如何迭代 Python 中的字节对象？

Question

I am doing a POST request in Django and I am receiving a bytes object. I need to count the number of times a particular user appears on this object but I am receiving the following error TypeError: 'int' object is not subscriptable . This is what I have so far:

def filter_url(user):
    ''' do the POST request, this code works '''

    filters = {"filter": {
    "filters": [{
        "field": "Issue_Status",
        "operator": "neq",
        "value": "Queued"
    }],
    "logic": "and"}}

    url = "http://10.61.202.98:8081/Dev/api/rows/cat/tickets?"
    response = requests.post(url, json=filters)
    return response

def request_count():
    '''This code requests a POST method and then it stores the result of all the data 
    for user001 as a bytes object in the response variable. Afterwards, a call to the 
    perform_count function is made to count the number of times that user user001 appeared.'''

    user = "user001"
    response = filter_url(user).text.encode('utf-8')
    weeks_of_data = []     
    weeks_of_data.append(perform_count(response))

def perform_count(response):
    ''' This code does not work, int object is not subscriptable '''
    return Counter([k['user_id'] for k in response)

#structure of the bytes object
b'[{"id":1018002,"user_id":"user001","registered_first_time":"Yes", ...}]'

# This is the result that indicates that response is a bytes object.
print(type(response))
<class 'bytes'>

How can I count the number of times that user001 appears by using the the peform_count() function? Which modification does this function require to work?

Answer 1

You do receive bytes, yes, but you then have the requests library decode it (via the response.text attribute, which automatically decodes the data ), which you then re-encode yourself:

response = filter_url(user).text.encode('utf-8')

Apart from just using the response.content attribute instead to avoid the decode -> encode round-trip, you should really just decode the data as JSON :

data = filter_url(user).json()

Now data is a list of dictionaries, and your perform_count() function can operate on that directly.