在 NATURAL JOIN 上使用 WHERE 子句
Using WHERE clause on NATURAL JOIN
我正在学习数据库系统概念第 6 版这本书。它使用大学数据库的示例。
该数据库的架构如下。粗体代表主键：
department(dept_name, building, budget)
course(course_id, title, dept_name, credits)
instructor(ID, name, dept_name, salary)
section(course_id, sec_id, semester, year, building, room number, time_slot_id)
teaches(ID, course_id, sec_id, semester, year)
prereq(course_id, prereq_id)
书中陈述了以下查询：
"List the names of instructors along with the the titles of courses that they teach".
书上给出的答案是：
select name, title
from instructor natural join teaches, course
where teaches.course id = course.course id;
此外，作者写道：
"相比之下，以下 SQL 查询不会计算相同的结果：
select name, title
from instructor natural join teaches natural join course;
要了解原因，请注意 instructor 和 teaches 的自然连接包含属性
(ID, name, dept name, salary, course id, sec id)，而课程关系包含
属性（课程 ID、标题、部门名称、学分）。结果，这些的自然结合
两个将要求来自两个输入的部门名称属性值是
相同，除了要求课程 ID 值相同。这个查询
然后将省略所有（教师姓名，课程名称）对，其中教师
在教师所在部门以外的部门教授课程。
另一方面，先前的查询正确地输出了这样的对。 “
现在我尝试了这两个查询，看看自己的区别。
mysql> select * from instructor;
+-------+------------+------------+----------+
| ID | name | dept_name | salary |
+-------+------------+------------+----------+
| 10101 | Srinivasan | Comp. Sci. | 65000.00 |
| 12121 | Wu | Finance | 90000.00 |
| 15151 | Mozart | Music | 40000.00 |
| 22222 | Einstein | Physics | 95000.00 |
| 32343 | El Said | History | 60000.00 |
| 33456 | Gold | Physics | 87000.00 |
| 45565 | Katz | Comp. Sci. | 75000.00 |
| 58583 | Califieri | History | 62000.00 |
| 76543 | Singh | Finance | 80000.00 |
| 76766 | Crick | Biology | 72000.00 |
| 83821 | Brandt | Comp. Sci. | 92000.00 |
| 98345 | Kim | Elec. Eng. | 80000.00 |
+-------+------------+------------+----------+
12 rows in set (0.00 sec)
mysql> select * from course;
+-----------+----------------------------+------------+---------+
| course_id | title | dept_name | credits |
+-----------+----------------------------+------------+---------+
| BIO-101 | Intro. to Biology | Biology | 4 |
| BIO-301 | Genetics | Biology | 4 |
| BIO-399 | Computational Biology | Biology | 3 |
| CS-101 | Intro. to Computer Science | Comp. Sci. | 4 |
| CS-190 | Game Design | Comp. Sci. | 4 |
| CS-315 | Robotica | Comp. Sci. | 3 |
| CS-319 | Image Processing | Comp. Sci. | 3 |
| CS-347 | Database System Concepts | Comp. Sci. | 3 |
| EE-181 | Intro. to Digital Systems | Elec. Eng. | 3 |
| FIN-201 | Investment Banking | Finance | 3 |
| HIS-351 | World History | History | 3 |
| MU-199 | Music Video Production | Music | 3 |
| PHY-101 | Physical Principles | Physics | 4 |
+-----------+----------------------------+------------+---------+
mysql> select * from teaches;
+-------+-----------+--------+----------+------+
| ID | course_id | sec_id | semester | year |
+-------+-----------+--------+----------+------+
| 76766 | BIO-101 | 1 | Summer | 2009 |
| 76766 | BIO-301 | 1 | Summer | 2010 |
| 10101 | CS-101 | 1 | Fall | 2009 |
| 45565 | CS-101 | 1 | Spring | 2010 |
| 83821 | CS-190 | 1 | Spring | 2009 |
| 83821 | CS-190 | 2 | Spring | 2009 |
| 10101 | CS-315 | 1 | Spring | 2010 |
| 45565 | CS-319 | 1 | Spring | 2010 |
| 83821 | CS-319 | 2 | Spring | 2010 |
| 10101 | CS-347 | 1 | Fall | 2009 |
| 98345 | EE-181 | 1 | Spring | 2009 |
| 12121 | FIN-201 | 1 | Spring | 2010 |
| 32343 | HIS-351 | 1 | Spring | 2010 |
| 15151 | MU-199 | 1 | Spring | 2010 |
| 22222 | PHY-101 | 1 | Fall | 2009 |
+-------+-----------+--------+----------+------+
15 rows in set (0.00 sec)
mysql> select name, title
-> from instructor natural join teaches, course
-> where teaches.course_id = course.course_id;
+------------+----------------------------+
| name | title |
+------------+----------------------------+
| Crick | Intro. to Biology |
| Crick | Genetics |
| Srinivasan | Intro. to Computer Science |
| Katz | Intro. to Computer Science |
| Brandt | Game Design |
| Brandt | Game Design |
| Srinivasan | Robotica |
| Katz | Image Processing |
