Sequelize - where 子句中的子查询 [英] Sequelize - subquery in where clause

问题描述

我在我的 Express 应用中使用 Sequelize.我需要生成一个在 WHERE 子句中有一个子查询的查询.

I'm using Sequelize in my Express app. I need to generate a query that has a subquery in the WHERE clause.

SELECT *
  FROM MyTable
 WHERE id NOT IN (
       SELECT fkey
         FROM MyOtherTable
        WHERE field1 = 1
          AND field2 = 2
          AND field3 = 3
       )

我首先通过我的模型尝试了关系/关联，但无法让它发挥作用.类似的东西:

I first tried relations/associations through my models but couldn't get it to work. Something like:

MyTable.find( {
    where: {
        id: {
            $notIn: // <= what goes here? Can I somehow reference my include model?
        }
    },
    include: [ {
        model: MyOtherTable,
        where: {
          field1: 1,
          field2: 2,
          field3: 3
    } ]
} );

然后我尝试使用 Sequelize.where()，但没有成功.

Then I tried using Sequelize.where(), no luck there.

然后我尝试了 Sequelize.literal() 并且它有效，但不确定它是否是在 Sequelize 的 where 子句中执行子查询的正确" 方式，就像我一样刚接触它.

Then I tried Sequelize.literal() and that works but not sure if it's a "proper" way of doing a subquery in a where clause in Sequelize as I'm new to it.

MyTable.find( {
    where: {
        id: {
            $notIn: sequelize.literal( 
                '( SELECT fkey ' +
                    'FROM MyOtherTable ' +
                   'WHERE field1 = ' + field1 +
                    ' AND field2 = ' + field2 +
                    ' AND field3 = ' + field3 + 
                ')'
        }
    } 
} );

我也知道我可以使用 Sequelize.query() 但我真的不知道我是否应该使用它或者 literal() 是不是马上因为我觉得我忽略了一些东西.

I also know that I could use Sequelize.query() but don't really know if I should reach for it or if literal()is the right away as I feel like there's something I'm overlooking.

我真的很想知道如何在 WHERE 子句中使用 Sequelize 正确" 方式执行子查询.

I would really like to know how to perform a subquery in a WHERE clause with Sequelize the "proper" way.

感谢您的反馈！

推荐答案

我在我的项目中遇到了类似的问题.我选择实现它的方式有点不同，原因有两个:

I have encountered a similar issue in my project. The way I choose to implement it is a bit different for two reasons:

1. 如果在某个时间点 Sequelize 决定实施子查询 - 语法已准备就绪.
2. 使用 Sequelize 保护来防止 SQL 注入.

这是我的代码片段，希望对您有所帮助.

Here is my code snippet, hope it helps.

Sequelize v5

const tempSQL = sequelize.dialect.QueryGenerator.selectQuery('MyOtherTable',{
    attributes: ['fkey'],
    where: {
          field1: 1,
          field2: 2,
          field3: 3
    }})
    .slice(0,-1); // to remove the ';' from the end of the SQL

MyTable.find( {
    where: {
        id: {
              [Sequelize.Op.notIn]: sequelize.literal(`(${tempSQL})`)
        }
    } 
} );

续集 v6

const tempSQL = sequelize.dialect.queryGenerator.selectQuery('MyOtherTable',{
    attributes: ['fkey'],
    where: {
          field1: 1,
          field2: 2,
          field3: 3
    }})
    .slice(0,-1); // to remove the ';' from the end of the SQL

MyTable.find( {
    where: {
        id: {
              [Sequelize.Op.notIn]: sequelize.literal(`(${tempSQL})`)
        }
    } 
} );

有些人可能会选择不使用 tempSQL 变量，而只是在 find 结构中构建 SQL(也许使用辅助方法?)

Some people might choose to not use the tempSQL variable and simply build the SQL inside the find structure (maybe using a helper method?)

我也认为这可能是 sequelize 的子查询扩展的基础，因为它几乎使用相同的语法.

I also think this might be the basis for a sub queries extension for sequelize as it uses the same syntax almost.

这篇关于Sequelize - where 子句中的子查询的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！