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/requests/requests/issues/3578. 16import encodings.idna 17 18from pip._vendor.urllib3.fields import RequestField 19from pip._vendor.urllib3.filepost import encode_multipart_formdata 20from pip._vendor.urllib3.util import parse_url 21from pip._vendor.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 <PreparedRequest [GET]> 284 285 >>> s = requests.Session() 286 >>> s.send(r) 287 <Response [200]> 288 """ 289 290 def __init__(self): 291 #: HTTP verb to send to the server. 292 self.method = None 293 #: HTTP URL to send the request to. 294 self.url = None 295 #: dictionary of HTTP headers. 296 self.headers = None 297 # The `CookieJar` used to create the Cookie header will be stored here 298 # after prepare_cookies is called 299 self._cookies = None 300 #: request body to send to the server. 301 self.body = None 302 #: dictionary of callback hooks, for internal usage. 303 self.hooks = default_hooks() 304 #: integer denoting starting position of a readable file-like body. 305 self._body_position = None 306 307 def prepare(self, 308 method=None, url=None, headers=None, files=None, data=None, 309 params=None, auth=None, cookies=None, hooks=None, json=None): 310 """Prepares the entire request with the given parameters.""" 311 312 self.prepare_method(method) 313 self.prepare_url(url, params) 314 self.prepare_headers(headers) 315 self.prepare_cookies(cookies) 316 self.prepare_body(data, files, json) 317 self.prepare_auth(auth, url) 318 319 # Note that prepare_auth must be last to enable authentication schemes 320 # such as OAuth to work on a fully prepared request. 321 322 # This MUST go after prepare_auth. Authenticators could add a hook 323 self.prepare_hooks(hooks) 324 325 def __repr__(self): 326 return '<PreparedRequest [%s]>' % (self.method) 327 328 def copy(self): 329 p = PreparedRequest() 330 p.method = self.method 331 p.url = self.url 332 p.headers = self.headers.copy() if self.headers is not None else None 333 p._cookies = _copy_cookie_jar(self._cookies) 334 p.body = self.body 335 p.hooks = self.hooks 336 p._body_position = self._body_position 337 return p 338 339 def prepare_method(self, method): 340 """Prepares the given HTTP method.""" 341 self.method = method 342 if self.method is not None: 343 self.method = to_native_string(self.method.upper()) 344 345 @staticmethod 346 def _get_idna_encoded_host(host): 347 from pip._vendor import idna 348 349 try: 350 host = idna.encode(host, uts46=True).decode('utf-8') 351 except idna.IDNAError: 352 raise UnicodeError 353 return host 354 355 def prepare_url(self, url, params): 356 """Prepares the given HTTP URL.""" 357 #: Accept objects that have string representations. 358 #: We're unable to blindly call unicode/str functions 359 #: as this will include the bytestring indicator (b'') 360 #: on python 3.x. 361 #: https://github.com/requests/requests/pull/2238 362 if isinstance(url, bytes): 363 url = url.decode('utf8') 364 else: 365 url = unicode(url) if is_py2 else str(url) 366 367 # Remove leading whitespaces from url 368 url = url.lstrip() 369 370 # Don't do any URL preparation for non-HTTP schemes like `mailto`, 371 # `data` etc to work around exceptions from `url_parse`, which 372 # handles RFC 3986 only. 373 if ':' in url and not url.lower().startswith('http'): 374 self.url = url 375 return 376 377 # Support for unicode domain names and paths. 378 try: 379 scheme, auth, host, port, path, query, fragment = parse_url(url) 380 except LocationParseError as e: 381 raise InvalidURL(*e.args) 382 383 if not scheme: 384 error = ("Invalid URL {0!r}: No schema supplied. Perhaps you meant http://{0}?") 385 error = error.format(to_native_string(url, 'utf8')) 386 387 raise MissingSchema(error) 388 389 if not host: 390 raise InvalidURL("Invalid URL %r: No host supplied" % url) 391 392 # In general, we want to try IDNA encoding the hostname if the string contains 393 # non-ASCII characters. This allows users to automatically get the correct IDNA 394 # behaviour. For strings containing only ASCII characters, we need to also verify 395 # it doesn't start with a wildcard (*), before allowing the unencoded hostname. 396 if not unicode_is_ascii(host): 397 try: 398 host = self._get_idna_encoded_host(host) 399 except UnicodeError: 400 raise InvalidURL('URL has an invalid label.') 401 elif host.startswith(u'*'): 402 raise InvalidURL('URL has an invalid label.') 403 404 # Carefully reconstruct the network location 405 netloc = auth or '' 406 if netloc: 407 netloc += '@' 408 netloc += host 409 if port: 410 netloc += ':' + str(port) 411 412 # Bare domains aren't valid URLs. 413 if not path: 414 path = '/' 415 416 if is_py2: 417 if isinstance(scheme, str): 418 scheme = scheme.encode('utf-8') 419 if isinstance(netloc, str): 420 netloc = netloc.encode('utf-8') 421 if isinstance(path, str): 422 path = path.encode('utf-8') 423 if isinstance(query, str): 424 query = query.encode('utf-8') 425 if isinstance(fragment, str): 426 fragment = fragment.encode('utf-8') 427 428 if isinstance(params, (str, bytes)): 429 params = to_native_string(params) 430 431 enc_params = self._encode_params(params) 432 if enc_params: 433 if query: 434 query = '%s&%s' % (query, enc_params) 435 else: 436 query = enc_params 437 438 url = requote_uri(urlunparse([scheme, netloc, path, None, query, fragment])) 439 self.url = url 440 441 def prepare_headers(self, headers): 442 """Prepares the given HTTP headers.""" 443 444 self.headers = CaseInsensitiveDict() 445 if headers: 446 for header in headers.items(): 447 # Raise exception on invalid header value. 448 check_header_validity(header) 449 name, value = header 450 self.headers[to_native_string(name)] = value 451 452 def prepare_body(self, data, files, json=None): 453 """Prepares the given HTTP body data.""" 454 455 # Check if file, fo, generator, iterator. 456 # If not, run through normal process. 457 458 # Nottin' on you. 459 body = None 460 content_type = None 461 462 if not data and json is not None: 463 # urllib3 requires a bytes-like body. Python 2's json.dumps 464 # provides this natively, but Python 3 gives a Unicode string. 465 content_type = 'application/json' 466 body = complexjson.dumps(json) 467 if not isinstance(body, bytes): 468 body = body.encode('utf-8') 469 470 is_stream = all([ 471 hasattr(data, '__iter__'), 472 not isinstance(data, (basestring, list, tuple, Mapping)) 473 ]) 474 475 try: 476 length = super_len(data) 477 except (TypeError, AttributeError, UnsupportedOperation): 478 length = None 479 480 if is_stream: 481 body = data 482 483 if getattr(body, 'tell', None) is not None: 484 # Record the current file position before reading. 485 # This will allow us to rewind a file in the event 486 # of a redirect. 487 try: 488 self._body_position = body.tell() 489 except (IOError, OSError): 490 # This differentiates from None, allowing us to catch 491 # a failed `tell()` later when trying to rewind the body 492 self._body_position = object() 493 494 if files: 495 raise NotImplementedError('Streamed bodies and files are mutually exclusive.') 496 497 if length: 498 self.headers['Content-Length'] = builtin_str(length) 499 else: 500 self.headers['Transfer-Encoding'] = 'chunked' 501 else: 502 # Multi-part file uploads. 503 if files: 504 (body, content_type) = self._encode_files(files, data) 505 else: 506 if data: 507 body = self._encode_params(data) 508 if isinstance(data, basestring) or hasattr(data, 'read'): 509 content_type = None 510 else: 511 content_type = 'application/x-www-form-urlencoded' 512 513 self.prepare_content_length(body) 514 515 # Add content-type if it wasn't explicitly provided. 516 if content_type and ('content-type' not in self.headers): 517 self.headers['Content-Type'] = content_type 518 519 self.body = body 520 521 def prepare_content_length(self, body): 522 """Prepare Content-Length header based on request method and body""" 523 if body is not None: 524 length = super_len(body) 525 if length: 526 # If length exists, set it. Otherwise, we fallback 527 # to Transfer-Encoding: chunked. 528 self.headers['Content-Length'] = builtin_str(length) 529 elif self.method not in ('GET', 'HEAD') and self.headers.get('Content-Length') is None: 530 # Set Content-Length to 0 for methods that can have a body 531 # but don't provide one. (i.e. not GET or HEAD) 532 self.headers['Content-Length'] = '0' 533 534 def prepare_auth(self, auth, url=''): 535 """Prepares the given HTTP auth data.""" 536 537 # If no Auth is explicitly provided, extract it from the URL first. 538 if auth is None: 539 url_auth = get_auth_from_url(self.url) 540 auth = url_auth if any(url_auth) else None 541 542 if auth: 543 if isinstance(auth, tuple) and len(auth) == 2: 544 # special-case basic HTTP auth 545 auth = HTTPBasicAuth(*auth) 546 547 # Allow auth to make its changes. 548 r = auth(self) 549 550 # Update self to reflect the auth changes. 551 self.__dict__.update(r.