In snowflake, you can do something like:
SELECT
my_event_time,
DATE_TRUNC('minute',my_event_time)::TIME AS minute
FROM table
And it would return something like:
my_event_time | minute
-------------------------------------
2020-08-17 13:23:49.227. | 13:23:00
Removing everything except the actual minute, can this be done in Pyspark df? The date_trunc('minute', ...)
in Pyspark does something else, it doesnt remove the date part.
CodePudding user response:
Use date_format function and pass required time format.
spark.sql("select date_format(current_timestamp,'HH:mm:ss') time").show()
--------
| time|
--------
|10:48:13|
--------
CodePudding user response:
Try this
spark.sql("current_timestamp,minute(current_timestamp)").show()