I don't even know if I am doing this query the right way.
There is a Sandwiches
table that has some 7 fields and 2 of them are comboboxes (Type
and Bread
).
So I made a query that combines all of the comboboxes values into one query, like this:
SELECT TypesAndBreads.TBName, TypesAndBreads.Type
FROM (SELECT [Sandwiches Types].[Sandwich Type] As TBName, "Sandwich Type" As Type
FROM [Sandwiches Types]
UNION ALL
SELECT Breads.Bread As TBName, "Bread" As Type
FROM Breads) AS TypesAndBreads;
I get the flat values of the tables now I want to count all the sandwiches under each TypesAndBreads.TBName
. I have this, just to make sure it works with all the Sandwiches:
SELECT TypesAndBread.Type, TypesAndBread.TBName,
(SELECT Count(Sandwiches.[SandwichID]) As SandwichCount
FROM Sandwiches) As SandwichCount
FROM TypesAndBread;
But I want to reference the current Type and TBName inside the subquery. Something like this:
SELECT TypesAndBread.Type, TypesAndBread.TBName,
(SELECT Count(Sandwiches.[SandwichID]) As SandwichCount
FROM Sandwiches
WHERE Sandwiches.[TypesAndBread.Type] = Sandwiches.[TypesAndBread.TBName]) As SandwichCount
FROM TypesAndBread;
But of course this doesn't work. I didn't think it will, just thought of giving it a try. I was thinking of maybe constructing the query with VBA when they open the Report that this query is going to be based of.
So I guess my question is: Is there a way to reference the current selected fields in a subquery? Or is there a different way to approach this?
Thanks for the help
EDIT: My table structure is like this:
Sandwiches
's fields
| SandwichID | Name | Date Added | Chef | Sandwich Type | Bread | Reviewed By |
where Sandwich Type
and Bread
are Lookup fields for these tables:
Sandwiches Types
's fields
| Sandwich Type |
Breads
's fields
| Bread |
The TypesAndBreads query combined the Sandwiches Types and Breads tables, but the reason for that is so that I can get the count of all the sandwiches that have that Type or bread. A result like this:
+=============================================+
| Type | TBName | SandwichCount |
+=============================================+
| Sandwich Type | Turkey Club | 10 |
| Bread | Italian | 5 |
| Bread | Garlic | 8 |
+---------------------------------------------+
the example result's first row basicly says there are 10 sandwiches in record with the Sandwich Type field equal to Turkey Club.
I hope that explains it better.
Not sure if Access supports it, but in most engines (including SQL Server
) this is called a correlated subquery and works fine:
SELECT TypesAndBread.Type, TypesAndBread.TBName,
(
SELECT Count(Sandwiches.[SandwichID]) As SandwichCount
FROM Sandwiches
WHERE (Type = 'Sandwich Type' AND Sandwiches.Type = TypesAndBread.TBName)
OR (Type = 'Bread' AND Sandwiches.Bread = TypesAndBread.TBName)
) As SandwichCount
FROM TypesAndBread
This can be made more efficient by indexing Type
and Bread
and distributing the subqueries over the UNION
:
SELECT [Sandwiches Types].[Sandwich Type] As TBName, "Sandwich Type" As Type,
(
SELECT COUNT(*) As SandwichCount
FROM Sandwiches
WHERE Sandwiches.Type = [Sandwiches Types].[Sandwich Type]
)
FROM [Sandwiches Types]
UNION ALL
SELECT [Breads].[Bread] As TBName, "Bread" As Type,
(
SELECT COUNT(*) As SandwichCount
FROM Sandwiches
WHERE Sandwiches.Bread = [Breads].[Bread]
)
FROM [Breads]