Hi All,

I found a strange bug which is related with reading data from a updated path and cache operation.
Please consider the following code:

import org.apache.spark.sql.DataFrame

def f(data: DataFrame): DataFrame = {
  val df = data.filter("id>10")
  df.cache
  df.count
  df
}

f(spark.range(100).asInstanceOf[DataFrame]).count // output 89 which is correct
f(spark.range(1000).asInstanceOf[DataFrame]).count // output 989 which is correct

val dir = "/tmp/test"
spark.range(100).write.mode("overwrite").parquet(dir)
val df = spark.read.parquet(dir)
df.count // output 100 which is correct
f(df).count // output 89 which is correct

spark.range(1000).write.mode("overwrite").parquet(dir)
val df1 = spark.read.parquet(dir)
df1.count // output 1000 which is correct, in fact other operation expect df1.filter("id>10") return correct result.
f(df1).count // output 89 which is incorrect

In fact when we use df1.filter("id>10"), spark will however use old cached dataFrame

Any idea? Thanks a lot

Cheers
Gen