UNION函數的使用以及SQL文的高效編輯方法

查找SELECT 相同列的狀況下，例如選擇一個月每一天的某幾項數據，至關於查詢一個月三十行數據，爲了在一個表中顯示就要使用union函數。

eg：

SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-01%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-02%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-03%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-04%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-05%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-06%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-07%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-08%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-09%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-10%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-11%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-12%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-13%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-14%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-15%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-16%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-17%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-18%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-19%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-20%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-21%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-22%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-23%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-24%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-25%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-26%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-27%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-28%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-29%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-30%' AND act = '1005001001' UNION ALL
SELECT COUNT(DISTINCT uid) AS c FROM da_track WHERE data_time LIKE '2018-05-31%' AND act = '1005001001'

View Code

1. UNION 表示去重的結果；

2. UNION ALL 表示不進行去重的結果；

 UNION函數就是用來拼表的，條件是每一個選擇語句的列相同，這個列的內容可能相同或者不一樣，不一樣的時候以第一個選擇語句爲準。

需求在哪？好比要使用拼接以後的表格，就頗有用了，場景：一個月30天，求每一天去重的日活量，而後求和再作平均，這就須要將去重的日活拼接成一張30行的表，而後再作SUM而後再求平均。

根據實際須要使用UNION函數，使用SQL最重要的技能就是理解表格之間的關係，先後層次，你須要什麼表格就使用鏈接JOIN、UNION等方法人爲實現什麼表格。

在重複查詢的時候，通常都是修改一段SQL文的小部分語句，好比日期的天的數字，這種狀況就須要掌握高級編輯方法來提升編輯效率，使用excel和高級文本編輯器(editplus等)以及正則化手段。

方法總結：

1. 寫好第一句SQL文；

2. 使用excel拼接SQL文，即複製重複部分，利用excel自動生成不重複部分（好比日期自動+1）；

3. 使用文本編輯器的正則化功能實現SQL文的修飾。

注意：使用Ctrl + D 來複制（向下填充）Ctrl + R 向右填充；修改excel單元格格式；excel的CONCATENATE()函數能夠拼接多個字符串；

　　    學會使用excel的單元格引用，$B3中的$符號表示絕對引用B列，非絕對引用3行，行數能夠拖動自動改變，用來引用單元格的內容，這在須要使用單元格內容的SQL文中很重要。

　　　editplus 使用 Ctrl + H 進入正則化替換。