Question

Scenario

I need to update a SQL 2008 database daily via a spreadsheet (the only option available). The format is pretty basic, however there are potentially millions of records. Column1 and Column3 will have many predefined duplicate values, already pulled out into separate tables.

Spreadsheet Sample

Column1 Column2 Column3
Apple   10      Red
Apple   20      Red
Apple   15      Blue
Apple   21      Green
Orange  10      Orange
Orange  7       Orange
Orange  9       Red
Orange  70      Blue
Orange  10      Blue

DB Setup

My database is set up with three separate tables:

//Lookup_Column1
id type
1  Apple
2  Orange

//Lookup_Column3
id type
1  Red
2  Blue
3  Green
4  Orange

//Main - this is what should be inserted, after Column1
//and Column2 are matched to their respective ID's
key Column1 Column2 Column3
1   1       10      1
2   1       20      1
3   1       15      2
4   1       21      3
5   2       10      4
6   2       7       4
7   2       9       1
8   2       70      2
9   2       10      2

Question

How can I write the SQL to insert records that match the information from the lookup tables? How can I go from this:

INSERT INTO Main(Column1, Column2) VALUES ('Apple', 10, 'Red');

To this:

INSERT INTO Main(Column1, Column2) VALUES (1, 10, 1);
//pulled from lookup tables, where Apple = 1 and Red = 1

Answer 1

You could try something like this:

    INSERT INTO Main(Column1, Column2, Column3) VALUES 
    (
    (SELECT id FROM Lookup_Column1 WHERE type = 'Apple'),
    10, 
    (SELECT id FROM Lookup_Column3 WHERE type = 'Red')
    );

There isn’t any fault-tolerance, but it would work as long as you could parse your spreadsheet values into SELECT statements.