__dict__) 552 553 # Recompute Content-Length 554 self.prepare_content_length(self.body) 555 556 def prepare_cookies(self, cookies): 557 """Prepares the given HTTP cookie data. 558 559 This function eventually generates a ``Cookie`` header from the 560 given cookies using cookielib. Due to cookielib's design, the header 561 will not be regenerated if it already exists, meaning this function 562 can only be called once for the life of the 563 :class:`PreparedRequest <PreparedRequest>` object. Any subsequent calls 564 to ``prepare_cookies`` will have no actual effect, unless the "Cookie" 565 header is removed beforehand. 566 """ 567 if isinstance(cookies, cookielib.CookieJar): 568 self._cookies = cookies 569 else: 570 self._cookies = cookiejar_from_dict(cookies) 571 572 cookie_header = get_cookie_header(self._cookies, self) 573 if cookie_header is not None: 574 self.headers['Cookie'] = cookie_header 575 576 def prepare_hooks(self, hooks): 577 """Prepares the given hooks.""" 578 # hooks can be passed as None to the prepare method and to this 579 # method. To prevent iterating over None, simply use an empty list 580 # if hooks is False-y 581 hooks = hooks or [] 582 for event in hooks: 583 self.register_hook(event, hooks[event]) 584 585 586class Response(object): 587 """The :class:`Response <Response>` object, which contains a 588 server's response to an HTTP request. 589 """ 590 591 __attrs__ = [ 592 '_content', 'status_code', 'headers', 'url', 'history', 593 'encoding', 'reason', 'cookies', 'elapsed', 'request' 594 ] 595 596 def __init__(self): 597 self._content = False 598 self._content_consumed = False 599 self._next = None 600 601 #: Integer Code of responded HTTP Status, e.g. 404 or 200. 602 self.status_code = None 603 604 #: Case-insensitive Dictionary of Response Headers. 605 #: For example, ``headers['content-encoding']`` will return the 606 #: value of a ``'Content-Encoding'`` response header. 607 self.headers = CaseInsensitiveDict() 608 609 #: File-like object representation of response (for advanced usage). 610 #: Use of ``raw`` requires that ``stream=True`` be set on the request. 611 # This requirement does not apply for use internally to Requests. 612 self.raw = None 613 614 #: Final URL location of Response. 615 self.url = None 616 617 #: Encoding to decode with when accessing r.text. 618 self.encoding = None 619 620 #: A list of :class:`Response <Response>` objects from 621 #: the history of the Request. Any redirect responses will end 622 #: up here. The list is sorted from the oldest to the most recent request. 623 self.history = [] 624 625 #: Textual reason of responded HTTP Status, e.g. "Not Found" or "OK". 626 self.reason = None 627 628 #: A CookieJar of Cookies the server sent back. 629 self.cookies = cookiejar_from_dict({}) 630 631 #: The amount of time elapsed between sending the request 632 #: and the arrival of the response (as a timedelta). 633 #: This property specifically measures the time taken between sending 634 #: the first byte of the request and finishing parsing the headers. It 635 #: is therefore unaffected by consuming the response content or the 636 #: value of the ``stream`` keyword argument. 637 self.elapsed = datetime.timedelta(0) 638 639 #: The :class:`PreparedRequest <PreparedRequest>` object to which this 640 #: is a response. 641 self.request = None 642 643 def __enter__(self): 644 return self 645 646 def __exit__(self, *args): 647 self.close() 648 649 def __getstate__(self): 650 # Consume everything; accessing the content attribute makes 651 # sure the content has been fully read. 652 if not self._content_consumed: 653 self.content 654 655 return {attr: getattr(self, attr, None) for attr in self.__attrs__} 656 657 def __setstate__(self, state): 658 for name, value in state.items(): 659 setattr(self, name, value) 660 661 # pickled objects do not have .raw 662 setattr(self, '_content_consumed', True) 663 setattr(self, 'raw', None) 664 665 def __repr__(self): 666 return '<Response [%s]>' % (self.status_code) 667 668 def __bool__(self): 669 """Returns True if :attr:`status_code` is less than 400. 670 671 This attribute checks if the status code of the response is between 672 400 and 600 to see if there was a client error or a server error. If 673 the status code, is between 200 and 400, this will return True. This 674 is **not** a check to see if the response code is ``200 OK``. 675 """ 676 return self.ok 677 678 def __nonzero__(self): 679 """Returns True if :attr:`status_code` is less than 400. 