SINGLE 列上的多行 SELECT 查询

Question

In my MySQL database I have two columns setup like this (for example):

series_id   category_id
    1            1
    1            2
    1            3
    2            1
    2            3
    3            2
    4            1
    4            2

What is the proper select query I can use to select the series_id when the category_id = 1 and 2?

Current query is this:

SELECT series_id 
FROM table 
WHERE category_id = '1' 
OR category_id = '2' 
GROUP BY series_id"

This does not retrieve any results:

SELECT series_id 
FROM table 
WHERE category_id = '1' 
AND category_id = '2' 
GROUP BY series_id

In regards to this example I would like it to return the two rows where the series_id is 1 and 2 so the result looks like this:

series_id
     1
     4

(Optional / Bonus) If it is possible to come up with another solution or utilize:

GROUP_CONCAT(category_id SEPARATOR ', ') AS category

for the returned series_id (s) from this query so the returned result is:

series_id    category
     1       1, 2, 3
     4       1, 2

Answer 1

SELECT   series_id
FROM     my_table
WHERE    category_id IN (1,2)
GROUP BY series_id
HAVING   COUNT(*) = 2

See it on sqlfiddle .

Note that if (series_id, category_id) are not guaranteed to be unique, you should replace COUNT(*) with the less performant COUNT(DISTINCT category_id) .

To return the grouped results, you can join against your table again:

SELECT series_id, GROUP_CONCAT(category_id)
FROM   my_table NATURAL JOIN (
  SELECT   series_id
  FROM     my_table
  WHERE    category_id IN (1,2)
  GROUP BY series_id
  HAVING   COUNT(*) = 2
) t
GROUP BY series_id

See it on sqlfiddle .

Answer 2

You need to have SubQuery to get the series_id so that later on you can use `GROUP_CONCAT.

SELECT series_ID, GROUP_CONCAT(category_id)
FROM tableName
WHERE series_ID IN
        (
            SELECT series_id
            FROM tableName
            WHERE Category_ID IN (1,2)
            GROUP BY seriesID
            HAVING COUNT(series_id) = 2
        )
GROUP BY series_ID

SQLFiddle Demo

Hope this helps you.