Parquet writer produced a corrupted file due to page value count overflow

Parquet writer only checks the number of rows and the page size to decide whether it needs to fit a content to be written in a single page. 

In the case of a composite column (ex: array/map) with a lot of nulls, it is possible to create 2billions+ values while under the default page-size & row-count threshold (1MB, 20000rows)

Repro using Spark:

      val dir = "/tmp/anyrandomDirectory"

      spark.range(0, 20000, 1, 1)
        .selectExpr("array_repeat(cast(null as binary), 110000) as n")
        .write
        .mode("overwrite")
        .save(dir)

      val result = spark
        .sql(s"select * from parquet.`$dir` limit 1000")
        .collect() // This will break