CODEWITHSUNDEEP
pythonpython-3.xcrudpymongo-3.x
Souvik Ray
Souvik Ray

Reputation: 3028

Unable to update data in pymongo

I have a JSON POST data that a user is going to send me every time to fetch some data from a third party service.I plan to cache the data based on a scope id so that I don't keep inserting the data each time the user requests for something.Futhermore I am keeping a time stamp for each user request.Below is the POST data that user is going to send me everytime.

{
        "scope_id": "user1",
        "tool_id": "appdynamics",
        "api_id": "get metrics",
        "input_params": {"user": "myuser", "pwd": "mypwd", "acc_id": "myaccount", "app_id": "TestApp", "metric-path": "ars", 
        "time-range-type": "BEFORE_NOW", "duration-in-mins": 10},
        "output_filters": {}
}

Below is the code snippet to handle the insertion of data

def post(self):
    data = ServiceAPI.parser.parse_args()
    print("First data", data)
    scope_id = data["scope_id"]
    tool_id = data["tool_id"]
    api_id = data["api_id"]
    input_params = data["input_params"]
    output_filter = data["output_filter"]

    if all([scope_id, tool_id, api_id]) and all(input_params.values()): 
        check_id = [j for i in users.find({}) for j in i if j == scope_id]
        if check_id and check_id[0] == scope_id:
            users.update({scope_id: [tool_id, api_id, input_params]},
                         {scope_id: [tool_id, api_id, input_params], 
                              "timestamp": datetime.now().strftime('%Y-%m-%d %H:%M:%S')}, upsert=True)
        else:
            users.insert_one(
                {scope_id: [tool_id, api_id, input_params],
                "timestamp": datetime.now().strftime('%Y-%m-%d %H:%M:%S')})

Here the update statement works great if the user request is exactly the same as last time but makes a new entry if the user demands a new information for example in the POST request api_id = "get logs" when ideally it should have updated the user's data with the latest one.

For the first time when the user makes a POST request, below is the data that gets stored in my database

[{'user1': ['appdynamics', 'get metrics', {'pwd': 'mypwd', 'metric-path': 'ars', 'user': 'myuser', 'time-range-type': 'BEFORE_NOW', 'acc_id': 'myaccount', 'app_id': 'TestApp', 'duration-in-mins': 10}], 'timestamp': '2018-03-24 21:49:28', '_id': ObjectId('5ab67a901899db6d8a266558')}]

Now I make the same request again, it ensures no new entry is made since its made by the same scope id

However now if the user requests some new information for example

{
        "scope_id": "user1",
        "tool_id": "appdynamics",
        "api_id": "get logs",
        "input_params": {"user": "myuser", "pwd": "mypwd", "acc_id": "myaccount", "app_id": "TestApp", "metric-path": "ars", 
        "time-range-type": "BEFORE_NOW", "duration-in-mins": 10},
        "output_filters": {}
}

Notice I have changed "api_id": "get logs", it makes a new entry instead of just modifying the existing data in my database.Here is the data now

[{'user1': ['appdynamics', 'get metrics', {'pwd': 'mypwd', 'metric-path': 'ars', 'user': 'myuser', 'time-range-type': 'BEFORE_NOW', 'acc_id': 'myaccount', 'app_id': 'TestApp', 'duration-in-mins': 10}], 'timestamp': '2018-03-24 21:49:28', '_id': ObjectId('5ab67a901899db6d8a266558')}, {'user1': ['appdynamics', 'get logs', {'pwd': 'mypwd', 'metric-path': 'ars', 'user': 'myuser', 'time-range-type': 'BEFORE_NOW', 'acc_id': 'myaccount', 'app_id': 'TestApp', 'duration-in-mins': 10}], 'timestamp': '2018-03-24 21:55:29', '_id': ObjectId('5ab67bf9089b16e9e77037f4')}]

So here the update seems to fail.What could be going wrong?

Note: This is a flask app and I suggest not to get into the details of the implementation.I just need to update the given data based on the scope id each time a user makes a request irrespective of whether it is the same request or a different one.

Upvotes: 0

Views: 136

Answers (1)

Bernie Hackett
Bernie Hackett

Reputation: 8909

You are passing upsert=True to update(). Upsert tells MongoDB to update an existing document if one matching the query is found, insert a new document otherwise. The first parameter to update() is a query filter to find documents to apply the update to. The update query filter where api_id == "get logs" isn't matching any existing document, so a new document is being created.

Upvotes: 1

Related Questions