# Copyright 2017 Google Inc. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """Support for downloading media from Google APIs.""" import urllib3.response # type: ignore import http from google.resumable_media import _download from google.resumable_media import common from google.resumable_media import _helpers from google.resumable_media.requests import _request_helpers _CHECKSUM_MISMATCH = """\ Checksum mismatch while downloading: {} The X-Goog-Hash header indicated an {checksum_type} checksum of: {} but the actual {checksum_type} checksum of the downloaded contents was: {} """ _STREAM_SEEK_ERROR = """\ Incomplete download for: {} Error writing to stream while handling a gzip-compressed file download. Please restart the download. """ _RESPONSE_HEADERS_INFO = """\ The X-Goog-Stored-Content-Length is {}. The X-Goog-Stored-Content-Encoding is {}. The download request read {} bytes of data. If the download was incomplete, please check the network connection and restart the download. """ class Download(_request_helpers.RequestsMixin, _download.Download): """Helper to manage downloading a resource from a Google API. "Slices" of the resource can be retrieved by specifying a range with ``start`` and / or ``end``. However, in typical usage, neither ``start`` nor ``end`` is expected to be provided. Args: media_url (str): The URL containing the media to be downloaded. stream (IO[bytes]): A write-able stream (i.e. file-like object) that the downloaded resource can be written to. start (int): The first byte in a range to be downloaded. If not provided, but ``end`` is provided, will download from the beginning to ``end`` of the media. end (int): The last byte in a range to be downloaded. If not provided, but ``start`` is provided, will download from the ``start`` to the end of the media. headers (Optional[Mapping[str, str]]): Extra headers that should be sent with the request, e.g. headers for encrypted data. checksum Optional([str]): The type of checksum to compute to verify the integrity of the object. The response headers must contain a checksum of the requested type. If the headers lack an appropriate checksum (for instance in the case of transcoded or ranged downloads where the remote service does not know the correct checksum) an INFO-level log will be emitted. Supported values are "md5", "crc32c" and None. The default is "md5". Attributes: media_url (str): The URL containing the media to be downloaded. start (Optional[int]): The first byte in a range to be downloaded. end (Optional[int]): The last byte in a range to be downloaded. """ def _write_to_stream(self, response): """Write response body to a write-able stream. .. note: This method assumes that the ``_stream`` attribute is set on the current download. Args: response (~requests.Response): The HTTP response object. Raises: ~google.resumable_media.common.DataCorruption: If the download's checksum doesn't agree with server-computed checksum. """ # Retrieve the expected checksum only once for the download request, # then compute and validate the checksum when the full download completes. # Retried requests are range requests, and there's no way to detect # data corruption for that byte range alone. if self._expected_checksum is None and self._checksum_object is None: # `_get_expected_checksum()` may return None even if a checksum was # requested, in which case it will emit an info log _MISSING_CHECKSUM. # If an invalid checksum type is specified, this will raise ValueError. expected_checksum, checksum_object = _helpers._get_expected_checksum( response, self._get_headers, self.media_url, checksum_type=self.checksum ) self._expected_checksum = expected_checksum self._checksum_object = checksum_object else: expected_checksum = self._expected_checksum checksum_object = self._checksum_object with response: # NOTE: In order to handle compressed streams gracefully, we try # to insert our checksum object into the decompression stream. If # the stream is indeed compressed, this will delegate the checksum # object to the decoder and return a _DoNothingHash here. local_checksum_object = _add_decoder(response.raw, checksum_object) body_iter = response.iter_content( chunk_size=_request_helpers._SINGLE_GET_CHUNK_SIZE, decode_unicode=False ) for chunk in body_iter: self._stream.write(chunk) self._bytes_downloaded += len(chunk) local_checksum_object.update(chunk) # Don't validate the checksum for partial responses. if ( expected_checksum is not None and response.status_code != http.client.PARTIAL_CONTENT ): actual_checksum = _helpers.prepare_checksum_digest(checksum_object.digest()) if actual_checksum != expected_checksum: headers = self._get_headers(response) x_goog_encoding = headers.get("x-goog-stored-content-encoding") x_goog_length = headers.get("x-goog-stored-content-length") content_length_msg = _RESPONSE_HEADERS_INFO.format( x_goog_length, x_goog_encoding, self._bytes_downloaded ) if ( x_goog_length and self._bytes_downloaded < int(x_goog_length) and x_goog_encoding != "gzip" ): # The library will attempt to trigger a retry by raising a ConnectionError, if # (a) bytes_downloaded is less than response header x-goog-stored-content-length, and # (b) the object is not gzip-compressed when stored in Cloud Storage. raise ConnectionError(content_length_msg) else: msg = _CHECKSUM_MISMATCH.format( self.media_url, expected_checksum, actual_checksum, checksum_type=self.checksum.upper(), ) msg += content_length_msg raise common.DataCorruption(response, msg) def consume( self, transport, timeout=( _request_helpers._DEFAULT_CONNECT_TIMEOUT, _request_helpers._DEFAULT_READ_TIMEOUT, ), ): """Consume the resource to be downloaded. If a ``stream`` is attached to this download, then the downloaded resource will be written to the stream. Args: transport (~requests.Session): A ``requests`` object which can make authenticated requests. timeout (Optional[Union[float, Tuple[float, float]]]): The number of seconds to wait for the server response. Depending on the retry strategy, a request may be repeated several times using the same timeout each time. Can also be passed as a tuple (connect_timeout, read_timeout). See :meth:`requests.Session.request` documentation for details. Returns: ~requests.Response: The HTTP response returned by ``transport``. Raises: ~google.resumable_media.common.DataCorruption: If the download's checksum doesn't agree with server-computed checksum. ValueError: If the current :class:`Download` has already finished. """ method, _, payload, headers = self._prepare_request() # NOTE: We assume "payload is None" but pass it along anyway. request_kwargs = { "data": payload, "headers": headers, "timeout": timeout, } if self._stream is not None: request_kwargs["stream"] = True # Assign object generation if generation is specified in the media url. if self._object_generation is None: self._object_generation = _helpers._get_generation_from_url(self.media_url) # Wrap the request business logic in a function to be retried. def retriable_request(): url = self.media_url # To restart an interrupted download, read from the offset of last byte # received using a range request, and set object generation query param. if self._bytes_downloaded > 0: _download.add_bytes_range( self._bytes_downloaded, self.end, self._headers ) request_kwargs["headers"] = self._headers # Set object generation query param to ensure the same object content is requested. if ( self._object_generation is not None and _helpers._get_generation_from_url(self.media_url) is None ): query_param = {"generation": self._object_generation} url = _helpers.add_query_parameters(self.media_url, query_param) result = transport.request(method, url, **request_kwargs) # If a generation hasn't been specified, and this is the first response we get, let's record the # generation. In future requests we'll specify the generation query param to avoid data races. if self._object_generation is None: self._object_generation = _helpers._parse_generation_header( result, self._get_headers ) self._process_response(result) # With decompressive transcoding, GCS serves back the whole file regardless of the range request, # thus we reset the stream position to the start of the stream. # See: https://cloud.google.com/storage/docs/transcoding#range if self._stream is not None: if _helpers._is_decompressive_transcoding(result, self._get_headers): try: self._stream.seek(0) except Exception as exc: msg = _STREAM_SEEK_ERROR.format(url) raise Exception(msg) from exc self._bytes_downloaded = 0 self._write_to_stream(result) return result return _request_helpers.wait_and_retry( retriable_request, self._get_status_code, self._retry_strategy ) class RawDownload(_request_helpers.RawRequestsMixin, _download.Download): """Helper to manage downloading a raw resource from a Google API. "Slices" of the resource can be retrieved by specifying a range with ``start`` and / or ``end``. However, in typical usage, neither ``start`` nor ``end`` is expected to be provided. Args: media_url (str): The URL containing the media to be downloaded. stream (IO[bytes]): A write-able stream (i.e. file-like object) that the downloaded resource can be written to. start (int): The first byte in a range to be downloaded. If not provided, but ``end`` is provided, will download from the beginning to ``end`` of the media. end (int): The last