Stream support for query_data_frame

[kaheicanaan]

I am working on a dataset which contains >10k fields in a measurement. I would like to query data the whole dataset of a given day (or several days) which makes the size of returned dataset very large. Here is my sample flux query:

from(bucket: "_bucket")
  |> range(start: _start, stop: _stop)
  |> filter(fn: (r) =>
    r._measurement == "_measurement"
)

Is it a good idea to add a kwarg stream: bool = False

influxdb-client-python/influxdb_client/client/query_api.py

Line 95 in 5ddf643

def query_data_frame(self, query: str, org=None, data_frame_index: List[str] = None):

and then the API return the generator directly

influxdb-client-python/influxdb_client/client/query_api.py

Line 116 in 5ddf643

_dataFrames = list(_parser.generator())

when stream=True?

[bednar]

Hi @kaheicanaan,

thanks for question.

I think the better way will be add query_data_frame_stream method. It will be consistent with:

• def query(self, query: str, org=None) -> List['FluxTable']:
• def query_stream(self, query: str, org=None) -> Generator['FluxRecord', Any, None]:

Regards