base64 encoding outputs incorrect ending byte

Question

One mobile app I am trying to scrape accepts b64 encoded requests. I tried the following:

post_data = {
"id": iv, 
"command": "queryDoc",
 "params": {
     "pageNum": "1", 
     "sortFields": "s50:desc", 
     "ciphertext": make_ciphertext(), 
     "devid": "d607f6d3de0f4f68b44aae416592f559", 
     "devtype": "1", 
     "pageSize": "20", 
     "queryCondition": [{"key":"s21","value":"2020"}]
     }
}

post_data_encoded = base64.b64encode(json.dumps(post_data).encode())
print (post_data_encoded)

with requests.post(url, headers=headers, data=post_data_encoded) as req:
    print (req.json())

However, the response reads:

{
    "data": {},
    "ret": {
        "code": 9,
        "msg": "Input byte array has incorrect ending byte at 900"
    }
}

I tried to decode the real requests from my phone browser, and it yielded a dictionary with all the same parameters. What could be going wrong here?

I did look at a similar issue here: Input byte array has incorrect ending byte at 40

Could it be that b64encode in Python outputs differently than java.util.Base64? I appreciate any suggestion.

Can you show a link to the documentation of that API you are using? — mkrieger1, Jun 20 '20 at 15:46
I suspect that you should use `json.dumps(post_data)`, not `str(post_data)`. — mkrieger1, Jun 20 '20 at 15:47
`headers=headers` are you setting content type and encoding? — tdelaney, Jun 20 '20 at 15:52
I tried `json.dumps(post_data)`, but the result is the same. — Louis, Jun 20 '20 at 15:54
Just added the full code and request URL. Thanks for looking into this! — Louis, Jun 20 '20 at 15:56

score 0 · Answer 1 · answered Jun 20 '20 at 18:19

0

I found the issue. Base64 works as expected. The problem is my request. Instead of posting post_data_encoded as data in the request, I created a dictionary {'request':post_data_encoded} and the result came out as expected.

answered Jun 20 '20 at 18:19

Louis

63
5

base64 encoding outputs incorrect ending byte

1 Answers1