1# -*- coding: utf-8 -*- 2 3""" 4requests.models 5~~~~~~~~~~~~~~~ 6 7This module contains the primary objects that power Requests. 8""" 9 10import datetime 11import sys 12 13# Import encoding now, to avoid implicit import later. 14# Implicit import within threads may cause LookupError when standard library is in a ZIP, 15# such as in Embedded Python. See https://github.com/psf/requests/issues/3578. 16import encodings.idna 17 18from urllib3.fields import RequestField 19from urllib3.filepost import encode_multipart_formdata 20from urllib3.util import parse_url 21from urllib3.exceptions import ( 22 DecodeError, ReadTimeoutError, ProtocolError, LocationParseError) 23 24from io import UnsupportedOperation 25from .hooks import default_hooks 26from .structures import CaseInsensitiveDict 27 28from .auth import HTTPBasicAuth 29from .cookies import cookiejar_from_dict, get_cookie_header, _copy_cookie_jar 30from .exceptions import ( 31 HTTPError, MissingSchema, InvalidURL, ChunkedEncodingError, 32 ContentDecodingError, ConnectionError, StreamConsumedError) 33from ._internal_utils import to_native_string, unicode_is_ascii 34from .utils import ( 35 guess_filename, get_auth_from_url, requote_uri, 36 stream_decode_response_unicode, to_key_val_list, parse_header_links, 37 iter_slices, guess_json_utf, super_len, check_header_validity) 38from .compat import ( 39 Callable, Mapping, 40 cookielib, urlunparse, urlsplit, urlencode, str, bytes, 41 is_py2, chardet, builtin_str, basestring) 42from .compat import json as complexjson 43from .status_codes import codes 44 45#: The set of HTTP status codes that indicate an automatically 46#: processable redirect. 47REDIRECT_STATI = ( 48 codes.moved, # 301 49 codes.found, # 302 50 codes.other, # 303 51 codes.temporary_redirect, # 307 52 codes.permanent_redirect, # 308 53) 54 55DEFAULT_REDIRECT_LIMIT = 30 56CONTENT_CHUNK_SIZE = 10 * 1024 57ITER_CHUNK_SIZE = 512 58 59 60class RequestEncodingMixin(object): 61 @property 62 def path_url(self): 63 """Build the path URL to use.""" 64 65 url = [] 66 67 p = urlsplit(self.url) 68 69 path = p.path 70 if not path: 71 path = '/' 72 73 url.append(path) 74 75 query = p.query 76 if query: 77 url.append('?') 78 url.append(query) 79 80 return ''.join(url) 81 82 @staticmethod 83 def _encode_params(data): 84 """Encode parameters in a piece of data. 85 86 Will successfully encode parameters when passed as a dict or a list of 87 2-tuples. Order is retained if data is a list of 2-tuples but arbitrary 88 if parameters are supplied as a dict. 89 """ 90 91 if isinstance(data, (str, bytes)): 92 return data 93 elif hasattr(data, 'read'): 94 return data 95 elif hasattr(data, '__iter__'): 96 result = [] 97 for k, vs in to_key_val_list(data): 98 if isinstance(vs, basestring) or not hasattr(vs, '__iter__'): 99 vs = [vs] 100 for v in vs: 101 if v is not None: 102 result.append( 103 (k.encode('utf-8') if isinstance(k, str) else k, 104 v.encode('utf-8') if isinstance(v, str) else v)) 105 return urlencode(result, doseq=True) 106 else: 107 return data 108 109 @staticmethod 110 def _encode_files(files, data): 111 """Build the body for a multipart/form-data request. 112 113 Will successfully encode files when passed as a dict or a list of 114 tuples. Order is retained if data is a list of tuples but arbitrary 115 if parameters are supplied as a dict. 116 The tuples may be 2-tuples (filename, fileobj), 3-tuples (filename, fileobj, contentype) 117 or 4-tuples (filename, fileobj, contentype, custom_headers). 118 """ 119 if (not files): 120 raise ValueError("Files must be provided.") 121 elif isinstance(data, basestring): 122 raise ValueError("Data must not be a string.") 