I also suspect that you need to do this:

payload = {"key1": "value1", "key2": 2}
json_payload = json.dumps(payload)
response = <http://session.post|session.post>(url, data=json_payload)

especially given the HTTP 400 you're getting.

I almost always use Postman as my first step in debugging for any Python API project I do.

From looking at your query one issue is your identifiers are single-quoted but they should be double-quoted, like this:

"stocks_intervals"

and

"__time"

instead of

'stocks_intervals'

and

'__time'

The new asynchronous Python function that partially works is as follows, with the problematic lines in bold: async def query_druid_async(ticker, interval): druid_host = 'localhost' druid_port = 8888 # Adjust the port as needed druid_scheme = 'http' # or 'https' if using SSL url = f"{druid_scheme}//{druid host}{druid_port}/druid/v2/sql" headers = { 'Accept': 'application/json', 'Content-Type': 'application/json' } # Query does not work if I use the f-string to dynamically change "ticker" and "interval" as below *query_string = f"SELECT * FROM \"stocks_intervals\" WHERE \"symbol\" = '{ticker}' AND \"__time\" > TIMESTAMP '{interval}'"* # Query does work if I have static query parameters as below *query_string = "SELECT * FROM \"stocks_intervals\" WHERE \"symbol\" = 'AAPL' AND \"__time\" > TIMESTAMP '2024-01-22 125000'"* query = json.dumps({ "query": query_string, "header": 'false', "typesHeader": 'false', "sqlTypesHeader": 'false' }) async with aiohttp.ClientSession() as session: print(f"Query String: {query}") async with session.post(url, headers=headers, data=query) as response: print(f"Response status: {response.status}") if response.content_type == 'application/json': data = await response.json() #print(data) return pd.DataFrame(data) else: response_text = await response.text() print(f"Unexpected content type: {response.content_type}") print(f"Response text: {response_text}") return pd.DataFrame()

I just whipped up this query to run in the router against the example datasource

wikipedia

SELECT * FROM "wikipedia" WHERE "page" = 'Salo Toraut' AND "__time" > TIMESTAMP '2016-06-01'

and that seems to work, so not sure what error you might be seeing with your example. I usually have POSTMAN set up to run queries and I heavily rely on the router to help with debugging, especially complex issues. I try to get something working, then make changes to isolate an issue.