Re: Database introduction

On Mon, 27 Jan 2014 13:56:04 +0100
Max <maxfalsa_at_despammed.com> wrote:

> Unfortunately I can't grasp division in a relational database..

Division is harder than multiplication. We learn it later in school, and it takes more cycles for a CPU. It's the basis for cryptography, and after 30 years SQL DBMSs still can't do it.

I find the easiest way to think about it is as what it is: the inverse of multiplication. Given any product

        insert into C select * from A cross join B

division produces

		A  = C ÷ B
	and
		B = C ÷ A

It doesn't come up a lot in my experience but, like any tool, when you need it you really need it.

The examples I remember involve whether a canonical set has been met. Say you have a type of task that consists of steps A, B, and C.

	insert into task_types 
		values ( ( 1, 'A' ), (2, 'B'), (3, 'C') )

As each step is done for each actual task, you add a row for the task and the step

        insert into status values (task_id, 'A')

Then eventually you want a list of completed tasks. That is, you want

        task_id = status ÷ task_types.step

SQL version left as exercise for the reader!

HTH. --jkl Received on Wed Jan 29 2014 - 02:26:37 CET