def _parse_extensions_header(self, headers):
extensions = headers.get("Sec-WebSocket-Extensions", '')
if extensions:
return [
httputil._parse_header(e.strip())
for e in extensions.split(',')
]
return []
def my_parse_multipart_form_data(boundary, data, arguments, files):
"""Parses a ``multipart/form-data`` body.
The ``boundary`` and ``data`` parameters are both byte strings.
The dictionaries given in the arguments and files parameters
will be updated with the contents of the body.
"""
# The standard allows for the boundary to be quoted in the header,
# although it's rare (it happens at least for google app engine
# xmpp). I think we're also supposed to handle backslash-escapes
# here but I'll save that until we see a client that uses them
# in the wild.
if boundary.startswith(b'"') and boundary.endswith(b'"'):
boundary = boundary[1:-1]
final_boundary_index = data.rfind(b"--" + boundary + b"--")
if final_boundary_index == -1:
warning("Invalid multipart/form-data: no final boundary")
return
parts = data[:final_boundary_index].split(b"--" + boundary + b"\r\n")
for part in parts:
if not part:
continue
eoh = part.find(b"\r\n\r\n")
if eoh == -1:
warning("multipart/form-data missing headers")
continue
headers = HTTPHeaders.parse(part[:eoh].decode("utf-8"))
disp_header = headers.get("Content-Disposition", "")
disposition, disp_params = _parse_header(disp_header)
if len(disp_header)==0:
# This is the part before first delimiter and in the original tornado.httputil it's validated for proper Content-Disposition,
# which is wrong
pass
else:
if len(disp_header) and disposition != "form-data" or not part.endswith(b"\r\n"):
warning("Invalid multipart/form-data")
continue
value = part[eoh + 4:-2]
if not disp_params.get("name"):
warning("multipart/form-data value missing name")
continue
name = disp_params["name"]
if disp_params.get("filename"):
ctype = headers.get("Content-Type", "application/unknown")
files.setdefault(name, []).append(HTTPFile( # type: ignore
filename=disp_params["filename"], body=value,
content_type=ctype))
else:
arguments.setdefault(name, []).append(value)
def data_received(self, chunk):
b = 0
if chunk.startswith(self.boundary):
i = chunk.find(b'\r\n\r\n')
if i != -1:
b = i + 4
headers = HTTPHeaders.parse(
chunk[len(self.boundary):i].decode("utf-8"))
disp_header = headers.get("Content-Disposition", "")
_, disp_params = _parse_header(disp_header)
filename = disp_params["filename"]
ext = filename.split('.')[-1]
self.filename = filename
self.temp_file_path = os.path.join(
self.tmp_path,
'uploading_file_%s.%s' % (str(uuid.uuid4()), ext))
self.file = open(self.temp_file_path, 'wb')
e = chunk.rfind(self.final_boundary_index)
if e == -1:
e = len(chunk)
if e > (self.len_final - 1):
temp = self.last + chunk[:self.len_final - 1]
else:
temp = self.last + chunk[:e]
last_index = temp.find(self.final_boundary_index)
if last_index != -1:
e = last_index - self.len_final + 1
if len(chunk) > self.len_final:
self.last = chunk[-self.len_final + 1:]
else:
self.last = chunk
if self.file:
self.file.write(chunk[b:e])
if e < len(chunk):
self.file.close()
self.uploaded_done()
def _parse_extensions_header(self, headers):
extensions = headers.get("Sec-WebSocket-Extensions", '')
if extensions:
return [httputil._parse_header(e.strip())
for e in extensions.split(',')]
return []
def data_received(self, data):
if self._buffer is not None:
data = self._buffer + data
self._buffer = None
boundary = data.find(self._boundary)
if boundary != 0 and self._disp_buffer is None:
self._boundary_length += boundary - self._boundary_padding
self._boundary_padding = boundary
self._sep = b''.join([data[:boundary], data[:boundary]])
elif boundary != 0:
# boundary not at the begining
value = data if boundary == -1 else data[:boundary - self._boundary_padding]
if not self.execute_handle(self.HANDLE_DATA_SUFFIX, value):
self._disp_buffer += value
if boundary == -1:
# boundary not found, streaming in progress
return
# boundary found, terminate current disposition
self.execute_handle(self.HANDLE_END_SUFFIX)
# process all disposition found in current stream
while boundary != -1:
app_log.debug('processing boundary')
data = data[boundary:]
# find next boundary
boundary = data.find(self._boundary, self._boundary_length)
eoh = data.find(self._sep)
if eoh == -1:
if boundary == -1:
# header and boundary not found, stream probably cut in the midle of header
self._buffer = data
break
# disposition not found because header not found
app_log.debug('invalid disposition header')
continue
# process header
data_header = data[self._boundary_length:eoh]
app_log.debug('header data: %r', data_header)
self._disp_header = HTTPHeaders.parse(data_header.decode('utf-8'))
disp_header = self._disp_header.get('Content-Disposition', '')
disposition, self._disp_params = _parse_header(disp_header)
if disposition != 'form-data':
app_log.warning('invalid multipart/form-data')
continue
self._disp_name = self._disp_params.get('name')
if self._disp_name is None:
app_log.warning('multipart/form-data value missing name')
continue
app_log.debug('disposition name %s', self._disp_name)
# get disposition value and execute begin handler
bod = eoh + len(self._sep)
eod = boundary - self._boundary_padding
if boundary == -1:
value = data[bod:]
else:
value = data[bod:eod]
self._disp_buffer = value
self.execute_handle(self.HANDLE_BEGIN_SUFFIX, value)
if boundary != -1:
# next boundary found, execute end handler
self.execute_handle(self.HANDLE_END_SUFFIX)
def data_received(self, chunk):
"""
Receive chunk of multipart/form-data
:arg chunk: chunk of data
"""
if not self._buffer:
self._buffer = chunk
else:
self._buffer += chunk
while True:
if self.current_phase == PHASE_BOUNDARY:
if len(self._buffer) > len(self._boundary_delimiter):
if self._buffer.startswith(self._boundary_delimiter):
self.current_phase = PHASE_HEADERS
self._buffer = self._buffer[len(self.
