Python API - Create/Push New Dataset

[jmcgi2]

I am pulling 2 different tables from Google Big Query. Then I am joining them together and manipulating the data. Finally, I need to push this new data to Power BI using my Python Script. I do NOT want to put my Pyhton Script into Power BI and run it that way.

Thank you in advance for any information!

[lbendlin]

All you need to do is call this API from your Python script. Doesn't even require authentication.

curl --include \
--request POST \
--header "Content-Type: application/json" \
--data-binary "[
{
\"Host\" :\"AAAAA555555\",
\"Timestamp\" :\"2024-09-24T17:15:51.558Z\",
\"Available Memory in MB\" :98.6,
\"Processor Load %\" :98.6,
\"% Free on C:\" :98.6,
\"img\" :\"AAAAA555555\",
\"Mashups\" :98.6,
\"Mashup Memory\" :98.6,
\"Gateway Memory\" :98.6,
\"Proxy ok\" :98.6,
\"% Free on D:\" :98.6,
\"Users\" :98.6
}
]" \
"https://api.powerbi.com/beta/<tenant id>/datasets/<dataset id>/rows?experience=power-bi&key=<api key>"

Note there is a slight difference between Push datasets and Streaming Datasets.

[jmcgi2]

Thank you for the reponse. I work for Suabru and I believe I have to go by their security, would what you provided still work? I want to do it the correct way so IT does not get mad. Below is my code from reading the Power BI Python API. Let me know what you think!

[lbendlin]

you are missing the /rows  part of the URL.

[jmcgi2]

 Ok I see that now. Could you let me know where to insert it into my string? I do not have an API Key, unless you are calling the API Key like the client_id or something. Thanks again!

dataset_url = f'https://api.powerbi.com/v1.0/myorg/groups/{workspace_id}/datasets'

[lbendlin · Accepted solution]

You get the API key when you create your Streaming dataset.   Push datasets work a little different.