MySQLは試合

Question

私のデータベースなしで2つのクエリに参加するには、両者がそれぞれの時間の価格 "の列のSUMを収集しようとしているいくつかの異なる値が、イムを持って

Wardrobe_CloakTable & Wardrobe_ServiceTable 

と呼ばれる2つの異なるテーブルで存在します。

このクエリは私にちょうどWardrobe_CloakTable

SELECT DATE(delivered) as date, sum(price) as sum, HOUR(delivered) AS hour 
      FROM Wardrobe_CloakTable 
     WHERE payingcustomerID = '$payingcustomerID' 
      AND delivered >= CONCAT(CURDATE() , ' 07:00:00') 
      AND delivered < CONCAT(CURDATE() , ' 07:00:00') + INTERVAL 1 DAY 
     GROUP BY DATE(delivered), HOUR(delivered) 

これに対して正しい結果を与えるには、私がやりたいものを実際にデータが存在するので、唯一の時間は、ある

--------------------------------- 
| date | sum | hour  | 
| 2016-09-15 | 100 | 9  | 
| 2016-09-15 | 200 | 10  | 
| 2016-09-15 | 100 | 12  | 
| 2016-09-15 | 100 | 18  | 
--------------------------------- 

のようなものを提供します

----------------------------------------------------------------- 
| date | sum(CloakTable) | sum(ServiceTable) | hour  | 
| 2016-09-15 | 100    | 200    | 9  | 
| 2016-09-15 | 200    | 500    | 10  | 
| 2016-09-15 |     | 400    | 11  | 
| 2016-09-15 | 200    |     | 14  | 
| 2016-09-15 | 100    | 400    | 15  | 
| 2016-09-15 | 100    | 200    | 18  | 
------------------------------------------------------------------ 

これまでのところ、私はUNIONを使用して作成しようとしていますとFULL JOINですが、私が気づいたように、完全な結合はMySQLでは不可能です。あなたが左をしない限り、右。

iは便利なものになってきた最も近い：

SELECT DATE(delivered) as date, 
    sum(price) as sum, 
    HOUR(delivered) as hour 

      FROM Wardrobe_CloakTable 
      WHERE payingcustomerID = 2 
      AND Wardrobe_CloakTable.delivered >= CONCAT(CURDATE() , ' 07:00:00') 
      AND Wardrobe_CloakTable.delivered < CONCAT(CURDATE() , ' 07:00:00') + INTERVAL 1 DAY 
      GROUP BY DATE(delivered), HOUR(delivered) 
    UNION ALL 
SELECT DATE(time) as date, 
    sum(price) as sum, 
    HOUR(time) as hour 
    FROM Wardrobe_ServiceTable   
     WHERE payingcustomerID = 2 
     AND Wardrobe_ServiceTable.time >= CONCAT(CURDATE() , ' 07:00:00') 
     AND Wardrobe_ServiceTable.time < CONCAT(CURDATE() , ' 07:00:00') + INTERVAL 1 DAY 
     GROUP BY DATE(time), HOUR(time) 

これは私に与える：

--------------------------------- 
| date | sum | hour  | 
| 2016-09-15 | 100 | 9  | (Cloaktable) 
| 2016-09-15 | 200 | 10  | (Cloaktable) 
| 2016-09-15 | 100 | 9  | (service) 
| 2016-09-15 | 100 | 11  | (service) 
--------------------------------- 

スキーマ

CREATE TABLE `Wardrobe_CloakTable` (
    `ID` int(64) NOT NULL, 
    `payingcustomerID` int(11) NOT NULL, 
    `deviceID` varchar(100) COLLATE utf8mb4_unicode_ci NOT NULL, 
    `terminalnumber` int(11) NOT NULL, 
    `qrcode` varchar(64) CHARACTER SET latin1 COLLATE latin1_danish_ci NOT NULL, 
    `cloakroomsection` varchar(20) COLLATE utf8mb4_unicode_ci NOT NULL, 
    `cloakroomnumber` int(11) NOT NULL, 
    `isbag` tinyint(1) NOT NULL, 
    `price` int(11) NOT NULL, 
    `delivered` datetime NOT NULL, 
    `collected` datetime DEFAULT NULL, 
    `reservedtime` datetime DEFAULT CURRENT_TIMESTAMP 
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci; 


CREATE TABLE `Wardrobe_ServiceTable` (
    `ID` int(11) NOT NULL, 
    `payingcustomerID` int(11) NOT NULL, 
    `qrcode` varchar(64) COLLATE utf8mb4_unicode_ci NOT NULL, 
    `time` datetime NOT NULL, 
    `price` int(11) NOT NULL 
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci; 

Answer 1

ソニー、

を私が作業を得ました要求。 100％確実ではない、それはそれを処理するための最良の方法ですが、少なくとも、それが働いている：

SELECT date_buy, hour_buy, sum(sum_cloak) AS `sum(CloakTable)`, sum(sum_service) AS `sum(ServiceTable)` FROM (
    SELECT DATE(wct.delivered) as date_buy, HOUR(wct.delivered) AS hour_buy, sum(wct.price) AS sum_cloak, 0 as sum_service FROM Wardrobe_CloakTable wct 
    WHERE wct.payingcustomerID = 2 
     AND wct.delivered >= CONCAT(CURDATE(), ' 07:00:00') 
     AND wct.delivered < CONCAT(CURDATE(), ' 07:00:00') + INTERVAL 1 DAY 
    GROUP BY date_buy, hour_buy 

    UNION ALL 

    SELECT DATE(wst.time) as date_buy, HOUR(wst.time) AS hour_buy, 0 AS sum_cloak, sum(wst.price) as sum_service 
    FROM Wardrobe_ServiceTable wst 
    WHERE wst.payingcustomerID = 2 
     AND wst.time >= CONCAT(CURDATE(), ' 07:00:00') 
     AND wst.time < CONCAT(CURDATE(), ' 07:00:00') + INTERVAL 1 DAY 
    GROUP BY date_buy, hour_buy 
) fullTab 
GROUP BY date_buy, hour_buy