_boundary_delimiter):]
elif self._buffer.startswith(self._end_boundary):
result = self.parser_delegate.finish_file()
if is_future(result):
yield result
return
else:
gen_log.warning("Invalid multipart/form-data")
return
else:
# wait for next chunk
return
if self.current_phase == PHASE_HEADERS:
if b"\r\n\r\n" in self._buffer:
headers, remaining_part = self._buffer.split(
b"\r\n\r\n", 1)
if headers:
headers = HTTPHeaders.parse(headers.decode("utf-8"))
else:
gen_log.warning("multipart/form-data missing headers")
return
disp_header = headers.get("Content-Disposition", "")
disposition, disp_params = _parse_header(disp_header)
if disposition != "form-data":
gen_log.warning("Invalid multipart/form-data")
return
self._buffer = remaining_part
self.current_phase = PHASE_BODY
result = self.parser_delegate.start_file(
headers, disp_params)
if is_future(result):
yield result
else:
# wait for all headers for current file
return
if self.current_phase == PHASE_BODY:
if self._boundary_delimiter in self._buffer:
data, remaining_data = self._buffer.split(
self._boundary_delimiter, 1)
self._buffer = remaining_data
result = self.parser_delegate.file_data_received(data[:-2])
if is_future(result):
yield result
self.current_phase = PHASE_HEADERS
result = self.parser_delegate.finish_file()
if is_future(result):
yield result
continue
elif self._end_boundary in self._buffer:
result = self.parser_delegate.file_data_received(
self._buffer.split(self._end_boundary)[0])
if is_future(result):
yield result
result = self.parser_delegate.finish_file()
if is_future(result):
yield result
return
else:
if self._buffer:
result = self.parser_delegate.file_data_received(
self._buffer)
if is_future(result):
yield result
self._buffer = b""
return
async def data_received(self, chunk):
"""Receive chunk of multipart/form-data."""
self._buffer += chunk
while True:
if self.current_phase == PHASE_BOUNDARY:
if len(self._buffer) > len(self._boundary_delimiter):
if self._buffer.startswith(self._boundary_delimiter):
self.current_phase = PHASE_HEADERS
self._buffer = self._buffer[len(self._boundary_delimiter):]
elif self._buffer.startswith(self._end_boundary):
# TODO: Is it possible?
return
else:
gen_log.warning('Invalid multipart/form-data')
return
else:
# Wait for next chunk
return
if self.current_phase == PHASE_HEADERS:
if b"\r\n\r\n" in self._buffer:
headers, remaining_part = self._buffer.split(b"\r\n\r\n", 1)
if headers:
headers = HTTPHeaders.parse(headers.decode(self.encoding))
else:
gen_log.warning('multipart/form-data missing headers')
return
if 'Content-Disposition' in headers:
self.current_field_type = FIELD
disposition_header = headers.get('Content-Disposition', '')
disposition, disposition_params = _parse_header(disposition_header)
if disposition != 'form-data':
gen_log.warning('Invalid multipart/form-data')
return
self.current_phase = PHASE_BODY
self._buffer = remaining_part
self._data_size = 0 # Reset data size counter before enter PHASE_BODY phase
try:
field_name = disposition_params['name'].strip()
except (KeyError, IndexError, AttributeError):
return
field_name = force_text(field_name, self.encoding, errors='replace')
self._field_name = field_name
self._transfer_encoding = headers.get('Content-Transfer-Encoding', '')
if 'filename' in disposition_params:
self.current_field_type = FILE
file_name = disposition_params.get('filename')
if file_name:
file_name = force_text(file_name, self.encoding, errors='replace')
self._file_name = file_name
if file_name:
content_type = headers.get('Content-Type', '')
content_type, content_type_extra = _parse_header(content_type)
charset = content_type_extra.get('charset')
try:
content_length = int(headers.get('Content-Length', 0))
except (TypeError, ValueError):
content_length = None
await self.new_file(
field_name, file_name, content_type, content_length,
charset, content_type_extra)
else:
# Wait for all headers for current file
return
if self.current_phase == PHASE_BODY:
if self._boundary_delimiter in self._buffer:
data, remaining_data = self._buffer.split(self._boundary_delimiter, 1)
self._buffer = remaining_data
await self.receive_data_chunk(data[:-2])
await self.complete_part()
self.current_phase = PHASE_HEADERS
continue
elif self._end_boundary in self._buffer:
remaining_data = self._buffer.split(self._end_boundary)[0]
await self.receive_data_chunk(remaining_data)
await self.complete_part()
return
else:
if self._buffer:
await self.receive_data_chunk(self._buffer)
self._buffer = b""
return
def get_content_type(self):
content_type = self.request.headers.get('Content-Type', 'text/plain')
# noinspection PyProtectedMember
return httputil._parse_header(content_type)