YouTube Comment Scraper

About

This is the web client for my Youtube Comment Scraper project. It is written in Node.js and uses the youtube-comment-api module to gain access to the comments. Given a YouTube video URL the client will request all comments for that video from the API. The results are displayed as nicely formatted JSON and CSV and can also be downloaded in those formats.

The results include the following information:

• Comment ID
• Username
• Date
• Timestamp
• Number of Likes
• Comment Text
• Replies

This fork saves the JSON result as "[video_title].comments([comment_count]).json"

Dependencies

A number of third party resources are required and need to be placed in the indicated directories.

Name	Directory
jQuery	static/libs/jquery.min.js
Font Awesome	static/font-awesome
Bootflat	static/bootflat
Papa Parse	static/libs/papaparse.min.js
json.human.js	static/libs/json.human.js, static/libs/json.human.css
download.js	static/libs/download.js

Deployment

A Dockerfile and deploy script are included. A running mongo docker instance is required to collect the stats.

First, create and start the mongo container.

$ docker pull mongo
$ docker run --name="ytcomments-mongo" --restart="always" -v /host/path/to/data/dir:/data/db -d mongo

Then run the deploy script.

$ ./deploy

The application will be listening for incoming connections on http://localhost:49161.

For subsequent deploys only run the deploy script, as long as the mongo container is still running.

Licensing

youtube-comment-scraper is licensed under ISC. See the included LICENSE file for details.

This project is in no way affiliated with YouTube.