur]
105
106
107
def parseWWWAuthenticate(tokenized):
108
headers = []
109
110
tokenList = list(tokenized)
111
112
while tokenList:
113
scheme = tokenList.pop(0)
114
challenge = {}
115
last = None
116
kvChallenge = False
117
118
while tokenList:
119
token = tokenList.pop(0)
120
if token == Token('='):
121
kvChallenge = True
122
challenge[last] = tokenList.pop(0)
123
last = None
124
125
elif token == Token(','):
126
if kvChallenge:
127
if len(tokenList) > 1 and tokenList[1] != Token('='):
128
break
129
130
else:
131
break
132
133
else:
134
last = token
135
136
if last and scheme and not challenge and not kvChallenge:
137
challenge = last
138
last = None
139
140
headers.append((scheme, challenge))
141
142
if last and last not in (Token('='), Token(',')):
143
if headers[-1] == (scheme, challenge):
144
scheme = last
145
challenge = {}
146
headers.append((scheme, challenge))
147
148
return headers
149
150
151
def parse(url, defaultPort=None):
152
"""
153
Split the given URL into the scheme, host, port, and path.
154
155
@type url: C{str}
156
@param url: An URL to parse.
157
158
@type defaultPort: C{int} or C{None}
159
@param defaultPort: An alternate value to use as the port if the URL does
160
not include one.
161
162
@return: A four-tuple of the scheme, host, port, and path of the URL. All
163
of these are C{str} instances except for port, which is an C{int}.
164
"""
165
url = url.strip()
166
parsed = http.urlparse(url)
167
scheme = parsed[0]
168
path = urlparse.urlunparse(('','')+parsed[2:])
169
if defaultPort is None:
170
if scheme == 'https':
171
defaultPort = 443
172
else:
173
defaultPort = 80
174
host, port = parsed[1], defaultPort
175
if ':' in host:
176
host, port = host.split(':')
177
port = int(port)
178
if path == "":
179
path = "/"
180
return scheme, host, port, path
181
182
183
def makeGetterFactory(url, factoryFactory, contextFactory=None,
184
*args, **kwargs):
185
"""
186
Create and connect an HTTP page getting factory.
187
188
Any additional positional or keyword arguments are used when calling
189
C{factoryFactory}.
190
191
@param factoryFactory: Factory factory that is called with C{url}, C{args}
192
and C{kwargs} to produce the getter
193
194
@param contextFactory: Context factory to use when creating a secure
195
connection, defaulting to C{None}
196
197
@return: The factory created by C{factoryFactory}
198
"""
199
scheme, host, port, path = parse(url)
200
factory = factoryFactory(url, *args, **kwargs)
201
if scheme == 'https':
202
from twisted.internet import ssl
203
if contextFactory is None:
204
contextFactory = ssl.ClientContextFactory()
205
reactor.connectSSL(host, port, factory, contextFactory)
206
else:
207
reactor.connectTCP(host, port, factory)
208
return factory
209
210
211
def getPage(url, contextFactory=None, *args, **kwargs):
212
"""
213
Download a web page as a string.
214
215
Download a page. Return a deferred, which will callback with a
216
page (as a string) or errback with a description of the error.
217
218
See HTTPClientFactory to see what extra args can be passed.
219
"""
220
return makeGetterFactory(
221
url,
222
client.HTTPClientFactory,
223
contextFactory=contextFactory,
224
*args, **kwargs)
225
226
algorithms = {
227
'md5': md5.new,
228
229
# md5-sess is more complicated than just another algorithm. It requires
230
# H(A1) state to be remembered from the first WWW-Authenticate challenge
231
# issued and re-used to process any Authorization header in response to
232
# that WWW-Authenticate challenge. It is *not* correct to simply
233
# recalculate H(A1) each time an Authorization header is received. Read
234
# RFC 2617, section 3.2.2.2 and do not try to make DigestCredentialFactory
235
# support this unless you completely understand it. -exarkun
236
'md5-sess': md5.new,
237
238
'sha': sha.new,
239
}
240
241
# DigestCalcHA1
242
def calcHA1(pszAlg, pszUserName, pszRealm, pszPassword, pszNonce, pszCNonce,
243
preHA1=None):
244
"""
245
Compute H(A1) from RFC 2617.
246
247
@param pszAlg: The name of the algorithm to use to calculate the digest.
248
Currently supported are md5, md5-sess, and sha.
249
@param pszUserName: The username
250
@param pszRealm: The realm
251
@param pszPassword: The password
252
@param pszNonce: The nonce
253
@param pszCNonce: The cnonce
254
255
@param preHA1: If available this is a str containing a previously
256
calculated H(A1) as a hex string. If this is given then the values for
257
pszUserName, pszRealm, and pszPassword must be C{None} and are ignored.
258
"""
259
260
if (preHA1 and (pszUserName or pszRealm or pszPassword)):
261
raise TypeError(("preHA1 is incompatible with the pszUserName, "
262
"pszRealm, and pszPassword arguments"))
263
264
if preHA1 is None:
265
# We need to calculate the HA1 from the username:realm:password
266
m = algorithms[pszAlg]()
267
m.update(pszUserName)
268
m.update(":")
269
m.update(pszRealm)
270
m.update(":")
271
m.update(pszPassword)
272
HA1 = m.digest()
273
else:
274
# We were given a username:realm:password
275
HA1 = preHA1.decode('hex')
276
277
if pszAlg == "md5-sess":
278
m = algorithms[pszAlg]()
279
m.update(HA1)
280
m.update(":")
281
m.update(pszNonce)
282
m.update(":")
283
m.update(pszCNonce)
284
HA1 = m.digest()
285
286
return HA1.encode('hex')
287
288
289
def calcHA2(algo, pszMethod, pszDigestUri, pszQop, pszHEntity):
290
"""
291
Compute H(A2) from RFC 2617.
292
293
@param pszAlg: The name of the algorithm to use to calculate the digest.
294
Currently supported are md5, md5-sess, and sha.
295
@param pszMethod: The request method.
296
@param pszDigestUri: The request URI.
297
@param pszQop: The Quality-of-Protection value.
298
@param pszHEntity: The hash of the entity body or C{None} if C{pszQop} is
299
not C{'auth-int'}.
300
@return: The hash of the A2 value for the calculation of the response
301
digest.
302
"""
303
m = algorithms[algo]()
304
m.update(pszMethod)
305
m.update(":")
306
m.update(pszDigestUri)
307
if pszQop == "auth-int":
308
m.update(":")
309
m.update(pszHEntity)
310
return m.digest().encode('hex')
311
312
313
def calcResponse(HA1, HA2, algo, pszNonce, pszNonceCount, pszCNonce, pszQop):
314
"""
315
Compute the digest for the given parameters.
316
317
@param HA1: The H(A1) value, as computed by L{calcHA1}.
318
@param HA2: The H(A2) value, as computed by L{calcHA2}.
319
@param pszNonce: The challenge nonce.
320
@param pszNonceCount: The (client) nonce count value for this response.
321
@param pszCNonce: The client nonce.
322
@param pszQop: The Quality-of-Protection value.
323
"""
324
m = algorithms[algo]()
325
m.update(HA1)
326
m.update(":")
327
m.update(pszNonce)
328
m.update(":")
329
if pszNonceCount and pszCNonce:
330
m.update(pszNonceCount)
331
m.update(":")
332
m.update(pszCNonce)
333
m.update(":")
334
m.update(pszQop)
335
m.update(":")
336
m.update(HA2)
337
respHash = m.digest().encode('hex')
338
return respHash
Loggerhead is a web-based interface for Breezy
Version: 2.0.1