what is difference UNIQUE …, in postgresql?

what is difference UNIQUE, UNIQUE NOT NULL, UNIQUE NULLS DISTINCT, UNIQUE NULLS NOT DISTINCT, in postgresql ?.

CREATE TABLE cus (
  customer_id SERIAL PRIMARY KEY,
  first VARCHAR(20) UNIQUE,
  third VARCHAR(20) UNIQUE NOT NULL,
  fourth VARCHAR(20) UNIQUE NULLS DISTINCT,
  fifth VARCHAR(20) UNIQUE NULLS NOT DISTINCT  
); 

>Solution :

There are a few variations at play here…

• UNIQUE means values in that column must be unique (e.g. they can’t be equal). You can then specify how the database should treat uniques:
  • UNIQUE NULLS DISTINCT (default) – treats NULL values as distinct (so you can still have multiple rows with a NULL value in this column). This is the default behavior, so it’s the same as writing UNIQUE alone.
  • UNIQUE NULLS NOT DISTINCT (default) – treats NULL values as if they are equal, so you can at most have one row with a NULL value in that column.
• NOT NULL is a different modifier that just means that column won’t accept NULL values.

So with those in mind, here is the explained behavior for your examples:

CREATE TABLE cus (
  customer_id SERIAL PRIMARY KEY,
  first VARCHAR(20) UNIQUE,
  -- All values in this column must be unique, but can have multiple nulls
  third VARCHAR(20) UNIQUE NOT NULL,
  -- All values in this column must be unique, nulls not allowed in this column at att
  fourth VARCHAR(20) UNIQUE NULLS DISTINCT,
  -- All values in this column must be unique, but can have multiple nulls (same as first)
  fifth VARCHAR(20) UNIQUE NULLS NOT DISTINCT  
  -- All values in this column must be unique, at most one NULL value allowed
);