Comments on: select "order by occurrence" mysql

Question: select "order by occurrence" mysql

outofpaper — Thu, 28 Jul 2005 14:35:20 -0800

Does any one know how to find the most common result for a column returned from a mysql select request?

By: wackybrit

wackybrit — Thu, 28 Jul 2005 14:40:12 -0800

Untested, but something like..

SELECT *, COUNT(desiredcolumn) AS occurances FROM table GROUP BY occurances ORDER BY occurances DESC LIMIT 1;

should work (may need a bit of fixing).

By: wackybrit

wackybrit — Thu, 28 Jul 2005 14:40:46 -0800

Correction already:

SELECT *, COUNT(desiredcolumn) AS occurances FROM table GROUP BY desiredcolumn ORDER BY occurances DESC LIMIT 1;

By: wackybrit

wackybrit — Thu, 28 Jul 2005 14:45:11 -0800

Tested. It works. If you have trouble, let us know.

By: wackybrit

wackybrit — Thu, 28 Jul 2005 14:46:13 -0800

Of course, if I were to be really pendantic to myself, I'd point out it's "occurrences".. duh :) I'll stop posting for a minute now...

By: oliyoung

oliyoung — Thu, 28 Jul 2005 16:27:44 -0800

An if you want to tack on a limiting clause to an aggregated SQL query, you'll have to use "HAVING x" instead of "WHERE x" - took me a while to remember that last week ..

so it'd become
SELECT *, COUNT(desiredcolumn) AS occurances FROM table GROUP BY desiredcolumn HAVING occurances > 5 ORDER BY occurances DESC LIMIT 1

By: matildaben

matildaben — Thu, 28 Jul 2005 18:40:09 -0800

But you don't need the HAVING occurrences > 5 unless you're sure that there is a case where there are more than 5 repeats, otherwise you'll exclude your result if the maxcount happens to be only 4. Also that clause induces additional calculation overhead. The LIMIT 1 will do what you originally asked.