shashwatblack/justmovies

TAMU CSCE 608 Project 1

JUST MOVIES

What is it?

Just Movies is a simple movie and celebrity directory. The cherry on the cake is the data feature, which runs analytical queries on the available data set and presents them in beautiful charts.

This work is done as a course project for Database Systems (TAMU CSCE 608). The author is Shashwat Shashi Mehta. UIN: 827008698

What are its features?

1. Just Movies
   • Grid of movie posters and names
   • Paginated 20 per page
   • Ability to apply following filters
     • Title (supports substring search)
     • Company (supports substring search)
     • Year (supports ranges from-to)
     • IMDB Rating
   • Link to IMDB page under each movie poster
   • Link to edit page
2. Just People
   • Search any celebrity by their name (supports substring search)
   • Will show data for the best match
   • Date of Birth, Roles, Introduction, and Filmography
3. Just Data
   • Lots of beautiful and insightful charts
   • General bird's-eye view stats
     • Total # of movies in system
     • Total # of celebrities in system
     • Total budget of all movies
     • Total reviews received (Note: Reviews feature hasn't been developed yet)
   • Pie Charts
     • Movie count distribution by genre
     • Movie count distribution by country
     • Movie count distribution by language
     • Movie budget distribution by genre
     • Movie budget distribution by country
     • Movie budget distribution by language
   • Budgets by year
     • Bar graph showing the total budget pattern from year to year
   • Celebrity Stats
     • Top 10 celebrities involved in at least 10 movies.
       • The ranking is based on average imdb rating for each of the movies they worked in.
     • Celebrity Role Distribution Venn Diagram
       • How many actors+writers are there?
       • Is writer+director role more common than actor+director?
4. Movie Edit
   • Edit any of the movie's fields
   • Fixed fields are provided as select dropdown
   • Ability to delete the movie entirely from the system.
     • This will also delete all the associated celebrity involvements and reviews from the system.

How was the data collected?

1. The initial data was downloaded from kaggle - https://www.kaggle.com/danielgrijalvas/movies
2. Additional data, such as posters, imdb id and so on were fetched using the OMDB API - omdbapi.com
3. A python script omdb_fetch.py was written to fetch data in batches over a period of few days (due to API limitations).
4. All the data was collected in two CSVs - data/movies.csv and data/movies_with_omdb.csv
5. Further tests were written to verify the consistency of the data.
6. Finally fixed values like country, language, and genre was extracted and SQL insert scripts were created for these.
7. For remaining tables, another script data/jumbo_load.py was written to run over the CSVs and insert everything into the database directly.

What technologies does it use?

1. PostgreSQL
2. Django
3. psycopg2
4. HTML
5. JavaScript
6. charts.js
7. d3.js
8. venn.js
9. Bootflat
10. CSS

How to set it up?

1. git clone https://github.tamu.edu/sswt/justmovies.git
2. cd path/to/project
3. create virtualenv if you want. Recommended!!
4. Install requirements with pip install -r requirements.txt
5. Now create a fresh database in postgres.
6. Add its credentials to utils/credentials.py
7. Run data/data_load.py. This will take around 2 minutes to complete. If everything goes well, you will have the complete database after it ends.
8. Alternatively, you can load data into postgres directly using the data/dump.sql pg_dump.
9. Migrate python manage.py migrate. Although the application logic uses postgres, Django uses sqlite here. I did not change this because in a course project, I want to keep these separate.
10. Start up the server python manage.py runserver

Screenshots

How to reach me if there's some issue?

If you have access, create an issue in this repo. Otherwise send me an email at sswt at tamu dot edu.