hive 自定义函数、GenericUDF、GenericUDTF（自定义函数计算给定字符串的长度、将一个任意分割符的字符串切割成独立的单词）

前言

  当Hive提供的内置函数无法满足业务处理需要时，可以考虑使用用户自定义函数

---

1. 自定义函数

官方文档

1.1 分类

  UDF（User-Defined-Function） 一进一出
  UDAF（User-Defined Aggregation Function）聚集函数，多进一出
  UDTF（User-Defined Table-Generating Functions）一进多出

1.2 编程步骤

  （1）继承Hive提供的类

org.apache.hadoop.hive.ql.udf.generic.GenericUDF  
org.apache.hadoop.hive.ql.udf.generic.GenericUDTF;

  （2）实现类中的抽象方法
  （3）在hive的命令行窗口创建函数
添加jar

add jar jar路径

创建function

create [temporary] function [dbname.]function_name AS class_name;

  （4）在hive的命令行窗口删除函数

drop [temporary] function [if exists] [dbname.]function_name;

---

2. 自定义UDF函数

2.1 参考内置函数concat

/**
 * GenericUDFConcat.
 */
@Description(name = "concat",
value = "_FUNC_(str1, str2, ... strN) - returns the concatenation of str1, str2, ... strN or "+
        "_FUNC_(bin1, bin2, ... binN) - returns the concatenation of bytes in binary data " +
        " bin1, bin2, ... binN",
extended = "Returns NULL if any argument is NULL.\n"
+ "Example:\n"
+ "  > SELECT _FUNC_('abc', 'def') FROM src LIMIT 1;\n"
+ "  'abcdef'")
@VectorizedExpressions({
    StringGroupConcatColCol.class,
    StringGroupColConcatStringScalar.class,
    StringScalarConcatStringGroupCol.class})
public class GenericUDFConcat extends GenericUDF {
  private transient ObjectInspector[] argumentOIs;
  private transient StringConverter[] stringConverters;
  private transient PrimitiveCategory returnType = PrimitiveCategory.STRING;
  private transient BytesWritable[] bw;
  private transient GenericUDFUtils.StringHelper returnHelper;

  @Override
  public ObjectInspector initialize(ObjectInspector[] arguments) throws UDFArgumentException {

    // Loop through all the inputs to determine the appropriate return type/length.
    // Return type:
    //  All CHAR inputs: return CHAR
    //  All VARCHAR inputs: return VARCHAR
    //  All CHAR/VARCHAR inputs: return VARCHAR
    //  All BINARY inputs: return BINARY
    //  Otherwise return STRING
    argumentOIs = arguments;

    PrimitiveCategory currentCategory;
    PrimitiveObjectInspector poi;
    boolean fixedLengthReturnValue = true;
    int returnLength = 0;  // Only for char/varchar return types
    for (int idx = 0; idx < arguments.length; ++idx) {
      if (arguments[idx].getCategory() != Category.PRIMITIVE) {
        throw new UDFArgumentException("CONCAT only takes primitive arguments");
      }
      poi = (PrimitiveObjectInspector)arguments[idx];
      currentCategory = poi.getPrimitiveCategory();
      if (idx == 0) {
        returnType = currentCategory;
      }
      switch (currentCategory) {
        case BINARY:
          fixedLengthReturnValue = false;
          if (returnType != currentCategory) {
            // mix of binary/non-binary args
            returnType = PrimitiveCategory.STRING;
          }
          break;
        case CHAR:
        case VARCHAR:
          if (!fixedLengthReturnValue) {
            returnType = PrimitiveCategory.STRING;
          }
          if (fixedLengthReturnValue && currentCategory == PrimitiveCategory.VARCHAR) {
            returnType = PrimitiveCategory.VARCHAR;
          }
          break;
        default:
          returnType = PrimitiveCategory.STRING;
          fixedLengthReturnValue = false;
          break;
      }

