1. rd行分割 -fd列分割 -charset编码

--建临时表
odps@ work_test_1>create table if not exists t_rd_fd(id int,name string);

--源数据data_t_rd_fd.txt
--666$$张三||555$$李四||

-- -rd行分割 -fd列分割 -charset编码
odps@ work_test_1>tunnel upload D:\#ODPS_D2\odps_data\data_t_rd_fd.txt work_test_1.t_rd_fd -rd "||" -fd "$$" -charset gbk;
Upload session: 20200220231800a0dbdb0b12e98660
Start upload:D:\#ODPS_D2\odps_data\data_t_rd_fd.txt
Using || to split records
Upload in strict schema mode: true
Total bytes:26   Split input to 1 blocks
0000-00-00 00:00:00     scan block: '1'
0000-00-00 00:00:00     scan block complete, block id: 1
0000-00-00 00:00:00     upload block: '1'
0000-00-00 00:00:00     upload block complete, block id: 1
OK

--检查一下
odps@ work_test_1>select * from work_test_1.t_rd_fd;
+------------+------------+
| id         | name       |
+------------+------------+
| 666        | 瀵姳绗?      |
| 555        | 閺夊骸娲?      |
+------------+------------+

2.脏数据-dbr true -s only ；tunnel show bad 0000；

--源数据data_t_rd_fd.txt
--666,张三
--555,李四 
--1.34,王五
--AA

-- -dbr true -s only;审查一下有哪些脏数据 仅读
odps@ work_test_1>tunnel upload D:\#ODPS_D2\odps_data\data_t_rd_fd.txt work_test_1.t_rd_fd -dbr true -s only;
Upload session: 000000000009151647df0b12dd8e9e  --注意这个session
Start upload:D:\#ODPS_D2\odps_data\data_t_rd_fd.txt
Using \r\n to split records
Upload in strict schema mode: true
Total bytes:39   Split input to 1 blocks
0000-00-00 00:00:00     scan block: '1'
0000-00-00 00:00:00     ERROR: format error - :1, BIGINT:'1.34'  For input string: "1.34"content: 1.34,鐜嬩簲
offset: 29
0000-00-00 00:00:00     ERROR: column mismatch, expected 2 columns, 1 columns found, please check data or delimiter
content: AA
offset: 34
0000-00-00 00:00:00     scan block complete, block id: 1 [bad 2]
OK
odps@ work_test_1>
--tunnel show bad 具体session;查看一下具体是哪些脏数据
--通过这种方法可以快速找到错误信息对源文件修正
odps@ work_test_1>tunnel show bad 0000000000329151647df0b12dd8e9e;
1.34,鐜嬩簲
AA
odps@ work_test_1>
--然后可以去源文件修正或删除，再上传