Question

I’ve hit a wall with my mySQL knowledge.

To explain it more clearly here are my tables (a pretty standard setup for tags):

keywords
---
keyword_id (ai)
keyword

–

topics_keywords
---
t_k_id (ai)
topics_id
keyword_id

–

topics
---
topics_id
etc

Now, lets say that keywords table has the following entries:

keyword_id || keyword
1 || baseball
2 || sports

And topics_keywords has the following entries:

t_k_id || topics_id || keyword_id
1 || 1 || 1 
2 || 1 || 2
3 || 2 || 1

So maybe you can see what im trying to do.

If someone inputs ‘baseball’ I want to get a COUNT for how many topics that share a related keyword. In the above example, baseball and sports are both shared keywords for topics_id: 1.

So if someone inputs baseball how would I get the count() that displays there is 1 other topic out there, named sports that shares a topic with baseball?

Answer 1

If I understand you correctly you might want to look into running something like this:

SQL Fiddle: http://sqlfiddle.com/#!2/9e6be/6

SELECT COUNT(*)                                                             
  FROM keywords
  JOIN topics_keywords
 USING (keyword_id)
 WHERE keyword != 'value'
   AND topics_id IN (
         SELECT topics_id
           FROM keywords 
           JOIN topics_keywords 
          USING (keyword_id)
          WHERE keyword = 'value' );

That is, you’re selecting the count of topics_keywords that do not have the keyword you’re looking for but do have the topics_id of the keyword you’re looking for.