      // If all arguments are of known length then we can keep track of the max
      // length of the return type. However if the return length exceeds the
      // max length for the char/varchar, then the return type reverts to string.
      if (fixedLengthReturnValue) {
        returnLength += GenericUDFUtils.StringHelper.getFixedStringSizeForType(poi);
        if ((returnType == PrimitiveCategory.VARCHAR
                && returnLength > HiveVarchar.MAX_VARCHAR_LENGTH)
            || (returnType == PrimitiveCategory.CHAR
                && returnLength > HiveChar.MAX_CHAR_LENGTH)) {
          returnType = PrimitiveCategory.STRING;
          fixedLengthReturnValue = false;
        }
      }
    }

    if (returnType == PrimitiveCategory.BINARY) {
      bw = new BytesWritable[arguments.length];
      return PrimitiveObjectInspectorFactory.writableBinaryObjectInspector;
    } else {
      // treat all inputs as string, the return value will be converted to the appropriate type.
      createStringConverters();
      returnHelper = new GenericUDFUtils.StringHelper(returnType);
      BaseCharTypeInfo typeInfo;
      switch (returnType) {
        case STRING:
          return PrimitiveObjectInspectorFactory.writableStringObjectInspector;
        case CHAR:
          typeInfo = TypeInfoFactory.getCharTypeInfo(returnLength);
          return PrimitiveObjectInspectorFactory.getPrimitiveWritableObjectInspector(typeInfo);
        case VARCHAR:
          typeInfo = TypeInfoFactory.getVarcharTypeInfo(returnLength);
          return PrimitiveObjectInspectorFactory.getPrimitiveWritableObjectInspector(typeInfo);
        default:
          throw new UDFArgumentException("Unexpected CONCAT return type of " + returnType);
      }
    }
  }

  private void createStringConverters() {
    stringConverters = new StringConverter[argumentOIs.length];
    for (int idx = 0; idx < argumentOIs.length; ++idx) {
      stringConverters[idx] = new StringConverter((PrimitiveObjectInspector) argumentOIs[idx]);
    }
  }

  @Override
  public Object evaluate(DeferredObject[] arguments) throws HiveException {
    if (returnType == PrimitiveCategory.BINARY) {
      return binaryEvaluate(arguments);
    } else {
      return returnHelper.setReturnValue(stringEvaluate(arguments));
    }
  }

  public Object binaryEvaluate(DeferredObject[] arguments) throws HiveException {
    int len = 0;
    for (int idx = 0; idx < arguments.length; ++idx) {
      bw[idx] = ((BinaryObjectInspector)argumentOIs[idx])
          .getPrimitiveWritableObject(arguments[idx].get());
      if (bw[idx] == null){
        return null;
      }
      len += bw[idx].getLength();
    }

    byte[] out = new byte[len];
    int curLen = 0;
    // Need to iterate twice since BytesWritable doesn't support append.
    for (BytesWritable bytes : bw){
      System.arraycopy(bytes.getBytes(), 0, out, curLen, bytes.getLength());
      curLen += bytes.getLength();
    }
    return new BytesWritable(out);
  }

  public String stringEvaluate(DeferredObject[] arguments) throws HiveException {
    StringBuilder sb = new StringBuilder();
    for (int idx = 0; idx < arguments.length; ++idx) {
      String val = null;
      if (arguments[idx] != null) {
        val = (String) stringConverters[idx].convert(arguments[idx].get());
      }
      if (val == null) {
        return null;
      }
      sb.append(val);
    }
    return sb.toString();
  }

  @Override
  public String getDisplayString(String[] children) {
    return getStandardDisplayString("concat", children);
  }

}

2.2 自定义一个UDF实现计算给定字符串的长度

  引入依赖

<dependencies>
		<dependency>
			<groupId>org.apache.hive</groupId>
			<artifactId>hive-exec</artifactId>
			<version>3.1.2</version>
		</dependency>
</dependencies>

  继承抽象类GenericUDF

public class MyStringLength extends GenericUDF {
    @Override
    public ObjectInspector initialize(ObjectInspector[] arguments) throws UDFArgumentException {
        if(arguments.length!=1){
            throw new UDFArgumentLengthException("输入参数个数不为1");
        }
        if(!arguments[0].getCategory().equals( ObjectInspector.Category.PRIMITIVE)){
            throw  new UDFArgumentTypeException(0,"输入参数类型错误");
        }
        return PrimitiveObjectInspectorFactory.javaIntObjectInspector;
    }

    @Override
    public Object evaluate(DeferredObject[] arguments) throws HiveException {
        if(arguments[0].get() == null){
            return 0 ;
        }
        return arguments[0].get().toString().length();
    }

    @Override
    public String getDisplayString(String[] children) {
        return "htdata MyStringLength";
    }
}

