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 Generated from either a :class:`Request <Request>` object or manually. 277 278 Usage:: 279 280 >>> import requests 281 >>> req = requests.Request('GET', 'https://httpbin.org/get') 282 >>> r = req.prepare() 283 >>> r 284 <PreparedRequest [GET]> 285 286 >>> s = requests.Session() 287 >>> s.send(r) 288 <Response [200]> 289 """ 290 291 def __init__(self): 292 #: HTTP verb to send to the server. 293 self.method = None 294 #: HTTP URL to send the request to. 295 self.url = None 296 #: dictionary of HTTP headers. 297 self.headers = None 298 # The `CookieJar` used to create the Cookie header will be stored here 299 # after prepare_cookies is called 300 self._cookies = None 301 #: request body to send to the server. 302 self.body = None 303 #: dictionary of callback hooks, for internal usage. 304 self.hooks = default_hooks() 305 #: integer denoting starting position of a readable file-like body. 306 self._body_position = None 307 308 def prepare(self, 309 method=None, url=None, headers=None, files=None, data=None, 310 params=None, auth=None, cookies=None, hooks=None, json=None): 311 """Prepares the entire request with the given parameters.""" 312 313 self.prepare_method(method) 314 self.prepare_url(url, params) 315 self.prepare_headers(headers) 316 self.prepare_cookies(cookies) 317 self.prepare_body(data, files, json) 318 self.prepare_auth(auth, url) 319 320 # Note that prepare_auth must be last to enable authentication schemes 321 # such as OAuth to work on a fully prepared request. 322 323 # This MUST go after prepare_auth. Authenticators could add a hook 324 self.prepare_hooks(hooks) 325 326 def __repr__(self): 327 return '<PreparedRequest [%s]>' % (self.method) 328 329 def copy(self): 330 p = PreparedRequest() 331 p.method = self.method 332 p.url = self.url 333 p.headers = self.headers.copy() if self.headers is not None else None 334 p._cookies = _copy_cookie_jar(self._cookies) 335 p.body = self.body 336 p.hooks = self.hooks 337 p._body_position = self._body_position 338 return p 339 340 def prepare_method(self, method): 341 """Prepares the given HTTP method.""" 342 self.method = method 343 if self.method is not None: 344 self.method = to_native_string(self.method.upper()) 345 346 @staticmethod 347 def _get_idna_encoded_host(host): 348 import idna 349 350 try: 351 host = idna.encode(host, uts46=True).decode('utf-8') 352 except idna.IDNAError: 353 raise UnicodeError 354 return host 355 356 def prepare_url(self, url, params): 357 """Prepares the given HTTP URL.""" 358 #: Accept objects that have string representations. 359 #: We're unable to blindly call unicode/str functions 360 #: as this will include the bytestring indicator (b'') 361 #: on python 3.x. 362 #: https://github.com/psf/requests/pull/2238 363 if isinstance(url, bytes): 364 url = url.decode('utf8') 365 else: 366 url = unicode(url) if is_py2 else str(url) 367 368 # Remove leading whitespaces from url 369 url = url.lstrip() 370 371 # Don't do any URL preparation for non-HTTP schemes like `mailto`, 372 # `data` etc to work around exceptions from `url_parse`, which 373 # handles RFC 3986 only. 374 if ':' in url and not url.lower().startswith('http'): 375 self.url = url 376 return 377 378 # Support for unicode domain names and paths. 379 try: 380 scheme, auth, host, port, path, query, fragment = parse_url(url) 381 except LocationParseError as e: 382 raise InvalidURL(*e.args) 383 384 if not scheme: 385 error = ("Invalid URL {0!r}: No schema supplied. Perhaps you meant http://{0}?") 386 error = error.format(to_native_string(url, 'utf8')) 387 388 raise MissingSchema(error) 389 390 if not host: 391 raise InvalidURL("Invalid URL %r: No host supplied" % url) 392 393 # In general, we want to try IDNA encoding the hostname if the string contains 394 # non-ASCII characters. This allows users to automatically get the correct IDNA 395 # behaviour. For strings containing only ASCII characters, we need to also verify 396 # it doesn't start with a wildcard (*), before allowing the unencoded hostname. 397 if not unicode_is_ascii(host): 398 try: 399 host = self._get_idna_encoded_host(host) 400 except UnicodeError: 401 raise InvalidURL('URL has an invalid label.') 402 elif host.startswith(u'*'): 403 raise InvalidURL('URL has an invalid label.') 