123 124 new_fields = [] 125 fields = to_key_val_list(data or {}) 126 files = to_key_val_list(files or {}) 127 128 for field, val in fields: 129 if isinstance(val, basestring) or not hasattr(val, '__iter__'): 130 val = [val] 131 for v in val: 132 if v is not None: 133 # Don't call str() on bytestrings: in Py3 it all goes wrong. 134 if not isinstance(v, bytes): 135 v = str(v) 136 137 new_fields.append( 138 (field.decode('utf-8') if isinstance(field, bytes) else field, 139 v.encode('utf-8') if isinstance(v, str) else v)) 140 141 for (k, v) in files: 142 # support for explicit filename 143 ft = None 144 fh = None 145 if isinstance(v, (tuple, list)): 146 if len(v) == 2: 147 fn, fp = v 148 elif len(v) == 3: 149 fn, fp, ft = v 150 else: 151 fn, fp, ft, fh = v 152 else: 153 fn = guess_filename(v) or k 154 fp = v 155 156 if isinstance(fp, (str, bytes, bytearray)): 157 fdata = fp 158 elif hasattr(fp, 'read'): 159 fdata = fp.read() 160 elif fp is None: 161 continue 162 else: 163 fdata = fp 164 165 rf = RequestField(name=k, data=fdata, filename=fn, headers=fh) 166 rf.make_multipart(content_type=ft) 167 new_fields.append(rf) 168 169 body, content_type = encode_multipart_formdata(new_fields) 170 171 return body, content_type 172 173 174class RequestHooksMixin(object): 175 def register_hook(self, event, hook): 176 """Properly register a hook.""" 177 178 if event not in self.hooks: 179 raise ValueError('Unsupported event specified, with event name "%s"' % (event)) 180 181 if isinstance(hook, Callable): 182 self.hooks[event].append(hook) 183 elif hasattr(hook, '__iter__'): 184 self.hooks[event].extend(h for h in hook if isinstance(h, Callable)) 185 186 def deregister_hook(self, event, hook): 187 """Deregister a previously registered hook. 188 Returns True if the hook existed, False if not. 189 """ 190 191 try: 192 self.hooks[event].remove(hook) 193 return True 194 except ValueError: 195 return False 196 197 198class Request(RequestHooksMixin): 199 """A user-created :class:`Request <Request>` object. 200 201 Used to prepare a :class:`PreparedRequest <PreparedRequest>`, which is sent to the server. 202 203 :param method: HTTP method to use. 204 :param url: URL to send. 205 :param headers: dictionary of headers to send. 206 :param files: dictionary of {filename: fileobject} files to multipart upload. 207 :param data: the body to attach to the request. If a dictionary or 208 list of tuples ``[(key, value)]`` is provided, form-encoding will 209 take place. 210 :param json: json for the body to attach to the request (if files or data is not specified). 211 :param params: URL parameters to append to the URL. If a dictionary or 212 list of tuples ``[(key, value)]`` is provided, form-encoding will 213 take place. 214 :param auth: Auth handler or (user, pass) tuple. 215 :param cookies: dictionary or CookieJar of cookies to attach to this request. 216 :param hooks: dictionary of callback hooks, for internal usage. 217 218 Usage:: 219 220 >>> import requests 221 >>> req = requests.Request('GET', 'https://httpbin.org/get') 222 >>> req.prepare() 223 <PreparedRequest [GET]> 224 """ 225 226 def __init__(self, 227 method=None, url=None, headers=None, files=None, data=None, 228 params=None, auth=None, cookies=None, hooks=None, json=None): 229 230 # Default empty dicts for dict params. 231 data = [] if data is None else data 232 files = [] if files is None else files 233 headers = {} if headers is None else headers 234 params = {} if params is None else params 235 hooks = {} if hooks is None else hooks 236 237 self.hooks = default_hooks() 238 for (k, v) in list(hooks.items()): 239 self.register_hook(event=k, hook=v) 240 241 self.method = method 242 self.url = url 243 self.headers = headers 244 self.files = files 245 self.data = data 246 self.json = json 247 self.params = params 248 self.auth = auth 249 self.cookies = cookies 250 251 def __repr__(self): 252 return '<Request [%s]>' % (self.method) 253 254 def prepare(self): 255 """Constructs