Hi All, Is there any limitation on converting pyfl...
# troubleshooting
j
Hi All, Is there any limitation on converting pyflink table to pandas? https://nightlies.apache.org/flink/flink-docs-release-1.17/docs/dev/python/table/conversion_of_pandas/
I am consuming the events from kaka topic and doing the window function on the data to capture and aggregate it. After that I can see my results by printing it, you can find it on below screenshot.
However when I convert pyflink table to pandas dataframe and print the dataframe I don't see any results. I wonder where it is going wrong and how to overcome this problem. please see below screenshot of my code.
e
As far as I understand, pandas works only in batch mode and if you connect to kafka, you probably use streaming mode
j
Okay Thank you for your clarifying.
But flink documentation they didn't specify anywhere like that.
I referred same example and able to print and convert to pandas. However when I read the data from kafka topic and transforming the data by storing the results to pyflink table and print it works. But it doesn't convert to pandas dataframe.
e
@Jagan Nalla Hi, I am not really sure, but I think sql tables and tables are different things, check the use of
from_path
here, I believe it passes sql table to table, which then you can pass to pandas here also here
Copy code
orders = t_env.from_path("Orders"
@Jagan Nalla Did that worked?
j
No @Eugenio Gastelum my bad I forgot to reply here.