MySQL 替换——查询还是在 Shell 中？

Question

我有很多表，正在进行一些时髦的字符替换。我找到了一些关于运行更新来替换这些信息的信息，但它存在于许多表和许多字段中。

update <table_name>
set <field> = replace(<field>,'’','\'');
update <table_name>
set <field>= replace(<field>,'…','...');
update <table_name>
set <field>= replace(<field>,'–','-');
update <table_name>
set <field>= replace(<field>,'“','"');
update <table_name>
set <field>= replace(<field>,'â€','"');
update <table_name>
set <field>= replace(<field>,'‘','\'');
update <table_name>
set <field>= replace(<field>,'•','-');
update <table_name>
set <field>= replace(<field>,'‡','c');

对于在所有表的每个字段中运行此操作，你们有什么建议吗？我想过做一个 mysql 转储并使用 sed 来替换所有内容，但我不确定如何构建命令，因为它们都是特殊字符。

任何帮助将不胜感激。

Answer 1

您可以尝试查询表information_schema.columns并查找 MySQL 数据库实例中具有字符字段的每个表。这是表格的布局：

mysql> show create table information_schema.columns\G
*************************** 1. row ***************************
       Table: COLUMNS
Create Table: CREATE TEMPORARY TABLE `COLUMNS` (
  `TABLE_CATALOG` varchar(512) NOT NULL DEFAULT '',
  `TABLE_SCHEMA` varchar(64) NOT NULL DEFAULT '',
  `TABLE_NAME` varchar(64) NOT NULL DEFAULT '',
  `COLUMN_NAME` varchar(64) NOT NULL DEFAULT '',
  `ORDINAL_POSITION` bigint(21) unsigned NOT NULL DEFAULT '0',
  `COLUMN_DEFAULT` longtext,
  `IS_NULLABLE` varchar(3) NOT NULL DEFAULT '',
  `DATA_TYPE` varchar(64) NOT NULL DEFAULT '',
  `CHARACTER_MAXIMUM_LENGTH` bigint(21) unsigned DEFAULT NULL,
  `CHARACTER_OCTET_LENGTH` bigint(21) unsigned DEFAULT NULL,
  `NUMERIC_PRECISION` bigint(21) unsigned DEFAULT NULL,
  `NUMERIC_SCALE` bigint(21) unsigned DEFAULT NULL,
  `CHARACTER_SET_NAME` varchar(32) DEFAULT NULL,
  `COLLATION_NAME` varchar(32) DEFAULT NULL,
  `COLUMN_TYPE` longtext NOT NULL,
  `COLUMN_KEY` varchar(3) NOT NULL DEFAULT '',
  `EXTRA` varchar(27) NOT NULL DEFAULT '',
  `PRIVILEGES` varchar(80) NOT NULL DEFAULT '',
  `COLUMN_COMMENT` varchar(1024) NOT NULL DEFAULT ''
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.01 sec)

以下是获取所有用户定义表的每个基于字符的列类型的查询：

mysql> select DISTINCT COLUMN_TYPE from information_schema.columns
    -> WHERE table_schema NOT IN ('mysql','information_schema','performance_schema')
    -> AND (COLUMN_TYPE REGEXP 'text$'
    -> OR COLUMN_TYPE REGEXP '^varchar'
    -> OR COLUMN_TYPE REGEXP '^char');
+--------------+
| COLUMN_TYPE  |
+--------------+
| mediumtext   |
| varchar(512) |
| varchar(20)  |
| longtext     |
| text         |
| varchar(200) |
| varchar(255) |
| varchar(100) |
| tinytext     |
| varchar(25)  |
| varchar(64)  |
| varchar(32)  |
| varchar(60)  |
| varchar(50)  |
| varchar(250) |
| varchar(150) |
| varchar(10)  |
| varchar(45)  |
| char(5)      |
+--------------+
19 rows in set (0.09 sec)