a :class:`PreparedRequest <PreparedRequest>` for transmission and returns it.""" 256 p = PreparedRequest() 257 p.prepare( 258 method=self.method, 259 url=self.url, 260 headers=self.headers, 261 files=self.files, 262 data=self.data, 263 json=self.json, 264 params=self.params, 265 auth=self.auth, 266 cookies=self.cookies, 267 hooks=self.hooks, 268 ) 269 return p 270 271 272class PreparedRequest(RequestEncodingMixin, RequestHooksMixin): 273 """The fully mutable :class:`PreparedRequest <PreparedRequest>` object, 274 containing the exact bytes that will be sent to the server. 275 276 Instances are generated from a :class:`Request <Request>` object, and 277 should not be instantiated manually; doing so may produce undesirable 278 effects. 279 280 Usage:: 281 282 >>> import requests 283 >>> req = requests.Request('GET', 'https://httpbin.org/get') 284 >>> r = req.prepare() 285 >>> r 286 <PreparedRequest [GET]> 287 288 >>> s = requests.Session() 289 >>> s.send(r) 290 <Response [200]> 291 """ 292 293 def __init__(self): 294 #: HTTP verb to send to the server. 295 self.method = None 296 #: HTTP URL to send the request to. 297 self.url = None 298 #: dictionary of HTTP headers. 299 self.headers = None 300 # The `CookieJar` used to create the Cookie header will be stored here 301 # after prepare_cookies is called 302 self._cookies = None 303 #: request body to send to the server. 304 self.body = None 305 #: dictionary of callback hooks, for internal usage. 306 self.hooks = default_hooks() 307 #: integer denoting starting position of a readable file-like body. 308 self._body_position = None 309 310 def prepare(self, 311 method=None, url=None, headers=None, files=None, data=None, 312 params=None, auth=None, cookies=None, hooks=None, json=None): 313 """Prepares the entire request with the given parameters.""" 314 315 self.prepare_method(method) 316 self.prepare_url(url, params) 317 self.prepare_headers(headers) 318 self.prepare_cookies(cookies) 319 self.prepare_body(data, files, json) 320 self.prepare_auth(auth, url) 321 322 # Note that prepare_auth must be last to enable authentication schemes 323 # such as OAuth to work on a fully prepared request. 324 325 # This MUST go after prepare_auth. Authenticators could add a hook 326 self.prepare_hooks(hooks) 327 328 def __repr__(self): 329 return '<PreparedRequest [%s]>' % (self.method) 330 331 def copy(self): 332 p = PreparedRequest() 333 p.method = self.method 334 p.url = self.url 335 p.headers = self.headers.copy() if self.headers is not None else None 336 p._cookies = _copy_cookie_jar(self._cookies) 337 p.body = self.body 338 p.hooks = self.hooks 339 p._body_position = self._body_position 340 return p 341 342 def prepare_method(self, method): 343 """Prepares the given HTTP method.""" 344 self.method = method 345 if self.method is not None: 346 self.method = to_native_string(self.method.upper()) 347 348 @staticmethod 349 def _get_idna_encoded_host(host): 350 import idna 351 352 try: 353 host = idna.encode(host, uts46=True).decode('utf-8') 354 except idna.IDNAError: 355 raise UnicodeError 356 return host 357 358 def prepare_url(self, url, params): 359 """Prepares the given HTTP URL.""" 360 #: Accept objects that have string representations. 361 #: We're unable to blindly call unicode/str functions 362 #: as this will include the bytestring indicator (b'') 363 #: on python 3.x. 364 #: https://github.com/psf/requests/pull/2238 365 if isinstance(url, bytes): 366 url = url.decode('utf8') 367 else: 368 url = unicode(url) if is_py2 else str(url) 369 370 # Remove leading whitespaces from url 371 url = url.lstrip() 372 373 # Don't do any URL preparation for non-HTTP schemes like `mailto`, 374 # `data` etc to work around exceptions from `url_parse`, which 375 # handles RFC 3986 only. 376 if ':' in url and not url.lower().startswith('http'): 377 self.url = url 378 return 379 380 # Support for unicode domain names and paths. 381 try: 382 scheme, auth, host, port, path, query, fragment = parse_url(url) 383 except LocationParseError as e: 384 raise InvalidURL(*e.args) 385 386 if not scheme: 387 error = ("Invalid URL {0!r}: No schema supplied. Perhaps you meant http://{0}?") 388 error = error.format(to_native_string(url, 'utf8')) 389 390 raise MissingSchema(error) 391 392 if not host: 393 raise InvalidURL("Invalid URL %r: No host supplied" % url) 394 395 # In general, we want to try IDNA encoding the hostname if the string contains 396 # non-ASCII characters. This allows users to automatically get the correct IDNA 397 # behaviour. For strings containing only ASCII characters, we need to also verify 398 # it doesn't start with a wildcard (*), before allowing the unencoded hostname. 399 if not unicode_is_ascii(host): 400 try: 401 host = self._get_idna_encoded_host(host) 402 except UnicodeError: 403 raise InvalidURL('URL has an invalid label.') 404 elif host.startswith(u'*'): 405 raise InvalidURL('URL has an invalid label.') 406 407 # Carefully reconstruct the network location 408 netloc = auth or '' 409 if netloc: 410 netloc += '@' 411 netloc += host 412 if port: 413 netloc += ':' + str(port) 414 415 # Bare domains aren't valid URLs. 416 if not path: 417 path = '/' 418 419 if is_py2: 420 if isinstance(scheme, str): 421 scheme = scheme.encode('utf-8') 422 if isinstance(netloc, str): 423 netloc = netloc.encode('utf-8') 424 if isinstance(path, str): 425 path = path.encode('utf-8') 426 if isinstance(query, str): 427 query = query.encode('utf-8') 428 if isinstance(fragment, str): 429 fragment = fragment.encode('utf-8') 430 431 if isinstance(params, (str, bytes)): 432 params = to_native_string(params) 433 434 enc_params = self._encode_params(params) 435 if enc_params: 436 if query: 437 query = '%s&%s' % (query, enc_params) 438 else: 439 query = enc_params 440 441 url = requote_uri(urlunparse([scheme, netloc, path, None, query, fragment])) 442 self.url = url 443 444 def prepare_headers(self, headers): 445 """Prepares the given HTTP headers.""" 446 447 self.headers = CaseInsensitiveDict() 448 if headers: 449 for header in headers.items(): 450 # Raise exception on invalid header value. 451 check_header_validity(header) 452 name, value = header 453 self.headers[to_native_string(name)] = value 454 455 def prepare_body(self, data, files, json=None): 456 """Prepares the given HTTP body data.""" 457 458 # Check if file, fo, generator, iterator. 459 # If not, run through normal process. 460 461 # Nottin' on you. 462 body = None 463 content_type = None 464 465 if not data and json is not None: 466 # urllib3 requires a bytes-like body. Python 2's json.dumps 467 # provides this natively, but Python 3 gives a Unicode string. 468 content_type = 'application/json' 469 body = complexjson.dumps(json) 470 if not isinstance(body, bytes): 471 body = body.encode('utf-8') 472 473 is_stream = all([ 474 hasattr(data, '__iter__'), 475 not isinstance(data, (basestring, list, tuple, Mapping)) 476 ]) 477 478 if is_stream: 479 try: 480 length = super_len(data) 481 except (TypeError, AttributeError, UnsupportedOperation): 482 length = None 483 484 body = data 485 486 if getattr(body, 'tell', None) is not None: 487 # Record the current file position before reading. 488 # This will allow us to rewind a file in the event 489 # of a redirect. 490 try: 491 self._body_position = body.tell() 492 except (IOError, OSError): 493 # This differentiates from None, allowing us to catch 494 # a failed `tell()` later when trying to rewind the body 495 self._body_position = object() 496 497 if files: 498 raise NotImplementedError('Streamed bodies and files are mutually exclusive.') 499 500 if length: 501 self.headers['Content-Length'] = builtin_str(length) 502 else: 503 self.headers['Transfer-Encoding'] = 'chunked' 504 else: 505 # Multi-part file uploads. 506 if files: 507 (body, content_type) = self._encode_files(files, data) 508 else: 509 if data: 510 body = self._encode_params(data) 511 if isinstance(data, basestring) or hasattr(data, 'read'): 512 content_type = None 513 else: 514 content_type = 'application/x-www-form-urlencoded' 515 516 self.prepare_content_length(body) 517 518 # Add content-type if it wasn't explicitly provided. 519 if content_type and ('content-type' not in self.headers): 520 self.headers['Content-Type'] = content_type 521 522 self.body = body 523 524 def prepare_content_length(self, body): 525 """Prepare Content-Length header based on request method and body""" 526 if body is not None: 527 length = super_len(body) 528 if length: 529 # If length exists, set it. Otherwise, we fallback 530 # to Transfer-Encoding: chunked. 531 self.headers['Content-Length'] = builtin_str(length) 532 elif self.method not in ('GET', 'HEAD') and self.headers.get('Content-Length') is None: 533 # Set Content-Length to 0 for methods that can have a body 534 # but don't provide one. (i.e. not GET or HEAD) 535 self.headers['Content-Length'] = '0' 536 537 def prepare_auth(self, auth, url=''): 538 """Prepares the given HTTP auth data.""" 539 540 # If no Auth is explicitly provided, extract it from the URL first. 541 if auth is None: 542 url_auth = get_auth_from_url(self.url) 543 auth = url_auth if any(url_auth) else None 544 545 if auth: 546 if isinstance(auth, tuple) and len(auth) == 2: 547 # special-case basic HTTP auth 548 auth = HTTPBasicAuth(*auth) 549 550 # Allow auth to make its changes. 551 r = auth(self) 552 553 # Update self to reflect the auth changes. 554 self.__dict__.update(r.__dict__) 555 556 # Recompute Content-Length 557 self.prepare_content_length(self.body) 558 559 def prepare_cookies(self, cookies): 560 """Prepares the given HTTP cookie data. 561 562 This function eventually generates a ``Cookie`` header from the 563 given cookies using cookielib. Due to cookielib's design, the header 564 will not be regenerated if it already exists, meaning this function 565 can only be called once for the life of the 566 :class:`PreparedRequest <PreparedRequest>` object. Any subsequent calls 567 to ``prepare_cookies`` will have no actual effect, unless the "Cookie" 568 header is removed beforehand. 569 """ 570 if isinstance(cookies, cookielib.CookieJar): 571 self._cookies = cookies 572 else: 573 self._cookies = cookiejar_from_dict(cookies) 574 575 cookie_header = get_cookie_header(self._cookies, self) 576 if cookie_header is not None: 577 self.headers['Cookie'] = cookie_header 578 579 def prepare_hooks(self, hooks): 580 """Prepares the given hooks.""" 581 # hooks can be passed as None to the prepare method and to this 582 # method. To prevent iterating over None, simply use an empty list 583 # if hooks is False-y 584 hooks = hooks or [] 585 for event in hooks: 586 self.register_hook(event, hooks[event]) 587 588 589class Response(object): 590 """The :class:`Response <Response>` object, which contains a 591 server's response to an HTTP request. 592 """ 593 594 __attrs__ = [ 595 '_content', 'status_code', 'headers', 'url', 'history', 596 'encoding', 'reason', 'cookies', 'elapsed', 'request' 597 ] 598 599 def __init__(self): 600 self._content = False 601 self._content_consumed = False 602 self._next = None 603 604 #: Integer Code of responded HTTP Status, e.g. 404 or 200. 605 self.status_code = None 606 607 #: Case-insensitive Dictionary of Response Headers. 608 #: For example, ``headers['content-encoding']`` will return the 609 #: value of a ``'Content-Encoding'`` response header. 610 self.headers = CaseInsensitiveDict() 611 612 #: File-like object representation of response (for advanced usage). 