Umut Tekin picks up on a thread:
Recently, I read Laurenz Albe’s blog about case insensitive string search. He recommended case insensitive collations saying, “it won’t be any more expensive (and maybe more intuitive) than the current techniques of using
upper()
or thecitext
extension are today”. In other words, a custom collation can be used instead of citext to achieve case insensivity with a lower performance penalty. Today we will be testing it.
Click through for the results of those tests.