404 405 # Carefully reconstruct the network location 406 netloc = auth or '' 407 if netloc: 408 netloc += '@' 409 netloc += host 410 if port: 411 netloc += ':' + str(port) 412 413 # Bare domains aren't valid URLs. 414 if not path: 415 path = '/' 416 417 if is_py2: 418 if isinstance(scheme, str): 419 scheme = scheme.encode('utf-8') 420 if isinstance(netloc, str): 421 netloc = netloc.encode('utf-8') 422 if isinstance(path, str): 423 path = path.encode('utf-8') 424 if isinstance(query, str): 425 query = query.encode('utf-8') 426 if isinstance(fragment, str): 427 fragment = fragment.encode('utf-8') 428 429 if isinstance(params, (str, bytes)): 430 params = to_native_string(params) 431 432 enc_params = self._encode_params(params) 433 if enc_params: 434 if query: 435 query = '%s&%s' % (query, enc_params) 436 else: 437 query = enc_params 438 439 url = requote_uri(urlunparse([scheme, netloc, path, None, query, fragment])) 440 self.url = url 441 442 def prepare_headers(self, headers): 443 """Prepares the given HTTP headers.""" 444 445 self.headers = CaseInsensitiveDict() 446 if headers: 447 for header in headers.items(): 448 # Raise exception on invalid header value. 449 check_header_validity(header) 450 name, value = header 451 self.headers[to_native_string(name)] = value 452 453 def prepare_body(self, data, files, json=None): 454 """Prepares the given HTTP body data.""" 455 456 # Check if file, fo, generator, iterator. 457 # If not, run through normal process. 458 459 # Nottin' on you. 460 body = None 461 content_type = None 462 463 if not data and json is not None: 464 # urllib3 requires a bytes-like body. Python 2's json.dumps 465 # provides this natively, but Python 3 gives a Unicode string. 466 content_type = 'application/json' 467 body = complexjson.dumps(json) 468 if not isinstance(body, bytes): 469 body = body.encode('utf-8') 470 471 is_stream = all([ 472 hasattr(data, '__iter__'), 473 not isinstance(data, (basestring, list, tuple, Mapping)) 474 ]) 475 476 try: 477 length = super_len(data) 478 except (TypeError, AttributeError, UnsupportedOperation): 479 length = None 480 481 if is_stream: 482 body = data 483 484 if getattr(body, 'tell', None) is not None: 485 # Record the current file position before reading. 486 # This will allow us to rewind a file in the event 487 # of a redirect. 488 try: 489 self._body_position = body.tell() 490 except (IOError, OSError): 491 # This differentiates from None, allowing us to catch 492 # a failed `tell()` later when trying to rewind the body 493 self._body_position = object() 494 495 if files: 496 raise NotImplementedError('Streamed bodies and files are mutually exclusive.') 497 498 if length: 499 self.headers['Content-Length'] = builtin_str(length) 500 else: 501 self.headers['Transfer-Encoding'] = 'chunked' 502 else: 503 # Multi-part file uploads. 504 if files: 505 (body, content_type) = self._encode_files(files, data) 506 else: 507 if data: 508 body = self._encode_params(data) 509 if isinstance(data, basestring) or hasattr(data, 'read'): 510 content_type = None 511 else: 512 content_type = 'application/x-www-form-urlencoded' 513 514 self.prepare_content_length(body) 515 516 # Add content-type if it wasn't explicitly provided. 517 if content_type and ('content-type' not in self.headers): 518 self.headers['Content-Type'] = content_type 519 520 self.body = body 521 522 def prepare_content_length(self, body): 523 """Prepare Content-Length header based on request method and body""" 524 if body is not None: 525 length = super_len(body) 526 if length: 527 # If length exists, set it. Otherwise, we fallback 528 # to Transfer-Encoding: chunked. 529 self.headers['Content-Length'] = builtin_str(length) 530 elif self.method not in ('GET', 'HEAD') and self.headers.get('Content-Length') is None: 531 # Set Content-Length to 0 for methods that can have a body 532 # but don't provide one. (i.e. not GET or HEAD) 533 self.headers['Content-Length'] = '0' 534 535 def prepare_auth(self, auth, url=''): 536 """Prepares the given HTTP auth data.""" 537 538 # If no Auth is explicitly provided, extract it from the URL first. 539 if auth is None: 540 url_auth = get_auth_from_url(self.url) 541 auth = url_auth if any(url_auth) else None 542 543 if auth: 544 if isinstance(auth, tuple) and len(auth) == 2: 545 # special-case basic HTTP auth 546 auth = HTTPBasicAuth(*auth) 547 548 # Allow auth to make its changes. 549 r = auth(self) 550 551 # Update self to reflect the auth changes. 552 self.__dict__.update(r.