千家信息网

Oracle使用sqluldr2导出数据

发表于:2025-02-01 作者:千家信息网编辑
千家信息网最后更新 2025年02月01日,sqluldr2是一个基于OCI的文本导出工具,小巧方便,使用方法类似于Oracle自带的exp,支持自定义SQL、本地和客户端的导出,速度快,效率高。下载地址,最下面有个小工具里面就有http://
千家信息网最后更新 2025年02月01日Oracle使用sqluldr2导出数据

sqluldr2是一个
基于OCI的文本导出工具,小巧方便,使用方法类似于Oracle自带的exp,支持自定义SQL、本地和客户端的导出,速度快,效率高。

下载地址,最下面有个小工具里面就有

  1. http://www.onexsoft.com/download.html
也可以直接在系统上wget

  1. wget http://www.onexsoft.com/software/sqluldr2linux64.zip
解压之后就是一个bin文件

  1. -rwxrwxrwx 1 oracle oinstall 185894 1月 27 2016 sqluldr2linux64.bin
直接执行就能显示帮助

  1. [oracle@hecomoracleserver2 sqluldr2]$ ./sqluldr2linux64.bin

  2. SQL*UnLoader: Fast Oracle Text Unloader (GZIP, Parallel), Release 4.0.1
  3. (@) Copyright Lou Fangxin (AnySQL.net) 2004 - 2010, all rights reserved.

  4. License: Free for non-commercial useage, else 100 USD per server.

  5. Usage: SQLULDR2 keyword=value [,keyword=value,...]

  6. Valid Keywords:
  7. user = username/password@tnsname
  8. sql = SQL file name
  9. query = select statement
  10. field = separator string between fields
  11. record = separator string between records
  12. rows = print progress for every given rows (default, 1000000)
  13. file = output file name(default: uldrdata.txt)
  14. log = log file name, prefix with + to append mode
  15. fast = auto tuning the session level parameters(YES)
  16. text = output type (MYSQL, CSV, MYSQLINS, ORACLEINS, FORM, SEARCH).
  17. charset = character set name of the target database.
  18. ncharset= national character set name of the target database.
  19. parfile = read command option from parameter file

  20. for field and record, you can use '0x' to specify hex character code,
  21. \r=0x0d \n=0x0a |=0x7c ,=0x2c, \t=0x09, :=0x3a, #=0x23, "=0x22
上面的参数中,比较常用的是:

query:用这个参数可以自定义导出语句
file:导出的文件名字和目录
log:日志
parfile:可以把参数放到parfile文件里,这个参数对于复杂sql很有用。
field:设置导出文件里的分隔符

连接数据的方法:


本地连接方式:

  1. ./sqluldr2linux64.bin user=gltest/gltest query="select * from T_H_BK_CUSTOMER_20180114_0;" file=/home/oracle/sqluldr2/T_H_BK_CUSTOMER_20180114_0.csv
客户端连接:TNS方式

  1. ./sqluldr2linux64.bin user=gltest/gltest@orcl query="select * from T_H_BK_CUSTOMER_20180114_0;" file=/home/oracle/sqluldr2/T_H_BK_CUSTOMER_20180114_0.csv
客户端连接:直接连接方式

  1. ./sqluldr2linux64.bin user=gltest/gltest@192.168.128.98:1521/orcl query="select * from T_H_BK_CUSTOMER_20180114_0;" file=/home/oracle/sqluldr2/T_H_BK_CUSTOMER_20180114_0.csv

测试导出速度:

因为我的sql比较长,所以用了parfile:

  1. [oracle@gloracleserver2 sqluldr2]$ more customer.txt
  2. user=gltest/gltest
  3. query=select ent_code,code,name,cust_levels,dept_code,poi_id,status,createon,createby,lastupdateon from v30_md_customer;
看看导出输出,1100万数据不到30秒,速度很快。

  1. [oracle@gloracleserver2 sqluldr2]$ ./sqluldr2linux64.bin parfile=/home/oracle/sqluldr2/customer.txt file=/home/oracle/sqluldr2/customer.csv
  2. 0 rows exported at 2018-01-16 20:25:09, size 0 MB.
  3. 1000000 rows exported at 2018-01-16 20:25:11, size 96 MB.
  4. 2000000 rows exported at 2018-01-16 20:25:14, size 196 MB.
  5. 3000000 rows exported at 2018-01-16 20:25:16, size 296 MB.
  6. 4000000 rows exported at 2018-01-16 20:25:18, size 392 MB.
  7. 5000000 rows exported at 2018-01-16 20:25:21, size 492 MB.
  8. 6000000 rows exported at 2018-01-16 20:25:23, size 592 MB.
  9. 7000000 rows exported at 2018-01-16 20:25:26, size 692 MB.
  10. 8000000 rows exported at 2018-01-16 20:25:28, size 788 MB.
  11. 9000000 rows exported at 2018-01-16 20:25:31, size 888 MB.
  12. 10000000 rows exported at 2018-01-16 20:25:34, size 988 MB.
  13. 11000000 rows exported at 2018-01-16 20:25:36, size 1088 MB.
  14. 11317061 rows exported at 2018-01-16 20:25:37, size 1121 MB.
  15. output file /home/oracle/sqluldr2/customer.csv closed at 11317061 rows, size 1121 MB.

遇到的错误:

1. 导出文件里有中文显示乱码,需要设置参数charset=UTF8

2. ORA-24345: A Truncation or null fetch error occurred,设置参数safe=yes








0