YTMASC(YouTube Music Audio Scraper & synChronizer) in a nutshell, aims to get your music library off of YouTube and provide you an offline backup of it along with other maintenance niceties.
Grab the latest alpha version here!
It's features are:
- Automatic downloading, converting and tagging
- CLI (shipped as a binary for Windows!)
- Import a CSV of your own (columns are: *
watch_id,artist,title) - Importing favorites from a RiMusic database
- Maintaining a data file for your music for an easily reproducible collection
- Scraping your library page from YouTube
- Some helper functions to modify your data file easier and for easy migration
The project just keeps expanding as I learn more stuff and want to implement niche things. So it's currently in alpha stages, you may see new features come and go every now and then.
Show the help:
ytmasc|ytmasc -h
Run a singular file operation using the fire interface:
ytmasc --fire download | convert | tag <watch_id>
Import (Changes in file tags, CSV, RiMusic database export):
ytmasc --import-from files | path/to/your.csv | path/to/your.db --import-level soft
Import library page
ytmasc --import-library-page fetch-soft | fetch-soft-no-overwrite | no-fetch-softno-fetch-soft-no-overwrite
Export the library as CSV to the path you specify
ytmasc --export-to path-of-your-choosing
Tip
You can chain all these commands in a meaningful manner, see the flowchart.
ytmasc --import-library-page fetch-soft --delete-library-page-files-afterwards --refetch-metadata --import-from rimusic_xxxxxxxxxxxxxx.db --import-from data/x.csv --download --convert --tag --lib-find-same
- You need
ffmpegbinaries for conversion. - YouTube blocks API requests if you exceed the amount they classify you as a bot (around 200 requests). You can either use a VPN, proxy or just wait to bypass this. See related
yt-dlpissue. Currently the download loop breaks and skips to the next task. - While downloading, some changes may occur in YouTube which results in an error.
You can use
ytmasc --db-replace-failsto replace thesewatch_ids to something else.
This part is a little duct taped, I couldn't find a good way to get the libraryPage formerly known as likesPage so I just emulated user input. It's written for a Windows computer that has firefox or zen as the default browser and file explorer. Shouldn't be hard to tinker and get it to work for your configuration. You can do this manually too, shouldn't take much of your time.
Change resendAmount based on your internet connection, page length. Rule of thumb is to divide your like amount by 10 for this.
Change openingDelay based on your internet connection.
Change savePageAsIndexOnRightClick to which index your save as is on your browser.
The rest is fine if you don't have a really old computer.
~=python3.11python3-tk(Linux)ffmpeg
In the monstrosity of a flowchart below, you can find the chaining logic for all the arguments. This is in an effort for the user to run seperate commands and do the whole shebang in one fell swoop. Strange results may occur, it's completely up to the user's usage.
graph TD;
convert-->Library_tools
convert-->tag
delete-library-page-files-afterwards--->refetch-metadata
delete-library-page-files-afterwards-->export
delete-library-page-files-afterwards-->Tasks
download-->convert
download-->Library_tools
download-->tag
export-->Tasks
imports-->delete-library-page-files-afterwards
imports-->export
imports-->refetch-metadata
imports-->Tasks
lib-find-same-->lib-compare
lib-find-same-->lib-find-unpaired
lib-find-same-->lib-replace-fails
lib-find-unpaired-->lib-compare
lib-find-unpaired-->lib-replace-fails
lib-replace-fails-->lib-compare
Library_operations-->export
Library_operations-->imports
Library_operations-->Library_tools
Library_operations-->refetch-metadata
Library_operations-->Tasks
Library_tools-->lib-compare-->FINISH
Library_tools-->lib-find-same-->FINISH
Library_tools-->lib-find-unpaired-->FINISH
Library_tools-->lib-replace-fails-->FINISH
refetch-metadata-->export
refetch-metadata-->Tasks
START-->help-->FINISH
START-->fire-->FINISH
START-->Library_operations
START-->verbosity
tag-->Library_tools
Tasks-->convert
Tasks-->download
Tasks-->Library_tools
Tasks-->tag
verbosity-->Library_operations
Python ~=3.11 - Core language. Licensed under PSFL license.
FFmpeg - Required for the conversion of files. Licensed under LGPLv2.1 license.
beautifulsoup4 - Used in parsing the user's likes page HTML. Licensed under MIT license.
click - Used in CLI, Licensed under BSD-3-Clause license.
eyed3 - Used in tagging the files. Licensed under GPL-3.0 license.
fire - Used in exposing some base functions for a basic CLI. Licensed under Apache-2.0 license.
ffmpeg-python - Used in converting files to desired format(s) as a wrapper. Licensed under Apache-2.0 license.
mutagen* - Subdependency, planned to replace eyed3. Licensed under GPL-2.0 license.
pandas - Used in CSV/JSON helper utilities. Licensed under BSD-3-Clause license.
pillow - Used in modification of covers. Licensed under MIT-CMU license.
pyautogui - Used in fetcher to get the user's likes page. Licensed under BSD-3-Clause license.
pyinstaller - Used in compilation. Licensed under a GPLv2 license.
yt-dlp - Used in downloading user's library. Licensed under Unlicense license.
ytmusicapi - Used in metadata fetching. Licensed under MIT license.
fuzzywuzzy - Used in comparison utilities for migration for fuzzy matching. Licensed under GPLv2 license.
python-levenshtein* - Subdependency for fuzzywuzzy. Licensed under GPL-2.0 license.
keyboard - Used in comparison utilities for migration to handle keypresses. Licensed under MIT license.
prettytable - Used in comparison utilities as a part of UX. Licensed under a custom license.
This project is not in any way, shape or form affiliated with YouTube, Google or any of their subsidiaries and affiliates.