byte in a range to be downloaded. If not provided, but ``start`` is provided, will download from the ``start`` to the end of the media. headers (Optional[Mapping[str, str]]): Extra headers that should be sent with the request, e.g. headers for encrypted data. checksum Optional([str]): The type of checksum to compute to verify the integrity of the object. The response headers must contain a checksum of the requested type. If the headers lack an appropriate checksum (for instance in the case of transcoded or ranged downloads where the remote service does not know the correct checksum) an INFO-level log will be emitted. Supported values are "md5", "crc32c" and None. The default is "md5". Attributes: media_url (str): The URL containing the media to be downloaded. start (Optional[int]): The first byte in a range to be downloaded. end (Optional[int]): The last byte in a range to be downloaded. """ def _write_to_stream(self, response): """Write response body to a write-able stream. .. note: This method assumes that the ``_stream`` attribute is set on the current download. Args: response (~requests.Response): The HTTP response object. Raises: ~google.resumable_media.common.DataCorruption: If the download's checksum doesn't agree with server-computed checksum. """ # Retrieve the expected checksum only once for the download request, # then compute and validate the checksum when the full download completes. # Retried requests are range requests, and there's no way to detect # data corruption for that byte range alone. if self._expected_checksum is None and self._checksum_object is None: # `_get_expected_checksum()` may return None even if a checksum was # requested, in which case it will emit an info log _MISSING_CHECKSUM. # If an invalid checksum type is specified, this will raise ValueError. expected_checksum, checksum_object = _helpers._get_expected_checksum( response, self._get_headers, self.media_url, checksum_type=self.checksum ) self._expected_checksum = expected_checksum self._checksum_object = checksum_object else: expected_checksum = self._expected_checksum checksum_object = self._checksum_object with response: body_iter = response.raw.stream( _request_helpers._SINGLE_GET_CHUNK_SIZE, decode_content=False ) for chunk in body_iter: self._stream.write(chunk) self._bytes_downloaded += len(chunk) checksum_object.update(chunk) response._content_consumed = True # Don't validate the checksum for partial responses. if ( expected_checksum is not None and response.status_code != http.client.PARTIAL_CONTENT ): actual_checksum = _helpers.prepare_checksum_digest(checksum_object.digest()) if actual_checksum != expected_checksum: headers = self._get_headers(response) x_goog_encoding = headers.get("x-goog-stored-content-encoding") x_goog_length = headers.get("x-goog-stored-content-length") content_length_msg = _RESPONSE_HEADERS_INFO.format( x_goog_length, x_goog_encoding, self._bytes_downloaded ) if ( x_goog_length and self._bytes_downloaded < int(x_goog_length) and x_goog_encoding != "gzip" ): # The library will attempt to trigger a retry by raising a ConnectionError, if # (a) bytes_downloaded is less than response header x-goog-stored-content-length, and # (b) the object is not gzip-compressed when stored in Cloud Storage. raise ConnectionError(content_length_msg) else: msg = _CHECKSUM_MISMATCH.format( self.media_url, expected_checksum, actual_checksum, checksum_type=self.checksum.upper(), ) msg += content_length_msg raise common.DataCorruption(response, msg) def consume( self, transport, timeout=( _request_helpers._DEFAULT_CONNECT_TIMEOUT, _request_helpers._DEFAULT_READ_TIMEOUT, ), ): """Consume the resource to be downloaded. If a ``stream`` is attached to this download, then the downloaded resource will be written to the stream. Args: transport (~requests.Session): A ``requests`` object which can make authenticated requests. timeout (Optional[Union[float, Tuple[float, float]]]): The number of seconds to wait for the server response. Depending on the retry strategy, a request may be repeated several times using the same timeout each time. Can also be passed as a tuple (connect_timeout, read_timeout). See :meth:`requests.Session.request` documentation for details. Returns: ~requests.Response: The HTTP response returned by ``transport``. Raises: ~google.resumable_media.common.DataCorruption: If the download's checksum doesn't agree with server-computed checksum. ValueError: If the current :class:`Download` has already finished. """ method, _, payload, headers = self._prepare_request() # NOTE: We assume "payload is None" but pass it along anyway. request_kwargs = { "data": payload, "headers": headers, "timeout": timeout, "stream": True, } # Assign object generation if generation is specified in the media url. if self._object_generation is None: self._object_generation = _helpers._get_generation_from_url(self.media_url) # Wrap the request business logic in a function to be retried. def retriable_request(): url = self.media_url # To restart an interrupted download, read from the offset of last byte # received using a range request, and set object generation query param. if self._bytes_downloaded > 0: _download.add_bytes_range( self._bytes_downloaded, self.end, self._headers ) request_kwargs["headers"] = self._headers # Set object generation query param to ensure the same object content is requested. if ( self._object_generation is not None and _helpers._get_generation_from_url(self.media_url) is None ): query_param = {"generation": self._object_generation} url = _helpers.add_query_parameters(self.media_url, query_param) result = transport.request(method, url, **request_kwargs) # If a generation hasn't been specified, and this is the first response we get, let's record the # generation. In future requests we'll specify the generation query param to avoid data races. if self._object_generation is None: self._object_generation = _helpers._parse_generation_header( result, self._get_headers ) self._process_response(result) # With decompressive transcoding, GCS serves back the whole file regardless of the range request, # thus we reset the stream position to the start of the stream. # See: https://cloud.google.com/storage/docs/transcoding#range if self._stream is not None: if _helpers._is_decompressive_transcoding(result, self._get_headers): try: self._stream.seek(0) except Exception as exc: msg = _STREAM_SEEK_ERROR.format(url) raise Exception(msg) from exc self._bytes_downloaded = 0 self._write_to_stream(result) return result return _request_helpers.wait_and_retry( retriable_request, self._get_status_code, self._retry_strategy ) class ChunkedDownload(_request_helpers.RequestsMixin, _download.ChunkedDownload): """Download a resource in chunks from a Google API. Args: media_url (str): The URL containing the media to be downloaded. chunk_size (int): The number of bytes to be retrieved in each request. stream (IO[bytes]): A write-able stream (i.e. file-like object) that will be used to concatenate chunks of the resource as they are downloaded. start (int): The first byte in a range to be downloaded. If not provided, defaults to ``0``. end (int): The last byte in a range to be downloaded. If not provided, will download to the end of the media. headers (Optional[Mapping[str, str]]): Extra headers that should be sent with each request, e.g. headers for data encryption key headers. Attributes: media_url (str): The URL containing the media to be downloaded. start (Optional[int]): The first byte in a range to be downloaded. end (Optional[int]): The last byte in a range to be downloaded. chunk_size (int): The number of bytes to be retrieved in each request. Raises: ValueError: If ``start`` is negative. """ def consume_next_chunk( self, transport, timeout=( _request_helpers._DEFAULT_CONNECT_TIMEOUT, _request_helpers._DEFAULT_READ_TIMEOUT, ), ): """Consume the next chunk of the resource to be downloaded. Args: transport (~requests.Session): A ``requests`` object which can make authenticated requests. timeout (Optional[Union[float, Tuple[float, float]]]): The number of seconds to wait for the server response. Depending on the retry strategy, a request may be repeated several times using the same timeout each time. Can also be passed as a tuple (connect_timeout, read_timeout). See :meth:`requests.Session.request` documentation for details. Returns: ~requests.Response: The HTTP response returned by ``transport``. Raises: ValueError: If the current download has finished. """ method, url, payload, headers = self._prepare_request() # Wrap the request business logic in a function to be retried. def retriable_request(): # NOTE: We assume "payload is None" but pass it along anyway. result = transport.request( method, url, data=payload, headers=headers, timeout=timeout, ) self._process_response(result) return result return _request_helpers.wait_and_retry( retriable_request, self._get_status_code, self._retry_strategy ) class RawChunkedDownload(_request_helpers.RawRequestsMixin, _download.ChunkedDownload): """Download a raw resource in chunks from a Google API. Args: media_url (str): The URL containing the media to be downloaded. chunk_size (int): The number of bytes to be retrieved in each request. stream (IO[bytes]): A write-able stream (i.e. file-like object) that will be used to concatenate chunks of