|
@@ -0,0 +1,21 @@
|
|
|
+package com.winhc.bigdata.udf;
|
|
|
+
|
|
|
+import com.aliyun.odps.udf.UDF;
|
|
|
+import com.aliyun.odps.utils.StringUtils;
|
|
|
+
|
|
|
+import java.util.Arrays;
|
|
|
+import java.util.List;
|
|
|
+import java.util.regex.Pattern;
|
|
|
+
|
|
|
+/**
|
|
|
+ * @Author: XuJiakai
|
|
|
+ * @Date: 2020/7/8 11:13
|
|
|
+ * @Description:
|
|
|
+ */
|
|
|
+public class CompanySplit extends UDF {
|
|
|
+ private static final Pattern pattern = Pattern.compile("[^\\u4e00-\\u9fa5a-zA-Z \\(\\)().]+");
|
|
|
+
|
|
|
+ public List<String> evaluate(String val) {
|
|
|
+ return Arrays.asList(StringUtils.isNotBlank(val) ? pattern.split(val) : new String[]{""});
|
|
|
+ }
|
|
|
+}
|