iRakson opened a new pull request, #57119:
URL: https://github.com/apache/spark/pull/57119

   ### What changes were proposed in this pull request?
   Extend `StatFunctions.multipleApproxQuantiles` to accept TimeType columns. 
TimeType columns are casted to DecimalType(14,9) before being casted to 
DoubleType to preserve nanosecond-of-day precision. 
`StatFunctions.multipleApproxQuantiles` returns `Seq[Seq[Double]]`, hence 
TimeType column's quantiles are returned as seconds since midnight. 
   
   Extended `StatFunctions.summary` to accept TimeType columns. `avg` and 
`stddev` does not operate on TimeType columns and hence their output is `NULL`.
   
   
   ### Why are the changes needed?
   The DataFrame stat APIs do not handle TIME. 
StatFunctions.multipleApproxQuantiles requires NumericType and casts to 
DoubleType, and summary includes only numeric/string columns. SQL 
approx_percentile already supports TIME , but df.stat.approxQuantile / 
df.summary() / df.describe() do not route TIME there. 
   
   ### Does this PR introduce _any_ user-facing change?
   Yes. `df.stat. approxQuantile` now accepts TimeType columns. df.summary() 
and df.describe() also accepts TimeType columns.
   
   
   ### How was this patch tested?
   UTs added.
   
   
   ### Was this patch authored or co-authored using generative AI tooling?
   No


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to