the resource as they are downloaded. start (int): The first byte in a range to be downloaded. If not provided, defaults to ``0``. end (int): The last byte in a range to be downloaded. If not provided, will download to the end of the media. headers (Optional[Mapping[str, str]]): Extra headers that should be sent with each request, e.g. headers for data encryption key headers. Attributes: media_url (str): The URL containing the media to be downloaded. start (Optional[int]): The first byte in a range to be downloaded. end (Optional[int]): The last byte in a range to be downloaded. chunk_size (int): The number of bytes to be retrieved in each request. Raises: ValueError: If ``start`` is negative. """ def consume_next_chunk( self, transport, timeout=( _request_helpers._DEFAULT_CONNECT_TIMEOUT, _request_helpers._DEFAULT_READ_TIMEOUT, ), ): """Consume the next chunk of the resource to be downloaded. Args: transport (~requests.Session): A ``requests`` object which can make authenticated requests. timeout (Optional[Union[float, Tuple[float, float]]]): The number of seconds to wait for the server response. Depending on the retry strategy, a request may be repeated several times using the same timeout each time. Can also be passed as a tuple (connect_timeout, read_timeout). See :meth:`requests.Session.request` documentation for details. Returns: ~requests.Response: The HTTP response returned by ``transport``. Raises: ValueError: If the current download has finished. """ method, url, payload, headers = self._prepare_request() # Wrap the request business logic in a function to be retried. def retriable_request(): # NOTE: We assume "payload is None" but pass it along anyway. result = transport.request( method, url, data=payload, headers=headers, stream=True, timeout=timeout, ) self._process_response(result) return result return _request_helpers.wait_and_retry( retriable_request, self._get_status_code, self._retry_strategy ) def _add_decoder(response_raw, checksum): """Patch the ``_decoder`` on a ``urllib3`` response. This is so that we can intercept the compressed bytes before they are decoded. Only patches if the content encoding is ``gzip`` or ``br``. Args: response_raw (urllib3.response.HTTPResponse): The raw response for an HTTP request. checksum (object): A checksum which will be updated with compressed bytes. Returns: object: Either the original ``checksum`` if ``_decoder`` is not patched, or a ``_DoNothingHash`` if the decoder is patched, since the caller will no longer need to hash to decoded bytes. """ encoding = response_raw.headers.get("content-encoding", "").lower() if encoding == "gzip": response_raw._decoder = _GzipDecoder(checksum) return _helpers._DoNothingHash() # Only activate if brotli is installed elif encoding == "br" and _BrotliDecoder: # type: ignore response_raw._decoder = _BrotliDecoder(checksum) return _helpers._DoNothingHash() else: return checksum class _GzipDecoder(urllib3.response.GzipDecoder): """Custom subclass of ``urllib3`` decoder for ``gzip``-ed bytes. Allows a checksum function to see the compressed bytes before they are decoded. This way the checksum of the compressed value can be computed. Args: checksum (object): A checksum which will be updated with compressed bytes. """ def __init__(self, checksum): super().__init__() self._checksum = checksum def decompress(self, data): """Decompress the bytes. Args: data (bytes): The compressed bytes to be decompressed. Returns: bytes: The decompressed bytes from ``data``. """ self._checksum.update(data) return super().decompress(data) # urllib3.response.BrotliDecoder might not exist depending on whether brotli is # installed. if hasattr(urllib3.response, "BrotliDecoder"): class _BrotliDecoder: """Handler for ``brotli`` encoded bytes. Allows a checksum function to see the compressed bytes before they are decoded. This way the checksum of the compressed value can be computed. Because BrotliDecoder's decompress method is dynamically created in urllib3, a subclass is not practical. Instead, this class creates a captive urllib3.requests.BrotliDecoder instance and acts as a proxy. Args: checksum (object): A checksum which will be updated with compressed bytes. """ def __init__(self, checksum): self._decoder = urllib3.response.BrotliDecoder() self._checksum = checksum def decompress(self, data): """Decompress the bytes. Args: data (bytes): The compressed bytes to be decompressed. Returns: bytes: The decompressed bytes from ``data``. """ self._checksum.update(data) return self._decoder.decompress(data) def flush(self): return self._decoder.flush() else: # pragma: NO COVER _BrotliDecoder = None # type: ignore # pragma: NO COVER