damienzeng73 / redash-api-client Goto Github PK

For example, for create_query:

It is probably good that the payload always contains a few parameters (e.g., data_source_id, name, etc...), but what about all other potential (and useful) parameters? For example tags, etc, Using kwargs would be a great way to make your API much more general.

For example for create_query:

def create_query(self, ds_id: int, name: str, qry: str, desc: str="", with_results: bool=True, **kwargs):
        payload = {
            "data_source_id": ds_id,
            "name": name,
            "query": qry,
            "description": desc,
            **kwargs
        }

        res = self.post('queries', payload)

        if with_results:
            self.generate_query_results(ds_id, qry)

        return res

Then you can use it with

redash.create_query(1, "my query", "select 1 from table", "description", with_results=False, tags=["mytag"], options={} )

If you remove options (i would) that would be a breaking change, so that's probably your decision.
But all that would be very easy to add that to all create_XXX functions.

When using create_visualizations() with some kind of visualizations type (eg: counter, funnel, etc), the API gives errors regarding x and y axis, even when they are not supposed to be given for that kind of visualizations.

https://github.com/damnee562/redash-api-client/blob/master/redashAPI/client.py#L220

I think Auto Detect is easier to use than Category.
For Auto Detect, specify -.
Can I change it if there is no special reason?

Currently there is no ability to set the parameterMappings field when adding a widget to a dashboard (i.e. setting whether the parameter is a dashboard or widget-level parameter) with the add_widget function. I would propose the attached changes to the function to add that capability.
add_widget.txt

Can you share an example of how to pass a parameter to the query? This is done using the {{ }} in the UI.

Hi,
I am using Redash.generate_query_result to fetch data from Redash.
It behaves randomly that sometimes it gets the result and sometimes throws an error with Status: 2. Attaching screenshot.

The same query runs in Redash successfully hence, no issues with server.

Currently using Python's Try, Exception to complete the job, but it sometimes take forever to execute.
Is it a known issue? What can be the possible reasons for it and what can be the solution?

Hey,

is there any max payload that someone can retrieve using the API you have created?
I have been using the library and have the following issue:

We define a start_date and an end_date:

• start = convert_str_datetime_to_utc("2022-05-01 00:00:00")
• end = convert_str_datetime_to_utc("2022-12-01 23:59:59")

The weird thing is that the query runs ok when the start_date is 2022-06-01 00:00:00 but when the start_date is 2022-05-01 00:00:00 the response you get is the following:

File ~/miniconda3/envs/opap-kpis/lib/python3.9/site-packages/redashAPI/client.py:125, in RedashAPIClient.query_and_wait_result(self, ds_id, query, timeout)
121 raise Exception('Polling timeout.')
123 time.sleep(0.2)
--> 125 return self.get(f'query_results/{query_result_id}')

File ~/miniconda3/envs/opap-kpis/lib/python3.9/site-packages/redashAPI/client.py:18, in RedashAPIClient.get(self, uri)
15 res = self.s.get(f"{self.host}/api/{uri}")
17 if res.status_code != 200:
---> 18 raise Exception(f"[GET] /api/{uri} ({res.status_code})")
20 return res

Exception: [GET] /api/query_results/2196159 (502)

Also I have noticed the same behaviour when adding some columns to a query that is running ok.

I was thinking that this is due to a max payload or a maximum number of lines the redash_api can fetch.

I should also mention here that the query is running on the redash client we have set up on a Kubernetes cluster.

Could you please look into this?

Thanks in advance!