SQL选择n到m的关系

tar*_*aki 7 sql

我有一个n对米之间关系AuthorBook.

表作者

ID       Name
1        Follett  
2        Rowling
3        Martin
Run Code Online (Sandbox Code Playgroud)

表书

ID     Title                       Category 
1        A Dance with Dragons      Fantasy
2        Harry Potter              Fantasy
3        The Key to Rebecca        Thriller
4        World without end         Drama
Run Code Online (Sandbox Code Playgroud)

表book_author

authorId       bookId
1        3  
2        2
3        1
1        4
Run Code Online (Sandbox Code Playgroud)

系统中有更多的作者和书籍.现在我想选择所有拥有" 幻想 " 类型书籍的作者.

这是我到目前为止提出的:

   select distinct a.id 
   from author a, book b, written w 
   where w.authorId = a.id and w.bookId = b.id and b.category = "Fantasy";
Run Code Online (Sandbox Code Playgroud)

我想知道如何优化这个查询,因为特别是桌面书真的很大.

Mic*_*ski 7

建议使用显式JOIN而不是当前具有的隐式(逗号分隔表列表)连接,因为如果需要引入左连接,它将提高灵活性.

SELECT
  DISTINCT a.id
FROM
  author a
  JOIN book_author ba ON a.id = ba.authorId
  JOIN books b ON b.id = ba.bookId
WHERE b.category = 'Fantasy'
Run Code Online (Sandbox Code Playgroud)

如果book_author已将FOREIGN KEY 关系定义回authorbooks表,则将强制执行索引.同样,id这些表中的各列应定义为PRIMARY KEY.除此之外,您可以做的唯一潜在优化是创建索引books.category.

CREATE TABLE book_author (
  authorId INT NOT NULL, /* or whatever the data type... */
  bookId INT NOT NULL,
  /* define FK constraints in book_author */
  FOREIGN KEY (authorId) REFERENCES author (id),
  FOREIGN KEY (bookId) REFERENCES books (id)
);
Run Code Online (Sandbox Code Playgroud)