Question

I’m coding for SQL Server 2008 R2. I have two databases that contains dozens of tables.

Each table in one database contains a table with the same name in the other database.

Tables with the same name have identical format (fields and data types). The contents of the two tables are similar but not identical. I need to update one based on changes made to the other, but only under certain circumstances. I think I want to use a cursor for this, but I can’t find a good example to go by.

So far, the MSDN examples are reading one field at a time into a variable. I do need to be able to read/modify two columns which are identical in each table, but I gotta believe there’s something less tedious than declaring variables for every column of every table.

I would like to be able to FETCH an entire row, check a couple of columns and then make a decision of whether I want to write the entire row to the other table after changing two columns – but do I have to declare variables for EVERY field I want to fetch / write? There’s no way to just FETCH an entire row and write an entire row?

Answer 1

you could create a temporary table and insert the row there, but I think the MERGE command, which is very configurable and fast, should help you here. http://technet.microsoft.com/en-us/library/bb510625.aspx