Question

使用spark数据框，我需要根据用户ID将行值转换为列和分区，并创建一个csv文件。


val someDF = Seq(
  ("user1", "math","algebra-1","90"),
  ("user1", "physics","gravity","70"),
  ("user3", "biology","health","50"),
  ("user2", "biology","health","100"),
  ("user1", "math","algebra-1","40"),
  ("user2", "physics","gravity-2","20")
).toDF("user_id", "course_id","lesson_name","score")

someDF.show(false)

+-------+---------+-----------+-----+
|user_id|course_id|lesson_name|score|
+-------+---------+-----------+-----+
|  user1|     math|  algebra-1|   90|
|  user1|  physics|    gravity|   70|
|  user3|  biology|     health|   50|
|  user2|  biology|     health|  100|
|  user1|     math|  algebra-1|   40|
|  user2|  physics|  gravity-2|   20|
+-------+---------+-----------+-----+

val result = someDF.groupBy("user_id", "course_id").pivot("lesson_name").agg(first("score"))

result.show(false)

+-------+---------+---------+-------+---------+------+
|user_id|course_id|algebra-1|gravity|gravity-2|health|
+-------+---------+---------+-------+---------+------+
|  user3|  biology|     null|   null|     null|    50|
|  user1|     math|       90|   null|     null|  null|
|  user2|  biology|     null|   null|     null|   100|
|  user2|  physics|     null|   null|       20|  null|
|  user1|  physics|     null|     70|     null|  null|
+-------+---------+---------+-------+---------+------+

使用上面的代码，我可以将行值（lesson_name）转换为列名。但是我需要在course_wise

中将其保存在csv中

在csv中预期的格式如下所示。

biology.csv // Expected Output

+-------+---------+------+
|user_id|course_id|health|
+-------+---------+------+
|  user3|  biology|  50  |
|  user2|  biology| 100  |
+-------+---------+-------

physics.csv // Expected Output

+-------+---------+---------+-------
|user_id|course_id|gravity-2|gravity|
+-------+---------+---------+-------+
|  user2|  physics|  50     |  null |
|  user1|  physics| 100     |  70   | 
+-------+---------+---------+-------+

**注意：csv中的每门课程都应仅包含其特定的课程名称，并且不应包含任何不相关的课程名称。

实际上，在csv中，我可以在formate以下**

result.write
  .partitionBy("course_id")
  .mode("overwrite")
  .format("com.databricks.spark.csv")
  .option("header", "true")
  .save(somepath)

例如：

biology.csv // Wrong output, Due to it is containing non-relevant course lesson's(algebra-1,gravity-2,algebra-1)
+-------+---------+---------+-------+---------+------+
|user_id|course_id|algebra-1|gravity|gravity-2|health|
+-------+---------+---------+-------+---------+------+
|  user3|  biology|     null|   null|     null|    50|
|  user2|  biology|     null|   null|     null|   100|
+-------+---------+---------+-------+---------+------+

任何人都可以帮助解决此问题吗？

Answer 1

只需按顺序过滤一下，即可：

val result = someDF.filter($"course_id" === "physics").groupBy("user_id", "course_id").pivot("lesson_name").agg(first("score"))

+-------+---------+-------+---------+
|user_id|course_id|gravity|gravity-2|
+-------+---------+-------+---------+
|user2  |physics  |null   |20       |
|user1  |physics  |70     |null     |

+ ------- + --------- + ------- + --------- +

Answer 2

我假设您的意思是您想通过course_id将数据保存到单独的目录中。您可以使用这种方法。

scala> val someDF = Seq(
("user1", "math","algebra-1","90"),
("user1", "physics","gravity","70"),
("user3", "biology","health","50"),
("user2", "biology","health","100"),
("user1", "math","algebra-1","40"),
("user2", "physics","gravity-2","20")
).toDF("user_id", "course_id","lesson_name","score")


scala> val result = someDF.groupBy("user_id", "course_id").pivot("lesson_name").agg(first("score"))

scala>     val eventNames = result.select($"course_id").distinct().collect() 
var eventlist =eventNames.map(x => x(0).toString)



for (eventName <- eventlist) {
val course = result.where($"course_id" === lit(eventName))
//remove null column

val row = course
.select(course.columns.map(c => when(col(c).isNull, 0).otherwise(1).as(c)): _*)
.groupBy().max(course.columns.map(c => c): _*)
.first

val colKeep = row.getValuesMap[Int](row.schema.fieldNames)
.map{c => if (c._2 == 1) Some(c._1) else None }
.flatten.toArray


var final_df = course.select(row.schema.fieldNames.intersect(colKeep)
.map(c => col(c.drop(4).dropRight(1))): _*)


final_df.show()

final_df.coalesce(1).write.mode("overwrite").format("csv").save(s"${eventName}")
}


+-------+---------+------+
|user_id|course_id|health|
+-------+---------+------+
|  user3|  biology|    50|
|  user2|  biology|   100|
+-------+---------+------+

+-------+---------+-------+---------+
|user_id|course_id|gravity|gravity-2|
+-------+---------+-------+---------+
|  user2|  physics|   null|       20|
|  user1|  physics|     70|     null|
+-------+---------+-------+---------+

+-------+---------+---------+
|user_id|course_id|algebra-1|
+-------+---------+---------+
|  user1|     math|       90|
+-------+---------+---------+

如果它满足您的目的，请接受答案。HappyHadoop

spark数据框，将行值转换为列名

2 个答案: