J
JimS
I am importing a table from another app. It is passed as an excel
spreadsheet. I have a routine that "cleans up" the sheet, then imports it
into a temp table. Then, I need to match it with an existing table to either
append new records or update existing ones... It's a compund key. To make
things worse, Excel insists one of the columns is numeric, even though I want
it to be seen as text.
I could set the resident table to have this key as a unique compound index
and just do an insert, letting all dups be "errored out". I suppose this is
the best way. But I started by setting up a simple block of code that would
rifle through the import table, matching with the resident table, and either
inserting or updating it as it goes. But ADO won't accept a compound
(multi-column) argument for its "find" method.
How does everyone else get around this?
spreadsheet. I have a routine that "cleans up" the sheet, then imports it
into a temp table. Then, I need to match it with an existing table to either
append new records or update existing ones... It's a compund key. To make
things worse, Excel insists one of the columns is numeric, even though I want
it to be seen as text.
I could set the resident table to have this key as a unique compound index
and just do an insert, letting all dups be "errored out". I suppose this is
the best way. But I started by setting up a simple block of code that would
rifle through the import table, matching with the resident table, and either
inserting or updating it as it goes. But ADO won't accept a compound
(multi-column) argument for its "find" method.
How does everyone else get around this?