fenway

Besides, exact values for pagination is just slow -- Google doesn't do it, either.

This has nothing to do with mysql.

The delimited list is only appropriate for summary tables.

Don't use these values as counters -- that's just wrong. Nothing will work properly if you reset them. Trust me.

MySQL will return the rows that match ANY of those values. Why do you want it 3 times?

Cleanest, yes; fastest, no. The most efficient way to is to self-join N times -- as long as N is small.

In general, use case-insensitive collations -- no performance hit.

Echo the query.

Drop the joins, get the desired grouped result first, then join.

Correct.

Good -- now post the output you get from each one, and what you want instead.

Store this as min/max values, with a text label.

That's because you can't mix JOINs and GROUP BY that way -- and you can't rely on any non-aggregated columns, either.

Yup, this is caching -- no reason to post the script in the mysql forum.

Don't mark your posts with "urgent" -- it's against the ToS. And besides, I don't see a question.

The table prefixes are not included -- use an explicit column alias.

Don't build the query, build the clauses -- push onto an array, join at the end.

If you really wanted, you could look up the index column in mysql's table, but it seems like a bother.

Don't generate users or view on-the-fly -- that's not a good approach.

You don't get a choice -- specify them -- typing is easy.

I'm not sure what you want us to do with that statement.

I mean it depends which DB class you're using.

What does the error log reveal?

I suppose you could simply add them as scalar subqueries....

It's not -- that's crazy talk. You're using the (default) case-insensitive collation -- that's a good thing. If you want a case-sensitive match, then collate as BINARY. -- see here.