当前位置：首页 > news >正文

SQL聚集函数与GROUP BY实战指南

news 2026/8/10 3:56:54

1. 聚集函数与GROUP BY基础概念解析

在数据处理和分析工作中，我们经常需要对数据进行汇总统计。SQL中的聚集函数(aggregate functions)和GROUP BY子句就是专门为此设计的黄金搭档。这对组合能够将海量数据按照特定维度分组，然后对每个组别进行数值计算，最终输出简洁有力的统计结果。

聚集函数主要包括以下五种核心函数：

COUNT()：计算行数
SUM()：计算数值总和
AVG()：计算平均值
MAX()：获取最大值
MIN()：获取最小值

这些函数之所以被称为"聚集"函数，是因为它们能够将多行数据"聚集"为一个汇总值。而GROUP BY子句则负责定义数据分组的维度，两者配合使用可以生成各种维度的统计报表。

2. 基础语法结构与执行顺序

2.1 标准语法格式

完整的GROUP BY查询通常包含以下结构：

SELECT 列名1, 列名2, 聚集函数(列名3) FROM 表名 WHERE 过滤条件 GROUP BY 列名1, 列名2 HAVING 分组后过滤条件 ORDER BY 排序字段;

2.2 关键执行顺序

理解SQL语句的执行顺序对于正确使用GROUP BY至关重要：

FROM子句：确定数据来源表
WHERE子句：对原始数据进行筛选
GROUP BY子句：按照指定列分组
聚集函数计算：对每个分组进行计算
HAVING子句：对分组结果进行筛选
SELECT子句：选择最终显示的列
ORDER BY子句：对结果进行排序

特别注意：WHERE和HAVING的区别在于前者在分组前过滤行，后者在分组后过滤组。

3. 五种聚集函数深度解析

3.1 COUNT函数的多面性

COUNT()函数有三种常见用法：

-- 计算所有行数(包括NULL) SELECT COUNT(*) FROM employees; -- 计算特定列的非NULL值数量 SELECT COUNT(department_id) FROM employees; -- 计算不重复值的数量 SELECT COUNT(DISTINCT department_id) FROM employees;

实际应用中，COUNT(*)通常比COUNT(列名)性能更好，因为不需要检查NULL值。

3.2 SUM函数的注意事项

SUM()函数专门用于数值型数据：

-- 基本用法 SELECT SUM(salary) FROM employees; -- 配合CASE语句实现条件求和 SELECT SUM(CASE WHEN gender = 'M' THEN salary ELSE 0 END) AS male_salary, SUM(CASE WHEN gender = 'F' THEN salary ELSE 0 END) AS female_salary FROM employees;

重要提示：SUM()会忽略NULL值，对非数值列使用SUM()会导致错误。

3.3 AVG函数的精度问题

AVG()函数计算平均值时需要注意：

-- 基本用法 SELECT AVG(salary) FROM employees; -- 等价于SUM()/COUNT() SELECT SUM(salary)/COUNT(salary) FROM employees;

浮点数精度问题：AVG()的结果可能会包含多位小数，可以使用ROUND()函数控制显示精度。

3.4 MAX/MIN函数的特殊用法

除了常规用法外，MAX/MIN还可以：

-- 获取最早/最晚日期 SELECT MIN(hire_date), MAX(hire_date) FROM employees; -- 配合DISTINCT使用 SELECT MAX(DISTINCT salary) FROM employees;

有趣的事实：MAX/MIN也可以用于文本数据，按照字典顺序比较。

4. GROUP BY高级应用技巧

4.1 多列分组统计

GROUP BY支持按多个列分组，生成更细致的统计维度：

SELECT department_id, job_id, COUNT(*) AS employee_count, AVG(salary) AS avg_salary FROM employees GROUP BY department_id, job_id;

这种多维分组在生成交叉报表时特别有用。

4.2 表达式分组

GROUP BY不仅限于列名，还可以使用表达式：

-- 按年份分组统计 SELECT EXTRACT(YEAR FROM hire_date) AS hire_year, COUNT(*) AS new_hires FROM employees GROUP BY EXTRACT(YEAR FROM hire_date); -- 按薪资区间分组 SELECT CASE WHEN salary < 5000 THEN '低薪' WHEN salary BETWEEN 5000 AND 10000 THEN '中薪' ELSE '高薪' END AS salary_level, COUNT(*) AS employee_count FROM employees GROUP BY salary_level;

4.3 ROLLUP与CUBE扩展

对于需要多层次汇总的场景，可以使用扩展功能：

-- ROLLUP生成小计和总计 SELECT department_id, job_id, COUNT(*) AS employee_count FROM employees GROUP BY ROLLUP(department_id, job_id); -- CUBE生成所有可能的组合 SELECT department_id, job_id, COUNT(*) AS employee_count FROM employees GROUP BY CUBE(department_id, job_id);

ROLLUP会生成从详细到汇总的层级结构，而CUBE会生成所有维度的组合。

5. 常见问题与性能优化

5.1 易犯错误集锦

SELECT列表不一致：

-- 错误：select列表包含非分组列 SELECT department_id, employee_name, AVG(salary) FROM employees GROUP BY department_id;

HAVING滥用：

-- 错误：对分组前过滤使用HAVING SELECT department_id, AVG(salary) FROM employees GROUP BY department_id HAVING salary > 5000; -- 应该用WHERE

NULL值分组： GROUP BY会将所有NULL值归为一组，这有时会导致意外结果。

5.2 性能优化建议

索引策略：
- 为GROUP BY列创建索引
- 复合索引顺序应与GROUP BY顺序一致
减少分组列数：分组列越多，性能开销越大，应只选择必要的分组维度。

先过滤后分组：

-- 更高效 SELECT department_id, AVG(salary) FROM employees WHERE hire_date > '2020-01-01' GROUP BY department_id; -- 低效 SELECT department_id, AVG(salary) FROM employees GROUP BY department_id HAVING MIN(hire_date) > '2020-01-01';

考虑使用物化视图：对于频繁执行的复杂分组查询，可以预先计算并存储结果。

6. 实际应用案例

6.1 销售数据分析

SELECT EXTRACT(YEAR FROM order_date) AS year, EXTRACT(MONTH FROM order_date) AS month, product_category, COUNT(DISTINCT customer_id) AS unique_customers, SUM(quantity) AS total_units_sold, SUM(quantity * unit_price) AS total_revenue, AVG(quantity * unit_price) AS avg_order_value FROM orders GROUP BY EXTRACT(YEAR FROM order_date), EXTRACT(MONTH FROM order_date), product_category ORDER BY year, month, product_category;

6.2 网站访问统计

SELECT DATE_TRUNC('day', visit_time) AS visit_date, traffic_source, COUNT(*) AS page_views, COUNT(DISTINCT user_id) AS unique_visitors, AVG(time_spent) AS avg_time_spent, SUM(CASE WHEN converted THEN 1 ELSE 0 END) AS conversions FROM website_visits GROUP BY DATE_TRUNC('day', visit_time), traffic_source HAVING COUNT(*) > 100 -- 只统计有足够样本的组 ORDER BY visit_date DESC, conversions DESC;

6.3 员工绩效报表

SELECT d.department_name, e.job_title, COUNT(*) AS headcount, ROUND(AVG(e.salary), 2) AS avg_salary, MIN(e.hire_date) AS oldest_hire, MAX(e.hire_date) AS newest_hire, SUM(CASE WHEN p.rating >= 4 THEN 1 ELSE 0 END) AS high_performers, ROUND(100.0 * SUM(CASE WHEN p.rating >= 4 THEN 1 ELSE 0 END) / COUNT(*), 1) AS high_performer_pct FROM employees e JOIN departments d ON e.department_id = d.department_id LEFT JOIN performance_reviews p ON e.employee_id = p.employee_id GROUP BY d.department_name, e.job_title ORDER BY d.department_name, high_performer_pct DESC;

7. 与其他SQL特性的结合使用

7.1 窗口函数对比

虽然GROUP BY进行数据聚合，但窗口函数可以保留原始行：

-- GROUP BY聚合 SELECT department_id, AVG(salary) FROM employees GROUP BY department_id; -- 窗口函数 SELECT employee_id, department_id, salary, AVG(salary) OVER (PARTITION BY department_id) AS dept_avg_salary FROM employees;

7.2 与JOIN结合

GROUP BY经常与多表连接一起使用：

SELECT d.department_name, l.city, COUNT(e.employee_id) AS employee_count FROM employees e JOIN departments d ON e.department_id = d.department_id JOIN locations l ON d.location_id = l.location_id GROUP BY d.department_name, l.city;

7.3 子查询中的GROUP BY

GROUP BY结果可以作为子查询：

SELECT department_id, avg_salary FROM ( SELECT department_id, AVG(salary) AS avg_salary FROM employees GROUP BY department_id ) dept_stats WHERE avg_salary > (SELECT AVG(salary) FROM employees);