# Read JSON files with automatic schema inference df = spark.read.json("logs.json") df.where("age > 21").select("name.first").show()