問題文
A team queries a folder of CSV files directly from SQL and finds that every column comes back as a string. What is the most accurate explanation in Apache Spark?
選択肢
- Direct queries against a path always return strings, and types appear only through the reader API.
- Spark SQL treats all direct file queries as text until a table is created over them.
- The files must be converted to a columnar format before types can be used.
- CSV carries no type information, so types must either be inferred with an option or supplied as an explicit schema.