613 #: Use of ``raw`` requires that ``stream=True`` be set on the request. 614 #: This requirement does not apply for use internally to Requests. 615 self.raw = None 616 617 #: Final URL location of Response. 618 self.url = None 619 620 #: Encoding to decode with when accessing r.text. 621 self.encoding = None 622 623 #: A list of :class:`Response <Response>` objects from 624 #: the history of the Request. Any redirect responses will end 625 #: up here. The list is sorted from the oldest to the most recent request. 626 self.history = [] 627 628 #: Textual reason of responded HTTP Status, e.g. "Not Found" or "OK". 629 self.reason = None 630 631 #: A CookieJar of Cookies the server sent back. 632 self.cookies = cookiejar_from_dict({}) 633 634 #: The amount of time elapsed between sending the request 635 #: and the arrival of the response (as a timedelta). 636 #: This property specifically measures the time taken between sending 637 #: the first byte of the request and finishing parsing the headers. It 638 #: is therefore unaffected by consuming the response content or the 639 #: value of the ``stream`` keyword argument. 640 self.elapsed = datetime.timedelta(0) 641 642 #: The :class:`PreparedRequest <PreparedRequest>` object to which this 643 #: is a response. 644 self.request = None 645 646 def __enter__(self): 647 return self 648 649 def __exit__(self, *args): 650 self.close() 651 652 def __getstate__(self): 653 # Consume everything; accessing the content attribute makes 654 # sure the content has been fully read. 655 if not self._content_consumed: 656 self.content 657 658 return {attr: getattr(self, attr, None) for attr in self.__attrs__} 659 660 def __setstate__(self, state): 661 for name, value in state.items(): 662 setattr(self, name, value) 663 664 # pickled objects do not have .raw 665 setattr(self, '_content_consumed', True) 666 setattr(self, 'raw', None) 667 668 def __repr__(self): 669 return '<Response [%s]>' % (self.status_code) 670 671 def __bool__(self): 672 """Returns True if :attr:`status_code` is less than 400. 673 674 This attribute checks if the status code of the response is between 675 400 and 600 to see if there was a client error or a server error. If 676 the status code, is between 200 and 400, this will return True. This 677 is **not** a check to see if the response code is ``200 OK``. 678 """ 679 return self.ok 680 681 def __nonzero__(self): 682 """Returns True if :attr:`status_code` is less than 400. 683 684 This attribute checks if the status code of the response is between 685 400 and 600 to see if there was a client error or a server error. If 686 the status code, is between 200 and 400, this will return True. This 687 is **not** a check to see if the response code is ``200 OK``. 688 """ 689 return self.ok 690 691 def __iter__(self): 692 """Allows you to use a response as an iterator.""" 693 return self.iter_content(128) 694 695 @property 696 def ok(self): 697 """Returns True if :attr:`status_code` is less than 400, False if not. 698 699 This attribute checks if the status code of the response is between 700 400 and 600 to see if there was a client error or a server error. If 701 the status code is between 200 and 400, this will return True. This 702 is **not** a check to see if the response code is ``200 OK``. 703 """ 704 try: 705 self.raise_for_status() 706 except HTTPError: 707 return False 708 return True 709 710 @property 711 def is_redirect(self): 712 """True if this Response is a well-formed HTTP redirect that could have 713 been processed automatically (by :meth:`Session.resolve_redirects`). 714 """ 715 return ('location' in self.headers and self.status_code in REDIRECT_STATI) 716 717 @property 718 def is_permanent_redirect(self): 719 """True if this Response one of the permanent versions of redirect.""" 