mysql：恢复使用mysqldump中的“--tab”选项创建的备份

Question

mysqldump has the "--tab" option to split up the dump separate files for each table. mysql creates a .sql file (with schema) and a .txt file (with data) for each table.

The .sql files work fine, but how do I import the data from the .txt files?

Answer 1

我自己找到了解决方案，请看这里： http ： //dev.mysql.com/doc/refman/5.1/en/reloading-delimited-text-dumps.html

Answer 2

Official documentation doesn't covert the case of importing these *.txt and foreign key constrains. There is still open long-lived bug#19996 to implement foreign key ignore in mysqlimport . So it has to be done manually.

#!/bin/bash -e

DIR=/path/to/csv/backup
DATABASE=database
USER=user
PASSWORD=password

for filename in $DIR/*.txt
do
tablename=`basename $filename .txt`
mysql --user=$USER --password=$PASSWORD $DATABASE <<EOF
  SET FOREIGN_KEY_CHECKS=0;
  LOAD DATA INFILE '$filename' INTO TABLE \`$tablename\`;
EOF
done

Also I would like to note that there's no much sense in storing these per-table *.sql files produced by mysqldump --tab , because there's also a foreign key issue on fields. As a schema is always known I suggest just to delete them after backup is completed.