__dict__) 553 554 # Recompute Content-Length 555 self.prepare_content_length(self.body) 556 557 def prepare_cookies(self, cookies): 558 """Prepares the given HTTP cookie data. 559 560 This function eventually generates a ``Cookie`` header from the 561 given cookies using cookielib. Due to cookielib's design, the header 562 will not be regenerated if it already exists, meaning this function 563 can only be called once for the life of the 564 :class:`PreparedRequest <PreparedRequest>` object. Any subsequent calls 565 to ``prepare_cookies`` will have no actual effect, unless the "Cookie" 566 header is removed beforehand. 567 """ 568 if isinstance(cookies, cookielib.CookieJar): 569 self._cookies = cookies 570 else: 571 self._cookies = cookiejar_from_dict(cookies) 572 573 cookie_header = get_cookie_header(self._cookies, self) 574 if cookie_header is not None: 575 self.headers['Cookie'] = cookie_header 576 577 def prepare_hooks(self, hooks): 578 """Prepares the given hooks.""" 579 # hooks can be passed as None to the prepare method and to this 580 # method. To prevent iterating over None, simply use an empty list 581 # if hooks is False-y 582 hooks = hooks or [] 583 for event in hooks: 584 self.register_hook(event, hooks[event]) 585 586 587class Response(object): 588 """The :class:`Response <Response>` object, which contains a 589 server's response to an HTTP request. 590 """ 591 592 __attrs__ = [ 593 '_content', 'status_code', 'headers', 'url', 'history', 594 'encoding', 'reason', 'cookies', 'elapsed', 'request' 595 ] 596 597 def __init__(self): 598 self._content = False 599 self._content_consumed = False 600 self._next = None 601 602 #: Integer Code of responded HTTP Status, e.g. 404 or 200. 603 self.status_code = None 604 605 #: Case-insensitive Dictionary of Response Headers. 606 #: For example, ``headers['content-encoding']`` will return the 607 #: value of a ``'Content-Encoding'`` response header. 608 self.headers = CaseInsensitiveDict() 609 610 #: File-like object representation of response (for advanced usage). 611 #: Use of ``raw`` requires that ``stream=True`` be set on the request. 612 #: This requirement does not apply for use internally to Requests. 613 self.raw = None 614 615 #: Final URL location of Response. 616 self.url = None 617 618 #: Encoding to decode with when accessing r.text. 619 self.encoding = None 620 621 #: A list of :class:`Response <Response>` objects from 622 #: the history of the Request. Any redirect responses will end 623 #: up here. The list is sorted from the oldest to the most recent request. 624 self.history = [] 625 626 #: Textual reason of responded HTTP Status, e.g. "Not Found" or "OK". 627 self.reason = None 628 629 #: A CookieJar of Cookies the server sent back. 630 self.cookies = cookiejar_from_dict({}) 631 632 #: The amount of time elapsed between sending the request 633 #: and the arrival of the response (as a timedelta). 634 #: This property specifically measures the time taken between sending 635 #: the first byte of the request and finishing parsing the headers. It 636 #: is therefore unaffected by consuming the response content or the 637 #: value of the ``stream`` keyword argument. 638 self.elapsed = datetime.timedelta(0) 639 640 #: The :class:`PreparedRequest <PreparedRequest>` object to which this 641 #: is a response. 642 self.request = None 643 644 def __enter__(self): 645 return self 646 647 def __exit__(self, *args): 648 self.close() 649 650 def __getstate__(self): 651 # Consume everything; accessing the content attribute makes 652 # sure the content has been fully read. 653 if not self._content_consumed: 654 self.content 655 656 return {attr: getattr(self, attr, None) for attr in self.__attrs__} 657 658 def __setstate__(self, state): 659 for name, value in state.items(): 660 setattr(self, name, value) 661 662 # pickled objects do not have .raw 663 setattr(self, '_content_consumed', True) 664 setattr(self, 'raw', None) 665 666 def __repr__(self): 667 return '<Response [%s]>' % (self.status_code) 668 669 def __bool__(self): 670 """Returns True if :attr:`status_code` is less than 400. 671 672 This attribute checks if the status code of the response is between 673 400 and 600 to see if there was a client error or a server error. If 674 the status code, is between 200 and 400, this will return True. This 675 is **not** a check to see if the response code is ``200 OK``. 676 """ 677 return self.ok 678 679 def __nonzero__(self): 680 """Returns True if :attr:`status_code` is less than 400. 