SUMIF - last n number of rows

[Excelllllllllllllllllllll]

Greetings!

I've recently been given a sizable chunk of my company's finance sheets to go through the processed payments only (bought X qty - sold X qty)
I can do this manually, but am having a trouble creating a formula or pivot that can help me with this.

COLUMN D	COLUMN E	COLUMN F	COLUMN R	COLUMN S
Client	Transaction	Quantity	Order ID (always unique number)	Final Price (formula)
Company1	Purchase	6	J512f3151f1sfhg4t23r	-600
Company1	Sale	-6	OIHKLNhbu2g8ihor	660
Company1	Purchase	56	*(TH"OIGNEP6D	-5600
Company2	Purchase	-12	LGPNJWHEI4GHjdkn	1200
Company2	Sale	12	IPHEIL£n4kkrENG	1800
Company1	Purchase	1	IOH£*g23fgopj	100

Basically I need a way to calculate the Purchase from row 1+ sale from row 2. Then ignore rows 3&6 completely until we sell 57 units (or 56 vs 56....and 1 vs 1 - because we buy each order separately so we wouldn't sell less than we bough, but we may sell all that we hold if it's for the same client (but this is only by request so it doesn't happen all the time)).

And then I need to do the same for Company 2.

SO - I need a formula or pivot for the above example that shows:
Company 1 = 60
Company 2 = 600

 

[lokayee]

 

[Excelllllllllllllllllllll]

View attachment 29132

Thank you very much for the idea.
I'm actually doing something similar in having these as a table with a totals row on the bottom and filtering company1 them manually leaving only the finished transactions.

That's why I'm looking for a formula that can automate this process and self-populate (if you will) the Y on a helper column perhaps (like in your example)?
Where I'm falling short is when we buy 56 & 1 - then sell 57 - I can't tie those up to match through a formula, or indeed ignore the 56+1 before the sale.

 

[lokayee]