最安全的方法是先将列修改为二进制类型，然后使用所需的字符集将其修改回二进制类型。

每种列类型都有其各自的二进制类型，如下所示:

>二进制 文本=> BLOB 丁文字=> TINYBLOB 文本=> memblob LONGTEXT => LONGBLOB VARCHAR => VARBINARY

Eg.:

ALTER TABLE [TABLE_SCHEMA].[TABLE_NAME] MODIFY [COLUMN_NAME] VARBINARY;

ALTER TABLE [TABLE_SCHEMA].[TABLE_NAME] MODIFY [COLUMN_NAME] VARCHAR(140) CHARACTER SET utf8mb4;

我尝试了几个拉丁表，它保留了所有的变音符。

你可以为所有列提取这个查询:

SELECT
CONCAT('ALTER TABLE ', TABLE_SCHEMA,'.', TABLE_NAME,' MODIFY ', COLUMN_NAME,' VARBINARY;'),
CONCAT('ALTER TABLE ', TABLE_SCHEMA,'.', TABLE_NAME,' MODIFY ', COLUMN_NAME,' ', COLUMN_TYPE,' CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci;')
FROM information_schema.columns
WHERE TABLE_SCHEMA IN ('[TABLE_SCHEMA]')
AND COLUMN_TYPE LIKE 'varchar%'
AND (COLLATION_NAME IS NOT NULL AND COLLATION_NAME NOT LIKE 'utf%');

在所有列上执行此操作后，再对所有表执行此操作:

ALTER TABLE [TABLE_SCHEMA].[TABLE_NAME] CONVERT TO CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci;

要为你所有的表生成这个查询，使用下面的查询:

SELECT
CONCAT('ALTER TABLE ', TABLE_SCHEMA, '.', TABLE_NAME, ' CONVERT TO CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci;')
FROM INFORMATION_SCHEMA.TABLES
WHERE TABLE_COLLATION NOT LIKE 'utf8%'
and TABLE_SCHEMA in ('[TABLE_SCHEMA]');

现在你修改了所有的列和表，在数据库上做同样的事情:

ALTER DATABASE [DATA_BASE_NAME] CHARSET = utf8mb4 COLLATE = utf8mb4_general_ci;

使用HeidiSQL。它是免费的，而且是一个非常好的数据库工具。

从工具菜单，进入批量表编辑器

选择完整的数据库或选择表进行转换，

tick更改默认排序规则:utf8mb4_general_ci tick转换为字符集:utf8

执行

这将在短短几秒钟内将完整的数据库从拉丁语转换为utf8。

工作就像一个魅力:)

HeidiSQL默认连接为utf8，因此任何特殊字符现在都应该被视为字符(æ ø å)，而不是在检查表数据时被编码。

从latin到utf8的真正陷阱是确保pdo与utf8字符集连接。如果不是，你会得到垃圾数据插入到utf8表和问号在你的网页上到处都是，让你认为表数据不是utf8…

受@sdfor注释的启发，这里有一个bash脚本可以完成这项工作

#!/bin/bash

printf "### Converting MySQL character set ###\n\n"

printf "Enter the encoding you want to set: "
read -r CHARSET

# Get the MySQL username
printf "Enter mysql username: "
read -r USERNAME

# Get the MySQL password
printf "Enter mysql password for user %s:" "$USERNAME"
read -rs PASSWORD

DBLIST=( mydatabase1 mydatabase2 )

printf "\n"


for DB in "${DBLIST[@]}"
do
(
    echo 'ALTER DATABASE `'"$DB"'` CHARACTER SET utf8 COLLATE `'"$CHARSET"'`;'
    mysql "$DB" -u"$USERNAME" -p"$PASSWORD" -e "SHOW TABLES" --batch --skip-column-names \
    | xargs -I{} echo 'ALTER TABLE `'{}'` CONVERT TO CHARACTER SET utf8 COLLATE `'"$CHARSET"'`;'
) \
| mysql "$DB" -u"$USERNAME" -p"$PASSWORD"

echo "$DB database done..."
done

echo "### DONE ###"
exit

如果你不能让你的表转换，或者你的表总是设置为一些非utf8字符集，但你想要utf8，你最好的办法可能是擦掉它，重新开始，并显式地指定:

create database database_name character set utf8;

DELIMITER $$  

CREATE PROCEDURE `databasename`.`update_char_set`()  

BEGIN  
 DECLARE done INT DEFAULT 0;  
 DECLARE t_sql VARCHAR(256);  
 DECLARE tableName VARCHAR(128);  
 DECLARE lists CURSOR FOR SELECT table_name FROM `information_schema`.`TABLES` WHERE table_schema = 'databasename';  
 DECLARE CONTINUE HANDLER FOR SQLSTATE '02000' SET done = 1;  
 OPEN lists;  
 FETCH lists INTO tableName;  
 REPEAT  
    SET @t_sql = CONCAT('ALTER TABLE ', tableName, ' CONVERT TO CHARACTER SET utf8 COLLATE utf8_general_ci');  
    PREPARE stmt FROM @t_sql;  
    EXECUTE stmt;  
    DEALLOCATE PREPARE stmt;  
 FETCH lists INTO tableName;  
 UNTIL done END REPEAT;  
 CLOSE lists;  
END$$  

DELIMITER ;  

CALL databasename.update_char_set();

对于有大量表的数据库，你可以使用一个简单的php脚本来更新数据库和所有表的字符集，使用以下方法:

$conn = mysqli_connect($host, $username, $password, $database);

if ($conn->connect_error) {
  die("Connection failed: " . $conn->connect_error);
}

$alter_database_charset_sql = "ALTER DATABASE ".$database." CHARACTER SET utf8 COLLATE utf8_unicode_ci";
mysqli_query($conn, $alter_database_charset_sql);

$show_tables_result = mysqli_query($conn, "SHOW TABLES");
$tables  = mysqli_fetch_all($show_tables_result);

foreach ($tables as $index => $table) {
  $alter_table_sql = "ALTER TABLE ".$table[0]." CONVERT TO CHARACTER SET utf8  COLLATE utf8_unicode_ci";
  $alter_table_result = mysqli_query($conn, $alter_table_sql);
  echo "<pre>";
  var_dump($alter_table_result);
  echo "</pre>";
}