详解MySQL中DISTINCT去重的核心注意事项

SELECT DISTINCT a, b FROM table;  -- 对(a,b)组合整体去重

-- 错误理解：以为只对name去重
SELECT DISTINCT name, class FROM students;  

INSERT INTO students (name, class, score) VALUES (NULL, '三班', 90);

​​​​​​​SELECT DISTINCT name FROM students;

-- 当表有百万行时慎用
SELECT DISTINCT text_column FROM huge_table; 

-- 先通过WHERE缩小范围再去重
SELECT DISTINCT text_column 
FROM huge_table 
WHERE create_time > '2023-01-01';

-- 或添加索引（对text类型有限制）
ALTER TABLE huge_table ADD INDEX idx_text(text_column(20)); -- 前缀索引

SELECT DISTINCT class 
FROM students 
ORDER BY score DESC; -- 错误！score不在SELECT中

-- 方案1：排序字段必须在SELECT中
SELECT DISTINCT class, MAX(score) AS max_score 
FROM students 
GROUP BY class 
ORDER BY max_score DESC;

-- 方案2：子查询
SELECT DISTINCT class FROM (
  SELECT class, score FROM students ORDER BY score DESC
) AS tmp;

-- 统计不重复的班级数量
SELECT COUNT(DISTINCT class) FROM students; 

-- 错误用法（语法无效）：
SELECT DISTINCT COUNT(class) FROM students; 

-- 尝试计算不同班级的平均分（错误！）
SELECT DISTINCT class, AVG(score) FROM students; 

SELECT class, AVG(score) 
FROM students 
GROUP BY class;  -- 这才是标准解法

SELECT DISTINCT class FROM students LIMIT 2;

SHOW VARIABLES LIKE 'tmp_table_size'; -- 默认16MB

CREATE TABLE orders (
    id INT PRIMARY KEY,
    product_id INT,
    user_id INT,
    amount DECIMAL(10,2),
    coupon_code VARCHAR(20) -- 允许为NULL
);

-- 方案1：基础写法
SELECT DISTINCT user_id, coupon_code 
FROM orders 
WHERE coupon_code IS NOT NULL; -- 若需包含NULL则去掉WHERE

​​​​​​​-- 方案2：大数据量优化（添加联合索引）
ALTER TABLE orders ADD INDEX idx_user_coupon(user_id, coupon_code);
SELECT DISTINCT user_id, coupon_code FROM orders;