/.venv
.DS_Store
links.txt
+*.egg-info/
+*.pyc
+
+ytdl.egg-info/dependency_links.txt
+ytdl.egg-info/dependency_links.txt
+ytdl.egg-info/entry_points.txt
+ytdl.egg-info/SOURCES.txt
+ytdl.egg-info/top_level.txt
+ytdl/__main__.py
+ytdl/__pycache__/__init__.cpython-312.pyc
+ytdl/__pycache__/__main__.cpython-312.pyc
+ytdl/__pycache__/classmodule.cpython-312.pyc
+ytdl/__pycache__/funcmodule.cpython-312.pyc
+/ytdl.egg-info
+/ytdl/__pycache__
+*.mp4
-# youtube-downloader
+# Youtube Downloader - ytdl
-HOW TO USE
-- Enter links into the links_file.txt file each on a new line.
- - You can even put links to playlists too!
-- Run the main.py file using python
-- Enjoy your newly downloaded videos in the "downloaded" folder. (audio and video streams are available in the respective folders in downloaded)
+## usage
+```shell
+ytdl "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
+```
+downloads the audio and video and stitches it together in the current directory. Automatically detects playlists.
+
+- `-a` - audio only
+- `-v` - video only
+- `-av` - audio + video separate
+- `-f` - force replace if file exists
+
+# TODO:
+- [ ] figure out why -av takes so long compared to -a and -v
\ No newline at end of file
+++ /dev/null
-pytube
-ffmpeg-python
\ No newline at end of file
--- /dev/null
+pip install -e .
\ No newline at end of file
+++ /dev/null
-https://www.youtube.com/watch?app=desktop&v=lW4KseyDqcY
-https://www.youtube.com/watch?v=nJ7ZCN0m14A
-https://www.youtube.com/watch?app=desktop&v=VvWX3vRRLME
-https://www.youtube.com/watch?v=iCx2nBfu54I
-https://www.youtube.com/watch?app=desktop&v=V2hrTDS4Ml4
-https://www.youtube.com/watch?app=desktop&v=_Sd11FWbvZ8
-https://www.youtube.com/watch?v=HDOIQZZoABg
-https://www.youtube.com/watch?v=j-ttaqEzXKM
-https://www.youtube.com/watch?app=desktop&v=UMiW3G1USHg#dialog
-https://www.youtube.com/watch?v=uPG77Gtn0Ws
-https://www.youtube.com/watch?app=desktop&v=7cKGmeTY9eg
-https://www.youtube.com/watch?app=desktop&v=Pfs9yAsRbaU
-https://www.youtube.com/watch?v=WxKbU98GLRY
-https://www.youtube.com/watch?v=9MzCxt1QpWg
-https://www.youtube.com/watch?v=jCaug9SkKEI
-https://www.youtube.com/watch?v=V19v3oNPixQ
-https://www.youtube.com/watch?v=NscXXbmAggI
-https://www.youtube.com/watch?v=iCx2nBfu54I
-https://www.youtube.com/watch?v=9pq-G57iSEQ
-https://www.youtube.com/watch?v=XVRno3Y1TX8
-https://www.youtube.com/watch?v=codyY_-AiXc
-https://www.youtube.com/watch?v=qZ2yWvholHw
-https://www.youtube.com/watch?v=wodcSTIxZtw
-https://www.youtube.com/watch?v=btKTE4eMPf4
-https://www.youtube.com/watch?v=kW-WE5zrJsg
-https://www.youtube.com/watch?v=i5M-WHDhQ4c
+++ /dev/null
-from pytubefix import YouTube, Playlist
-from pytubefix.exceptions import VideoUnavailable
-import ffmpeg
-
-
-RES = ["1440p", "1080p", "720p", "480p", "360p", "240p", "144p"]
-ABR = ["160kbps", "128kbps", "70kbps", "50kbps", "48kbps"]
-target_res = 0
-target_abr=0
-
-failed_download = set()
-
-if __name__ == "__main__":
-
- # get list of links from file
- links = []
- with open('links.txt', 'r') as f:
- links = f.read().split('\n')
- if links[-1] == "":
- links = links[:-1]
-
- for link in links:
- if "playlist" in link:
- p = Playlist(link)
- for url in p.video_urls:
- links.append(url)
- links.remove(link)
-
-
- # download links one by one
- for link in links:
- target_res = 0
- target_abr = 0
- video_success = True
- audio_success = True
-
- try:
- yt = YouTube(link)
- yt.streams
-
- except VideoUnavailable:
- print(f'Video {link} is unavaialable, skipping.')