720 return ('location' in self.headers and self.status_code in (codes.moved_permanently, codes.permanent_redirect)) 721 722 @property 723 def next(self): 724 """Returns a PreparedRequest for the next request in a redirect chain, if there is one.""" 725 return self._next 726 727 @property 728 def apparent_encoding(self): 729 """The apparent encoding, provided by the chardet library.""" 730 return chardet.detect(self.content)['encoding'] 731 732 def iter_content(self, chunk_size=1, decode_unicode=False): 733 """Iterates over the response data. When stream=True is set on the 734 request, this avoids reading the content at once into memory for 735 large responses. The chunk size is the number of bytes it should 736 read into memory. This is not necessarily the length of each item 737 returned as decoding can take place. 738 739 chunk_size must be of type int or None. A value of None will 740 function differently depending on the value of `stream`. 741 stream=True will read data as it arrives in whatever size the 742 chunks are received. If stream=False, data is returned as 743 a single chunk. 744 745 If decode_unicode is True, content will be decoded using the best 746 available encoding based on the response. 747 """ 748 749 def generate(): 750 # Special case for urllib3. 751 if hasattr(self.raw, 'stream'): 752 try: 753 for chunk in self.raw.stream(chunk_size, decode_content=True): 754 yield chunk 755 except ProtocolError as e: 756 raise ChunkedEncodingError(e) 757 except DecodeError as e: 758 raise ContentDecodingError(e) 759 except ReadTimeoutError as e: 760 raise ConnectionError(e) 761 else: 762 # Standard file-like object. 763 while True: 764 chunk = self.raw.read(chunk_size) 765 if not chunk: 766 break 767 yield chunk 768 769 self._content_consumed = True 770 771 if self._content_consumed and isinstance(self._content, bool): 772 raise StreamConsumedError() 773 elif chunk_size is not None and not isinstance(chunk_size, int): 774 raise TypeError("chunk_size must be an int, it is instead a %s." % type(chunk_size)) 775 # simulate reading small chunks of the content 776 reused_chunks = iter_slices(self._content, chunk_size) 777 778 stream_chunks = generate() 779 780 chunks = reused_chunks if self._content_consumed else stream_chunks 781 782 if decode_unicode: 783 chunks = stream_decode_response_unicode(chunks, self) 784 785 return chunks 786 787 def iter_lines(self, chunk_size=ITER_CHUNK_SIZE, decode_unicode=False, delimiter=None): 788 """Iterates over the response data, one line at a time. When 789 stream=True is set on the request, this avoids reading the 790 content at once into memory for large responses. 791 792 .. note:: This method is not reentrant safe. 793 """ 794 795 pending = None 796 797 for chunk in self.iter_content(chunk_size=chunk_size, decode_unicode=decode_unicode): 798 799 if pending is not None: 800 chunk = pending + chunk 801 802 if delimiter: 803 lines = chunk.split(delimiter) 804 else: 805 lines = chunk.splitlines() 806 807 if lines and lines[-1] and chunk and lines[-1][-1] == chunk[-1]: 808 pending = lines.pop() 809 else: 810 pending = None 811 812 for line in lines: 813 yield line 814 815 if pending is not None: 816 yield pending 817 818 @property 819 def content(self): 820 """Content of the response, in bytes.""" 821 822 if self._content is False: 823 # Read the contents. 824 if self._content_consumed: 825 raise RuntimeError( 826 'The content for this response was already consumed') 827 828 if self.status_code == 0 or self.raw is None: 829 self._content = None 830 else: 831 self._content = b''.join(self.iter_content(CONTENT_CHUNK_SIZE)) or b'' 832 833 self._content_consumed = True 834 # don't need to release the connection; that's been handled by urllib3 835 # since we exhausted the data. 836 return self._content 837 838 @property 839 def text(self): 840 """Content of the response, in unicode. 