在like运算符上应用字符串列表

Question

问题陈述:我需要传递一个字符串列表并在spark java中使用filter函数应用like运算符.

原因: 像运算符一样考虑单个字符串实体,因此代码中存在很多性能问题,因为我们需要应用于更大的数据集,为了克服这个问题,我需要传递字符串列表并使用过滤器函数应用类似运算符.

所以请让我知道如何使用类似的应用列表.因为我需要找到类似ID的模式

 JavaSparkContext sc = new JavaSparkContext(new SparkConf().setAppName("SparkJdbcDs").setMaster("local[*]"));
  SQLContext sqlContext = new SQLContext(sc);
  SparkSession spark = SparkSession.builder().appName("JavaTokenizerExample").getOrCreate();

  List<Row> data = Arrays.asList(
      RowFactory.create("J40504", "CRC Industries"),
      RowFactory.create("K630-0746777","Dixon value"),
      RowFactory.create("K444-4444","3M INdustries"),
      RowFactory.create("4333444","3M INdustries"),
      RowFactory.create("566-655","3M INdustries"),
      RowFactory.create("4444888","3M INdustries"),
      RowFactory.create("P477-7444","3M INdustries"),
      RowFactory.create("566655","Dixon coupling valve"));
   // In real time we have large dataset

  StructType schema = new StructType(new StructField[] {new StructField("label1", DataTypes.StringType, false,Metadata.empty()),
    new StructField("sentence1", DataTypes.StringType, false,Metadata.empty()) });

  Dataset<Row> sentenceDataFrame = spark.createDataFrame(data, schema);

  List<String> listStrings = new ArrayList<String>();
  listStrings.add("40504");
  listStrings.add("630-0746");
  listStrings.add("477-7444");
  listStrings.add("444-4444");
// In real time we have large list of string to be compared with
  sentenceDataFrame.show();
  System.out.println("Array list :"+listStrings);
  for(int i=0;i<listStrings.size();i++){
 sentenceDataFrame=sentenceDataFrame.filter(col("label1").like("%"+listStrings.get(i)+"%"));
  }
  sentenceDataFrame.show();

Answer 1

第一解决方案

而不是迭代数据集N(其中N是listStrings的大小)的次数,您可以构建表达式并仅过滤一次数据集:

    StringBuilder expressionBuilder = new StringBuilder();

    String separator = "";

    for (String s : listStrings) {
        expressionBuilder.append(separator + " label1 LIKE '%" + s + "%'");
        separator = " OR ";
    }

    String expression = expressionBuilder.toString();

    sentenceDataFrame = sentenceDataFrame.filter(expression);

二解决方案

我们可以将listStrings加载到数据集中:

 StructType schemaList = new StructType(new StructField[]{new StructField("labelToFind", DataTypes.StringType, false, Metadata.empty())});

 List<Row> listStrings = Arrays.asList(
                RowFactory.create("40504"),
                RowFactory.create("630-0746"),
                RowFactory.create("477-7444"),
                RowFactory.create("444-4444"));

 Dataset<Row>listDataset = sqlContext.createDataFrame(listStrings, schemaList);

然后我们可以加入两个数据集以过滤行:

   sentenceDataFrame = sentenceDataFrame.join(listDataset ,sentenceDataFrame.col("label1").contains(listDataset.col("labelToFind"))).select("label1","sentence1");