DBMS 2020 final project

Team Info

• Name: CCLab
• Members: 盧佳妤 LuJiaYu, 段寶鈞 Ivy, 陳先灝 Jacky, 王均捷 Rainboltz, 方文忠 LittleCube

---

Database Preparations

• We created a database based on the public dataset MovieLens-1M
• Instead of the public data, we also retreive some additional informations from websites by using the web crawlers.
• for reviewing the dataset on PostgresDB, we provide alternative raw data input using SQL file to import dataset into the PostgresDB (NOT RECOMMENDED)

  psql -f "ml1m" -U username

1. crawl the additional informations

• the presentation can be previewed here
• we crawled the additional informations such as:
  1. movie poster
  2. directors
  3. writers
  4. stars
  5. introductions & storylines

2. dataset preprocess

• we remain the full dataset in our github repo here :)
• usage:

  cd preprocess
  python get_csv.py && get_graph.py

3. setup the database

1. download data from google drive link
2. its a .sqlite file, just put it under directory sql_api_server

---

Website Service Instructions

• A pure handmade MVC-structured website. Tornado on Python provides the Controller, the web pages are Templates (Views), and the SQL-api server infers to Model.
• Frontend is composed of Bootstrap + jQuery (#javascript).
• Backend is provided by Tornado (#python)
• SQL queries are safe and cannot be SQL injected :)

0. install requirements

• python version >= 3.6
• tornado >= 6.0.0

pip3 install -r requirements.txt

1. setup web service

• for windows, simply click website_server/start.bat.
• for general OS systems:

  cd website_server
  python startserver.py

• the website URL is default at http://127.0.0.1:8889.
• To change service IP address (default=localhost):

  cd website_server
  python set_server_ip.py 123.456.789.012

  then the website will be on http://123.456.789.012:8889

2. setup SQL-api service

• for windows, simply click sql_api_server/start.bat.
• for general OS systems:

  cd sql_api_server
  python correct_md5.py && startserver.py

• notice that this sql api service can be only used on localhost for security issues. (http://127.0.0.1:8888)