| Brandt | Image Processing |
| Srinivasan | Database System Concepts |
| Kim | Intro. to Digital Systems |
| Wu | Investment Banking |
| El Said | World History |
| Mozart | Music Video Production |
| Einstein | Physical Principles |
+------------+----------------------------+
15 rows in set (0.00 sec)
mysql> select name, title
-> from instructor natural join teaches natural join course;
+------------+----------------------------+
| name | title |
+------------+----------------------------+
| Crick | Intro. to Biology |
| Crick | Genetics |
| Srinivasan | Intro. to Computer Science |
| Katz | Intro. to Computer Science |
| Brandt | Game Design |
| Brandt | Game Design |
| Srinivasan | Robotica |
| Katz | Image Processing |
| Brandt | Image Processing |
| Srinivasan | Database System Concepts |
| Kim | Intro. to Digital Systems |
| Wu | Investment Banking |
| El Said | World History |
| Mozart | Music Video Production |
| Einstein | Physical Principles |
+------------+----------------------------+
15 rows in set (0.00 sec)
从结果上看没有区别，而且我不明白为什么会有区别。
与书中所述不同，instructor 和 teaches 的自然连接包含属性 ID、name、dept name、salary , 课程 ID, sec id, 学期, 年份.
这本书是正确的。这些查询是不等价的。示例数据（detp_name值不同）：
INSERT INTO course(course_id, title, dept_name, credits)
VALUES (1, 'Course_1', 'DeptX', 10);
INSERT INTO instructor(ID, name, dept_name, salary)
VALUES (1, 'Prof X', 'DeptY', 10000);
INSERT INTO teaches(ID, course_id, sec_id, semester, year)
VALUES (1, 1, 10,10,2000);
就我个人而言，我不会使用任何建议的查询：
-- mixing natural join with old-comma syntax
select name, title
from instructor natural join teaches, course
where teaches.course_id = course.course_id;
--Output
--name title
--Prof X Course_1
-- natural join(potentially dangerous-depends on names of columns)
select name, title
from instructor natural join teaches natural join course;
-- 0 rows selected
并将其重写为：
SELECT i.name, c.title
FROM instructor i
JOIN teaches t
ON t.ID = i.ID
JOIN course c
ON t.course_id = c.course_id;
我正在学习数据库系统概念第 6 版这本书。它使用大学数据库的示例。 该数据库的架构如下。粗体代表主键：
department(dept_name, building, budget) course(course_id, title, dept_name, credits) instructor(ID, name, dept_name, salary) section(course_id, sec_id, semester, year, building, room number, time_slot_id) teaches(ID, course_id, sec_id, semester, year) prereq(course_id, prereq_id)
书中陈述了以下查询：
"List the names of instructors along with the the titles of courses that they teach".
书上给出的答案是：
select name, title
from instructor natural join teaches, course
where teaches.course id = course.course id;
此外，作者写道：
"相比之下，以下 SQL 查询不会计算相同的结果：
select name, title
from instructor natural join teaches natural join course;
要了解原因，请注意 instructor 和 teaches 的自然连接包含属性 (ID, name, dept name, salary, course id, sec id)，而课程关系包含 属性（课程 ID、标题、部门名称、学分）。结果，这些的自然结合 两个将要求来自两个输入的部门名称属性值是 相同，除了要求课程 ID 值相同。这个查询 然后将省略所有（教师姓名，课程名称）对，其中教师 在教师所在部门以外的部门教授课程。 另一方面，先前的查询正确地输出了这样的对。 “
现在我尝试了这两个查询，看看自己的区别。
mysql> select * from instructor;
+-------+------------+------------+----------+
| ID | name | dept_name | salary |
+-------+------------+------------+----------+
| 10101 | Srinivasan | Comp. Sci. | 65000.00 |
| 12121 | Wu | Finance | 90000.00 |
| 15151 | Mozart | Music | 40000.00 |
| 22222 | Einstein | Physics | 95000.00 |
| 32343 | El Said | History | 60000.00 |
| 33456 | Gold | Physics | 87000.00 |
| 45565 | Katz | Comp. Sci. | 75000.00 |
| 58583 | Califieri | History | 62000.00 |
| 76543 | Singh | Finance | 80000.00 |
| 76766 | Crick | Biology | 72000.00 |
| 83821 | Brandt | Comp. Sci. | 92000.00 |
| 98345 | Kim | Elec. Eng. | 80000.00 |
+-------+------------+------------+----------+
12 rows in set (0.00 sec)
mysql> select * from course;
+-----------+----------------------------+------------+---------+
| course_id | title | dept_name | credits |
+-----------+----------------------------+------------+---------+
| BIO-101 | Intro. to Biology | Biology | 4 |
