Question

我有一个项目，每周我会导入几个包含不正确数据的大型数据集，例如重复的员工ID，它们不应该是重复的。为了标记重复项，我尝试以下代码：

ALTER TABLE AccountDuplicates
  ADD UNIQUE INDEX EmployeeID (EmployeeID);

INSERT INTO AccountDuplicates
  SELECT
    EmployeeID,
    FirstName,
    LastName
  FROM AccountsWork
  ON DUPLICATE KEY UPDATE
    EmployeeID = CONCAT(VALUES(EmployeeID), '*');

INSERT语句给了我错误，我看不出我做错了什么：

[42000][1064] You have an error in your SQL syntax; check the manual that correspondsto your MySQL server version for the right syntax to use near 'FROM EAD_UserAccountsWork
           ON DUPLICATE KEY UPDATE EmployeeID = CONCAT(VALUES(E' at line 36

如果它是相关的，我在OS X 10.11.4，INNODB引擎和mysql_mode =''上运行MySQL 5.7.12。我的目的是识别重复的ID，以便我可以将它们转发给相应的DBA进行更正。

更新：我已按如下方式设置数据库默认值：

[client]
default-character-set = utf8mb4

[mysqld]
sql_mode=''
character-set-client-handshake = FALSE
character-set-server = utf8mb4
collation-server = utf8mb4_unicode_ci

[mysql]
default-character-set = utf8mb4

Answer 1

我认为这是因为您没有正确认定“来源”EmployeeID

INSERT INTO AccountDuplicates
  SELECT
    EmployeeID,
    FirstName,
    LastName
  FROM AccountsWork t
  ON DUPLICATE KEY UPDATE
    EmployeeID = CONCAT(t.EmployeeID, '*');

EmployeeID左侧的=引用AccountDuplicates表，右侧的AccountsWork引用from distutils.core import setup from distutils.extension import Extension from Cython.Build import cythonize import numpy as np setup( ext_modules = cythonize( [Extension("*",["*.pyx"], libraries =["MyLib"], extra_compile_args = ["-fopenmp","-O3"], extra_link_args=["-L/path/to/lib"]) ]), include_dirs = [np.get_include()], )表

Answer 2

你的语法看起来不错，也许有输入错误？这几乎有效：

-- drop table AccountsWork ;
-- drop table  AccountDuplicates;
CREATE TABLE AccountsWork (
    EmployeeID varchar(16),
    FirstName INT,
    LastName INT
);
CREATE TABLE AccountDuplicates (
    EmployeeID  varchar(16),
    FirstName INT,
    LastName INT
);
    alter table AccountDuplicates add unique index(EmployeeID); 
insert into AccountsWork values('a',2,3);
insert into AccountsWork values(1,2,3);
insert into AccountsWork values('b',2,3);
insert into AccountsWork values('c',2,3);
insert into AccountsWork values('c',2,3);
insert into AccountsWork values('c',2,3);
insert into AccountsWork values('c',2,3);
SELECT 
    *
FROM
    AccountsWork;
-- there is no syntax errors here (your original query):    
INSERT INTO AccountDuplicates
  SELECT
    EmployeeID,
    FirstName,
    LastName
  FROM AccountsWork
  ON DUPLICATE KEY UPDATE
    EmployeeID =  CONCAT(VALUES(EmployeeID), '*');

SELECT 
    *
FROM
    AccountDuplicates;

带有重复键的简单插入有效，但“从重复键更新表中选择”不起作用。

我认为您需要查看此问题：INSERT INTO ... SELECT FROM ... ON DUPLICATE KEY UPDATE

看起来mysql解析器对这种查询感到疯狂

使用DUPLICATE KEY UPDATE

2 个答案: