Mysql 在一个查询中使用多个具有不同条件的SELECT语句

我有以下表格：

“价格”

+------------+--------------+------+-----+---------+----------------+
| Field      | Type         | Null | Key | Default | Extra          |
+------------+--------------+------+-----+---------+----------------+
| id         | mediumint(9) | NO   | PRI | NULL    | auto_increment |
| product_id | int(11)      | YES  |     | NULL    |                |
| price      | text         | YES  |     | NULL    |                |
| date       | text         | YES  |     | NULL    |                |
| time       | text         | YES  |     | NULL    |                |
+------------+--------------+------+-----+---------+----------------+

“产品”

+-----------------+--------------+------+-----+---------+----------------+
| Field           | Type         | Null | Key | Default | Extra          |
+-----------------+--------------+------+-----+---------+----------------+
| id              | mediumint(9) | NO   | PRI | NULL    | auto_increment |
| category_id     | int(11)      | YES  |     | NULL    |                |
| product_urls    | text         | YES  |     | NULL    |                |
| product_title   | text         | YES  |     | NULL    |                |
| product_image   | text         | YES  |     | NULL    |                |
| product_content | text         | YES  |     | NULL    |                |
+-----------------+--------------+------+-----+---------+----------------+

这里的连接在

Products.id

和

Prices.Products\u id

之间

我有一个脚本在运行，它可以抓取URL，从特定网页中获取价格，然后每小时更新“价格”表

我希望通过一个数据库查询显示以下内容：

• 产品名称
• 产品·产品·形象
• prices.price（当前价格，即每个产品id的最后一次插入）
• 今天的平均价格与昨天的平均价格之间的百分比差异
• 当月平均价格与上月平均价格之间的百分比差异

我可以单独做这些事情，但我不能将它们合并到一个查询中

为了清楚起见，我想在一个特定的类别中一次抓住所有的产品-考虑到这一点，我一直在这样做：

...WHERE category_id=%s...GROUP BY product_id...

---

我直接编写SQL，特别是使用MySQLdb Python库。

第一个子选择的未测试示例（在此模板上进行其他子选择）：

---

可能不是最有效的方法…根据@Cosmin的建议，最后的片段

SELECT p.*,
     (SELECT prices.price
          FROM prices
          WHERE prices.product_id = p.id
          ORDER BY prices.id DESC
          LIMIT 1) as last_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id = p.id AND date = CURRENT_DATE()
          ORDER BY prices.id DESC
          LIMIT 1) as todays_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id = p.id AND date = SUBDATE(CURRENT_DATE, 1)
          ORDER BY prices.id DESC
          LIMIT 1) as yesterdays_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id=p.id AND MONTH(date)=MONTH(current_date)
          ORDER BY prices.id DESC
          LIMIT 1) as current_months_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id=p.id AND MONTH(date)=MONTH(CURRENT_DATE - INTERVAL 1 MONTH)
          ORDER BY prices.id DESC
          LIMIT 1) as previous_months_average_price
FROM products p
WHERE p.category_id=%s;

---

我不能100%确定如何在查询中计算百分比差异，但我将使用Python来保持查询的干净。

如果可以单独进行这些操作，则可以在一个查询中使用价格的子选择，通过匹配内部价格来实现。product\u id（来自子选择）对于outer products.id，是时候学习数据类型了。DATETIME表示DATETIME，DECIMAL表示price，VARCHAR和INT（可能）表示一切else@Cosmin你能举一些例子吗？@草莓有什么好处，内存分配？主要是关于优化。除了你的答案，再次感谢！）你能解释一下起始的p吗？我不明白它是怎么工作的？@Adders它的意思是“表p中的所有列”，这是一个坏习惯，在大多数情况下你应该避免。如果这些是您想要的列，请显式列出

p.product\u title、p.product\u image、p.price

。@Air是的，这是一个糟糕的习惯，而且由于查询仍然不完整，因此我没有写入所需的单元格，而只是编写了*，很抱歉造成混淆

SELECT p.*,
     (SELECT prices.price
          FROM prices
          WHERE prices.product_id = p.id
          ORDER BY prices.id DESC
          LIMIT 1) as last_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id = p.id AND date = CURRENT_DATE()
          ORDER BY prices.id DESC
          LIMIT 1) as todays_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id = p.id AND date = SUBDATE(CURRENT_DATE, 1)
          ORDER BY prices.id DESC
          LIMIT 1) as yesterdays_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id=p.id AND MONTH(date)=MONTH(current_date)
          ORDER BY prices.id DESC
          LIMIT 1) as current_months_average_price,

     (SELECT ROUND(AVG(prices.price), 2)
          FROM prices
          WHERE prices.product_id=p.id AND MONTH(date)=MONTH(CURRENT_DATE - INTERVAL 1 MONTH)
          ORDER BY prices.id DESC
          LIMIT 1) as previous_months_average_price
FROM products p
WHERE p.category_id=%s;