| BIO-301 | Genetics | Biology | 4 |
| BIO-399 | Computational Biology | Biology | 3 |
| CS-101 | Intro. to Computer Science | Comp. Sci. | 4 |
| CS-190 | Game Design | Comp. Sci. | 4 |
| CS-315 | Robotica | Comp. Sci. | 3 |
| CS-319 | Image Processing | Comp. Sci. | 3 |
| CS-347 | Database System Concepts | Comp. Sci. | 3 |
| EE-181 | Intro. to Digital Systems | Elec. Eng. | 3 |
| FIN-201 | Investment Banking | Finance | 3 |
| HIS-351 | World History | History | 3 |
| MU-199 | Music Video Production | Music | 3 |
| PHY-101 | Physical Principles | Physics | 4 |
+-----------+----------------------------+------------+---------+
mysql> select * from teaches;
+-------+-----------+--------+----------+------+
| ID | course_id | sec_id | semester | year |
+-------+-----------+--------+----------+------+
| 76766 | BIO-101 | 1 | Summer | 2009 |
| 76766 | BIO-301 | 1 | Summer | 2010 |
| 10101 | CS-101 | 1 | Fall | 2009 |
| 45565 | CS-101 | 1 | Spring | 2010 |
| 83821 | CS-190 | 1 | Spring | 2009 |
| 83821 | CS-190 | 2 | Spring | 2009 |
| 10101 | CS-315 | 1 | Spring | 2010 |
| 45565 | CS-319 | 1 | Spring | 2010 |
| 83821 | CS-319 | 2 | Spring | 2010 |
| 10101 | CS-347 | 1 | Fall | 2009 |
| 98345 | EE-181 | 1 | Spring | 2009 |
| 12121 | FIN-201 | 1 | Spring | 2010 |
| 32343 | HIS-351 | 1 | Spring | 2010 |
| 15151 | MU-199 | 1 | Spring | 2010 |
| 22222 | PHY-101 | 1 | Fall | 2009 |
+-------+-----------+--------+----------+------+
15 rows in set (0.00 sec)
mysql> select name, title
-> from instructor natural join teaches, course
-> where teaches.course_id = course.course_id;
+------------+----------------------------+
| name | title |
+------------+----------------------------+
| Crick | Intro. to Biology |
| Crick | Genetics |
| Srinivasan | Intro. to Computer Science |
| Katz | Intro. to Computer Science |
| Brandt | Game Design |
| Brandt | Game Design |
| Srinivasan | Robotica |
| Katz | Image Processing |
| Brandt | Image Processing |
| Srinivasan | Database System Concepts |
| Kim | Intro. to Digital Systems |
| Wu | Investment Banking |
| El Said | World History |
| Mozart | Music Video Production |
| Einstein | Physical Principles |
+------------+----------------------------+
15 rows in set (0.00 sec)
mysql> select name, title
-> from instructor natural join teaches natural join course;
+------------+----------------------------+
| name | title |
+------------+----------------------------+
| Crick | Intro. to Biology |
| Crick | Genetics |
| Srinivasan | Intro. to Computer Science |
| Katz | Intro. to Computer Science |
| Brandt | Game Design |
| Brandt | Game Design |
| Srinivasan | Robotica |
| Katz | Image Processing |
| Brandt | Image Processing |
| Srinivasan | Database System Concepts |
| Kim | Intro. to Digital Systems |
| Wu | Investment Banking |
| El Said | World History |
| Mozart | Music Video Production |
| Einstein | Physical Principles |
+------------+----------------------------+
15 rows in set (0.00 sec)
从结果上看没有区别，而且我不明白为什么会有区别。 与书中所述不同，instructor 和 teaches 的自然连接包含属性 ID、name、dept name、salary , 课程 ID, sec id, 学期, 年份.
这本书是正确的。这些查询是不等价的。示例数据（detp_name值不同）：
INSERT INTO course(course_id, title, dept_name, credits)
VALUES (1, 'Course_1', 'DeptX', 10);
INSERT INTO instructor(ID, name, dept_name, salary)
VALUES (1, 'Prof X', 'DeptY', 10000);
INSERT INTO teaches(ID, course_id, sec_id, semester, year)
VALUES (1, 1, 10,10,2000);
就我个人而言，我不会使用任何建议的查询：
-- mixing natural join with old-comma syntax
select name, title
from instructor natural join teaches, course
where teaches.course_id = course.course_id;
--Output
--name title
--Prof X Course_1
-- natural join(potentially dangerous-depends on names of columns)
select name, title
from instructor natural join teaches natural join course;
-- 0 rows selected
并将其重写为：
SELECT i.name, c.title
FROM instructor i
JOIN teaches t
ON t.ID = i.ID
JOIN course c
ON t.course_id = c.course_id;