Closed zyd14 closed 1 year ago
Thanks for reporting this. There is a work around for now:
df_spec = dg.DataGenerator(spark, name="test-data", rows=2)
df_spec = df_spec.withColumn(
"test",
StringType(), # can also use "string"
values=["A", "C", "T", "G"],
numFeatures=(1,3),
structType="array"
)
test_df = df_spec.build()
great, thanks for the workaround and quick response!
Expected Behavior
With version 0.2.1 I used to be able to make a column containing string arrays like this:
And I would receive a Dataframe with a column named "test", with 2 rows of values picked from [["A"], ["C"], ["T"], ["G"]]
Current Behavior
When I execute the same code with version 0.3.1+ (tried 0.3.1, 0.3.2, 0.3.3, 0.3.4, 0.3.4.post1), the resulting "test" column only contains None values.
Steps to Reproduce (for bugs)
Run the sample code provided in
Expected Behavior
Your Environment
dbldatagen
version used: 0.3.1-0.3.4