Question

我正在以下网站https://www.freecodecamp.org/news/project-1-analyzing-dvd-rentals-with-sql-fd12dd674a64/上进行第一个问题的研究，以确定“什么是最热门和租金最低（需求最多）的类型，它们的总销售额是多少？”给出了所需的SQL下面。

with q1 as (select c.name as genre, sum(p.amount) as total_spent from 
    film f 
    join film_category fc
    using (film_id)
    join category c
    using (category_id)
    join inventory i
    using (film_id)
    join rental r
    using (inventory_id)
    join payment p
    using (rental_id)
group by c.name
order by total_spent desc),
    q2 as (select c.name as genre, count(cu.customer_id) as number_of_rentals from 
    film f 
    join film_category fc
    using (film_id)
    join category c
    using (category_id)
    join inventory i
    using (film_id)
    join rental r
    using (inventory_id)
    join customer cu
    using (customer_id)
group by c.name
order by number_of_rentals desc)

select genre, number_of_rentals, total_spent
from q1 join q2
using (genre)
order by number_of_rentals desc;

现在我要复制的是Scala的Slick（3.1）ORM。该过程涉及进行两个查询q1和q2，然后将它们连接在一起以获得第三个查询。下面给出前两个查询

    val q1 = (for {
      (((((film, filmCategory), category), inventory), rental), payment) <- Film join
        FilmCategory on (_.filmId === _.filmId) join
        Category on (_._2.categoryId === _.categoryId) join
        Inventory on (_._1._1.filmId === _.filmId) join
        Rental on (_._2.inventoryId === _.inventoryId) join
        Payment on (_._2.rentalId === _.rentalId)
    } yield (category.name, payment.amount)) groupBy (_._1) map {
      case (name, ids) =>
        name -> ids.map(_._2).sum
    } sortBy (_._2 desc)

    val q2 = (for {
      (((((film, filmCategory), category), inventory), rental), customer) <- Film join
        FilmCategory on (_.filmId === _.filmId]) join
        Category on (_._2.categoryId === _.categoryId) join
        Inventory on (_._1._1.filmId === _.filmId) join
        Rental on (_._2.inventoryId === _.inventoryId) join
        Customer on (_._2 === _.customerId)
    } yield (category.name, customer.customerId)) groupBy (_._1) map {
      case (name, ids) =>
        name -> ids.map(_._2).length
    } sortBy (_._2 desc)

最后的连接由

给出

val result = for {
      ((genre, count), price) <- q1 join q2 on (_._1 === _._1)
    } yield (genre, count, price)

即将这两个表联接在category.name列上。但是，尽管这不会产生编译错误，但会给出

的运行时错误。

org.postgresql.util.PSQLException: ERROR: column x2.x20 does not exist

我可以解决此问题的唯一方法是用

替换联接

val result = for {
      (genre1, count) <- q1
      (genre2, price) <- q2 if genre2 === genre1
    } yield (genre1, count, price)

问题：第一次加入有什么问题？为什么我需要在后者中用双重理解代替它？在Slick Select查询返回的结果上进行联接的正确方法是什么。

滑动连接查询结果

0 个答案: