SQLite3 - Counting number of duplicate and non-duplicate books each user owns - sql

Question

I'm creating a database that keeps track of books, users, and what books each user owns. A user can have several copies of a certain book, specified by their book id. What I'm trying to calculate in particular is show for each username the number of all books they own and the number of non-duplicated books they own. I have an attempt below but the numbers do not appear to be correct. For example after doing my select statement, it says that Sammy's total number of duplicated books is 4 and his total number of non-duplicated books is 3.
When you actually look at the data in the owns table, you can see the real values are that Sammy's total duplicated books is 3+2+1+1 = 7 books, and his total number of non-duplicated books would just be the total number of unique book_ids he has in his collection which is just 4.
I'm not sure what's wrong with the logic of my query and would appreciate some help.
Schema:
CREATE TABLE IF NOT EXISTS books(
id integer NOT NULL primary key UNIQUE,
title text NOT NULL UNIQUE,
genre text NOT NULL,
price integer NOT NULL,
units_available integer NOT NULL
);
CREATE TABLE IF NOT EXISTS users(
username text primary key NOT NULL UNIQUE,
password text NOT NULL
);
CREATE TABLE IF NOT EXISTS owns(
owners_username integer NOT NULL,
book_id integer NOT NULL,
quantity integer NOT NULL,
PRIMARY KEY (owners_username, book_id),
FOREIGN KEY (owners_username) REFERENCES users (username),
FOREIGN KEY (book_id) REFERENCES books (id)
);
Everything in owns table / all books and their quantities owned by each user:
select * from owns;
owners_username book_id quantity
--------------- --------- --------
Bobby 47911 1
Bobby 49286 1
Bobby 55622 1
Sammy 50818 3
Sammy 49290 2
Sammy 55617 1
Sammy 6555 1
Andrew 50546 1
Andrew 49290 4
Andrew 48401 1
When I attempt to count the number of duplicated and non-duplicated books each user owns:
select owners_username, dup_count, nodup_count
from
(select owners_username, count(quantity) as dup_count
from (select owners_username, quantity from owns)
group by owners_username)
natural join
(select owners_username, count(quantity) as nodup_count
from (select distinct owners_username, quantity from owns)
group by owners_username);
owners_username dup_count nodup_count
--------------- --------- -----------
Andrew 3 2
Bobby 3 1
Sammy 4 3

Answer 1

You need for each user the sum of the column quantity and the number of distinct book_ids:
SELECT owners_username,
SUM(quantity) dup_count,
COUNT(DISTINCT book_id) nodup_count
FROM owns
GROUP BY owners_username
See the demo.
Results:
owners_username
dup_count
nodup_count
Andrew
6
3
Bobby
3
3
Sammy
7
4