'commit'

src/main/java/com/dsideal/base/Tools/DeleteNotUnique.java

@@ -0,0 +1,119 @@
+package com.dsideal.base.Tools;
+
+import java.sql.SQLOutput;
+import java.util.List;
+import java.util.Map;
+import java.util.HashMap;
+import java.util.Set;
+import java.util.HashSet;
+import java.util.stream.Collectors;
+
+import com.dsideal.base.Tools.Util.LocalMysqlConnectUtil;
+import com.jfinal.plugin.activerecord.Record;
+import com.jfinal.plugin.activerecord.Db;
+
+import static com.dsideal.base.DataEase.Model.DataEaseModel.DB_NAME;
+
+public class DeleteNotUnique {
+
+    public static void main(String[] args) {
+        LocalMysqlConnectUtil.Init();
+        DeleteNotUnique deleteNotUnique = new DeleteNotUnique();
+
+        String sql = "select * from t_dp_dataset where b_use=1";
+        List<Record> list = Db.find(sql);
+
+        for (Record record : list) {
+            String table_name = record.getStr("table_name");
+            deleteNotUnique.removeDuplicates(table_name);
+            System.out.println("删除完成:" + table_name);
+        }
+        System.out.println("所有表清理完成! ");
+    }
+
+    /**
+     * 删除重复数据，保留ID最小的记录
+     *
+     * @param tableName 表名
+     */
+    public void removeDuplicates(String tableName) {
+        // 1. 获取表的所有字段名（除id外）
+        List<String> columnNames = getColumnNames(tableName);
+        System.out.println("表字段：" + columnNames);
+
+        // 2. 获取所有数据
+        List<Record> allRecords = Db.use(DB_NAME).find("SELECT * FROM `" + tableName + "` ORDER BY id");
+
+        // 3. 使用Map存储唯一记录
+        Map<String, Record> uniqueRecords = new HashMap<>();
+        Set<Integer> idsToDelete = new HashSet<>();
+
+        for (Record record : allRecords) {
+            // 生成数据特征
+            String dataKey = generateDataKey(record, columnNames);
+
+            if (uniqueRecords.containsKey(dataKey)) {
+                // 比较ID，保留较小的
+                Record existingRecord = uniqueRecords.get(dataKey);
+                int existingId = existingRecord.getInt("id");
+                int currentId = record.getInt("id");
+
+                if (currentId < existingId) {
+                    idsToDelete.add(existingId);
+                    uniqueRecords.put(dataKey, record);
+                } else {
+                    idsToDelete.add(currentId);
+                }
+            } else {
+                uniqueRecords.put(dataKey, record);
+            }
+        }
+
+        // 4. 删除重复记录
+        if (!idsToDelete.isEmpty()) {
+            Db.use(DB_NAME).batch("DELETE FROM " + tableName + " WHERE id = ?",
+                    idsToDelete.stream()
+                            .map(id -> new Object[]{id})
+                            .collect(Collectors.toList()).toArray(new Object[0][]),
+                    100);
+
+            System.out.println("已删除 " + idsToDelete.size() + " 条重复数据");
+        } else {
+            System.out.println("没有发现重复数据");
+        }
+    }
+
+    /**
+     * 获取表的所有字段名（除id外）
+     */
+    private List<String> getColumnNames(String tableName) {
+        String sql = "SELECT COLUMN_NAME FROM INFORMATION_SCHEMA.COLUMNS " +
+                "WHERE TABLE_NAME = ? " +
+                "AND COLUMN_NAME != 'id' " +
+                "ORDER BY ORDINAL_POSITION";
+
+        return Db.use(DB_NAME).find(sql, tableName)
+                .stream()
+                .map(record -> record.getStr("COLUMN_NAME"))
+                .collect(Collectors.toList());
+    }
+
+    /**
+     * 生成数据特征
+     */
+    private String generateDataKey(Record record, List<String> columnNames) {
+        StringBuilder key = new StringBuilder();
+
+        for (String columnName : columnNames) {
+            // 添加分隔符避免值的混淆
+            key.append("#").append(columnName).append("=");
+
+            // 获取字段值，处理null情况
+            Object value = record.get(columnName);
+            key.append(value == null ? "NULL" : value.toString());
+        }
+
+        return key.toString();
+    }
+
+}