mysql - sql 将选择计算声明为变量以在查询中进一步使用

Question

我想将选择查询内的特定计算更改为变量，以便重新处理查询内其他计算的结果。在此示例中，我需要作为变量的值是 dt_raw。到目前为止，我还没有成功实现解决方案。

SELECT
cpe.entity_id AS product_id,
cpe.sku,
CASE
    WHEN (csi.qty - csi.stock_ordered_qty) >= cped2.value AND csi.stock_id = 3  /* on stock burgsdorgstrasse */
        THEN 3
    WHEN (csi.qty - csi.stock_ordered_qty) >= cped2.value AND csi.stock_id = 4  /* on stock hoppegarten */
        THEN 7
    WHEN sum(sm_qty) > 2 AND csi.stock_id = 3                       /* history burgsdorfstrasse */
        THEN round(avg(DATEDIFF(sm.sm_date, po.po_date)) + 3)
    WHEN sum(sm_qty) > 2 AND csi.stock_id = 4                       /* history hoppegarten */
        THEN round(avg(DATEDIFF(sm.sm_date, po.po_date)) + 4)
        ELSE cped.value                                             /* default_delivery_time */
    END AS dt_raw


FROM `test_live`.`catalog_product_entity` cpe

LEFT JOIN `test_live`.`cataloginventory_stock_item` csi
ON cpe.entity_id = csi.product_id

LEFT JOIN `test_live`.`catalog_product_entity_decimal` cped     /* default_delivery_time */
ON cpe.entity_id = cped.entity_id
AND cped.attribute_id = 392

LEFT JOIN `test_live`.`catalog_product_entity_decimal` cped2    /* min_qty_delivery_time */
ON cpe.entity_id = cped2.entity_id
AND cped2.attribute_id = 393

LEFT JOIN `test_live`.`stock_movement` sm
ON cpe.entity_id = sm.sm_product_id
AND sm.sm_type = "supply"
AND sm.sm_date > NOW() - Interval 90 DAY

LEFT JOIN `test_live`.`purchase_order` po
ON po.po_num = sm.sm_po_num


WHERE
csi.is_favorite_warehouse = 1
AND (csi.stock_id = 3 OR csi.stock_id = 4)

GROUP BY cpe.entity_id

我想在进一步计算中使用dt_raw的结果。

例如:

concat ("ca. ", round(dt_raw), " weeks") as delivery_time

CASE
    WHEN dt_raw <= 30
    THEN 50
    ELSE 0
END AS amazon_qty

Answer 1

如果您想不受限制地使用查询结果，您应该将其保存在表中。之后使用选择来获取您想要的数据。

创建

CREATE TABLE your_schema.dt_raw_temp
SELECT
cpe.entity_id AS product_id,
cpe.sku,
CASE
    WHEN (csi.qty - csi.stock_ordered_qty) >= cped2.value AND csi.stock_id = 3  /* on stock burgsdorgstrasse */
        THEN 3
    WHEN (csi.qty - csi.stock_ordered_qty) >= cped2.value AND csi.stock_id = 4  /* on stock hoppegarten */
        THEN 7
    WHEN sum(sm_qty) > 2 AND csi.stock_id = 3                       /* history burgsdorfstrasse */
        THEN round(avg(DATEDIFF(sm.sm_date, po.po_date)) + 3)
    WHEN sum(sm_qty) > 2 AND csi.stock_id = 4                       /* history hoppegarten */
        THEN round(avg(DATEDIFF(sm.sm_date, po.po_date)) + 4)
        ELSE cped.value                                             /* default_delivery_time */
    END AS dt_raw


FROM `test_live`.`catalog_product_entity` cpe

LEFT JOIN `test_live`.`cataloginventory_stock_item` csi
ON cpe.entity_id = csi.product_id

LEFT JOIN `test_live`.`catalog_product_entity_decimal` cped     /* default_delivery_time */
ON cpe.entity_id = cped.entity_id
AND cped.attribute_id = 392

LEFT JOIN `test_live`.`catalog_product_entity_decimal` cped2    /* min_qty_delivery_time */
ON cpe.entity_id = cped2.entity_id
AND cped2.attribute_id = 393

LEFT JOIN `test_live`.`stock_movement` sm
ON cpe.entity_id = sm.sm_product_id
AND sm.sm_type = "supply"
AND sm.sm_date > NOW() - Interval 90 DAY

LEFT JOIN `test_live`.`purchase_order` po
ON po.po_num = sm.sm_po_num


WHERE
csi.is_favorite_warehouse = 1
AND (csi.stock_id = 3 OR csi.stock_id = 4)

GROUP BY cpe.entity_id;

选择

SELECT CONCAT('ca. ', round(dt_raw), ' weeks') AS delivery_time, 
       IF(dt_raw <= 30, 50, 0) AS amazon_qty
FROM your_schema.dt_raw_temp;

The advantage of using tables is that you can reproduce your calculations at any given moment as long as you keep your tables.