Wrong datatype for timestamp fields

[Khelied]

I use Mosaic to load a geopackage into a Spark Dataframe. When loading timestamp fields, something goes wrong and I get the date 1899-11-30 everywhere. Mosaic automatically makes this a "datetime" while this should be a "timestamp". There is no option to load everything as a "string" or define the schema.

This is wat I get:

And this is what it need to be:

Thank you

[milos-colic]

@Khelied thank you for raising the issue, could you share the code snippet please so that I can reproduce.
I will open a PR to address this for the next release.

[Khelied]

@milos-colic hereby my code:

import mosaic as mos
mos.enable_mosaic(spark, dbutils)       # Enable Mosaic
mos.enable_gdal(spark)                  # Enable GDAL
import fiona
import datetime
import os
from pyspark.sql.functions import *
from pyspark.sql.types import *

FilePath = f'/mnt/my_geopackage.gpkg'
gpkg_layer = 'layer_one'

#layers = fiona.listlayers("/dbfs"+FilePath)
#print(layers)

path, filename = os.path.split(FilePath)
print(filename)
print(path)

spark_gpkg_df = (mos.read().format("multi_read_ogr")
    .option("driverName", "GPKG")
    .option("layerName", gpkg_layer)
    .option("asWKB", "true")
    .option("chunkSize", "500000")
    .load(FilePath)
)

[Khelied]

@milos-colic Is there an indication when this will be resolved? Thank you.