UDF No TypeTag available for String

写 spark sql时踩到一个巨坑

val markLabel   = udf((userId: String, orderId: String) => if (blackBroadcast.value.contains(userId) || orderBroadcast.value.contains(orderId)) 1.0 else 0.0);
val labelData   = extData.withColumn("label", markLabel(extData("user_accountId"), extData("order_orderId")));

Run状态是OK的，结果也对，但是打包的时候就老报下面的错误，网上找了一圈也没找到答案，

[ERROR] E:\Code\scalaCode\RiskManager\05_DataAnaly\src\main\scala\RiskManager\DataAnaly.scala:51: error: No TypeTag available for String
[INFO]         val markLabel   = udf((userId: String, orderId: String) => if (blackBroadcast.value.contains(userId) || orderBroadcast.value.contains(orderId)) 1.0 else 0.0);

最后发现了原来是这个原因，引入了一个叫play_json的包

<dependency>
    <groupId>com.typesafe.play</groupId>
    <artifactId>play-json_2.10</artifactId>
    <version>2.4.0-M1</version>
</dependency>

去掉就能正常打包了！！！！