I am trying to write a macro to query from our database using the IN
clause except with one problem. I am reaching the limit of the IN
clause for SQL Server.
My macro looks like this:
Dim row_count As Double
row_count = ActiveSheet.UsedRange.Rows.Count - 1
half_row_count = row_count
Dim i As Double
Dim products As String
For i = 2 To half_row_count
Dim product_id As String
product_id = Cells(i, 1).Value
'test = sixtyDays(product_id, conn)
'Cells(i, 10).Value = test
products = products & "'" & product_id & "'" & ", "
Next i
Dim sample As New ADODB.Recordset
products = Left(products, Len(products) - 2)
Set sample = sixtyDays(products, conn)
Sheets(1).Range("K2").CopyFromRecordset sample
conn.Close
Function sixtyDays(ProductID As String, new_conn As ADODB.Connection) As ADODB.Recordset
Dim sConnString As String
Dim rst As New ADODB.Recordset
Dim recordsAffecfted As Long
StrQuery = "SELECT ProductAnalysisByMonth.SalesQty FROM ProductAnalysisByMonth WHERE ProductAnalysisByMonth.ProductID IN (" + ProductID + ") AND ProductAnalysisByMonth.Month = " + CStr(Month(Date) - 2)
rst.Open StrQuery, new_conn
Set sixtyDays = rst
End Function
So I need to some how split the query into smaller chunks, except, the number of arguments passed to the SQL query will vary from week to week.
What is the most efficient way of handling this problem?
Create a table function that will return your string results into a data-set that can be inserted into a CTE, temp table, or used directly in a join. This has been the most effective way for me to get around this limitation. Below is a link to Ole Michelsen's website who provides a simple but flexible solution.
Link: http://ole.michelsen.dk/blog/split-string-to-table-using-transact-sql.html