681 682 This attribute checks if the status code of the response is between 683 400 and 600 to see if there was a client error or a server error. If 684 the status code, is between 200 and 400, this will return True. This 685 is **not** a check to see if the response code is ``200 OK``. 686 """ 687 return self.ok 688 689 def __iter__(self): 690 """Allows you to use a response as an iterator.""" 691 return self.iter_content(128) 692 693 @property 694 def ok(self): 695 """Returns True if :attr:`status_code` is less than 400, False if not. 696 697 This attribute checks if the status code of the response is between 698 400 and 600 to see if there was a client error or a server error. If 699 the status code is between 200 and 400, this will return True. This 700 is **not** a check to see if the response code is ``200 OK``. 701 """ 702 try: 703 self.raise_for_status() 704 except HTTPError: 705 return False 706 return True 707 708 @property 709 def is_redirect(self): 710 """True if this Response is a well-formed HTTP redirect that could have 711 been processed automatically (by :meth:`Session.resolve_redirects`). 712 """ 713 return ('location' in self.headers and self.status_code in REDIRECT_STATI) 714 715 @property 716 def is_permanent_redirect(self): 717 """True if this Response one of the permanent versions of redirect.""" 718 return ('location' in self.headers and self.status_code in (codes.moved_permanently, codes.permanent_redirect)) 719 720 @property 721 def next(self): 722 """Returns a PreparedRequest for the next request in a redirect chain, if there is one.""" 723 return self._next 724 725 @property 726 def apparent_encoding(self): 727 """The apparent encoding, provided by the chardet library.""" 728 return chardet.detect(self.content)['encoding'] 729 730 def iter_content(self, chunk_size=1, decode_unicode=False): 731 """Iterates over the response data. When stream=True is set on the 732 request, this avoids reading the content at once into memory for 733 large responses. The chunk size is the number of bytes it should 734 read into memory. This is not necessarily the length of each item 735 returned as decoding can take place. 736 737 chunk_size must be of type int or None. A value of None will 738 function differently depending on the value of `stream`. 739 stream=True will read data as it arrives in whatever size the 740 chunks are received. If stream=False, data is returned as 741 a single chunk. 742 743 If decode_unicode is True, content will be decoded using the best 744 available encoding based on the response. 745 """ 746 747 def generate(): 748 # Special case for urllib3. 749 if hasattr(self.raw, 'stream'): 750 try: 751 for chunk in self.raw.stream(chunk_size, decode_content=True): 752 yield chunk 753 except ProtocolError as e: 754 raise ChunkedEncodingError(e) 755 except DecodeError as e: 756 raise ContentDecodingError(e) 757 except ReadTimeoutError as e: 758 raise ConnectionError(e) 759 else: 760 # Standard file-like object. 761 while True: 762 chunk = self.raw.read(chunk_size) 763 if not chunk: 764 break 765 yield chunk 766 767 self._content_consumed = True 768 769 if self._content_consumed and isinstance(self._content, bool): 770 raise StreamConsumedError() 771 elif chunk_size is not None and not isinstance(chunk_size, int): 772 raise TypeError("chunk_size must be an int, it is instead a %s." % type(chunk_size)) 773 # simulate reading small chunks of the content 774 reused_chunks = iter_slices(self._content, chunk_size) 775 776 stream_chunks = generate() 777 778 chunks = reused_chunks if self._content_consumed else stream_chunks 779 780 if decode_unicode: 781 chunks = stream_decode_response_unicode(chunks, self) 782 783 return chunks 784 785 def iter_lines(self, chunk_size=ITER_CHUNK_SIZE, decode_unicode=False, delimiter=None): 786 """Iterates over the response data, one line at a time. When 787 stream=True is set on the request, this avoids reading the 788 content at once into memory for large responses. 789 790 .. note:: This method is not reentrant safe. 791 """ 792 793 pending = None 794 795 for chunk in self.iter_content(chunk_size=chunk_size, decode_unicode=decode_unicode): 796 797 if pending is not None: 798 chunk = pending + chunk 799 800 if delimiter: 801 lines = chunk.split(delimiter) 802 else: 803 lines = chunk.splitlines() 804 805 if lines and lines[-1] and chunk and lines[-1][-1] == chunk[-1]: 806 pending = lines.pop() 807 else: 808 pending = None 809 810 for line in lines: 811 yield line 812 813 if pending is not None: 814 yield pending 815 816 @property 817 def content(self): 818 """Content of the response, in bytes.""" 