Originally used to build this dataset on Kaggle, which has about 6 months worth of trending YouTube videos on it. This script will scrape the most relevant information from videos that are currently trending on YouTube in a specified set of countries. You can find example output files in the output directory.
Trending YouTube videos for whatever country you are in currently can be found here.
In order to use this script, you will need a valid API key for the YouTube Data API. It is free and the instructions for doing so are here. It is slightly awkward to get a key, but if you follow the instructions you should be ok.
Once you have the key, put it inside a text file named api_key.txt in the same directory as the script, or if it's not in the same directory you can target it with the --key_path parameter.
The only module needed that is not in the standard library is the requests module.
In order to run, the script needs country codes for the countries to collect trending videos from. These are 2 letter country abbreviations according to ISO 3166-1. A list of all existing ones can be found here, however not all of these are assured to work with the YouTube API. This project comes with a list of 10 inside the country_codes.txt file.
The script is fairly simple to run, it takes the following optional parameters:
--key_pathwhich takes a path argument that targets the text file containing your API key. By default this isapi_key.txtin the current directory.--country_code_pathwhich takes a path argument that targets the text file containing the list of country codes to target. By default this iscountry_codes.txtin the current directory.--output_dirwhich takes a path argument that specifies the folder to create the output CSV files for each country. By default this isoutput/in the current directory.
This project is licensed under the BSD 2-Clause License - see the LICENSE.md file for details