Forum Discussion
Solved
Delete Duplicates (leave 1)
Hello, I have duplicates in my table and I want to delete only 1 of them. I have searched on the net for some code and found the below and fashioned it to my case. I have an error though and it appe...
I don't follow; how do you set an index in MS Access across multiple fields? For example, if I want to avoid duplicates and I have records with First Name, Last Name, City, and State, but I don't want that same person at that location entered twice, how do I set up an index to prevent that?
Also, you can create a composite PK just temporarily to append the records and weed out the dupes, then change back to using a different PK.
Can you explain why a composite primary key should be avoided? Is it a processing thing? Does it matter with record sets below a certain size? Does it appreciably slow down processing when adding records? I'd like to see documentation to determine when this is or isn't a concern.
Also, you can create a composite PK just temporarily to append the records and weed out the dupes, then change back to using a different PK.
Can you explain why a composite primary key should be avoided? Is it a processing thing? Does it matter with record sets below a certain size? Does it appreciably slow down processing when adding records? I'd like to see documentation to determine when this is or isn't a concern.
George has already shown you how to create a unique index for multiple fields.
I have used composite primary keys in the past but eventually stopped doing so (except in rare cases) because they tended to make managing queries & code more unwieldy. If you run an online search you will find articles both for & against their use. A fairly balanced (if rather complex) article can be found at https://medium.com/@pablodalloglio/7-reasons-not-to-use-composite-keys-1d5efd5ec2e5 though I've only skimmed the article
Lets take a simple reason for avoiding them. As well as not containing duplicates, primary keys cannot ever contain null values. So all those fields must ALWAYS contain data which at times may prevent a new record being saved. However, with a unique index, you can choose to ignore nulls.
Hope that helps