添加 WHERE 后出现语法错误。 R 中的 SQLite

Question

我有一个航班数据，包括出发地、目的地和月份。我试图在一年中的一个月内获得从起点到终点的航班频率。

我完成了：

dbGetQuery(conn,"
                 SELECT
                  Month AS Month,
                  COUNT(OriDes) AS Freq,
                  OriDes 
                 FROM(
                  SELECT Origin || '-' || Dest AS OriDes,
                  Month AS Month
                  FROM flights
                  )
                 GROUP BY OriDes, Month
                 ")

首先结合数据库中的 Origin 和 Dest，然后计算它的频率以获得输出：

   Month Freq  OriDes
1      1  123 ABE-ATL
2      2  140 ABE-ATL
3      3  117 ABE-ATL
4      4  112 ABE-ATL
5      5  111 ABE-ATL
6      6  120 ABE-ATL
7      7  113 ABE-ATL
8      8  124 ABE-ATL
9      9   94 ABE-ATL
10    10  119 ABE-ATL
11    11  116 ABE-ATL
12    12  124 ABE-ATL
13     6    1 ABE-AVP
14     1   66 ABE-CLE
15     2   72 ABE-CLE
16     3   99 ABE-CLE
17     4   89 ABE-CLE
18     5   91 ABE-CLE
19     6   97 ABE-CLE
20     7   92 ABE-CLE

但是因为这是一个大数据我希望过滤并且只处理Freq > 500。所以我尝试了：

dbGetQuery(conn,"
                 SELECT
                  Month AS Month,
                  COUNT(OriDes) AS Freq,
                  OriDes 
                 FROM(
                  SELECT Origin || '-' || Dest AS OriDes,
                  Month AS Month
                  FROM flights
                  )
                 GROUP BY OriDes, Month
                 WHERE Freq > 500
                 ")

但我会得到一个错误：

Error: near "WHERE": syntax error

有人可以向我解释我做错了什么吗？

Answer 1

正如评论中所建议的：

SELECT
Month AS Month,
COUNT(OriDes) AS Freq,
OriDes 
FROM(
     SELECT Origin || '-' || Dest AS OriDes,
     Month AS Month
     FROM flights
     )
GROUP BY OriDes, Month
HAVING COUNT(OriDes) > 500

Answer 2

您可以直接在 table 中进行聚合而无需子查询，您只需要一个 HAVING 子句来过滤聚合结果，而不是 WHERE 子句来过滤聚合结果table聚合前：

SELECT Month,
       COUNT(*) AS Freq,
       Origin || '-' || Dest AS OriDes
FROM flights
GROUP BY Month, OriDes
HAVING Freq > 500;

请注意，由于您在 GROUP BY 子句中使用了 OriDes，因此没有必要在 COUNT() 中使用它。你可以用 COUNT(*).

来计算

添加 WHERE 后出现语法错误。 R 中的 SQLite

Syntax error after adding WHERE. SQLite in R

sql

sqlite

group-by

count

having