Has YouTube unrightfully removed your beloved playlist you've been curating for the past 10 years? Don't worry, you are most likely able to recover it!
This project aims to provide a way to extract and save a list of videos from your removed playlist. It can do this because YouTube shows you all the videos from a removed playlist in the content review website.
What you get at the end is a human- and computer-readable file containing titles, links (and more) of videos on your playlist. For example, you can open the resulting csv file in Google Spreadsheets and it will even let you click links!
The goal of this project though is not uploading the recovered playlist back to your YouTube account. I can't recommend doing so as this might trigger YouTube's deletion evasion detection and get you in even more trouble...
However if you were to risk it, you have two choices:
- paid but convenient service like soundiiz or tunemymusic (a little birdie told me that soundiiz works quite well with the
--just-linksoption and is free for up to 200 videos...) - uploading using someone else's project (cumbersome, requires a Google cloud app and shuffling passcodes)
This software is provided under the MIT license and the author does not guarantee it works correctly. Full license
Now after you've read the important stuff, you can move on to recovering your playlist by following the ✨instructions✨.
The FAQ is here.
Python version 3.9 or later.
After cloning, execute make to install dependencies and set up scripts
and git hooks. Alternatively, execute pip3 install -r requirements.txt
.
python3 recover_playlist.py filename [-h]
[--format {csv,excel-csv,json}]
[--extractor {auto,v1}]
[-o OUTPUT_FILENAME]
[--extra-info]
[--just-links]
[--ids-only]
[--no-lxml]
[--write-info]
[--log {DEBUG,INFO,WARNING,ERROR,CRITICAL}]
positional arguments:
filename HTML file to be processed
options:
-h, --help show this help message and exit
--format {csv,excel-csv,json}
choose output format (default: csv)
--extractor {auto,v1}
choose extractor version (default: auto)
-o OUTPUT_FILENAME, --output-filename OUTPUT_FILENAME
output file name (default: None)
--extra-info store extra information available in the file (default: False)
--just_links only export links to videos (may be needed for some services) (default: False)
--ids_only store links as video ids (default: False)
--no-lxml don't use lxml (default: False)
--write-info write playlist info to a separate file (default: False)
--log {DEBUG,INFO,WARNING,ERROR,CRITICAL}
choose logging level (default: CRITICAL)
This project is just an overengineered version of what could have been two-pages worth of Python; and I'm sure there are better ways to write most of what I've used.
What we gain from all this though is easy extendability. YouTube updated their interface and the extractor no longer works? Write a new one, add it to the factory and you're set.
You can run tests by executing commands:
cd test
chmod +x ./text.sh
./test.sh
This script performs simple output tests for files defined in the testfiles directory.
You can run the script with --no-remove argument to stop it from removing files produced by the tests.
Issues and pull requests are very welcome, it would be awesome if this project could help people for as long as possible! Just please don't mind waiting a bit for a response, I can't guarantee that I'll always be available. :)