為什麼我的 MySQL `NOT IN` 查詢使用 `COUNT()` 會導致「運算元應包含 1 列」錯誤？

NOT IN 具有聚合函數的子查詢中的 MySQL「運算元應包含 1 列」錯誤

在 MySQL 中使用 NOT IN 子查詢需要仔細注意列計數。 當子查詢使用 COUNT() 等聚合函數傳回多列時，會出現一個常見錯誤「運算元應包含 1 列」。 MySQL 的 NOT IN 運算子需要單列比較。

根本原因：

此錯誤源自於主查詢和子查詢傳回的列數不符。 包含聚合函數的 NOT IN 子查詢會產生包含多個欄位的結果集，與主查詢的 id 子句中的單列 WHERE 衝突。

範例：

考慮這個有問題的查詢：

SELECT *
FROM campaigns
WHERE id NOT IN
(
    SELECT
        e.id_campaign,  
        d.name,        
        d.frequency,   
        d.country,     
        d.referral,    
        d.bid,         
        d.status,      
        COUNT(e.id) AS countcap  
    FROM campaigns d
    LEFT JOIN served e
        ON d.id = e.id_campaign
    WHERE
        d.status = 'Active'
    GROUP BY
        e.id_campaign
    HAVING
        countcap < p>The intention is to select campaigns *not* included in the subquery's results. The subquery, however, returns eight columns, causing the "Operand should contain 1 column" error because `NOT IN` expects a single-column comparison against the `id` column in the `campaigns` table.</p><p>**Resolution:**</p><p>The solution involves restructuring the subquery to return only the `id_campaign` column:</p><pre class="brush:php;toolbar:false"><code class="language-sql">SELECT *
FROM campaigns
WHERE id NOT IN
(
    SELECT e.id_campaign
    FROM campaigns d
    LEFT JOIN served e ON d.id = e.id_campaign
    WHERE d.status = 'Active'
    GROUP BY e.id_campaign
    HAVING COUNT(e.id) < 

Alternatively, for situations requiring multiple columns, use `EXISTS` or `NOT EXISTS` for a more efficient and accurate solution:

SELECT *
FROM campaigns c
WHERE NOT EXISTS
(
    SELECT 1
    FROM campaigns d
    INNER JOIN served e ON d.id = e.id_campaign
    WHERE d.id = c.id
      AND d.status = 'Active'
      AND COUNT(e.id) < 

This revised approach avoids the column count mismatch and provides a cleaner solution for scenarios involving aggregate functions within subqueries used with `NOT IN`.

以上是為什麼我的 MySQL `NOT IN` 查詢使用 `COUNT()` 會導致「運算元應包含 1 列」錯誤？的詳細內容。更多資訊請關注PHP中文網其他相關文章！