val words
= List("spark","hive","hadoop","hbase","zookeeper","hive","hive",
"hive","hbase","hbase","spark","hive","spark","hive")
val wordsplit
:List
[(String
,Int
)] = words
.flatMap(w
=>w
.split(",")
.map((_
,1)))
val wordcount
= wordsplit
.groupBy(identity
).mapValues(_
.size
)
Map((hive
,1) -> 6, (zookeeper
,1) -> 1, (hbase
,1) -> 3, (spark
,1) -> 3, (hadoop
,1) -> 1)
val wordcount1
= wordsplit
.groupBy(x
=>x
).mapValues(_
.size
)
Map((hive
,1) -> 6, (zookeeper
,1) -> 1, (hbase
,1) -> 3, (spark
,1) -> 3, (hadoop
,1) -> 1)
identity 函数的作用就是将传入的参数“直接”当做返回值回传给调用者，这在正常使用中，可以说是毫无作用，但他在groupBy和sortBy等函数中的作用，在于避免程序员书写相同且容易出错的逻辑
