def test_tolerant_magic_dict_get_first(self):
magic_dict = TolerantMagicDict()
test_list = ["b", "c", "d"]
for item in test_list:
magic_dict.add(random.choice(["a", "A"]), item)
self.assertEqual(magic_dict.get_first("A"), test_list[0])
def test_tolerant_magic_dict_delitem_method(self):
magic_dict = TolerantMagicDict()
magic_dict.add("a", "b")
magic_dict.add("A", "c")
magic_dict.add("a", "d")
del magic_dict["a"]
self.assertNotIn("a", magic_dict)
self.assertNotIn("A", magic_dict)
def _parse_origin_path(self):
parsed_url = urllib.parse.urlparse(self.origin_path)
self._path = parsed_url.path
link_args = TolerantMagicDict()
for (query_name, query_value) in urllib.parse.parse_qsl(
parsed_url.query):
link_args.add(query_name, query_value)
self._link_args = link_args
def parse(content_type: str, data: bytes) -> "HTTPMultipartBody":
"""
Parse HTTP v1 Multipart Body.
It will raise an Error during the parse period if parse failed.
"""
body_args = HTTPMultipartBody()
if not content_type.lower().startswith("multipart/form-data"):
raise ProtocolError("Unknown content-type.")
for field in content_type.split(";"): # Search Boundary
if field.find("boundary=") == -1:
continue
boundary = ensure_bytes(field.split("=")[1])
if boundary.startswith(b'"') and boundary.endswith(b'"'):
boundary = boundary[1:-1]
break
else:
raise ProtocolError("Cannot Find Boundary.")
full_boundary = b"--" + boundary
body_content = data.split(full_boundary + b"--")[0]
full_boundary += _CRLF_BYTES_MARK
splitted_body_content = body_content.split(full_boundary)
for part in splitted_body_content:
if not part:
continue
initial, content = part.split(_CRLF_BYTES_MARK * 2)
headers = HTTPHeaders.parse(initial)
disposition = headers.get_first("content-disposition")
disposition_list = []
disposition_dict = TolerantMagicDict()
for field in disposition.split(";"): # Split Disposition
field = field.strip() # Remove Useless Spaces.
if field.find("=") == -1: # This is not a key-value pair.
disposition_list.append(field)
continue
key, value = field.split("=")
if value.startswith('"') and value.endswith('"'):
value = value[1:-1]
disposition_dict.add(key.strip().lower(), value.strip())
if disposition_list[0] != "form-data":
raise ProtocolError("Cannot Parse Body.")
# Mixed form-data will be supported later.
content = content[:-2] # Drop CRLF Mark
if "filename" in disposition_dict.keys():
body_args.files.add(
disposition_dict.get_first("name", ""),
HTTPMultipartFileField(
fieldname=disposition_dict.get_first("name", ""),
filename=disposition_dict.get_first("filename", ""),
content=content,
content_type=headers.get_first(
"content-type", "application/octet-stream"),
headers=headers,
encoding=headers.get_first("content-transfer-encoding",
"binary")))
else:
try:
content = content.decode()
except UnicodeDecodeError:
pass
body_args.add(disposition_dict.get_first("name", ""), content)
return body_args
def __init__(self, *args, **kwargs):
self.files = TolerantMagicDict()
TolerantMagicDict.__init__(self, *args, **kwargs)
class HTTPMultipartBody(TolerantMagicDict):
"""
HTTPBody class, based on TolerantMagicDict.
It has not only all the features from TolerantMagicDict, but also
can parse and make HTTP Body.
"""
def __init__(self, *args, **kwargs):
self.files = TolerantMagicDict()
TolerantMagicDict.__init__(self, *args, **kwargs)
@staticmethod
def parse(content_type: str, data: bytes) -> "HTTPMultipartBody":
"""
Parse HTTP v1 Multipart Body.
It will raise an Error during the parse period if parse failed.
