僅用一句SQL更新整張表的漲跌幅、漲跌率
問題場景
各大平台店鋪的三項評分（物流、服務、商品）變化情況；
商品每日價格的變化記錄；
股票的實時漲跌浮；
復現場景
表：主鍵ID，商品編號，記錄時的時間，記錄時的價格，創建時間。
問題：獲取每個商品每次的變化情況（漲跌幅、漲跌率）。
解決思路
1、要想高效率的更新漲跌，就肯定不能是逐條數據更新，要通過自連表建立起對應關係，將每一條數據關聯到上一次的價格數據。
2、由於數據庫非常龐大，所以可能存在很多垃圾數據，就比如說相關的字段值為NULL或者非有效值的，這些數據要先排除掉。
SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL;
3、然後在獲取每條數據的上一條數據，同樣也要先排除掉垃圾數據。
SELECT tmp_a.*, MAX(tmp_b.goods_date) AS last_date FROM
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_a
LEFT JOIN
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_b
ON tmp_a.goods_code = tmp_b.goods_code AND tmp_a.goods_date > tmp_b.goods_date GROUP BY tmp_a.id;
4、獲取到上一條數據後，獲取上條數據對應的商品價格。
SELECT tmp_ab.*,tmp_c.goods_price AS last_price FROM
(
SELECT tmp_a.*, MAX(tmp_b.goods_date) AS last_date FROM
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_a
LEFT JOIN
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_b
ON tmp_a.goods_code = tmp_b.goods_code AND tmp_a.goods_date > tmp_b.goods_date GROUP BY tmp_a.id
) AS tmp_ab
LEFT JOIN (SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_c
ON tmp_ab.goods_code = tmp_c.goods_code AND tmp_c.goods_date = tmp_ab.last_date ORDER BY tmp_ab.id;
5、獲取到上條數據以及對應的價格後，開始進行計算，獲取到最終的結果。
SELECT
*,
(CONVERT(goods_price, DECIMAL(10,2)) - CONVERT(last_price, DECIMAL(10,2))) AS '漲跌幅',
ROUND((CONVERT(goods_price, DECIMAL(10,2)) - CONVERT(last_price, DECIMAL(10,2)))/CONVERT(last_price, DECIMAL(10,2)), 2) AS '漲跌率'
FROM (
SELECT tmp_ab.*,tmp_c.goods_price AS last_price FROM
(
SELECT tmp_a.*, MAX(tmp_b.goods_date) AS last_date FROM
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_a
LEFT JOIN
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_b
ON tmp_a.goods_code = tmp_b.goods_code AND tmp_a.goods_date > tmp_b.goods_date GROUP BY tmp_a.id
) AS tmp_ab
LEFT JOIN (SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_c
ON tmp_ab.goods_code = tmp_c.goods_code AND tmp_c.goods_date = tmp_ab.last_date ORDER BY tmp_ab.id
) AS tmp
解決方案
-- 創建表SQL
CREATE TABLE `test_goods_price_change` (
`id` int(11) NOT NULL AUTO_INCREMENT COMMENT '主鍵ID',
`goods_code` varchar(50) NOT NULL COMMENT '商品編碼',
`goods_date` int(11) NOT NULL COMMENT '記錄時的時間',
`goods_price` decimal(10,2) NOT NULL COMMENT '記錄時的價格',
`created_at` int(11) NOT NULL COMMENT '創建時間',
PRIMARY KEY (`id`)
) ENGINE=InnoDB CHARSET=utf8mb4;
-- 獲取漲跌浮SQL
SELECT
*,
(CONVERT(goods_price, DECIMAL(10,2)) - CONVERT(last_price, DECIMAL(10,2))) AS '漲跌幅',
ROUND((CONVERT(goods_price, DECIMAL(10,2)) - CONVERT(last_price, DECIMAL(10,2)))/CONVERT(last_price, DECIMAL(10,2)), 2) AS '漲跌率'
FROM (
SELECT tmp_ab.*,tmp_c.goods_price AS last_price FROM
(
SELECT tmp_a.*, MAX(tmp_b.goods_date) AS last_date FROM
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_a
LEFT JOIN
( SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_b
ON tmp_a.goods_code = tmp_b.goods_code AND tmp_a.goods_date > tmp_b.goods_date GROUP BY tmp_a.id
) AS tmp_ab
LEFT JOIN (SELECT id,goods_code,goods_date,goods_price FROM test_goods_price_change WHERE goods_price IS NOT NULL AND goods_date IS NOT NULL ) AS tmp_c
ON tmp_ab.goods_code = tmp_c.goods_code AND tmp_c.goods_date = tmp_ab.last_date ORDER BY tmp_ab.id
) AS tmp
