def _build_response(self, result, request, protocol):
self.result = result
respcls = responsetypes.from_args(url=request.url)
protocol.close()
body = protocol.filename or protocol.body.read()
headers = {"local filename": protocol.filename or '', "size": protocol.size}
return respcls(url=request.url, status=200, body=body, headers=headers)
def retrieve_response(self, spider, request):
data = self._read_data(spider, request)
if data is None:
return # not cached
url = data['url']
status = data['status']
headers = Headers(data['headers'])
body = data['body']
respcls = responsetypes.from_args(headers=headers, url=url)
response = respcls(url=url, headers=headers, status=status, body=body)
return response
def retrieve_response(self, spider, request):
"""Return response if present in cache, or None otherwise."""
metadata = self._read_meta(spider, request)
if metadata is None:
return # not cached
rpath = self._get_request_path(spider, request)
with open(os.path.join(rpath, 'response_body'), 'rb') as f:
body = f.read()
with open(os.path.join(rpath, 'response_headers'), 'rb') as f:
rawheaders = f.read()
url = metadata.get('response_url')
status = metadata['status']
headers = Headers(headers_raw_to_dict(rawheaders))
respcls = responsetypes.from_args(headers=headers, url=url)
response = respcls(url=url, headers=headers, status=status, body=body)
return response
def process_response(self, request, response, spider):
if isinstance(response, Response):
content_encoding = response.headers.getlist("Content-Encoding")
if content_encoding and not is_gzipped(response):
encoding = content_encoding.pop()
decoded_body = self._decode(response.body, encoding.lower())
respcls = responsetypes.from_args(headers=response.headers, url=response.url)
kwargs = dict(cls=respcls, body=decoded_body)
if issubclass(respcls, TextResponse):
# force recalculating the encoding until we make sure the
# responsetypes guessing is reliable
kwargs["encoding"] = None
response = response.replace(**kwargs)
if not content_encoding:
del response.headers["Content-Encoding"]
return response
def _cb_bodydone(self, result, request, url):
txresponse, body, flags = result
status = int(txresponse.code)
headers = Headers(txresponse.headers.getAllRawHeaders())
respcls = responsetypes.from_args(headers=headers, url=url)
return respcls(url=url, status=status, headers=headers, body=body, flags=flags)
def download_request(self, request, spider):
filepath = file_uri_to_path(request.url)
body = open(filepath, 'rb').read()
respcls = responsetypes.from_args(filename=filepath, body=body)
return respcls(url=request.url, body=body)
def _build_response(self, body, request):
request.meta['download_latency'] = self.headers_time-self.start_time
status = int(self.status)
headers = Headers(self.response_headers)
respcls = responsetypes.from_args(headers=headers, url=self.url)
return respcls(url=self.url, status=status, headers=headers, body=body)