1# -*- coding: utf-8 -*- 2 3""" 4requests.cookies 5~~~~~~~~~~~~~~~~ 6 7Compatibility code to be able to use `cookielib.CookieJar` with requests. 8 9requests.utils imports from here, so be careful with imports. 10""" 11 12import copy 13import time 14import calendar 15 16from ._internal_utils import to_native_string 17from .compat import cookielib, urlparse, urlunparse, Morsel, MutableMapping 18 19try: 20 import threading 21except ImportError: 22 import dummy_threading as threading 23 24 25class MockRequest(object): 26 """Wraps a `requests.Request` to mimic a `urllib2.Request`. 27 28 The code in `cookielib.CookieJar` expects this interface in order to correctly 29 manage cookie policies, i.e., determine whether a cookie can be set, given the 30 domains of the request and the cookie. 31 32 The original request object is read-only. The client is responsible for collecting 33 the new headers via `get_new_headers()` and interpreting them appropriately. You 34 probably want `get_cookie_header`, defined below. 35 """ 36 37 def __init__(self, request): 38 self._r = request 39 self._new_headers = {} 40 self.type = urlparse(self._r.url).scheme 41 42 def get_type(self): 43 return self.type 44 45 def get_host(self): 46 return urlparse(self._r.url).netloc 47 48 def get_origin_req_host(self): 49 return self.get_host() 50 51 def get_full_url(self): 52 # Only return the response's URL if the user hadn't set the Host 53 # header 54 if not self._r.headers.get('Host'): 55 return self._r.url 56 # If they did set it, retrieve it and reconstruct the expected domain 57 host = to_native_string(self._r.headers['Host'], encoding='utf-8') 58 parsed = urlparse(self._r.url) 59 # Reconstruct the URL as we expect it 60 return urlunparse([ 61 parsed.scheme, host, parsed.path, parsed.params, parsed.query, 62 parsed.fragment 63 ]) 64 65 def is_unverifiable(self): 66 return True 67 68 def has_header(self, name): 69 return name in self._r.headers or name in self._new_headers 70 71 def get_header(self, name, default=None): 72 return self._r.headers.get(name, self._new_headers.get(name, default)) 73 74 def add_header(self, key, val): 75 """cookielib has no legitimate use for this method; add it back if you find one.""" 76 raise NotImplementedError("Cookie headers should be added with add_unredirected_header()") 77 78 def add_unredirected_header(self, name, value): 79 self._new_headers[name] = value 80 81 def get_new_headers(self): 82 return self._new_headers 83 84 @property 85 def unverifiable(self): 86 return self.is_unverifiable() 87 88 @property 89 def origin_req_host(self): 90 return self.get_origin_req_host() 91 92 @property 93 def host(self): 94 return self.get_host() 95 96 97class MockResponse(object): 98 """Wraps a `httplib.HTTPMessage` to mimic a `urllib.addinfourl`. 99 100 ...what? Basically, expose the parsed HTTP headers from the server response 101 the way `cookielib` expects to see them. 102 """ 103 104 def __init__(self, headers): 105 """Make a MockResponse for `cookielib` to read. 106 107 :param headers: a httplib.HTTPMessage or analogous carrying the headers 108 """ 109 self._headers = headers 110 111 def info(self): 112 return self._headers 113 114 def getheaders(self, name): 115 self._headers.getheaders(name) 116 117 118def extract_cookies_to_jar(jar, request, response): 119 """Extract the cookies from the response into a CookieJar. 120 121 :param jar: cookielib.CookieJar (not necessarily a RequestsCookieJar) 122 :param request: our own requests.Request object 123 :param response: urllib3.HTTPResponse object 124 """ 125 if not (hasattr(response, '_original_response') and 126 response._original_response): 127 return 128 # the _original_response field is the wrapped httplib.HTTPResponse object, 129 req = MockRequest(request) 130 # pull out the HTTPMessage with the headers and put it in the mock: 131 res = MockResponse(response._original_response.msg) 132 jar.extract_cookies(res, req) 133 134 135def get_cookie_header(jar, request): 136 """ 137 Produce an appropriate Cookie header string to be sent with `request`, or None. 138 139 :rtype: str 140 """ 141 r = MockRequest(request) 142 jar.add_cookie_header(r) 143 return r.get_new_headers().get('Cookie') 144 145 146def remove_cookie_by_name(cookiejar, name, domain=None, path=None): 147 """Unsets a cookie by name, by default over all domains and paths. 