819 820 if self._content is False: 821 # Read the contents. 822 if self._content_consumed: 823 raise RuntimeError( 824 'The content for this response was already consumed') 825 826 if self.status_code == 0 or self.raw is None: 827 self._content = None 828 else: 829 self._content = b''.join(self.iter_content(CONTENT_CHUNK_SIZE)) or b'' 830 831 self._content_consumed = True 832 # don't need to release the connection; that's been handled by urllib3 833 # since we exhausted the data. 834 return self._content 835 836 @property 837 def text(self): 838 """Content of the response, in unicode. 839 840 If Response.encoding is None, encoding will be guessed using 841 ``chardet``. 842 843 The encoding of the response content is determined based solely on HTTP 844 headers, following RFC 2616 to the letter. If you can take advantage of 845 non-HTTP knowledge to make a better guess at the encoding, you should 846 set ``r.encoding`` appropriately before accessing this property. 847 """ 848 849 # Try charset from content-type 850 content = None 851 encoding = self.encoding 852 853 if not self.content: 854 return str('') 855 856 # Fallback to auto-detected encoding. 857 if self.encoding is None: 858 encoding = self.apparent_encoding 859 860 # Decode unicode from given encoding. 861 try: 862 content = str(self.content, encoding, errors='replace') 863 except (LookupError, TypeError): 864 # A LookupError is raised if the encoding was not found which could 865 # indicate a misspelling or similar mistake. 866 # 867 # A TypeError can be raised if encoding is None 868 # 869 # So we try blindly encoding. 870 content = str(self.content, errors='replace') 871 872 return content 873 874 def json(self, **kwargs): 875 r"""Returns the json-encoded content of a response, if any. 876 877 :param \*\*kwargs: Optional arguments that ``json.loads`` takes. 878 :raises ValueError: If the response body does not contain valid json. 879 """ 880 881 if not self.encoding and self.content and len(self.content) > 3: 882 # No encoding set. JSON RFC 4627 section 3 states we should expect 883 # UTF-8, -16 or -32. Detect which one to use; If the detection or 884 # decoding fails, fall back to `self.text` (using chardet to make 885 # a best guess). 886 encoding = guess_json_utf(self.content) 887 if encoding is not None: 888 try: 889 return complexjson.loads( 890 self.content.decode(encoding), **kwargs 891 ) 892 except UnicodeDecodeError: 893 # Wrong UTF codec detected; usually because it's not UTF-8 894 # but some other 8-bit codec. This is an RFC violation, 895 # and the server didn't bother to tell us what codec *was* 896 # used. 897 pass 898 return complexjson.loads(self.text, **kwargs) 899 900 @property 901 def links(self): 902 """Returns the parsed header links of the response, if any.""" 903 904 header = self.headers.get('link') 905 906 # l = MultiDict() 907 l = {} 908 909 if header: 910 links = parse_header_links(header) 911 912 for link in links: 913 key = link.get('rel') or link.get('url') 914 l[key] = link 915 916 return l 917 918 def raise_for_status(self): 919 """Raises stored :class:`HTTPError`, if one occurred.""" 920 921 http_error_msg = '' 922 if isinstance(self.reason, bytes): 923 # We attempt to decode utf-8 first because some servers 924 # choose to localize their reason strings. If the string 925 # isn't utf-8, we fall back to iso-8859-1 for all other 926 # encodings. (See PR #3538) 927 try: 928 reason = self.reason.decode('utf-8') 929 except UnicodeDecodeError: 930 reason = self.reason.decode('iso-8859-1') 931 else: 932 reason = self.reason 933 934 if 400 <= self.status_code < 500: 935 http_error_msg = u'%s Client Error: %s for url: %s' % (self.status_code, reason, self.url) 936 937 elif 500 <= self.status_code < 600: 938 http_error_msg = u'%s Server Error: %s for url: %s' % (self.status_code, reason, self.url) 939 940 if http_error_msg: 941 raise HTTPError(http_error_msg, response=self) 942 943 def close(self): 944 """Releases the connection back to the pool. Once this method has been 945 called the underlying ``raw`` object must not be accessed again. 946 947 *Note: Should not normally need to be called explicitly.* 948 """ 949 if not self._content_consumed: 950 self.raw.close() 951 952 release_conn = getattr(self.raw, 'release_conn', None) 953 if release_conn is not None: 954 release_conn() 955