Thanks Mich figured that might be the case, regardless, appreciate the help
:)
On Thu, Sep 7, 2023 at 8:36 PM Mich Talebzadeh
wrote:
> Hi,
>
> As far as I am aware there is no Spark or JVM setting that can make Spark
> assume a different timezone during the initial load from Parquet as Parquet
>
Hi,
As far as I am aware there is no Spark or JVM setting that can make Spark
assume a different timezone during the initial load from Parquet as Parquet
files store timestamps in UTC. The timezone conversion can be done (as I
described before) after the load.
HTH
Mich Talebzadeh,
Distinguished
Thanks Mich, sorry, I might have been a bit unclear in my original email.
The timestamps are getting loaded as 2003-11-24T09:02:32+ for example
but I want it loaded as 2003-11-24T09:02:32+1300 I know how to do this with
various transformations however I'm wondering if there's any spark or jvm
s
Hi Jack,
You may use from_utc_timestamp and to_utc_timestamp to see if they help.
from pyspark.sql.functions import from_utc_timestamp
You can read your Parquet file into DF
df = spark.read.parquet('parquet_file_path')
# Convert timestamps (assuming your column name) from UTC to
Pacific/Auckla
