从两个不同的表中连续减去两列

Question

I have two tables, twentyBuyer and twentySeller , and in both of the tables, I have multiple columns but the one I want to pull out is totalCost . I have been trying to figure out a way to do so that I can subtract the twentyBuyer.totalCost from twentySeller.totalCost successively. The tables are like this below.

TABLE twentyBuyer

ID, Date,        Name,   TotalCost, Type
1,  "today",     "John", 46,        "ONE"
2,  "yesterday", "Doe",  506,       "ONE"

TABLE twentySeller

ID, Date,        Name,     TotalCost, TYPE
1,  "today",     "Franko", 273,       "ONE"
2,  "yesterday", "Bob",    207,       "ONE"

At first, I tried to pull the information using SELECT statements so I could just do the subtraction very easily doing SELECT twentyBuyer.totalCost - twenty_seller.totalCost FROM twentyBuyer, twenty_seller; to which I got

diff
299
-161
233
-227

I thought that this way was wrong so I Googled some and came across joins. When I started using INNER JOIN , I almost got the right results but there were two more results

CODE:

SELECT DISTINCT twenty_seller.totalCost, twentyBuyer.totalCost
FROM twenty_seller
LEFt OUTER JOIN twentyBuyer
ON 1=1;


totalCost | totalCost
207       | 506
273       | 506
207       | 46
273       | 46 

I tried to add UNIQUE to it and nothing seemed to work. I then switched to views but was having similar problems. I deleted the views code but I was running into similar issues.

I want the results to be like

difference
-227
299 

I apologize if this is a trivial question but I just cannot figure out how to do this and I think there is a simple solution that I cannot think of yet. Any help appreciated

Answer 1

The database doesn't attempt to match rows from the two tables - it performs a Cartesian product (ie, matching every row from the first table with every row of the second table), and it's up to you to provide a condition telling the database how to join these rows.

In the first example, you don't have a condition, so you're getting 2*2 = 4 rows. This is called a cross-join. In the second example, the condition of 1=1 always evaluates as true, so although it doesn't like it, it's also a cross-join, and produces the same results as the first query.

Assuming you meant to match the rows with the same IDs, you need to provide this logic as a condition:

SELECT td.id, ts.totalCost - tb.totalCost
FROM   twenty_seller ts
JOIN   twentyBuyer tb ON ts.id = tb.id;

Answer 2

You're looking for the "AS" clause with some math thrown in.

You're also not giving it any columns to match, which means that ALL the rows from BOTH tables are just going to be returned as is.

You can fix this by giving it a specific column to match.

Like so:

SELECT 
  twenty_seller.totalCost AS 'cost1',
  twentyBuyer.totalCost AS 'cost2',
  twenty_seller.totalCost-twentyBuyer.totalCost AS 'Difference'
FROM twenty_seller
LEFT OUTER JOIN twentyBuyer ON twenty_seller.ID=twentyBuyer.ID; 

Or, if you just want the difference:

SELECT 
  twenty_seller.totalCost-twentyBuyer.totalCost AS 'Difference'
FROM twenty_seller
LEFT OUTER JOIN twentyBuyer ON twenty_seller.ID=twentyBuyer.ID; 

And if you want the total over all the rows:

SELECT 
  SUM(twenty_seller.totalCost)-SUM(twentyBuyer.totalCost) AS 'Difference'
FROM twenty_seller
LEFT OUTER JOIN twentyBuyer ON twenty_seller.ID=twentyBuyer.ID;