notfilippo commented on issue #11513:
URL: https://github.com/apache/datafusion/issues/11513#issuecomment-2245251759
> Now, i have add_one() function that can take 64-bit integer values and add
+1 to them. The +1 operation is perfectly valid operation for i64 -- it's valid
for sql long/bigint type. It's valid also for my_duration(nanos), but it's not
valid for my_duration(micros), since it produces unaligned value (not divisible
by 1000).
@findepi -- If I understand correctly your example is about a the logical
type `my_duration({micro | nano})` with the underlying physical type `int64`.
In this case the type `my_duration(micro)` is logically different from
`my_duration(nano)` (both logically different from an `int64`).
Functions that have specialised implementations for `my_duration`, specified
in their signature, would have to handle either case differently. Instead,
functions such as `add_one(int64)` would have a signature accepting only
logical `int64`s and not `my_duration`.
Hypothetically, if `add_one` was to be edited to support `my_duration` (by
accepting any numeric type) it would either have to provide a specific
implementation for the `my_duration` logical type or it would've to materialise
the array to a known numeric type like `int64`. The materialisation wouldn't
necessarily match the underlying physical type and for this example I think it
would make sense to materialise as an `int64` array with the number of
nanoseconds (physical type) multiplied by the unit of the duration.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
