YouTube_Audio_library_extractor/media_files_extractor.py

import undetected_chromedriver.v2 as uc
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
import os, json, time

"""
As there is something looking as an anti-bot for downloading media files, we use a Selenium-based approach.
"""

AUDIO_LIBRARY_URL = 'https://studio.youtube.com/channel/UC/music'

options = Options()
options.add_argument("--user-data-dir=selenium")
browser = uc.Chrome(options=options)
browser.get(AUDIO_LIBRARY_URL)

"""
For `Music` tab, YouTube UI returns 3,000 entries while my reverse-engineering approach returns 5,819 entries.
For `Sound effects` tab, YouTube UI returns 400 entries while my reverse-engineering approach returns 2021 entries.

So I assume YouTube UI pagination doesn't work fine, so to retrieve all media files, the idea is to filter by `Track title` and download all entries, preferably only those that have the title we are looking for, as some tracks have the same titles.
As for `Sound effects`, even with `Sound effect`, `Duration`, `Category` and `Added` there is an ambiguity on which files do we refer to (for instance for `Truck Driving in Parking Structure`, as they all are different).
"""

with open('music.json') as json_file:
    tracks = json.load(json_file)

path = '/home/benjamin/Downloads'

os.chdir(path)

MAXIMAL_NUMBER_OF_RESULTS = 162

with open('rename.txt', 'w') as f:
    browser.find_element(By.XPATH, '//*[@id="trigger"]/ytcp-dropdown-trigger/div/div[2]/span').click()
    browser.find_element(By.CSS_SELECTOR, '#text-item-2 > ytcp-ve > div > div > yt-formatted-string').click()
    tracks = tracks[38:]
    for trackIndex, track in enumerate(tracks):
        seconds = track["duration"]["seconds"]
        cleanDuration = f'{seconds // 60}:{seconds % 60}'
        print(f'{trackIndex} / {len(tracks)}: {track["title"]} - {track["artist"]["name"]} - {cleanDuration}')
        browser.find_element(By.ID, 'text-input').send_keys(track['title'])
        browser.find_element(By.XPATH, '/html/body/ytcp-text-menu/tp-yt-paper-dialog/tp-yt-paper-listbox/tp-yt-paper-item[2]/ytcp-ve/div/div/yt-formatted-string/span[1]').click()

        numberOfResults = MAXIMAL_NUMBER_OF_RESULTS + 1
        while numberOfResults > MAXIMAL_NUMBER_OF_RESULTS:
            if numberOfResults != MAXIMAL_NUMBER_OF_RESULTS + 1:
                time.sleep(1)
            numberOfResults = int(browser.find_element(By.CSS_SELECTOR, '.page-description').get_attribute('innerHTML').split()[-1])
            print(f'Found {numberOfResults} results')

        # `DOWNLOAD`
        # Doesn't block.
        if numberOfResults > 100:
            print('More than 100 results')
            break

        if numberOfResults > 1:
            print('found multiple')
            occurrences = 0
            for resultsIndex in range(numberOfResults):
                row = f'/html/body/ytcp-app/ytcp-entity-page/div/div/main/div/ytcp-animatable[24]/ytmus-page/ytmus-library-table/div[1]/ytmus-library-row[{resultsIndex + 1}]/div/'
                title = browser.find_element(By.XPATH, row + 'div[2]/div').get_attribute('innerHTML')
                artistCommon = 'div[5]/div'
                try:
                    artist = browser.find_element(By.XPATH, row + artistCommon + '/ytcp-hover-anchor/a/span')
                except:
                    artist = browser.find_element(By.XPATH, row + artistCommon)
                artist = artist.get_attribute('innerHTML')
                print(title, artist)
                if title == track['title'] and artist == track['artist']['name']:
                    print("it's a match")
                    browser.find_element(By.XPATH, row + 'div[8]/div[2]/ytcp-button/div').click()
                    occurrences += 1
            if occurrences > 1:
                print('multiple occurrences')
                break
            #break
            #browser.get('chrome://downloads/')
            #browser.get(AUDIO_LIBRARY_URL)
        else:
            browser.find_element(By.CSS_SELECTOR, 'ytcp-button.style-scope:nth-child(1) > div:nth-child(2)').click()
        browser.find_element(By.CSS_SELECTOR, '#delete-icon > tp-yt-iron-icon').click()
        f.write(f"{track['title']} - {track['artist']['name']}|{track['viperId']}\n")
        #break

#browser.quit()