|
|
|
|
|
|
by nathan_long
3668 days ago
|
|
|
Suppose you're writing a system for vacation rentals. Here are two data consistency rules: 1) only one user may have a given email address 2) a given property can't have two overlapping rentals. Postgres can enforce both, in a way that's not subject to race conditions. 1) Is a unique constraint and 2) is an exclusion constraint. As far as I'm aware, the only way for application code to do this would be for it to 1) lock an entire table/collection of data 2) do a query to see what data is there 3) check the new data against the existing 4) write new data 5) unlock the database. Postgres probably has to do the same thing conceptually, but by using indexes and running the checks in C on the same machine where the data lives, it can do it very quickly. Are you saying you have a better solution? |
|
|