- failed_download.add(((yt.title, link)))
- else:
- video_streams = []
- while len(video_streams) == 0:
- video_streams = yt.streams.filter(file_extension='mp4', res=RES[target_res]) # find available streams
- if target_res + 1 < len(RES):
- target_res = target_res + 1
- else:
- video_success = False
- break
- if not video_success:
- print(f"Unable to find video stream for {yt.title}")
- failed_download.add(((yt.title, link)))
-
- break
- vstream = video_streams[0]
-
- # audio
- audio_streams = []
- while len(audio_streams) == 0:
- audio_streams = yt.streams.filter(only_audio=True, abr=ABR[target_abr]) # find available streams
-
- if target_abr + 1 < len(RES):
- target_abr = target_abr + 1
- else:
- audio_success = False
- break
- if not audio_success:
- print(f"Unable to find audio stream for {yt.title}")
- failed_download.add(((yt.title, link)))
-
- break
- astream = audio_streams[0]
-
- vstream.download(output_path="downloaded/video_only")
- astream.download(output_path="downloaded/audio_only")
-
- input_video = ffmpeg.input(f"downloaded/video_only/{vstream.default_filename}")
- input_audio = ffmpeg.input(f"downloaded/audio_only/{astream.default_filename}")
-
- ffmpeg.concat(input_video, input_audio, v=1, a=1).output(f'downloaded/{yt.title}.mp4').run()
-
-print("Failed Downloading:")
-print(failed_download)
-
--- /dev/null
+pytubefix~=8.0.0
+requests~=2.32.3
+ffmpeg
\ No newline at end of file
--- /dev/null
+from setuptools import setup
+setup(
+ name = 'ytdl',
+ version = '0.1.0',
+ packages = ['ytdl'],
+ entry_points = {
+ 'console_scripts': [
+ 'ytdl = ytdl.__main__:main'
+ ]
+ })
\ No newline at end of file
--- /dev/null
+Metadata-Version: 2.1
+Name: ytdl
+Version: 0.1.0
+License-File: LICENSE
--- /dev/null
+LICENSE
+README.md
+setup.py
+ytdl/__init__.py
+ytdl/__main__.py
+ytdl/funcmodule.py
+ytdl.egg-info/PKG-INFO
+ytdl.egg-info/SOURCES.txt
+ytdl.egg-info/dependency_links.txt
+ytdl.egg-info/entry_points.txt
+ytdl.egg-info/top_level.txt
\ No newline at end of file
--- /dev/null
+[console_scripts]
+ytdl = ytdl.__main__:main
--- /dev/null
+import sys
+from .funcmodule import check_playlist, download
+import concurrent.futures
+
+
+def main():
+ args = sys.argv[1:]
+ modes = ["-d", "-a", "-v", "-av"]
+
+ links = []
+ mode = "-d"
+ force = False
+ assert len(args) > 0, "no args :("
+ for arg in args:
+ if arg in modes:
+ mode = arg
+ if arg == '-f': # force
+ force = True
+ if "youtube" in arg or "youtu.be" in arg:
+ links.extend(arg.split(" "))
+
+ assert len(links) > 0, "Should pass at least one link as arg"
+ assert mode in modes, f"Mode should be one of {modes}"
+ print("Processing links")
+ # remove empty strings
+ links = list(filter(None, links))
+ assert len(links) > 0, "Should not remove all links"
+ print("Checking for playlists")
+ links = check_playlist(links)
+ assert len(links) > 0, "Should be at least one song in playlist"
+
+ # Use ThreadPoolExecutor to run downloads concurrently
+ with concurrent.futures.ThreadPoolExecutor() as executor:
+ # Schedule the download_audio_stream function for each audio stream
+ futures = {executor.submit(download, link, mode, force): link for link in links}
+
+
+if __name__ == '__main__':
+ main()
--- /dev/null
+import glob
+import os
+import subprocess
+
+import requests
+from pytubefix import YouTube, Playlist
+
+
+def check_playlist(links):
+ for link in links:
+ if "playlist" in link:
+ p = Playlist(link)
+ for url in p.video_urls:
+ links.append(url)
+ links.remove(link)
+ return links
+
+
+def big_num_format(num): # https://stackoverflow.com/a/579376
+ magnitude = 0
+ while abs(num) >= 1000:
+ magnitude += 1
+ num /= 1000.0
+ return '%.1f%s' % (num, ['', 'K', 'M', 'B'][magnitude])
+
+
+def fix_filename(filename):
+ for i in ['/', ':', '*', '?', '"', '<', '>', '|']:
+ filename = filename.replace(i, '')
+ return filename
+
+
+def download_thumbnail(thumbnail_url, thumbnail_filename):
+ data = requests.get(thumbnail_url).content
