Open mbecker opened 3 years ago
@Jackie-Jiang @npawar shall we consider providing built-in formats to avoid those date format conversion difficulties?
The TIMESTAMP
data type always use yyyy-MM-dd HH:mm:ss.SSS
, but this won't avoid the time format conversion if we need the result to be in another format.
@mbecker From the Presto SQL query, can you try setting the input format (second argument) to match the field spec format?
Hi,
thanks for your help!
Sorry, in my original post I mixed some tools like using for example "Presto" (vs. only "Superset").
So, my original use case is that the Kafka Message has this attribute "time" with the RFC 3339 format. I'm ingesting it as a string type and creating the "Pinot" table as described above. In "Superset" I'm defining the dataset and setting the attribute "time" as a time-field with the DATETIME format as described above and with the new format yyyy-MM-dd HH:mm:ss.SSS
as commented by @Jackie-Jiang .
Because this is not the repo for "Superset", I'm just asking how to check that the ingestion and the table-format with a STRING-field for the dateTimeFieldSpecs
in Pinot is correct :-)
Of course, any other help would be great too ;-)
Thank you very much.
@mbecker If you are just trying to validate if the event is ingested in Pinot correctly and the time format is store correctly in Pinot, then you can simply run a pinot query to do so:
select _time from <tableName>
Hi,
I'm having a time format in the Kafka Messages as follows:
I'm using the following schema for the date:
(I'm using a Java format here)
How can I check that the ingest in Pinot is working with the specified date time field?
Then in a Presto SQL the DATETIMECONVERT doesn't work as expected and returns an error:
(Here I'm using a python format? as described in superset)
The returned error is as follows:
How can I use the column "_time" with the RFC 3339 format as the date time field in Presto / Superset?
Thanks for your help and cheers!