Question

我会说我对SQL的理解是平均的，但我想知道是否有任何方法可以更有效地编写此语句？

表Company的主键是uniqentity，并且该行中的公司名称为nameof。

在表格Line中，主键为uniqline，我有entityCompanyBilling和entityCompanyIssuing列（uniqentity的两个外键）。

下面这段代码运行得很好，我只是想提高效率。有可能吗？

SELECT 
   l.uniqline, b.nameof AS billingcompany, 
   l.UniqEntityCompanyBilling, i.nameof AS issuingcompany, 
   l.UniqEntityCompanyIssuing
FROM 
   Line l 
INNER JOIN 
   Company b ON b.uniqentity = l.uniqentitycompanybilling
INNER JOIN 
   Company i ON i.uniqentity = l.uniqentitycompanyissuing

Answer 1

更改查询的结构可能无法提高效率（它似乎很简单）。但是，更改数据的结构可能会：

在UniqEntityCompanyBilling和UniqEntityCompanyIssuing
~~在Company.UniqEntity上添加索引~~ ^{没有发现该列是PK}
如果Company.UniqEntity上的索引尚未群集，请在nameof的索引中添加Company.UniqEntity字段，以便查询不必返回物理表。

Answer 2

此查询完全正常。

你可以把它缩短如下，但这会降低可读性，通常被认为是不好的做法，所以不要。

SELECT l.uniqline, b.nameof AS billingcompany, l.UniqEntityCompanyBilling, i.nameof AS issuingcompany, l.UniqEntityCompanyIssuing
FROM Line l, Company b, Company i
WHERE b.uniqentity = l.uniqentitycompanybilling
AND i.uniqentity = l.uniqentitycompanyissuing

Answer 3

如果可能，您可以在加入表格时加入过滤器，如下所示：

SELECT 
   l.uniqline, b.nameof AS billingcompany, 
   l.UniqEntityCompanyBilling, i.nameof AS issuingcompany, 
   l.UniqEntityCompanyIssuing
FROM 
   Line l 
INNER JOIN 
   (SELECT *
    FROM Company
    WHERE <REGION = 'xyz', etc.>
   ) b ON b.uniqentity = l.uniqentitycompanybilling
INNER JOIN 
   (SELECT *
    FROM Company 
    WHERE <REGION = 'xyz', etc.>
   ) i ON i.uniqentity = l.uniqentitycompanyissuing
WHERE <any filter condition>;

SQL结合两个连接

3 个答案: