Initial commit

Author: jfarmer

.gitignore

@@ -0,0 +1 @@
+.DS_Store

README.md

@@ -1,4 +1,119 @@
-sql-queries
-===========
+# A Toy Database
 
-Some exercises around SQL queries
+Here is a toy database which models an online music store.  In this database, we have falling into three categories: information about the company, information about customers and their orders, and information about the products being sold.
+
+You'll be editing the `queries.sql` file, which consists of a series of
+"plain English" sentences.  Your job is to translate those sentences into a
+single SQL `SELECT` query.  Don't worry if you can't do it — take your best shot!
+
+## Contents
+
+1. [The Plain English Queries](#the-plain-english-queries)
+2. [Configuring SQLite3](#configuring-sqlite3)
+3. [The Database Tables](#the-database-tables)
+  1. [Information About The Products](#information-about-the-products)
+  2. [Information About Customers and Their Orders](#information-about-customers-and-their-orders)
+  3. [Information About The Company](#information-about-the-company)
+4. [One Key Idea](#one-key-idea)
+
+## The Plain English Queries
+
+See queries.sql.  Open the database by running this in your terminal:
+
+```bash
+sqlite3 music-store.db
+```
+
+Once inside, type the following at the SQLite3 prompt
+
+```text
+sqlite> .schema
+```
+
+to list all the tables in the database.  To see the schema for a specific table,
+you can type, e.g., the `invoices` table
+
+```text
+sqlite> .schema invoices
+```
+
+Don't be afraid to explore the data in the tables to get a better feeling
+for how it's formatted.  It's impossible to damage your database as long
+as you're only running `SELECT` queries.  For example, what does the data
+in the invoices table look like?  Well, let's look at a 5 rows
+
+```sql
+SELECT * FROM invoices LIMIT 5;
+```
+
+That's what we mean by "explore."
+
+## Configuring SQLite3
+
+The default settings for SQLite3 can make it hard to read the output of `SELECT`
+queries.  To change these settings, run the following three commands once you're inside
+the SQLite3 shell.  You should just be able to copy and paste these three lines.
+
+```text
+.header on
+.mode column
+.width 20
+```
+
+It should look roughly like this: http://cl.ly/image/24431S2W0I3f
+
+If you want to make these settings permanent, copy and paste the following
+command into your **terminal** (not the sqlite shell). **Note**: this will not
+work on Windows.
+
+```bash
+cat <<EOF > ~/.sqliterc
+.header on
+.mode column
+.width 20
+EOF
+```
+
+You should copy and paste all 5 lines and it should look roughly like this: http://cl.ly/image/3w1Y0l1a0W38
+
+Don't worry about what all the parts of this command are.  The short of it is
+that this is adding lines to a file named `.sqliterc`, which SQLite3 uses to
+determine its default settings.
+
+## The Database Tables
+
+Remember: relational databases are organized into tables.  Each table has a "schema", which dictates the fields (aka columns) that the data in the table can or must contain.  The data itself is stored as records (aka rows).
+
+Here are the tables, organized by high-level purpose.
+
+### Information About The Products
+
+1. `albums` are the albums our store is selling
+2. `artists` are all the artists
+3. `tracks` are all the tracks, containing a column that tells us which album the track belongs to.
+4. `media_types` tell us what format a track is in, e.g., MP3, Apple Audio, etc.
+5. `genres` are a list of genres.
+6. `playlists` are, well, playlists.  A play list has many tracks and a single track can belong to many playlists.
+7. `playlist_tracks` tells us which tracks belong to which playlists
+
+### Information About Customers and Their Orders
+
+1. `customers` are our customer records and they may or may not have an assigned "customer support representative"
+2. `invoices` are a list of specific customer orders
+3. `invoice_lines` are the line-items on specific invoices
+
+### Information About The Company
+
+1. `employees` contains information about our company's employees.  This is mostly used to assign support representatives to customers.
+
+## One Key Idea
+
+One of the key ideas in how relational databases like SQLite3 and MySQL organize data is that we try to minimize redundancy by using references to other data rather than duplicating that data between different tables.
+
+For example, every track belongs to one (and only one) album.  An album has its own associated information, like album title, publication date, and so on.  Each track also has its own associated information, like track title, track position, and duration.  We want to be able to ask questions like "What is the title of the album on which track X appears?"
+
+If you imagine a spreadsheet with a bunch of track listings, one way to achieve this would be to have an "Album Title" column and to answer this question we would just look at the value in the "Album Title" column for track X.  Every track on the same album would have the same value in the "Album Title" column, although heaven help us if there are two separate albums with the same title!
+
+This is not how we store information in a relational database.  Rather than storing album-related information in the same table as track-related information, we store album-related information in an "albums" table and track-related information in a "tracks" table.  We assign each track and album a unique ID.  In the "tracks" table we would when have an `album_id` column containing the unique album ID as a *reference* or *pointer* to the relevant row in the "albums" table.
+
+Excel can do this, but it is too cumbersome for the most common tasks.  In a relational database like SQLite3 or MySQL, however, it is much easier to deal with.  In fact, SQL (the language) is counting on you storing your data this way.

music-store.db

@@ -0,0 +1,97 @@
+-- Below are a bunch of "plain English" sentences.  Your job is to take a stab
+-- at translating each one into a single SELECT query.  You can write your
+-- query below each "plain English" sentence.
+--
+-- We've written some example queries as a way to both show you how and where
+-- to type your work and also as a way to show you new SQL commands, clauses,
+-- or syntax where necessary.
+
+-- **Please** don't delete the "plain English" versions. Just write your query
+-- on the line below it. :)
+
+-- The lines in this file that begin with "--" are comments.  You're encouraged
+-- to leave your own comments, reflecting on things you're confused about,
+-- questons you have, or anything else.  The more we know about how you're
+-- thinking the better feedback we'll be able to give!
+
+-- "albums" table
+
+-- Every album
+SELECT * FROM albums;
+
+-- Every album title (and only the title)
+SELECT title FROM albums;
+
+-- Every album ordered by album title in ascending order (A-Z)
+-- Note: ORDER BY orders by lowest-to-highest by default
+SELECT * FROM albums ORDER BY title;
+
+-- Every album ordered by album title in descending order (Z-A)
+SELECT * FROM albums ORDER BY title DESC;
+
+-- Every album whose title starts with A in alphabetical order
+SELECT * FROM albums WHERE title LIKE 'A%' ORDER BY title;
+
+-- "invoices" table
+-- Remember: run ".schema invoices" to see the structure of the "invoices" table
+
+-- Every invoice
+
+-- Every invoice ordered by total invoice amount ("total")
+
+-- Every invoice with a total greater than 10
+
+-- The 10 least expensive invoices
+-- Remember: ORDER BY orders from lowest-to-highest by default
+SELECT * FROM invoices ORDER BY total LIMIT 10;
+
+-- The 10 most expensive invoices
+
+-- The 15 most recent invoices
+
+-- The 15 oldest invoices
+
+-- The 10 most expensive invoices from the US
+
+-- The 10 least expensive invoices from the US
+
+-- The 10 most expensive invoices from outside the US
+-- Hint: If "=" means equal, use "!=" to mean "not equal"
+
+-- Every invoice from Chicago, IL
+SELECT * FROM invoices WHERE billing_city = "Chicago" AND billing_state = "IL" AND billing_country="USA";
+
+-- A list of all the invoices worth more than $5.00 from Chicago, IL
+
+-- The billing addresses of the 5 most valuable invoices from Mountain View CA
+-- Gotta reward those big spenders!
+
+-- A list of the 10 most valuable invoices made before January 1st, 2010
+-- Hint: Dates are formatted like 'YYYY-MM-DD' and you can compare them using '<', '>', '<=' and '>='
+
+
+-- The number of invoices from Chicago, IL
+SELECT COUNT(*) FROM invoices WHERE billing_city = "Chicago" AND billing_state = "IL" AND billing_country = "USA";
+
+-- The number of invoices from the US, grouped by state
+SELECT billing_state, COUNT(*) FROM invoices WHERE billing_country = "USA" GROUP BY billing_state;
+
+-- The state in the US with the most invoices
+SELECT billing_state, COUNT(*) FROM invoices WHERE billing_country = "USA" GROUP BY billing_state ORDER BY COUNT(*) DESC LIMIT 1;
+
+-- The total invoice value from California
+SELECT billing_state, SUM(total) FROM invoices WHERE billing_country = "USA" and billing_state = "CA";
+
+-- The number of invoices and the invoice total from California
+SELECT billing_state, COUNT(*), SUM(total) FROM invoices WHERE billing_country = "USA" and billing_state = "CA";
+
+-- The count, total, and average of invoice totals from California
+SELECT billing_state, COUNT(*), SUM(total), AVG(total) FROM invoices WHERE billing_country = "USA" and billing_state = "CA";
+
+-- The count, total, and average of invoice totals, grouped by state, ordered by average invoice total from highest-to-lowest
+
+-- A list of the top 5 countries by number of invoices
+
+-- A list of the top 5 countries by gross/total invoice size
+
+-- A list of the top 5 countries by average invoice size