Benjamin_Loison/YouTube_Audio_library_extractor
1
0
Fork 0
Code Issues 8 Pull Requests Packages Projects Releases Wiki Activity

Make the media download work up to 100 results or duplicates

Browse Source 
The `MAXIMAL_NUMBER_OF_RESULTS` constant was computed thanks to:

```py
import os, json

path = '/home/benjamin/Desktop/bens_folder/dev/yt/audio_library'

os.chdir(path)

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

mostResults = 0
mostResultsTitle = None

for track in tracks:
    title = track['title']
    results = 0
    for otherTrack in tracks:
        if title in otherTrack['title']:
            results += 1
    if results > mostResults:
        mostResults = results
        mostResultsTitle = title

print(mostResults, mostResultsTitle)
```
This commit is contained in:
Benjamin Loison 2023-02-04 19:12:54 +01:00
parent 4f7e9ac336
commit 5540098e96
Signed by: Benjamin_Loison
SSH Key Fingerprint: SHA256:BtnEgYTlHdOg1u+RmYcDE0mnfz1rhv5dSbQ2gyxW8B8
1 changed files with 60 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

62
media_files_extractor.py
View File

@ -1,16 +1,18 @@
import undetected_chromedriver.v2 as uc
from selenium.webdriver.common.by import By
from selenium.webdriver.chrome.options import Options
import json
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('https://studio.youtube.com/channel/UC/music')
browser.get(AUDIO_LIBRARY_URL)
"""
For `Music` tab, YouTube UI returns 3,000 entries while my reverse-engineering approach returns 5,819 entries.
@ -23,15 +25,63 @@ As for `Sound effects`, even with `Sound effect`, `Duration`, `Category` and `Ad
with open('music.json') as json_file:
tracks = json.load(json_file)
for track in tracks:
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()
number_of_results = int(browser.find_element(By.CSS_SELECTOR, '.page-description').get_attribute('innerHTML').split()[-1])
print(number_of_results)
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`
browser.find_element(By.XPATH, 'div.overflow-actions:nth-child(12) > ytcp-button:nth-child(1) > div:nth-child(2)').click()
# 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()