+ with open(thumbnail_filename, 'wb') as f:
+ f.write(data)
+
+
+def download(link, mode, force=False):
+ yt = YouTube(link)
+ filename = fix_filename(yt.title)
+ if (
+ (
+ (mode == '-av' and
+ (filename + ' (audio only).mp4' in glob.glob("*.mp4") or
+ filename + ' (video only).mp4' in glob.glob("*.mp4"))) or
+ (mode != '-av' and filename + '.mp4' in glob.glob("*.mp4"))
+ ) and
+ (not force)
+ ):
+ print(f"{yt.title} is already downloaded")
+ return
+ yt.check_availability()
+
+ thumbnail_filename = f'{filename}.jpg'
+ download_thumbnail(yt.thumbnail_url, thumbnail_filename)
+
+ if mode == '-a' or mode == '-v':
+ download_single_stream(yt, filename, thumbnail_filename, mode)
+ elif mode == '-av' or mode == '-d':
+ download_double_stream(yt, filename, thumbnail_filename, mode)
+
+
+def convert_add_metadata(input1, input2, output, yt, m1=1, m2=0):
+ album = yt.title
+ if 'keywords' in yt.vid_info['videoDetails'].keys() and len(yt.vid_info['videoDetails']['keywords']) > 2:
+ album = yt.vid_info['videoDetails']['keywords'][-2]
+ command = [
+ 'ffmpeg',
+ '-i', input1,
+ '-i', input2,
+ '-map', f'{m1}',
+ '-map', f'{m2}',
+ '-c', 'copy',
+ f'-disposition:v:{m2}', 'attached_pic',
+ '-metadata', f'title={yt.title}',
+ '-metadata', f'artist={yt.author}',
+ '-metadata', f'comment={big_num_format(yt.views) + " views"}',
+ '-metadata', f'date={yt.publish_date}',
+ '-metadata', f'album={album}',
+ output + ".mp4",
+ '-y'
+ ]
+ subprocess.run(command)
+
+
+def download_single_stream(yt, filename, thumbnail_filename, mode):
+ print(f"Fetching stream for {yt.title}")
+ stream = None
+ if mode == "-a":
+ assert len(yt.streams.filter(only_audio=True)) > 0, "No available audio streams"
+ stream = yt.streams.filter(only_audio=True).order_by("abr").last()
+ if mode == "-v":
+ assert len(yt.streams.filter(only_video=True)) > 0, "No available video streams"
+ stream = yt.streams.filter(only_video=True).order_by("resolution").last()
+
+ assert stream is not None, "mode is not valid"
+ print(f"Downloading stream for {yt.title}")
+ default_filename = "default " + fix_filename(stream.default_filename)
+ stream.download(filename=default_filename, skip_existing=True)
+
+ print(f"Adding metadata to {yt.title}")
+ convert_add_metadata(default_filename, thumbnail_filename, filename, yt)
+
+ print("Removing temporary files")
+ os.remove(thumbnail_filename)
+ os.remove(default_filename)
+
+
+def download_double_stream(yt, filename, thumbnail_filename, mode):
+ print(f"Fetching streams for {yt.title}")
+ assert len(yt.streams.filter(only_audio=True)) > 0, "No available audio streams"
+ audio_stream = yt.streams.filter(only_audio=True).order_by("abr").last()
+ assert len(yt.streams.filter(only_video=True)) > 0, "No available video streams"
+ video_stream = yt.streams.filter(only_video=True).order_by("resolution").last()
+
+ print(f"Downloading streams for {yt.title}")
+ audio_default_filename = "default audio " + fix_filename(audio_stream.default_filename)
+ video_default_filename = "default video " + fix_filename(video_stream.default_filename)
+
+ audio_stream.download(filename=audio_default_filename, skip_existing=True)
+ video_stream.download(filename=video_default_filename, skip_existing=True)
+
+ print(f"Adding metadata to {yt.title}")
+ if mode == '-av':
+ for suffix, stream in [(" (audio only)", audio_default_filename), (" (video only)", video_default_filename)]:
+ convert_add_metadata(stream, thumbnail_filename, filename + suffix, yt)
+ elif mode == '-d':
+ convert_add_metadata(audio_default_filename, video_default_filename, filename + "tmp", yt, m1=0, m2=1)
+ convert_add_metadata(filename + "tmp.mp4", thumbnail_filename, filename, yt)
+
+ print("Removing temporary files")
+ os.remove(thumbnail_filename)
+ os.remove(audio_default_filename)
+ os.remove(video_default_filename)
+ if mode == '-d':
+ os.remove(filename + "tmp" + ".mp4")
projects / youtube-downloader.git / commitdiff