InvervalMatch

[#1]

Hi, this is a question about "IntervalMatch".

I have an inventory table which includes inventory qty only within each period, and I want to match dates with the periods so I can have inventory qty for everyday.

The following codes work for only several SKUs. Could anyone help with the codes? Thanks!!

For example,

Inventory Table

SKU FirstDate LastDate QTY

015341755069 2/11/2008 2/27/2008 0

015341755069 2/28/2008 6/4/2009 1

Dates

1/7/2008-6/4/2009

Desired Table

SKU Dates QTY

015341755069 2/11/2008 0

015341755069 2/12/2008 0

015341755069 2/13/2008 0

:

:

015341755069 2/27/2008 0

015341755069 2/28/2008 1

015341755069 2/29/2008 1

:

:

015341755069 6/4/2008 1

This is my code:

[#2]

Dates.qvd uploaded

[#3]

Inv.qvd uploaded

[johnw]

You may have to describe the problem in more detail. I downloaded your files, ran the load, and the results look fine to me. You have 0 up through 2/27/2008, then it switches to 1, just like you asked for. I chose another SKU, and while it was MUCH more complicated, the first set of values that I looked at looked correct.

What problem are you seeing? Is there some specific SKU I need to select to see the problem?

[#5]

My problem was when I limited SKU to only one or two items, the code worked as you saw.

But when I tried to load everything, the execuation of reload failed.

Even if I limit "Date" to only one day, it still failed. I am wondering if it's a problem of memory limit.

But since date is limited to 1 row, inventory is only read 40,000 row or so(if I am not mistaken, I don't have files with me right now),

there shouldn't be a memory problem.

What should we do to solve the problem?

Thank you.

(I'll reload more information when I get data with me)

[rwunderlich]

I ran your sample without the Join and it generated 84M rows. Fit ok in my RAM but when I tried to do the Join I ran out of RAM. I frequently use IntervalMatch for this kind of problem, but I can see it may not a great solution when you have a large number of rows.

Anyone suggest a better way that can handle a large product or rows?

-Rob