From spark dataframe when I run df.printSchema() all varchar(1), varchar(10), etc... becomes string Is there a way to differentiate varchars?

Doesn't have to be in pyspark. method to do it in spark sql is welcome as well.

CodePudding user response：

VarcharType(length) can only be used in table schema, not functions/operators. This means you can use it in create table sql, but not dataFrame schema.