Selecting from a Many to Many table [duplicate]

Question

This question already has an answer here:

SQL: Get Products from a category but also must be in another set of categories (1 answer)

Closed 8 years ago.

I want to get the post_ID and the thumbnail of the post(s) that have certain multiple tags. My tables are as follow:

table: post
-------------------
post_ID | thumbnail
1       | 1.jpg
2       | 2.jpg
3       | 3.jpg
4       | 4.jpg
5       | 5.jpg

table: tags
-----------
tag_ID | tag
1      | red
2      | yellow
3      | orange
4      | blue
5      | pink

table: post_tags
----------------
post_ID | tag_ID
1       | 1
1       | 2
1       | 4
1       | 5
2       | 2
2       | 3
3       | 4
3       | 1
3       | 2
4       | 5
4       | 4
5       | 1
5       | 3
5       | 5

I currently use something like this, but it is not working: I first get all the tags of posts in an array, and then compare them to find if a post has all the 3 tags mentioned.

 SELECT post_ID FROM post_tags 
 WHERE tag_ID IN ($thisTags[0], $thisTags[1], $thisTags[2], $thisTags[3])

Do I need to join or group or something? I am new to SQL and PHP. Thanks a lot.

Answer 1

If you have a UNIQUE constraint on post_tags(post_ID,tag_ID), and you can verify that the elements of thisTags are unique, then you could do something like this:

SELECT post_ID FROM post_tags 
 WHERE tag_ID IN ($thisTags[0], $thisTags[1], $thisTags[2], $thisTags[3])
 GROUP BY post_ID
HAVING COUNT(1) = 4

This approach REQUIRES that the elements in $thisTags are unique. You can easily work around the uniqueness issue in the post_tags table (but the query above requires that to be unique as well.)

There are other SQL statements, other approaches, that can achieve an equivalent result, but the query above is the most concise. (Some of the other approaches are to use EXISTS predicates, or to use JOIN operations on the sets of post_ID returned for each tag.)

Answer 2

This is an example of a "set-within-sets" subquery. I recommend using group by and having for maximum flexibility. The idea is to move the logic into the having clause. In your case, this looks like:

SELECT post_ID
FROM post_tags 
group by post_id
having sum(tag_id = $thisTags[0]) > 0 and
       sum(tag_id = $thisTags[1]) > 0 and
       sum(tag_id = $thisTags[2]) > 0 and
       sum(tag_id = $thisTags[3]) > 0;

If you wanted only these tags and no others:

SELECT post_ID
FROM post_tags 
group by post_id
having sum(tag_id = $thisTags[0]) > 0 and
       sum(tag_id = $thisTags[1]) > 0 and
       sum(tag_id = $thisTags[2]) > 0 and
       sum(tag_id = $thisTags[3]) > 0 and
       count(distinct tag_id) = 4;

If you wanted at least three of these four tags:

SELECT post_ID
FROM post_tags 
group by post_id
having (max(tag_id = $thisTags[0]) + and
        max(tag_id = $thisTags[1]) + and
        max(tag_id = $thisTags[2]) + and
        max(tag_id = $thisTags[3])
       ) >= 3;

Answer 3

Are you looking for something like this?

SELECT post_ID 
  FROM post_tags 
 WHERE tag_ID IN (1, 2) -- IN($thisTags[0], $thisTags[1], $thisTags[2], $thisTags[3])
 GROUP BY post_ID
 HAVING COUNT(DISTINCT tag_ID) = 2 -- COUNT(DISTINCT tag_ID) = 4

Here is SQLFiddle demo

HAVING COUNT(DISTINCT tag_ID) = 2 makes sure that resultset contains only those post_ID that have both tag_ID values 1 and 2.