youcos 0.0.2

youcos is a simple Python package for scraping YouTube videos and comments

youcos

youcos (youtube comment scraper) is a simple Python package for scraping YouTube comments!

:four_leaf_clover: Easy YouTube v3 API Wrapper - Simply provide your YouTube v3 API Key, and youcos will take care of the rest! No additional code needed to configure API and process fetched JSON data.

:earth_asia: Diverse Language Support - Search, scrape, and save videos & comments in different languges with youcos.

:test_tube: Configurable Search Settings - Adjust search location and search method.

:chart_with_upwards_trend: Minimized Quota Usage - Scrape comments without exhausting YouTube v3 quota through the built-in Selenium module with youcos!

Demo

There are two main ways to scrape comments:

1. Scraping video titles and comments together

from youcos import scrape_youtube

KEY = 'YOUR_YOUTUBE_V3_API_KEY'

# Call YouTube v3 API to request the first 30 videos search results for 'stocks'
# and scrape each video's top __ comments
scrape_youtube('stocks', KEY)

2. Scraping video titles and comments separately

from youcos import scrape_videos, scrape_comments

KEY = 'YOUR_YOUTUBE_V3_API_KEY'

# Call YouTube v3 API to request the first 30 videos search results for 'stocks'
videos = scrape_videos("stocks", KEY)

# Filter videos to avoid re-scraping comments
filtered_videos = foo(videos)

# Scrape comments for filtered videos
scrape_comments(filtered_videos)

def foo(videos):
    # function to filter videos

Installation

pip install youcos

Features

The following data are saved into the csv file:

Video Data	Comment Data
URL	text
title	author
channel name	date
upload date	no. of replies
no. of likes	no. of upvotes
no. of dislikes	-

Dependencies

• Selenium

Documentation

Check ___ for specific API documentation. This project was documented following the numpy docstring conventions, which are supported by common documentation tools like Sphynx while also maintaining readability.

Credits

• Author: Seunggun Lee
• Languages/Tools: Python3, Selenium

To Do

Functionalities

• search based on different filters
• selenium dependency support for all drivers
• choose to filter comments based on relevancy & top comments
• headless browser scraping (option)
• maximum number of videos to scrape
• maximum number of comments to scrape
• method to skip video authentication

Deployment

• use Sphynx to build documentation