StringUnionUDF

build.gradle

@@ -39,7 +39,7 @@ dependencies {
 //    compileOnly('org.apache.hive:hive-exec:1.1.0-cdh5.14.0')
 //    compile group: 'org.apache.hive', name: 'hive-exec', version: '1.1.0'
     compileOnly('org.apache.hive:hive-exec:1.1.0')
-//    compile group: 'cz.mallat.uasparser', name: 'uasparser', version: '0.6.2'
+    compile group: 'cz.mallat.uasparser', name: 'uasparser', version: '0.6.2'
     compile group: 'nl.basjes.parse.useragent', name: 'yauaa', version: '5.21'
 }
 

src/main/java/cn/com/duiba/udf/StringUnionUDF.java

+package cn.com.duiba.udf;
+import com.google.common.collect.Lists;
+import org.apache.commons.lang3.StringUtils;
+import org.apache.hadoop.hive.ql.exec.UDF;
+
+import java.util.HashSet;
+import java.util.List;
+import java.util.Set;
+
+/**
+ * 字符串求差值
+ */
+
+public class StringUnionUDF extends UDF{
+    public String evaluate(String stra, String strb, String delimiter) {
+        if (StringUtils.isBlank(stra)) {
+            return strb;
+        }
+        if (StringUtils.isBlank(strb)) {
+            return stra;
+        }
+
+        if (StringUtils.isBlank(delimiter)) {
+            return null;
+        }
+
+        List<String> listA = Lists.newArrayList(StringUtils.split(stra, delimiter));
+        List<String> listB = Lists.newArrayList(StringUtils.split(strb, delimiter));
+
+        Set<String> set = new HashSet<>();
+        //用set的特性去做
+        set.addAll(listA);
+        set.addAll(listB);
+        return StringUtils.join(set, delimiter);
+    }
+
+}