We know how to deal with one table, now we'll see how to deal with two tables that are related.
One Owner
has many dogs
but one dog
can only have one owner.
If we want one record to refer to another (one-to-many relationship) we can make a reference to that record by primary key id in the other table.
In the reffering table it's always named tablename_id
The dog table would have a column: owner_id
that would contain the primary key of that row.
Q. Why do we keep the relationship in the dog
table?
A: Each record/row can only have one value per column. (How would you store all of an owner's dog references in one column?)
tables.sql:
CREATE TABLE owners (
id SERIAL PRIMARY KEY,
name TEXT,
phone TEXT,
email TEXT
);
CREATE TABLE dogs (
id SERIAL PRIMARY KEY,
name TEXT,
owner_id INTEGER
);
Get all of an owner's dogs: Given you already know that owner is id=1;
SELECT * FROM dogs WHERE owner_id = 1;
Use the examples above. Create a database that records dogs and users.
Use psql to make all of your sql queries that affect the DB.
Insert user records, then insert dogs that belong to the users.
Further
Create a database that records students and teachers.
Further
Create a node command line program that takes the name of the owner and prints out the dogs that belong to that owner.
(Note that this should require 2 different SQL queries)
Seed your DB:
INSERT INTO owners (name, phone, email) VALUES ('Chee Kean','34562876','ck@ga.co');
INSERT INTO owners (name, phone, email) VALUES ('Scott','37562876','scott@ga.co');
INSERT INTO dogs (name, owner_id) VALUES ('Fido', 1);
INSERT INTO dogs (name, owner_id) VALUES ('Susan Chan', 1);
INSERT INTO dogs (name, owner_id) VALUES ('Alice', 2);
INSERT INTO dogs (name, owner_id) VALUES ('Mr. Winkle Pants', 2);
When you run:
node index.js "Chee Kean"
Your program should return:
Fido, Susan Chan