680 681 This attribute checks if the status code of the response is between 682 400 and 600 to see if there was a client error or a server error. If 683 the status code, is between 200 and 400, this will return True. This 684 is **not** a check to see if the response code is ``200 OK``. 685 """ 686 return self.ok 687 688 def __iter__(self): 689 """Allows you to use a response as an iterator.""" 690 return self.iter_content(128) 691 692 @property 693 def ok(self): 694 """Returns True if :attr:`status_code` is less than 400, False if not. 695 696 This attribute checks if the status code of the response is between 697 400 and 600 to see if there was a client error or a server error. If 698 the status code is between 200 and 400, this will return True. This 699 is **not** a check to see if the response code is ``200 OK``. 700 """ 701 try: 702 self.raise_for_status() 703 except HTTPError: 704 return False 705 return True 706 707 @property 708 def is_redirect(self): 709 """True if this Response is a well-formed HTTP redirect that could have 710 been processed automatically (by :meth:`Session.resolve_redirects`). 711 """ 712 return ('location' in self.headers and self.status_code in REDIRECT_STATI) 713 714 @property 715 def is_permanent_redirect(self): 716 """True if this Response one of the permanent versions of redirect.""" 717 return ('location' in self.headers and self.status_code in (codes.moved_permanently, codes.permanent_redirect)) 718 719 @property 720 def next(self): 721 """Returns a PreparedRequest for the next request in a redirect chain, if there is one.""" 722 return self._next 723 724 @property 725 def apparent_encoding(self): 726 """The apparent encoding, provided by the chardet library.""" 727 return chardet.detect(self.content)['encoding'] 728 729 def iter_content(self, chunk_size=1, decode_unicode=False): 730 """Iterates over the response data. When stream=True is set on the 731 request, this avoids reading the content at once into memory for 732 large responses. The chunk size is the number of bytes it should 733 read into memory. This is not necessarily the length of each item 734 returned as decoding can take place. 735 736 chunk_size must be of type int or None. A value of None will 737 function differently depending on the value of `stream`. 738 stream=True will read data as it arrives in whatever size the 739 chunks are received. If stream=False, data is returned as 740 a single chunk. 741 742 If decode_unicode is True, content will be decoded using the best 743 available encoding based on the response. 744 """ 745 746 def generate(): 747 # Special case for urllib3. 748 if hasattr(self.raw, 'stream'): 749 try: 750 for chunk in self.raw.stream(chunk_size, decode_content=True): 751 yield chunk 752 except ProtocolError as e: 753 raise ChunkedEncodingError(e) 754 except DecodeError as e: 755 raise ContentDecodingError(e) 756 except ReadTimeoutError as e: 757 raise ConnectionError(e) 758 else: 759 # Standard file-like object. 760 while True: 761 chunk = self.raw.read(chunk_size) 762 if not chunk: 763 break 764 yield chunk 765 766 self._content_consumed = True 767 768 if self._content_consumed and isinstance(self._content, bool): 769 raise StreamConsumedError() 770 elif chunk_size is not None and not isinstance(chunk_size, int): 771 raise TypeError("chunk_size must be an int, it is instead a %s." % type(chunk_size)) 772 # simulate reading small chunks of the content 773 reused_chunks = iter_slices(self._content, chunk_size) 774 775 stream_chunks = generate() 776 777 chunks = reused_chunks if self._content_consumed else stream_chunks 778 779 if decode_unicode: 780 chunks = stream_decode_response_unicode(chunks, self) 781 782 return chunks 783 784 def iter_lines(self, chunk_size=ITER_CHUNK_SIZE, decode_unicode=False, delimiter=None): 785 """Iterates over the response data, one line at a time. When 786 stream=True is set on the request, this avoids reading the 787 content at once into memory for large responses. 788 789 .. note:: This method is not reentrant safe. 790 """ 791 792 pending = None 793 794 for chunk in self.iter_content(chunk_size=chunk_size, decode_unicode=decode_unicode): 795 796 if pending is not None: 797 chunk = pending + chunk 798 799 if delimiter: 800 lines = chunk.split(delimiter) 801 else: 802 lines = chunk.splitlines() 803 804 if lines and lines[-1] and chunk and lines[-1][-1] == chunk[-1]: 805 pending = lines.pop() 806 else: 807 pending = None 808 809 for line in lines: 810 yield line 811 812 if pending is not None: 813 yield pending 814 815 @property 816 def content(self): 817 """Content of the response, in bytes.""" 