MyDramaScraper

This repository provides a classic Scrapy project with a spider named dramalist which allows one to scrap information about the dramas that are listed on MyDramaList.

It also allows to scrap the list of dramas a user completed and their associated ratings

Quick start

To run the spider, first install the dependencies running :

pip install -r requirements.txt

Then, navigate through the the scrapy project called dramascraper which is located in the root directory and from there

You can either scrape MyDramaList's list of drama by running :

scrapy crawl dramalist

To scrape data about user's list, you can run :

scrapy crawl userdramalist -a "<user1>,<user2>...<userN>"

Note : List of users must be comma separated and enclosed in double quotes. A single user can also be passed.

Data types

Drama information

Key	Type	Description
name	String	Name of the drama
synopsis	String	Synopsis of the drama
duration_in_minutes	Integer	Duration of each episodes (in minutes)
nb_episodes	Integer	Number of episodes
country_origin	String	Drama's country of origin
ratings	Float	Ratings from MyDramaList's users
ranking	Integer	Ranking based on the ratings
popularity_rank	Integer	Ranking based on the drama's popularity
nb_watchers	Integer	Number of MyDramaList's users currently watching the drama
nb_ratings	Integer	Number of users that have rated the drama
nb_reviews	Integer	Number of reviews written by MyDramaList's users
streamed_on	List	List of platforms on which the drama is broadcaster
genres	List	List of genres associated to the drama
tags	List	List of tags associated to the drama
mydramalist_url	String	URL of the drama on MyDramaList
director	List	List of the drama's director(s)
screenwriter	List	List of the drama's screenwriter(s)
main_roles	List	List of the actors having a main role in the drama
support_roles	List	List of the actors having a support role in the drama
guest_roles	List	List of the actors having a guest role in the drama

User's drama list

Key	Type	Description
title	str	Title of the drama
user	str	User's username
score	int	Rating given by the user to the drama

Insert in MySQL

This feature is only available when using the spider designed to scrape information about dramas.

This scrapy project comes with a pipeline allowing to insert the results in a MySQL database.

How to enable the pipeline ?

The pipeline is enabled by default but won't insert anything unless it's asked to.

It does require a small extra loading time when initializing the spider so I do recommend disabling it by emptying the ITEM_PIPELINES in the settings.py file located in the Scrapy project. It should then look like this :

ITEM_PIPELINES = {}

Where to provide the credentials ?

The credentials of the database are currently provided to the pipeline as environmental variables through a hidden .env file which is loaded using the dotenv python package.

Here is an example of how the file looks like :

HOST=xxxxxxxxxxxxx
DB=xxxxxxxxxxxxx
USERNAME=xxxxxxxxxxxxx
PASSWORD=xxxxxxxxxxxxx

The name of the environmental can be changed within this file but don't forget also to change them in the db_connection method from the InsertItem pipeline.

For simplicity, you could also hardcode your credentials in the db_connection method but this is not recommended.

How to ask the pipeline to insert the records ?

Even though the pipeline is enabled by default, it won't insert your records unless you ask him for it by specifying an argument when launching the spider.

To do so, you could run something like :

scrapy crawl dramalist -a sql=True

Note: The value of the sql argument is case insensitive

What columns should I create in my DB ?

Currently, the insertion is based on hardcoded column names that are arbitrarily chosen.

They can be found in the query variable (which refers to the SQL query) within the process_item method in the InsertItem pipeline.

Feel free to name them as you wish and to process the data the way it suits to your needs. However, don't forget to assign the VARCHAR or JSON type to the columns that should store data of type list.

If changing the predetermined names of the columns, please ensure that the order is still corresponding to the order of the keys from the item returned by the scrapy spider.

The insertion is here based on values wrapped into a tuple. The order of the values provided in the tuple should be corresponding to the column name they are associated to.

For example, name is provided as the first column to be filled which implies that its value will be equal to the first element of the tuple that will be provided.

When changing the name of the columns and their order, keep in mind that the order of the values provided as a tuple should be changed correspondingly too.

Motivation

This project is the brick of another upcoming project. Indeed, we are motivated in scraping information on MyDramaList so that we can later create a drama recommandation system based on the user's taste in terms of drama.

We are mainly interested in retrieving information that could be relevant when creating such a system which explains why we decided not to retrieve some types of information from MyDramaList.

To do

• Add a requirements.txt file
• Add docstrings
• Implement some user arguments :
  • Maximum pages to scrape
  • Minimum rating
  • Date range
• Scrape information located in the statistics tab of the drama's page
• Creating a spider to retrieve information about the actors
• Implement a pipeline to allow insertion into MySQL databases

Contact

If there is any missing information that you would be interested in retrieving through the spiders, please send me an e-mail at jordan.vuong96@gmail.com.