def test_from_response(self):
response = mock.NonCallableMagicMock(spec=requests.Response)
response.headers = {
'content-type': 'multipart/related; boundary="samp1"'
}
cnt = io.BytesIO()
cnt.write(b'\r\n--samp1\r\n')
cnt.write(b'Header-1: Header-Value-1\r\n')
cnt.write(b'Header-2: Header-Value-2\r\n')
cnt.write(b'\r\n')
cnt.write(b'Body 1, Line 1\r\n')
cnt.write(b'Body 1, Line 2\r\n')
cnt.write(b'--samp1\r\n')
cnt.write(b'\r\n')
cnt.write(b'Body 2, Line 1\r\n')
cnt.write(b'--samp1--\r\n')
response.content = cnt.getvalue()
decoder_2 = MultipartDecoder.from_response(response)
assert decoder_2.content_type == response.headers['content-type']
assert (
decoder_2.parts[0].content == b'Body 1, Line 1\r\nBody 1, Line 2'
)
assert decoder_2.parts[0].headers[b'Header-1'] == b'Header-Value-1'
assert len(decoder_2.parts[1].headers) == 0
assert decoder_2.parts[1].content == b'Body 2, Line 1'
def test_from_responsecaplarge(self):
response = mock.NonCallableMagicMock(spec=requests.Response)
response.headers = {
'content-type': 'Multipart/Related; boundary="samp1"'
}
cnt = io.BytesIO()
cnt.write(b'\r\n--samp1\r\n')
cnt.write(b'Header-1: Header-Value-1\r\n')
cnt.write(b'Header-2: Header-Value-2\r\n')
cnt.write(b'\r\n')
cnt.write(b'Body 1, Line 1\r\n')
cnt.write(b'Body 1, Line 2\r\n')
cnt.write(b'--samp1\r\n')
cnt.write(b'\r\n')
cnt.write(b'Body 2, Line 1\r\n')
cnt.write(b'--samp1--\r\n')
response.content = cnt.getvalue()
decoder_2 = MultipartDecoder.from_response(response)
assert decoder_2.content_type == response.headers['content-type']
assert (
decoder_2.parts[0].content == b'Body 1, Line 1\r\nBody 1, Line 2'
)
assert decoder_2.parts[0].headers[b'Header-1'] == b'Header-Value-1'
assert len(decoder_2.parts[1].headers) == 0
assert decoder_2.parts[1].content == b'Body 2, Line 1'
def _eval_adhoc(self, lang, line, cell):
self._check_for_session()
valid_lang = ("xquery", "javascript")
if lang not in valid_lang:
raise TypeError("lang can only be " + ", ".join(valid_lang))
args = shlex.split(line, comments = True)
self._check_invalid_args(args, ("-raw", "-dataframe"))
output_opt = "default"
if "-raw" in args:
output_opt = "raw"
if "-dataframe" in args:
output_opt = "dataframe"
# execute code on MarkLogic
url = self.session_baseurl + "/v1/eval"
body = {
lang: cell
}
output = []
response = self.session.post(url, data = body)
if response.status_code == 200:
headers = { header.lower(): value for (header, value) in response.headers.items() }
# as per docs, /v1/eval should always return multipart/mixed
if "content-type" in headers and headers["content-type"].startswith("multipart/mixed"):
mp_data = MultipartDecoder.from_response(response)
for part in mp_data.parts:
part_headers = { header.decode("utf-8").lower(): value.decode("utf-8") for (header, value) in part.headers.items() }
part_content_type = part_headers["content-type"]
part_x_primitive = part_headers["x-primitive"]
part_content = part.content.decode("utf-8")
data = part_content if output_opt == "raw" else self._parse_content(part_content, part_content_type, part_x_primitive)
output.append(data)
else:
raise EvalError("Request to %s returned %s %s: %s" % (url, response.status_code, response.reason, response.text))
output = output[0] if len(output) == 1 else output
if output_opt == "dataframe":
try:
output = pandas.DataFrame(output) # try to fit into a dataframe, may raise an exception depending on result's shape
except:
raise CoercionError("Unable to fit response into a DataFrame")
return {
"status_code": response.status_code,
"output": output,
"output_opt": output_opt,
"args": args
}
def parse_multipart(response: Response_) -> Sequence[Any]:
"""
RFC 2045 describes the format of an Internet message body containing a MIME message. The
body contains one or more body parts, each preceded by a boundary delimiter line, and the
last one followed by a closing boundary delimiter line. After its boundary delimiter line,
each body part then consists of a header area, a blank line, and a body area.
HTTP/1.1 200 OK
Server: Apache/2.0.13
Date: Fri, 22 OCT 2004 12:03:38 GMT
Cache-Control: private
RETS-Version: RETS/1.7.2
MIME-Version: 1.0
Content-Type: multipart/parallel; boundary="simple boundary"
--simple boundary
Content-Type: image/jpeg
Content-ID: 123456
Object-ID: 1
<binary data>
--simple boundary
Content-Type: text/xml
Content-ID: 123457
Object-ID: 1
<RETS ReplyCode="20403" ReplyText="There is no listing with that ListingID"/>
--simple boundary--
"""
multipart = MultipartDecoder.from_response(response, response.encoding)
# We need to decode the headers because MultipartDecoder returns bytes keys and values,
# while requests.Response.headers uses str keys and values.
for part in multipart.parts:
part.headers = _decode_headers(part.headers, response.encoding)
def parse_multipart(parts):
for part in parts:
try:
yield parse_response(part)
except RetsApiError as e:
if e.reply_code != 20403: # No object found
raise
return tuple(parse_multipart(multipart.parts))
def _parse_multipart(response: ResponseLike,
parser: Parser) -> Sequence[Object]:
"""
RFC 2045 describes the format of an Internet message body containing a MIME message. The
body contains one or more body parts, each preceded by a boundary delimiter line, and the
last one followed by a closing boundary delimiter line. After its boundary delimiter line,
each body part then consists of a header area, a blank line, and a body area.
HTTP/1.1 200 OK
Server: Apache/2.0.13
Date: Fri, 22 OCT 2004 12:03:38 GMT
Cache-Control: private
RETS-Version: RETS/1.7.2
MIME-Version: 1.0
Content-Type: multipart/parallel; boundary="simple boundary"
--simple boundary
Content-Type: image/jpeg
Content-ID: 123456
Object-ID: 1
<binary data>
--simple boundary
Content-Type: text/xml
Content-ID: 123457
Object-ID: 1
<RETS ReplyCode="20403" ReplyText="There is no listing with that ListingID"/>
--simple boundary--
"""
encoding = response.encoding or DEFAULT_ENCODING
multipart = MultipartDecoder.from_response(response, encoding)
# We need to decode the headers because MultipartDecoder returns bytes keys and values,
# while requests.Response.headers uses str keys and values.
for part in multipart.parts:
part.headers = _decode_headers(part.headers, encoding)
objects = (_parse_body_part(part, parser) for part in multipart.parts)
return tuple(object_ for object_ in objects if object_ is not None)
def test_non_multipart_response_fails(self):
jpeg_response = mock.NonCallableMagicMock(spec=requests.Response)
jpeg_response.headers = {'content-type': 'image/jpeg'}
with pytest.raises(NonMultipartContentTypeException):
MultipartDecoder.from_response(jpeg_response)
def test_non_multipart_response_fails(self):
jpeg_response = mock.NonCallableMagicMock(spec=requests.Response)
jpeg_response.headers = {'content-type': 'image/jpeg'}
with pytest.raises(NonMultipartContentTypeException):
MultipartDecoder.from_response(jpeg_response)
