Question

我在这里尝试的是过滤val sc1Rdd=parDf.select(parDf("token"),parDf("log_type")).rdd val sc2Rdd=sc1Rdd.filter(x=>x,log_type=="1")的行这是我的代码：

<div ng-controller="repeatPeople">
  <br>
  <p>
    <input type="text" id="search" ng-model="searchPeople" placeholder="Search">
  </p>
  <br>
  <br>
  <table border="0">
    <thead style="background-color: lightgray;">
      <tr>
        <td style="width: 30px;">Click</td>
        <td>Name</td>
        <td>Gender</td>
      </tr>
    </thead>
    <tbody>
      <tr ng-repeat-start="person in result = (people | filter:searchPeople)">
        <td>
          <button ng-if="person.expanded" ng-click="person.expanded = !person.expanded">-</button>
          <button ng-if="person.expanded == false" ng-click=" person.expanded = !person.expanded">+</button>
        </td>
        <td>{{person.name}}</td>
        <td>{{person.gender}}</td>
      </tr>
      <tr ng-if="person.expanded" ng-repeat-end="">
        <td colspan="3">{{person.details}}</td>
      </tr>
    </tbody>
  </table>
</div>

但错误代码显示：

parDf：org.apache.spark.sql.DataFrame = [action_time：bigint， action_type：bigint ...另外21个字段] sc1Rdd： org.apache.spark.rdd.RDD [org.apache.spark.sql.Row] = MapPartitionsRDD [865] at rdd at：186：188：error：方法过滤器的参数太多：（f：org.apache.spark.sql.Row =＆gt; 布尔值）org.apache.spark.rdd.RDD [org.apache.spark.sql.Row] val sc2Rdd = sc1Rdd.filter（x =＆gt; x，log_type ==＆＃34; 1＆＃34;）

任何帮助将不胜感激。

Answer 1

您无需将数据框更改为RDD 你可以简单地过滤或在哪里

val result = parDf.select(parDf("token"),parDf("log_type")).filter(parDf("long_type")===1)

parDf.select(p===1)arDf("token"),parDf("log_type")).where(parDf("long_type")

希望这有帮助！

方法过滤器的参数太多

1 个答案: