val tagsStandardizeTmp = tagsStandardize.collect()
val tagsSimi = tagsStandardize.map{
f=>
var retTag = f._2
if (f._2.toString.split(" ").size == 1) {
var simiTmp = ""
val tagsTmpStand = tagsStandardizeTmp
.filter(_._2.toString.split(" ").size != 1 )
.filter(f._2.toString.size < _._2.toString.size)
.sortBy(_._2.toString.size)
var x = 0
val loop = new Breaks
tagsTmpStand.map{
tagTmp=>
val flag = getEditSize(f._2.toString,tagTmp._2.toString)
if (flag == 1){
retTag = tagTmp._2
loop.break()
}
}
((f._1,retTag),1)
} else {
((f._1,f._2),1)
}
}
来源: https://mp.weixin.qq.com/s/zeQYrYkFrJfnFqWa34vm4g