docs: 调整 Upsert 示例代码和格式

Author: dingxin-tech

docs/docs/example-code/upsert.md

@@ -19,37 +19,34 @@ sidebar_position: 5
 首先，您需通过 `TableTunnel` 的 `buildUpsertSession` 方法初始化一个 Upsert 会话，并配置必要的参数，如项目名、表名、模式名（如果适用）以及分区规格等。
 ```java
 TableTunnel tunnel = odps.tableTunnel();
-UpsertSession upsert = tunnel.buildUpsertSession(projectName, tableName)
+    UpsertSession upsert = tunnel.buildUpsertSession(projectName, tableName)
     .setSchemaName(schemaName)
     .setPartitionSpec(partitionSpec)
     .build();
-System.out.println("Upsert Session ID: " + upsert.getId());
+    System.out.println("Upsert Session ID: " + upsert.getId());
 ```
 若需要重新加载已有的 Upsert 会话，可以使用会话ID进行重建。
 ```java
 if(reload) {
     String id = upsert.getId();
     upsert = tunnel.buildUpsertSession(projectName, tableName)
-        .setSchemaName(schemaName)
-        .setPartitionSpec(partitionSpec)
-        .setUpsertId(id)
-        .build();
-}
+    .setSchemaName(schemaName)
+    .setPartitionSpec(partitionSpec)
+    .setUpsertId(id)
+    .build();
+    }
 ```
 
 ## 执行 Upsert 操作
 接下来，创建 `UpsertStream` 实例，并利用它执行具体的插入或更新记录操作。
 ```java
-try (UpsertStream stream = upsert.buildUpsertStream().setListener(listener).build()) {
+    UpsertStream stream = upsert.buildUpsertStream().build()
     Record record = upsert.newRecord();
     record.setString("key", "0");
     record.setString("value", "v1");
     stream.upsert(record); // 执行 Upsert
     stream.delete(record); // 删除记录示例
     stream.flush(); // 刷新到缓冲区
-} catch (Exception e) {
-    e.printStackTrace();
-}
 ```
 
 ## 提交并关闭会话
@@ -71,38 +68,38 @@ upsert.close();
 import ...; // 导入必要的包
 
 public class UpsertDeltaTableDemo {
-    public static void main(String[] args) {
-        try {
-            TableTunnel tunnel = TableTunnel.newInstance("<Your_ODPS_Instance>");
-            String projectName = "<Your_Project_Name>";
-            String tableName = "<Your_Table_Name>";
-            UpsertSession upsert = tunnel.buildUpsertSession(projectName, tableName)
-                .setSchemaName("<Your_Schema_Name>")
-                .setPartitionSpec("<Your_Partition_Spec>")
-                .setLifecycle(1)
-                .build();
+  public static void main(String[] args) {
+    try {
+      Account account = new AliyunAccount("<Your_Access_ID>", "<Your_Access_Key>");
+      Odps odps = new Odps(account);
 
-            try (UpsertStream stream = upsert.buildUpsertStream().build()) {
-                Record record = upsert.newRecord();
-                record.setString("key", "exampleKey");
-                record.setString("value", "exampleValue");
-                stream.upsert(record);
-                stream.flush();
-            }
+      TableTunnel tunnel = odps.tableTunnel();
+      String projectName = "<Your_Project_Name>";
+      String tableName = "<Your_Table_Name>";
+      try (TableTunnel.UpsertSession upsert = tunnel.buildUpsertSession(projectName, tableName)
+          .setSchemaName("<Your_Schema_Name>")
+          .setPartitionSpec("<Your_Partition_Spec>")
+          .build()) {
+        UpsertStream stream = upsert.buildUpsertStream().build();
+        Record record = upsert.newRecord();
+        record.setString("key", "exampleKey");
+        record.setString("value", "exampleValue");
+        stream.upsert(record);
+        stream.flush();
 
-            upsert.commit(false);
-            System.out.println("Commit successful.");
-            upsert.close();
-        } catch (Exception e) {
-            e.printStackTrace();
-        }
+        upsert.commit(false);
+        System.out.println("Commit successful.");
+      }
+    } catch (Exception e) {
+      e.printStackTrace();
     }
+  }
 }
 ```
 
 ## 注意事项
 - 由于主键表的写入特性，我们在并发写入同一张表（分区）时，应当谨慎地控制写入逻辑。如果存在多个并发，同时对同一主键进行写入，则可能发生不可预期的行为。
-常用的方案是使用 shuffle by pk 操作，将相同主键的记录分配到同一个线程中进行写入。
+  常用的方案是使用 shuffle by pk 操作，将相同主键的记录分配到同一个线程中进行写入。
 
 ## 相关链接
 - [UpsertSession API 参考](../api-reference/tunnel/UpsertSession.md)