818 819 if self._content is False: 820 # Read the contents. 821 if self._content_consumed: 822 raise RuntimeError( 823 'The content for this response was already consumed') 824 825 if self.status_code == 0 or self.raw is None: 826 self._content = None 827 else: 828 self._content = b''.join(self.iter_content(CONTENT_CHUNK_SIZE)) or b'' 829 830 self._content_consumed = True 831 # don't need to release the connection; that's been handled by urllib3 832 # since we exhausted the data. 833 return self._content 834 835 @property 836 def text(self): 837 """Content of the response, in unicode. 838 839 If Response.encoding is None, encoding will be guessed using 840 ``chardet``. 841 842 The encoding of the response content is determined based solely on HTTP 843 headers, following RFC 2616 to the letter. If you can take advantage of 844 non-HTTP knowledge to make a better guess at the encoding, you should 845 set ``r.encoding`` appropriately before accessing this property. 846 """ 847 848 # Try charset from content-type 849 content = None 850 encoding = self.encoding 851 852 if not self.content: 853 return str('') 854 855 # Fallback to auto-detected encoding. 856 if self.encoding is None: 857 encoding = self.apparent_encoding 858 859 # Decode unicode from given encoding. 860 try: 861 content = str(self.content, encoding, errors='replace') 862 except (LookupError, TypeError): 863 # A LookupError is raised if the encoding was not found which could 864 # indicate a misspelling or similar mistake. 865 # 866 # A TypeError can be raised if encoding is None 867 # 868 # So we try blindly encoding. 869 content = str(self.content, errors='replace') 870 871 return content 872 873 def json(self, **kwargs): 874 r"""Returns the json-encoded content of a response, if any. 875 876 :param \*\*kwargs: Optional arguments that ``json.loads`` takes. 877 :raises ValueError: If the response body does not contain valid json. 878 """ 879 880 if not self.encoding and self.content and len(self.content) > 3: 881 # No encoding set. JSON RFC 4627 section 3 states we should expect 882 # UTF-8, -16 or -32. Detect which one to use; If the detection or 883 # decoding fails, fall back to `self.text` (using chardet to make 884 # a best guess). 885 encoding = guess_json_utf(self.content) 886 if encoding is not None: 887 try: 888 return complexjson.loads( 889 self.content.decode(encoding), **kwargs 890 ) 891 except UnicodeDecodeError: 892 # Wrong UTF codec detected; usually because it's not UTF-8 893 # but some other 8-bit codec. This is an RFC violation, 894 # and the server didn't bother to tell us what codec *was* 895 # used. 896 pass 897 return complexjson.loads(self.text, **kwargs) 898 899 @property 900 def links(self): 901 """Returns the parsed header links of the response, if any.""" 902 903 header = self.headers.get('link') 904 905 # l = MultiDict() 906 l = {} 907 908 if header: 909 links = parse_header_links(header) 910 911 for link in links: 912 key = link.get('rel') or link.get('url') 913 l[key] = link 914 915 return l 916 917 def raise_for_status(self): 918 """Raises stored :class:`HTTPError`, if one occurred.""" 919 920 http_error_msg = '' 921 if isinstance(self.reason, bytes): 922 # We attempt to decode utf-8 first because some servers 923 # choose to localize their reason strings. If the string 924 # isn't utf-8, we fall back to iso-8859-1 for all other 925 # encodings. (See PR #3538) 926 try: 927 reason = self.reason.decode('utf-8') 928 except UnicodeDecodeError: 929 reason = self.reason.decode('iso-8859-1') 930 else: 931 reason = self.reason 932 933 if 400 <= self.status_code < 500: 934 http_error_msg = u'%s Client Error: %s for url: %s' % (self.status_code, reason, self.url) 935 936 elif 500 <= self.status_code < 600: 937 http_error_msg = u'%s Server Error: %s for url: %s' % (self.status_code, reason, self.url) 938 939 if http_error_msg: 940 raise HTTPError(http_error_msg, response=self) 941 942 def close(self): 943 """Releases the connection back to the pool. Once this method has been 944 called the underlying ``raw`` object must not be accessed again. 945 946 *Note: Should not normally need to be called explicitly.* 947 """ 948 if not self._content_consumed: 949 self.raw.close() 950 951 release_conn = getattr(self.raw, 'release_conn', None) 952 if release_conn is not None: 953 release_conn() 954