Question

这是我的表结构

 CREATE TABLE branch_book_list (
   branch_name CHAR(10),
   book_ISBN CHAR (13),
   book_name CHAR(40),
   PRIMARY KEY (branch_name, book_ISBN)
   )

我在表格中有这些数据行：

branch_name book_ISBN     book_name                                
----------- ------------- ---------------------------------------- 
Branch A    1-56592-756-7 Transact-SQL Cookbook                   
Branch A    1-56592-401-0 Transact-SQL Programming                
Branch A    1-56592-401-0 Transact-SQL Programming                
Branch B    1-56592-756-7 Transact-SQL Cookbook                   
Branch B    1-56592-756-7 Transact-SQL Cookbook                   
Branch B    1-56592-401-0 Transact-SQL Programming

执行此查询时：

SELECT bbl1.*, COUNT(*) DupeCount
FROM branch_book_list bbl1
WHERE bbl1.branch_name = 'Branch A'
GROUP BY bbl1.branch_name, bbl1.book_ISBN, bbl1.book_name
HAVING NOT EXISTS (
      SELECT bbl2.*, COUNT(*)
      FROM branch_book_list bbl2
      WHERE branch_name = 'Branch B'
      GROUP BY bbl2.branch_name, bbl2.book_ISBN, bbl2.book_name
      HAVING bbl1.book_ISBN = bbl2.book_ISBN
         AND bbl1.book_name = bbl2.book_name
         AND COUNT(*) = COUNT(ALL bbl1.book_ISBN))

OUTPUT

branch_name book_ISBN     book_name                                DupeCount   
----------- ------------- ---------------------------------------- ----------- 
Branch A    1-56592-401-0 Transact-SQL Programming                 2
Branch A    1-56592-756-7 Transact-SQL Cookbook                    1

当我只执行查询的顶部

时

SELECT bbl1.*, COUNT(*) DupeCount
FROM branch_book_list bbl1
WHERE bbl1.branch_name = 'Branch A'
GROUP BY bbl1.branch_name, bbl1.book_ISBN, bbl1.book_name

我的输出是

branch_name book_ISBN   book_name                   DupeCount
Branch A    1-56592-401-0   Transact-SQL Programming    2
Branch A    1-56592-756-7   Transact-SQL Cookbook       1

但是当执行完整查询时，我该如何获得此输出

branch_name book_ISBN     book_name                                DupeCount   
----------- ------------- ---------------------------------------- ----------- 
Branch A    1-56592-401-0 Transact-SQL Programming                 2
Branch A    1-56592-756-7 Transact-SQL Cookbook                    1

任何人都可以告诉我如何获得以上输出吗？

Answer 1

你的困惑是什么？过滤器WHERE bbl1.branch_name = 'Branch A' 将所有三个查询限制为仅分支A记录... 因此，“不存在”是多余的...
没有被having子句检查的记录可以有Branch B数据，因为查询主要部分中的Where子句过滤器会消除它们。

Answer 2

您的NOT EXISTS子查询搜索不同分支中具有相同ISBN，相同名称和相同重复数的行。如果其他分支中的条件不匹配，则查询包括输出中的行。

现在，两个分支具有相同的项目，但重复的数量不同，这就是为什么返回两个分支A项目，即HAVING子句的they are not filtered out。但试试，例如删除完整查询和简化版之间的两个(Branch B, 1-56592-756-7)项之一和you'll see the difference。

SQL子查询：这是如何产生输出的？

2 个答案: