Typically one creates a separate table with a list of constants. These can be included in your query or report as you wish. But I'm a bit confused by what you mean as a "constant list" that is assigned to each record but then, "if another record exists" you want to replace some record.
Are you saying that you want to update the constant list based on information in a new record?
Are you saying that each record has it's own set of constants, but that the constants may change when a new record comes in? This would be the case where a person's address remains the same over an extended time, and multiple records, but is subject to being changed when a new one comes in.
In the last case you may want to extract a separate table of "constants" and check that with each new data set. When there is a change the the separate constants table would need to be updated. You do need to have a unique identifier for each individual, or entity, so that the constants table is being updated for that individual or entity. I have an application like that here where I'm tracking employees and maintaining a table of their location in our organization togather with their email information. I use it for reference when we receive a record missing information but update it when a new record comes in as occasional moves occur from one job location to another.
Does any of this address your question?
Moshe
Washington, DC