841 842 If Response.encoding is None, encoding will be guessed using 843 ``chardet``. 844 845 The encoding of the response content is determined based solely on HTTP 846 headers, following RFC 2616 to the letter. If you can take advantage of 847 non-HTTP knowledge to make a better guess at the encoding, you should 848 set ``r.encoding`` appropriately before accessing this property. 849 """ 850 851 # Try charset from content-type 852 content = None 853 encoding = self.encoding 854 855 if not self.content: 856 return str('') 857 858 # Fallback to auto-detected encoding. 859 if self.encoding is None: 860 encoding = self.apparent_encoding 861 862 # Decode unicode from given encoding. 863 try: 864 content = str(self.content, encoding, errors='replace') 865 except (LookupError, TypeError): 866 # A LookupError is raised if the encoding was not found which could 867 # indicate a misspelling or similar mistake. 868 # 869 # A TypeError can be raised if encoding is None 870 # 871 # So we try blindly encoding. 872 content = str(self.content, errors='replace') 873 874 return content 875 876 def json(self, **kwargs): 877 r"""Returns the json-encoded content of a response, if any. 878 879 :param \*\*kwargs: Optional arguments that ``json.loads`` takes. 880 :raises ValueError: If the response body does not contain valid json. 881 """ 882 883 if not self.encoding and self.content and len(self.content) > 3: 884 # No encoding set. JSON RFC 4627 section 3 states we should expect 885 # UTF-8, -16 or -32. Detect which one to use; If the detection or 886 # decoding fails, fall back to `self.text` (using chardet to make 887 # a best guess). 888 encoding = guess_json_utf(self.content) 889 if encoding is not None: 890 try: 891 return complexjson.loads( 892 self.content.decode(encoding), **kwargs 893 ) 894 except UnicodeDecodeError: 895 # Wrong UTF codec detected; usually because it's not UTF-8 896 # but some other 8-bit codec. This is an RFC violation, 897 # and the server didn't bother to tell us what codec *was* 898 # used. 899 pass 900 return complexjson.loads(self.text, **kwargs) 901 902 @property 903 def links(self): 904 """Returns the parsed header links of the response, if any.""" 905 906 header = self.headers.get('link') 907 908 # l = MultiDict() 909 l = {} 910 911 if header: 912 links = parse_header_links(header) 913 914 for link in links: 915 key = link.get('rel') or link.get('url') 916 l[key] = link 917 918 return l 919 920 def raise_for_status(self): 921 """Raises :class:`HTTPError`, if one occurred.""" 922 923 http_error_msg = '' 924 if isinstance(self.reason, bytes): 925 # We attempt to decode utf-8 first because some servers 926 # choose to localize their reason strings. If the string 927 # isn't utf-8, we fall back to iso-8859-1 for all other 928 # encodings. (See PR #3538) 929 try: 930 reason = self.reason.decode('utf-8') 931 except UnicodeDecodeError: 932 reason = self.reason.decode('iso-8859-1') 933 else: 934 reason = self.reason 935 936 if 400 <= self.status_code < 500: 937 http_error_msg = u'%s Client Error: %s for url: %s' % (self.status_code, reason, self.url) 938 939 elif 500 <= self.status_code < 600: 940 http_error_msg = u'%s Server Error: %s for url: %s' % (self.status_code, reason, self.url) 941 942 if http_error_msg: 943 raise HTTPError(http_error_msg, response=self) 944 945 def close(self): 946 """Releases the connection back to the pool. Once this method has been 947 called the underlying ``raw`` object must not be accessed again. 948 949 *Note: Should not normally need to be called explicitly.* 950 """ 951 if not self._content_consumed: 952 self.raw.close() 953 954 release_conn = getattr(self.raw, 'release_conn', None) 955 if release_conn is not None: 956 release_conn() 957