148 149 Wraps CookieJar.clear(), is O(n). 150 """ 151 clearables = [] 152 for cookie in cookiejar: 153 if cookie.name != name: 154 continue 155 if domain is not None and domain != cookie.domain: 156 continue 157 if path is not None and path != cookie.path: 158 continue 159 clearables.append((cookie.domain, cookie.path, cookie.name)) 160 161 for domain, path, name in clearables: 162 cookiejar.clear(domain, path, name) 163 164 165class CookieConflictError(RuntimeError): 166 """There are two cookies that meet the criteria specified in the cookie jar. 167 Use .get and .set and include domain and path args in order to be more specific. 168 """ 169 170 171class RequestsCookieJar(cookielib.CookieJar, MutableMapping): 172 """Compatibility class; is a cookielib.CookieJar, but exposes a dict 173 interface. 174 175 This is the CookieJar we create by default for requests and sessions that 176 don't specify one, since some clients may expect response.cookies and 177 session.cookies to support dict operations. 178 179 Requests does not use the dict interface internally; it's just for 180 compatibility with external client code. All requests code should work 181 out of the box with externally provided instances of ``CookieJar``, e.g. 182 ``LWPCookieJar`` and ``FileCookieJar``. 183 184 Unlike a regular CookieJar, this class is pickleable. 185 186 .. warning:: dictionary operations that are normally O(1) may be O(n). 187 """ 188 189 def get(self, name, default=None, domain=None, path=None): 190 """Dict-like get() that also supports optional domain and path args in 191 order to resolve naming collisions from using one cookie jar over 192 multiple domains. 193 194 .. warning:: operation is O(n), not O(1). 195 """ 196 try: 197 return self._find_no_duplicates(name, domain, path) 198 except KeyError: 199 return default 200 201 def set(self, name, value, **kwargs): 202 """Dict-like set() that also supports optional domain and path args in 203 order to resolve naming collisions from using one cookie jar over 204 multiple domains. 205 """ 206 # support client code that unsets cookies by assignment of a None value: 207 if value is None: 208 remove_cookie_by_name(self, name, domain=kwargs.get('domain'), path=kwargs.get('path')) 209 return 210 211 if isinstance(value, Morsel): 212 c = morsel_to_cookie(value) 213 else: 214 c = create_cookie(name, value, **kwargs) 215 self.set_cookie(c) 216 return c 217 218 def iterkeys(self): 219 """Dict-like iterkeys() that returns an iterator of names of cookies 220 from the jar. 221 222 .. seealso:: itervalues() and iteritems(). 223 """ 224 for cookie in iter(self): 225 yield cookie.name 226 227 def keys(self): 228 """Dict-like keys() that returns a list of names of cookies from the 229 jar. 230 231 .. seealso:: values() and items(). 232 """ 233 return list(self.iterkeys()) 234 235 def itervalues(self): 236 """Dict-like itervalues() that returns an iterator of values of cookies 237 from the jar. 238 239 .. seealso:: iterkeys() and iteritems(). 240 """ 241 for cookie in iter(self): 242 yield cookie.value 243 244 def values(self): 245 """Dict-like values() that returns a list of values of cookies from the 246 jar. 247 248 .. seealso:: keys() and items(). 249 """ 250 return list(self.itervalues()) 251 252 def iteritems(self): 253 """Dict-like iteritems() that returns an iterator of name-value tuples 254 from the jar. 255 256 .. seealso:: iterkeys() and itervalues(). 257 """ 258 for cookie in iter(self): 259 yield cookie.name, cookie.value 260 261 def items(self): 262 """Dict-like items() that returns a list of name-value tuples from the 263 jar. Allows client-code to call ``dict(RequestsCookieJar)`` and get a 264 vanilla python dict of key value pairs. 265 266 .. seealso:: keys() and values(). 267 """ 268 return list(self.iteritems()) 269 270 def list_domains(self): 271 """Utility method to list all the domains in the jar.""" 272 domains = [] 273 for cookie in iter(self): 274 if cookie.domain not in domains: 275 domains.append(cookie.domain) 276 return domains 277 278 def list_paths(self): 279 """Utility method to list all the paths in the jar.""" 280 paths = [] 281 for cookie in iter(self): 282 if cookie.path not in paths: 283 paths.append(cookie.path) 284 return paths 285 286 def multiple_domains(self): 287 """Returns True if there are multiple domains in the jar. 288 Returns False otherwise. 289 290 :rtype: bool 291 """ 292 domains = [] 293 for cookie in iter(self): 294 if cookie.domain is not None and cookie.domain in domains: 295 return True 296 domains.append(cookie.domain) 297 return False # there is only one domain in jar 298 299 def get_dict(self, domain=None, path=None): 300 """Takes as an argument an optional domain and path and returns a plain 301 old Python dict of name-value pairs of cookies that meet the 302 requirements. 303 304 :rtype: dict 305 """ 306 dictionary = {} 307 for cookie in iter(self): 308 if ( 309 (domain is None or cookie.domain == domain) and 310 (path is None or cookie.path == path) 311 ): 312 dictionary[cookie.name] = cookie.value 313 return dictionary 314 315 def __contains__(self, name): 316 try: 317 return super(RequestsCookieJar, self).__contains__(name) 318 except CookieConflictError: 319 return True 320 321 def __getitem__(self, name): 322 """Dict-like __getitem__() for compatibility with client code. Throws 323 exception if there are more than one cookie with name. In that case, 324 use the more explicit get() method instead. 325 326 .. warning:: operation is O(n), not O(1). 327 """ 328 return self._find_no_duplicates(name) 329 330 def __setitem__(self, name, value): 331 """Dict-like __setitem__ for compatibility with client code. Throws 332 exception if there is already a cookie of that name in the jar. In that 333 case, use the more explicit set() method instead. 334 """ 335 self.set(name, value) 336 337 def __delitem__(self, name): 338 """Deletes a cookie given a name. Wraps ``cookielib.CookieJar``'s 339 ``remove_cookie_by_name()``. 340 """ 341 remove_cookie_by_name(self, name) 342 343 def set_cookie(self, cookie, *args, **kwargs): 344 if hasattr(cookie.value, 'startswith') and cookie.value.startswith('"') and cookie.value.endswith('"'): 345 cookie.value = cookie.value.replace('\\"', '') 346 return super(RequestsCookieJar, self).set_cookie(cookie, *args, **kwargs) 347 348 def update(self, other): 349 """Updates this jar with cookies from another CookieJar or dict-like""" 350 if isinstance(other, cookielib.CookieJar): 351 for cookie in other: 352 self.set_cookie(copy.copy(cookie)) 353 else: 354 super(RequestsCookieJar, self).update(other) 355 356 def _find(self, name, domain=None, path=None): 357 """Requests uses this method internally to get cookie values. 358 359 If there are conflicting cookies, _find arbitrarily chooses one. 360 See _find_no_duplicates if you want an exception thrown if there are 361 conflicting cookies. 362 363 :param name: a string containing name of cookie 364 :param domain: (optional) string containing domain of cookie 365 :param path: (optional) string containing path of cookie 366 :return: cookie.value 367 """ 368 for cookie in iter(self): 369 if cookie.name == name: 370 if domain is None or cookie.domain == domain: 371 if path is None or cookie.path == path: 372 return cookie.value 373 374 raise KeyError('name=%r, domain=%r, path=%r' % (name, domain, path)) 375 376 def _find_no_duplicates(self, name, domain=None, path=None): 377 """Both ``__get_item__`` and ``get`` call this function: it's never 378 used elsewhere in Requests. 379 380 :param name: a string containing name of cookie 381 :param domain: (optional) string containing domain of cookie 382 :param path: (optional) string containing path of cookie 383 :raises KeyError: if cookie is not found 384 :raises CookieConflictError: if there are multiple cookies 385 that match name and optionally domain and path 386 :return: cookie.value 387 """ 388 toReturn = None 389 for cookie in iter(self): 390 if cookie.name == name: 391 if domain is None or cookie.domain == domain: 392 if path is None or cookie.path == path: 393 if toReturn is not None: # if there are multiple cookies that meet passed in criteria 394 raise CookieConflictError('There are multiple cookies with name, %r' % (name)) 395 toReturn = cookie.value # we will eventually return this as long as no cookie conflict 396 397 if toReturn: 398 return toReturn 399 raise KeyError('name=%r, domain=%r, path=%r' % (name, domain, path)) 400 401 def __getstate__(self): 402 """Unlike a normal CookieJar, this class is pickleable.""" 403 state = self.