下面的查询将生成ALTER查询，将所有表中所有适当列的排序规则更改为某种类型(在下面的示例中为utf8_general_ci)。

SELECT concat
        (
            'ALTER TABLE ', 
                t1.TABLE_SCHEMA, 
                '.', 
                t1.table_name, 
                ' MODIFY ', 
                t1.column_name, 
                ' ', 
                t1.data_type, 
                '(' , 
                    CHARACTER_MAXIMUM_LENGTH, 
                ')', 
                ' CHARACTER SET utf8 COLLATE utf8_general_ci;'
        )
from 
    information_schema.columns t1
where 
    t1.TABLE_SCHEMA like 'you_db_name_goes_here' AND
    t1.COLLATION_NAME IS NOT NULL AND
    t1.COLLATION_NAME NOT IN ('utf8_general_ci');

我使用了以下shell脚本。它将数据库名作为参数，并将所有表转换为另一个字符集和排序规则(由脚本中定义的另一个参数或默认值给出)。

#!/bin/bash

# mycollate.sh <database> [<charset> <collation>]
# changes MySQL/MariaDB charset and collation for one database - all tables and
# all columns in all tables

DB="$1"
CHARSET="$2"
COLL="$3"

[ -n "$DB" ] || exit 1
[ -n "$CHARSET" ] || CHARSET="utf8mb4"
[ -n "$COLL" ] || COLL="utf8mb4_general_ci"

echo $DB
echo "ALTER DATABASE $DB CHARACTER SET $CHARSET COLLATE $COLL;" | mysql

echo "USE $DB; SHOW TABLES;" | mysql -s | (
    while read TABLE; do
        echo $DB.$TABLE
        echo "ALTER TABLE $TABLE CONVERT TO CHARACTER SET $CHARSET COLLATE $COLL;" | mysql $DB
    done
)

我很惊讶地发现，所以我不得不回到这里报告，优秀和维护良好的Interconnect/it SAFE SEARCH and REPLACE ON DATABASE脚本有一些选项可以将表转换为utf8 / unicode，甚至转换为innodb。它是一个脚本，通常用于将数据库驱动的网站(Wordpress, Drupal, Joomla等)从一个域迁移到另一个域。

https://github.com/interconnectit/Search-Replace-DB https://interconnectit.com/products/search-and-replace-for-wordpress-databases/

只需运行这个SQL语句就可以一次转换所有数据库表。根据需要更改COLLATION和databaseName。

SELECT CONCAT("ALTER TABLE ", TABLE_SCHEMA, '.', TABLE_NAME," COLLATE utf8_general_ci;") AS    ExecuteTheString
FROM INFORMATION_SCHEMA.TABLES
WHERE TABLE_SCHEMA="databaseName"
AND TABLE_TYPE="BASE TABLE";

快速方法-导出到SQL文件，使用搜索和替换来更改您需要更改的文本。创建新数据库，导入数据，然后将旧数据库和新数据库重命名为旧名称。

我在这里读到，你需要手动转换每个表，这是不正确的。下面是一个如何使用存储过程的解决方案:

DELIMITER $$

DROP PROCEDURE IF EXISTS changeCollation$$

-- character_set parameter could be 'utf8'
-- or 'latin1' or any other valid character set
CREATE PROCEDURE changeCollation(IN character_set VARCHAR(255))
BEGIN
DECLARE v_finished INTEGER DEFAULT 0;
DECLARE v_table_name varchar(255) DEFAULT "";
DECLARE v_message varchar(4000) DEFAULT "No records";

-- This will create a cursor that selects each table,
-- where the character set is not the one
-- that is defined in the parameter

DECLARE alter_cursor CURSOR FOR SELECT DISTINCT TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE()
AND COLLATION_NAME NOT LIKE CONCAT(character_set, '_%');

-- This handler will set the value v_finished to 1
-- if there are no more rows

DECLARE CONTINUE HANDLER
FOR NOT FOUND SET v_finished = 1;

OPEN alter_cursor;

-- Start a loop to fetch each rows from the cursor
get_table: LOOP

-- Fetch the table names one by one
FETCH alter_cursor INTO v_table_name;

-- If there is no more record, then we have to skip
-- the commands inside the loop
IF v_finished = 1 THEN
LEAVE get_table;
END IF;

IF v_table_name != '' THEN

IF v_message = 'No records' THEN
SET v_message = '';
END IF;

-- This technic makes the trick, it prepares a statement
-- that is based on the v_table_name parameter and it means
-- that this one is different by each iteration inside the loop

SET @s = CONCAT('ALTER TABLE ',v_table_name,
' CONVERT TO CHARACTER SET ', character_set);
PREPARE stmt FROM @s;
EXECUTE stmt;
DEALLOCATE PREPARE stmt;

SET v_message = CONCAT('The table ', v_table_name ,
' was changed to the default collation of ', character_set,
'.\n', v_message);

SET v_table_name = '';

END IF;
-- Close the loop and the cursor
END LOOP get_table;
CLOSE alter_cursor;

-- Returns information about the altered tables or 'No records'
SELECT v_message;

END $$

DELIMITER ;

创建过程后，可以简单地调用它:

CALL changeCollation('utf8');

更多细节请阅读这篇博客。