Hi, I'm trying to use the Spark connector to load data into StarRocks array columns, but I'm encountering an issue with array types. According to the documentation(https://docs.starrocks.io/docs/3.5/loading/Spark-connector-starrocks/#load-data-into-columns-of-array-type), array types are supported. However, when I try to run the example from the documentation, I get this error: parse error. The value is array type in json document stream, you can set strip_outer_array=true to parse each element of the array as individual rows I've already set strip_outer_array=true but I'm still getting the same error. I am using starrocks 3.5.6. Has anyone successfully used array types with the Spark connector? Are there any additional configurations I'm missing? Thanks for your help!

I'm wondering why you use Spark Connector to load data? What is the data source?

I read data from HDFS, do some processing using spark and want to save the results in starrocks

Why not use starrocks to read HDFS directly and do the processing in StarRocks?