Question

I have a table which contains orders, and orders are being added to the table by users as time goes by.
I want to implement a service that checks if a row was added to the table.
Is there a specific way to do that?
thanks!

Answer 1

If you want to know which rows have been added since last time you checked, put a timestamp in each row, and keep track somewhere (separately) of the newest row you’ve seen so far. To find new rows, query for all rows whose timestamp is newer than newest one you’ve seen before. Then take the most recent timestamp from the result set, and use it to update your “newest row seen so far” variable.

The database itself doesn’t keep track of which rows have been newly-added because the meaning of “new” depends on who’s asking. A row that was added six months ago is “new” to someone who hasn’t checked since then. That’s why you have to use timestamps, and have the application keep track of which timestamp currently marks the boundary between “old” and “new”.

Edit: Actually, instead of timestamps, you might want to use an auto-increment integer column. With timestamps there’s a slight chance that two rows may be added so close together in time that they get the same timestamp, and if the application does its query at a moment when only one of those rows has been inserted, it’ll “miss” the other one next time it checks for new rows because it thinks that timestamp has been seen already. A value that always increases for every new row would avoid that problem, plus many tables have one already (for use as a primary key).