__dict__.copy() 404 # remove the unpickleable RLock object 405 state.pop('_cookies_lock') 406 return state 407 408 def __setstate__(self, state): 409 """Unlike a normal CookieJar, this class is pickleable.""" 410 self.__dict__.update(state) 411 if '_cookies_lock' not in self.__dict__: 412 self._cookies_lock = threading.RLock() 413 414 def copy(self): 415 """Return a copy of this RequestsCookieJar.""" 416 new_cj = RequestsCookieJar() 417 new_cj.set_policy(self.get_policy()) 418 new_cj.update(self) 419 return new_cj 420 421 def get_policy(self): 422 """Return the CookiePolicy instance used.""" 423 return self._policy 424 425 426def _copy_cookie_jar(jar): 427 if jar is None: 428 return None 429 430 if hasattr(jar, 'copy'): 431 # We're dealing with an instance of RequestsCookieJar 432 return jar.copy() 433 # We're dealing with a generic CookieJar instance 434 new_jar = copy.copy(jar) 435 new_jar.clear() 436 for cookie in jar: 437 new_jar.set_cookie(copy.copy(cookie)) 438 return new_jar 439 440 441def create_cookie(name, value, **kwargs): 442 """Make a cookie from underspecified parameters. 443 444 By default, the pair of `name` and `value` will be set for the domain '' 445 and sent on every request (this is sometimes called a "supercookie"). 446 """ 447 result = { 448 'version': 0, 449 'name': name, 450 'value': value, 451 'port': None, 452 'domain': '', 453 'path': '/', 454 'secure': False, 455 'expires': None, 456 'discard': True, 457 'comment': None, 458 'comment_url': None, 459 'rest': {'HttpOnly': None}, 460 'rfc2109': False, 461 } 462 463 badargs = set(kwargs) - set(result) 464 if badargs: 465 err = 'create_cookie() got unexpected keyword arguments: %s' 466 raise TypeError(err % list(badargs)) 467 468 result.update(kwargs) 469 result['port_specified'] = bool(result['port']) 470 result['domain_specified'] = bool(result['domain']) 471 result['domain_initial_dot'] = result['domain'].startswith('.') 472 result['path_specified'] = bool(result['path']) 473 474 return cookielib.Cookie(**result) 475 476 477def morsel_to_cookie(morsel): 478 """Convert a Morsel object into a Cookie containing the one k/v pair.""" 479 480 expires = None 481 if morsel['max-age']: 482 try: 483 expires = int(time.time() + int(morsel['max-age'])) 484 except ValueError: 485 raise TypeError('max-age: %s must be integer' % morsel['max-age']) 486 elif morsel['expires']: 487 time_template = '%a, %d-%b-%Y %H:%M:%S GMT' 488 expires = calendar.timegm( 489 time.strptime(morsel['expires'], time_template) 490 ) 491 return create_cookie( 492 comment=morsel['comment'], 493 comment_url=bool(morsel['comment']), 494 discard=False, 495 domain=morsel['domain'], 496 expires=expires, 497 name=morsel.key, 498 path=morsel['path'], 499 port=None, 500 rest={'HttpOnly': morsel['httponly']}, 501 rfc2109=False, 502 secure=bool(morsel['secure']), 503 value=morsel.value, 504 version=morsel['version'] or 0, 505 ) 506 507 508def cookiejar_from_dict(cookie_dict, cookiejar=None, overwrite=True): 509 """Returns a CookieJar from a key/value dictionary. 510 511 :param cookie_dict: Dict of key/values to insert into CookieJar. 512 :param cookiejar: (optional) A cookiejar to add the cookies to. 513 :param overwrite: (optional) If False, will not replace cookies 514 already in the jar with new ones. 515 :rtype: CookieJar 516 """ 517 if cookiejar is None: 518 cookiejar = RequestsCookieJar() 519 520 if cookie_dict is not None: 521 names_from_jar = [cookie.name for cookie in cookiejar] 522 for name in cookie_dict: 523 if overwrite or (name not in names_from_jar): 524 cookiejar.set_cookie(create_cookie(name, cookie_dict[name])) 525 526 return cookiejar 527 528 529def merge_cookies(cookiejar, cookies): 530 """Add cookies to cookiejar and returns a merged CookieJar. 531 532 :param cookiejar: CookieJar object to add the cookies to. 533 :param cookies: Dictionary or CookieJar object to be added. 534 :rtype: CookieJar 535 """ 536 if not isinstance(cookiejar, cookielib.CookieJar): 537 raise ValueError('You can only merge into CookieJar') 538 539 if isinstance(cookies, dict): 540 cookiejar = cookiejar_from_dict( 541 cookies, cookiejar=cookiejar, overwrite=False) 542 elif isinstance(cookies, cookielib.CookieJar): 543 try: 544 cookiejar.update(cookies) 545 except AttributeError: 546 for cookie_in_jar in cookies: 547 cookiejar.set_cookie(cookie_in_jar) 548 549 return cookiejar 550