sending compressed numpy array (zlib) to flask server with post request [python] - python-3.x

Question

I am trying to send a compressed numpy array (compressed with zlib) to the flask server with post request, but the compressed bytes object is getting changed in the server end. How to properly send the bytes object with requests post request so that I can decompress on the server end?
server.py
from flask import Flask
from flask_restful import Resource, Api, reqparse
import json
import numpy as np
import base64
# compression
import zlib
app = Flask(__name__)
api = Api(app)
parser = reqparse.RequestParser()
parser.add_argument('imgb64')
class Predict(Resource):
def post(self):
data = parser.parse_args()
if data['imgb64'] == "":
return {
'data':'',
'message':'No file found',
'status':'error'
}
img = data['imgb64']
print('rec')
# decompress
print(type(img))
print(img)
dec = zlib.decompress(img) # this gives me error
if img:
pass
return json.dumps({
'data': 'done',
'message':'darknet processed',
'status':'success'
})
return {
'data':'',
'message':'Something when wrong',
'status':'error'
}
api.add_resource(Predict,'/predict')
if __name__ == '__main__':
app.run(debug=True, host = '0.0.0.0', port = 5000, threaded=True)
client.py
import numpy as np
import base64
import zlib
import requests
frame = np.random.randint(0,255,(5,5,3)) # dummy rgb image
# compress
data = zlib.compress(frame)
print('b64 encoded')
print(data)
print(len(data))
print(type(data))
r = requests.post("http://127.0.0.1:5000/predict", data={'imgb64' : data}) # sending compressed numpy array
This gives me the following error:
TypeError: a bytes-like object is required, not 'str'
So, I tried to convert the string to bytes object:
dec = zlib.decompress(img.encode()) # this gives me error
But, this one also gives me an error:
zlib.error: Error -3 while decompressing data: incorrect header check
I tried with other encodings, they also failed.
One thing I noticed is, when I print the compressed bytes in the client end, it reads:
b'x\x9c-\xcf?J\x82q\x00\x06\xe0\x0ftrVP\\C\xf4\x06\x9268\t\xcdR\x8ej.F\xa0\xe0\xd0\xa6\xa3\xe0V\x07\x10\x1cL\xc8\xd1\x03\xd4\xe4\t\x0c\x12\x84\xb6D\x0c#\xbc\x80O\xf0\x1b\x9e\xf5\xfdS\x89\xa2h\xcf\x9a\x03\xef\xc4\xf8cF\x92\r\xbf4i\x11g\xc83\x0f\x8c\xb9\xa2#\x8e\x1bn\xd91g\xc0\x91%\xd7\xdc\xf3M\x83<i:L\xa8\xf1\x19\xfa\xffw\xfd\xf0\xc5\x94:O\x9cH\x85\xcc6#\x1e\xc3\xf6\x05\xe5\xa0\xc7\x96\x04]J\\\x90\xa1\x1f~Ty\xe1\x8d\x15w|P\xe4\x95K\xb2!\xe3\x0cw)%I'
But on the server end, the received string is completely different:
�4ig�3���#�n�1g��%���M�<i:L����w��Ŕ:O�H��6#���ǖ]J\��~Ty�w|P�K�!�w)%I
I also tried to send the bytes as string, by
r = requests.post("http://127.0.0.1:5000/predict", data={'imgb64' : str(data)})
But, I can't decompress the data on the server end.

Answer 1

It seems, I can't send the zlib compressed bytes directly, so I used base64 to encode the data into ascii string.
So, in summary this worked for me, numpy array/any non-string data -> zlib compression -> base64 encode -> post request -> flask -> base64 decode -> zlib decompress
client.py
import numpy as np
import base64
import zlib
import requests
frame = np.random.randint(0,255,(5,5,3)) # dummy rgb image
# compress
data = zlib.compress(frame)
data = base64.b64encode(data)
data_send = data
data2 = base64.b64decode(data)
data2 = zlib.decompress(data2)
fdata = np.frombuffer(data2, dtype=np.uint8)
print(fdata)
r = requests.post("http://127.0.0.1:5000/predict", data={'imgb64' : data_send})
server.py
from flask import Flask
from flask_restful import Resource, Api, reqparse
import json
import numpy as np
import base64
# compression
import zlib
import codecs
app = Flask(__name__)
api = Api(app)
parser = reqparse.RequestParser()
parser.add_argument('imgb64', help = 'type error')
class Predict(Resource):
def post(self):
data = parser.parse_args()
#print(data)
if data['imgb64'] == "":
return {
'data':'',
'message':'No file found',
'status':'error'
}
#img = open(data['imgb64'], 'r').read() # doesn't work
img = data['imgb64']
data2 = img.encode()
data2 = base64.b64decode(data2)
data2 = zlib.decompress(data2)
fdata = np.frombuffer(data2, dtype=np.uint8)
print(fdata)
if img:
return json.dumps({
'data': 'done',
'message':'darknet processed',
'status':'success'
})
return {
'data':'',
'message':'Something when wrong',
'status':'error'
}
api.add_resource(Predict,'/predict')
if __name__ == '__main__':
app.run(debug=True, host = '0.0.0.0', port = 5000, threaded=True)