  打成jar包上传到服务器，将jar包添加到hive的classpath

add jar /home/hdfs/hiveFunction-1.0-SNAPSHOT.jar;

  创建临时函数与java class关联

create temporary function htdata_length as "com.cz.MyStringLength";

---

3. 自定义UDTF函数

3.1 参考内置函数explode

public class GenericUDTFExplode extends GenericUDTF {

  private transient ObjectInspector inputOI = null;
  @Override
  public void close() throws HiveException {
  }

  @Override
  public StructObjectInspector initialize(ObjectInspector[] args) throws UDFArgumentException {
    if (args.length != 1) {
      throw new UDFArgumentException("explode() takes only one argument");
    }

    ArrayList<String> fieldNames = new ArrayList<String>();
    ArrayList<ObjectInspector> fieldOIs = new ArrayList<ObjectInspector>();

    switch (args[0].getCategory()) {
    case LIST:
      inputOI = args[0];
      fieldNames.add("col");
      fieldOIs.add(((ListObjectInspector)inputOI).getListElementObjectInspector());
      break;
    case MAP:
      inputOI = args[0];
      fieldNames.add("key");
      fieldNames.add("value");
      fieldOIs.add(((MapObjectInspector)inputOI).getMapKeyObjectInspector());
      fieldOIs.add(((MapObjectInspector)inputOI).getMapValueObjectInspector());
      break;
    default:
      throw new UDFArgumentException("explode() takes an array or a map as a parameter");
    }

    return ObjectInspectorFactory.getStandardStructObjectInspector(fieldNames,
        fieldOIs);
  }

  private transient final Object[] forwardListObj = new Object[1];
  private transient final Object[] forwardMapObj = new Object[2];

  @Override
  public void process(Object[] o) throws HiveException {
    switch (inputOI.getCategory()) {
    case LIST:
      ListObjectInspector listOI = (ListObjectInspector)inputOI;
      List<?> list = listOI.getList(o[0]);
      if (list == null) {
        return;
      }
      for (Object r : list) {
        forwardListObj[0] = r;
        forward(forwardListObj);
      }
      break;
    case MAP:
      MapObjectInspector mapOI = (MapObjectInspector)inputOI;
      Map<?,?> map = mapOI.getMap(o[0]);
      if (map == null) {
        return;
      }
      for (Entry<?,?> r : map.entrySet()) {
        forwardMapObj[0] = r.getKey();
        forwardMapObj[1] = r.getValue();
        forward(forwardMapObj);
      }
      break;
    default:
      throw new TaskExecutionException("explode() can only operate on an array or a map");
    }
  }

  @Override
  public String toString() {
    return "explode";
  }
}

3.2 自定义一个UDTF实现将一个任意分割符的字符串切割成独立的单词

  继承抽象类GenericUDTF

public class MyUDTF extends GenericUDTF {

    private ArrayList<String> outList = new ArrayList<>();

    @Override
    public StructObjectInspector initialize(StructObjectInspector argOIs) throws UDFArgumentException {
        //1.定义输出数据的列名和类型
        List<String> fieldNames = new ArrayList<>();
        List<ObjectInspector> fieldOIs = new ArrayList<>();
        //2.添加输出数据的列名和类型
        fieldNames.add("lineToWord");
        fieldOIs.add(PrimitiveObjectInspectorFactory.javaStringObjectInspector);

        return ObjectInspectorFactory.getStandardStructObjectInspector(fieldNames, fieldOIs);
    }

    @Override
    public void process(Object[] args) throws HiveException {
        //1.获取原始数据
        String arg = args[0].toString();
        //2.获取数据传入的第二个参数，此处为分隔符
        String splitKey = args[1].toString();
        //3.将原始数据按照传入的分隔符进行切分
        String[] fields = arg.split(splitKey);
        //4.遍历切分后的结果，并写出
        for (String field : fields) {
            //集合为复用的，首先清空集合
            outList.clear();
            //将每一个单词添加至集合
            outList.add(field);
            //将集合内容写出
            forward(outList);
        }
    }

    @Override
    public void close() throws HiveException {

    }
}

  打成jar包上传到服务器,将jar包添加到hive的classpath下,创建临时函数与java class关联

create temporary function htdata_split as "com.cz.MyUDTF";

---