您可以将该查询加入到第二个查询中。第二个查询将包含具有这些列类型的每个表的名称：

SELECT BB.table_schema,BB.table_name,BB.COLUMN_NAME FROM
(select DISTINCT COLUMN_TYPE from information_schema.columns
WHERE table_schema NOT IN ('mysql','information_schema','performance_schema')
AND (COLUMN_TYPE REGEXP 'text$'
OR COLUMN_TYPE REGEXP '^varchar'
OR COLUMN_TYPE REGEXP '^char')) AA
INNER JOIN
(SELECT table_schema,table_name,COLUMN_NAME,COLUMN_TYPE
FROM information_schema.columns
WHERE table_schema NOT IN
('mysql','information_schema','performance_schema')) BB
USING (COLUMN_TYPE);

现在，向输出发送消息，使其为您编写通用脚本：

SELECT CONCAT('UPDATE ',db,'.',tb,' SET ',fld,
'=REPLACE(',fld,',''oldchar'',''newchar'');')
UpdateCommand
FROM (SELECT db,tb,fld FROM
(
    SELECT table_schema db,table_name tb,COLUMN_NAME fld FROM
    (select DISTINCT COLUMN_TYPE from information_schema.columns
    WHERE table_schema NOT IN ('mysql','information_schema','performance_schema')
    AND (COLUMN_TYPE REGEXP 'text$'
    OR COLUMN_TYPE REGEXP '^varchar'
    OR COLUMN_TYPE REGEXP '^char')) AAA
    INNER JOIN
    (SELECT table_schema,table_name,COLUMN_NAME,COLUMN_TYPE
    FROM information_schema.columns
    WHERE table_schema NOT IN
    ('mysql','information_schema','performance_schema')) BBB
    USING (COLUMN_TYPE)
) AA) A;

输出应如下所示：

+-----------------------------------------------------------------------------------------------+
| UpdateCommand                                                                                 |
+-----------------------------------------------------------------------------------------------+
| UPDATE example.user SET user_name=REPLACE(user_name,'oldchar','newchar');                     |
| UPDATE garbage.rolando SET post_content=REPLACE(post_content,'oldchar','newchar');            |
| UPDATE garbage.rolando SET post_title=REPLACE(post_title,'oldchar','newchar');                |
| UPDATE garbage.rolando SET post_excerpt=REPLACE(post_excerpt,'oldchar','newchar');            |
| UPDATE garbage.rolando SET post_status=REPLACE(post_status,'oldchar','newchar');              |
| UPDATE garbage.rolando SET comment_status=REPLACE(comment_status,'oldchar','newchar');        |
| UPDATE garbage.rolando SET ping_status=REPLACE(ping_status,'oldchar','newchar');              |
| UPDATE garbage.rolando SET post_password=REPLACE(post_password,'oldchar','newchar');          |
| UPDATE garbage.rolando SET post_name=REPLACE(post_name,'oldchar','newchar');                  |

通用脚本生成的最后一步是将查询输出到文本文件，如下所示：

mysql -u... -p... -A --skip-column-names -e"SELECT CONCAT('UPDATE ',db,'.',tb,' SET ',fld, '=REPLACE(',fld,',''oldchar'',''newchar'');') UpdateCommand FROM (SELECT db,tb,fld FROM (     SELECT table_schema db,table_name tb,COLUMN_NAME fld FROM     (select DISTINCT COLUMN_TYPE from information_schema.columns     WHERE table_schema NOT IN ('mysql','information_schema','performance_schema')     AND (COLUMN_TYPE REGEXP 'text$'     OR COLUMN_TYPE REGEXP '^varchar'     OR COLUMN_TYPE REGEXP '^char')) AAA     INNER JOIN     (SELECT table_schema,table_name,COLUMN_NAME,COLUMN_TYPE     FROM information_schema.columns     WHERE table_schema NOT IN     ('mysql','information_schema','performance_schema')) BBB USING (COLUMN_TYPE) ) AA) A;" > CharUpdateScript.sql

您可以从那里自定义通用脚本。