Starting from December 2000, I kept track of all the movies I watched, writing reviews and posting them to the website I maintained at that time. The last entry was in April 2007. Back then, the addition of reviews was almost entirely manual. When I finally decided to decommission that site, I made a backup of the data.
This repository preserves that data, but also acts as the root for projects which utilize that data in one form or another.
The original database was MySQL. sql/eiga.sql is a slightly
updated version of the final data set. I've loaded the data into a more recent
version of MySQL and updated errors that I've come across while playing with the
data more recently. Because the original data did not maintain foreign keys
(though they effectively did exist), I have not added them there. The data
consists of two tables: movies and reviews. movies tracks the facts about
the movies watched, while reviews captures my review along with when and where
I watched the film.
eiga-lite.sql takes the MySQL format and strips it
down to a more ANSI-compliant format which can be read into SQLite with .read eiga-lite.sql. The tables can then be queried from the command line.
movie.dat and review.dat are pipe-delimited data import files, without headers, which are used to import the data into the applications.
The maru_sqlite project is written in Elixir and uses Maru and Sqlite.Ecto. It is a fairly manual approach to exposing the data, useful for me in exploring putting an Elixir project together without Phoenix.
The vase project is written using the cognitect/vase Clojure library. Vase pushes most of the work into configuration files. By default, vase uses Datomic to store data. Vase is built on top pedestal.
The pedestal project is written using the pedestal Clojure library. Pedestal provides the web service layer. I have not yet determined which database this project will use.
Copyright Michael S. Daines, 2000-2018. All Rights Reserved.