"""
body_args = HTTPMultipartBody()
if not content_type.lower().startswith("multipart/form-data"):
raise ProtocolError("Unknown content-type.")
for field in content_type.split(";"): # Search Boundary
if field.find("boundary=") == -1:
continue
boundary = ensure_bytes(field.split("=")[1])
if boundary.startswith(b'"') and boundary.endswith(b'"'):
boundary = boundary[1:-1]
break
else:
raise ProtocolError("Cannot Find Boundary.")
full_boundary = b"--" + boundary
body_content = data.split(full_boundary + b"--")[0]
full_boundary += _CRLF_BYTES_MARK
splitted_body_content = body_content.split(full_boundary)
for part in splitted_body_content:
if not part:
continue
initial, content = part.split(_CRLF_BYTES_MARK * 2)
headers = HTTPHeaders.parse(initial)
disposition = headers.get_first("content-disposition")
disposition_list = []
disposition_dict = TolerantMagicDict()
for field in disposition.split(";"): # Split Disposition
field = field.strip() # Remove Useless Spaces.
if field.find("=") == -1: # This is not a key-value pair.
disposition_list.append(field)
continue
key, value = field.split("=")
if value.startswith('"') and value.endswith('"'):
value = value[1:-1]
disposition_dict.add(key.strip().lower(), value.strip())
if disposition_list[0] != "form-data":
raise ProtocolError("Cannot Parse Body.")
# Mixed form-data will be supported later.
content = content[:-2] # Drop CRLF Mark
if "filename" in disposition_dict.keys():
body_args.files.add(
disposition_dict.get_first("name", ""),
HTTPMultipartFileField(
fieldname=disposition_dict.get_first("name", ""),
filename=disposition_dict.get_first("filename", ""),
content=content,
content_type=headers.get_first(
"content-type", "application/octet-stream"),
headers=headers,
encoding=headers.get_first("content-transfer-encoding",
"binary")))
else:
try:
content = content.decode()
except UnicodeDecodeError:
pass
body_args.add(disposition_dict.get_first("name", ""), content)
return body_args
def assemble(self) -> Tuple[bytes, str]:
"""
Generate HTTP v1 Body to bytes.
It will return the body in bytes and the content-type in str.
"""
body = b""
boundary = "----------FutureFinityFormBoundary"
boundary += ensure_str(security.get_random_str(8)).lower()
content_type = "multipart/form-data; boundary=" + boundary
full_boundary = b"--" + ensure_bytes(boundary)
for field_name, field_value in self.items():
body += full_boundary + _CRLF_BYTES_MARK
if isinstance(field_value, str):
body += b"Content-Disposition: form-data; "
body += ensure_bytes("name=\"%s\"\r\n" % field_name)
body += _CRLF_BYTES_MARK
body += ensure_bytes(field_value)
body += _CRLF_BYTES_MARK
else:
raise ProtocolError("Unknown Field Type")
for file_field in self.files.values():
body += full_boundary + _CRLF_BYTES_MARK
body += file_field.assemble()
body += full_boundary + b"--" + _CRLF_BYTES_MARK
return body, content_type
def __str__(self) -> str:
# Multipart Body is not printable.
return object.__str__(self)
def __repr__(self) -> str:
# Multipart Body is not printable.
return object.__repr__(self)
def copy(self) -> "HTTPMultipartBody":
raise ProtocolError("HTTPMultipartBody is not copyable.")
__copy__ = copy
def test_tolerant_magic_dict_copy(self):
magic_dict = TolerantMagicDict()
magic_dict.add("a", "b")
magic_dict.add("C", "D")
copied_magic_dict = magic_dict.copy()
self.assertEqual(magic_dict, copied_magic_dict)
def test_tolerant_magic_dict_str_method(self):
magic_dict = TolerantMagicDict()
magic_dict.add("A", "b")
self.assertEqual(str(magic_dict), "TolerantMagicDict([('a', 'b')])")
def test_tolerant_magic_dict_getitem_method(self):
magic_dict = TolerantMagicDict()
magic_dict.add("a", "b")
magic_dict.add("a", "c")
magic_dict.add("a", "d")
self.assertEqual(magic_dict["A"], "b")
def test_tolerant_magic_dict_setitem_method(self):
magic_dict = TolerantMagicDict()
magic_dict["A"] = "b"
self.assertListEqual(magic_dict.get_list("a"), ["b"])
def test_tolerant_magic_dict_add(self):
magic_dict = TolerantMagicDict()
self.assertNotIn("a", magic_dict)
magic_dict.add("A", "b")
self.assertIn("a", magic_dict)
self.assertEqual("b", magic_dict["a"])
def __init__(self, files: typing.List=[HTTPMultipartFileField],
*args, **kwargs):
self.files = TolerantMagicDict()
TolerantMagicDict.__init__(self, *args, **kwargs)
