详解 Hive UDF 开发之Java 实现步骤与代码调试技巧

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

import org.apache.hadoop.hive.ql.exec.UDF;
import org.apache.hadoop.io.Text;
public class StringReverseUDF extends UDF {
  public Text evaluate(Text input) {
    if (input == null) return null;
    return new Text(new StringBuilder(input.toString()).reverse().toString());
  }
}

mvn clean package  # 生成 JAR 文件（如 udf.jar）

ADD JAR /path/to/udf.jar;        -- 加载 JAR
CREATE TEMPORARY FUNCTION reverse AS 'com.example.StringReverseUDF';  -- 注册函数
SELECT reverse('hello');         -- 输出：olleh

@Test
public void testReverse() {
  StringReverseUDF udf = new StringReverseUDF();
  assertEquals("olleh", udf.evaluate(new Text("hello")).toString());
}

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
public class StringReverseUDF extends UDF {
  private static final Log LOG = LogFactory.getLog(StringReverseUDF.class);
  public Text evaluate(Text input) {
    LOG.info("Input: " + input);  // 输出到 Hive 日志
    // ...逻辑代码
  }
}

try {
  // 业务逻辑
} catch (Exception e) {
  return null;  // Hive 会处理为 NULL 值
}

SELECT reverse(NULL);      -- 应返回 NULL
SELECT reverse('');       -- 空字符串
SELECT reverse('123!@#'); -- 特殊字符

public ObjectInspector initialize(ObjectInspector[] arguments) {
  if (arguments.length != 1) throw new UDFArgumentLengthException("需1个参数");
  if (!arguments[0].getCategory().equals(PrimitiveCategory.STRING)) {
    throw new UDFArgumentTypeException(0, "参数需为字符串");
  }
  return PrimitiveObjectInspectorFactory.writableStringObjectInspector;
}