Question

我有以下查询：

select
    DATE_FORMAT(created_at,'%d %M') as day,
    count(*) as Total,
    count(content_sentiment = 0) as Neutral,
    count(content_sentiment < 0) as Negative,
    count(content_sentiment > 0) as Positive
from Master
where content_sentiment IS NOT NULL
GROUP BY DAY(created_at)
;

这让我：

day         total   Neautral    Negative    Positive
------------------------------------------------------------
01 August   9037    9037    9037    9037
02 August   6439    6439    6439    6439
03 August   5922    5922    5922    5922
04 August   5382    5382    5382    5382

总值应该是正数，中性和负数的总和，因为不是猜测，因为每个查询都得到整个表的总计，而不是按天分组。

如何为每个积极，中性和负值包含一个分组？

Answer 1

您需要条件聚合：

...
count(*) as Total,
SUM(CASE WHEN content_sentiment = 0 THEN 1 ELSE 0 END) as Neutral,
SUM(CASE WHEN content_sentiment < 0 THEN 1 ELSE 0 END) as Negative,
SUM(CASE WHEN content_sentiment > 0 THEN 1 ELSE 0 END) as Positive
...

Answer 2

    select
    DATE_FORMAT(created_at,'%d %M') as day,
    count(*) as Total,
    sum(if(content_sentiment = 0,1,0)) as Neutral,
    sum(if(content_sentiment < 0,1,0)) as Negative,
    sum(if(content_sentiment > 0,1,0)) as Positive
from Master
where content_sentiment IS NOT NULL
GROUP BY DAY(created_at)
;

MySQL - 组比较按天计算

2 个答案: