Added ENHANCED_SPLIT_INDEX function

Kyle Dong · Kyle Dong · commit 2756ce4a60dc · 2021-09-15T20:04:50.000+08:00
diff --git a/src/main/java/com/tencent/cloud/oceanus/udf/EnhancedSplitIndex.java b/src/main/java/com/tencent/cloud/oceanus/udf/EnhancedSplitIndex.java
@@ -0,0 +1,32 @@
+package com.tencent.cloud.oceanus.udf;
+
+import org.apache.flink.table.functions.ScalarFunction;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.util.regex.Pattern;
+
+/**
+ * 增强版的 ENHANCED_SPLIT_INDEX, 对连续的分隔符会如实处理, 例如输入 "1,2,,3" 会从 ["1","2","","3"] 中取字段
+ * Flink 自带的 SPLIT_INDEX 函数会把多个分隔符当作一个, 例如输入 "1,2,,3" 会从 ["1","2","3"] 中取字段
+ *
+ * SQL 代码声明方式:
+ * CREATE TEMPORARY SYSTEM FUNCTION ENHANCED_SPLIT_INDEX AS 'com.tencent.cloud.oceanus.udf.EnhancedSplitIndex' LANGUAGE JAVA;
+ */
+public class EnhancedSplitIndex extends ScalarFunction {
+	private static final Logger LOGGER = LoggerFactory.getLogger(EnhancedSplitIndex.class);
+
+	public String eval(String input, String separator, int index) {
+		String[] splits = input.split(Pattern.quote(separator));
+
+		if (index >= splits.length || index < 0) {
+			return null;
+		}
+		return splits[index];
+	}
+
+	public static void main(String[] args) {
+		EnhancedSplitIndex instance = new EnhancedSplitIndex();
+		System.out.println(instance.eval("a|b|c|d|e|f|0|||||||1|0|2|4", "|", 0));
+	}
+}
