From e844629839d17f7dc1a8758856ab6c401b2a7ebe Mon Sep 17 00:00:00 2001
From: Skullheadx <94652084+Skullheadx@users.noreply.github.com>
Date: Mon, 14 Oct 2024 00:37:25 -0400
Subject: [PATCH] fix file names with bad chars

---
 ytdl/__main__.py   | 15 +--------
 ytdl/funcmodule.py | 83 ++++++++++++----------------------------------
 2 files changed, 23 insertions(+), 75 deletions(-)

diff --git a/ytdl/__main__.py b/ytdl/__main__.py
index aa323d0..5f29931 100644
--- a/ytdl/__main__.py
+++ b/ytdl/__main__.py
@@ -1,5 +1,5 @@
 import sys
-from .funcmodule import check_playlist, get_audio_metadata_stream, download_audio_stream, get_and_download
+from .funcmodule import check_playlist, get_and_download
 import concurrent.futures
 
 
@@ -27,19 +27,6 @@ def main():
     links = check_playlist(links)
     assert len(links) > 0, "Should be at least one song in playlist"
 
-    audio_streams = []
-    metadata_list = []
-
-    for link in links:
-        stream, metadata = get_audio_metadata_stream(link)
-        assert stream is not None, "was not able to get audio stream"
-        assert metadata is not None, "no metadata found"
-        audio_streams.append(stream)
-        metadata_list.append(metadata)
-
-    assert len(audio_streams) > 0, "no audio streams found"
-    assert len(metadata_list) > 0, "no metadata found"
-
     if mode == "-d":
         pass
     elif mode == "-a":
diff --git a/ytdl/funcmodule.py b/ytdl/funcmodule.py
index 18c837e..124283c 100644
--- a/ytdl/funcmodule.py
+++ b/ytdl/funcmodule.py
@@ -2,6 +2,7 @@ from pytubefix import YouTube, Playlist
 import requests
 import subprocess
 import os
+import glob
 
 
 def check_playlist(links):
@@ -14,20 +15,6 @@ def check_playlist(links):
     return links
 
 
-def get_audio_metadata_stream(link):
-    yt = YouTube(link)
-    yt.check_availability()
-    print(f"Fetching stream for {yt.title}")
-    assert len(yt.streams.filter(only_audio=True)) > 0, "No available audio streams"
-    yield yt.streams.filter(only_audio=True).order_by("abr").last()
-    yield {
-            "title": yt.title,
-            "artist": yt.author,
-            "thumbnail_url": yt.thumbnail_url,
-            "publish_date": yt.publish_date,
-            "views": yt.views
-        }
-
 def big_num_format(num):  # https://stackoverflow.com/a/579376
     magnitude = 0
     while abs(num) >= 1000:
@@ -36,74 +23,48 @@ def big_num_format(num):  # https://stackoverflow.com/a/579376
     return '%.1f%s' % (num, ['', 'K', 'M', 'B'][magnitude])
 
 
-def download_audio_stream(audio_stream, metadata):
-    print(f"Downloading audio stream for {audio_stream.title}")
-    audio_stream.download()
-
-    # create thumbnail file
-    data = requests.get(metadata["thumbnail_url"]).content
-    thumbnail_filename = f'{audio_stream.title}.jpg'
-    with open(thumbnail_filename, 'wb') as f:
-        f.write(data)
-
-    command = [
-        'ffmpeg',
-        '-i', audio_stream.default_filename,
-        '-i', thumbnail_filename,
-        '-map', '0',
-        '-map', '1',
-        '-metadata', f'title={audio_stream.title}',
-        '-metadata', f'artist={metadata["artist"]}',
-        '-metadata', f'date={metadata["publish_date"]}',
-        '-metadata', f'comment={big_num_format(metadata["views"]) + " views"}',
-        audio_stream.title + ".mp4",
-        '-y'
-    ]
-    subprocess.run(command)
-
-    # clean up tmp files
-    os.remove(thumbnail_filename)
-    os.remove(audio_stream.default_filename)
+def fix_filename(filename):
+    for i in ['/', ':', '*', '?', '"', '<', '>', '|']:
+        filename = filename.replace(i, '')
+    return filename
 
 
 def get_and_download(link):
     yt = YouTube(link)
+    if fix_filename(yt.title) + '.mp4' in glob.glob("*.mp4"):
+        print(f"{yt.title} is already downloaded")
+        return
+
     yt.check_availability()
     print(f"Fetching stream 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()
-    metadata = {
-        "title": yt.title,
-        "artist": yt.author,
-        "thumbnail_url": yt.thumbnail_url,
-        "publish_date": yt.publish_date,
-        "views": yt.views
-    }
 
-    print(f"Downloading audio stream for {audio_stream.title}")
-    audio_stream.download()
+    print(f"Downloading audio stream for {yt.title}")
+    audio_stream.download(filename=fix_filename(audio_stream.default_filename), skip_existing=True)
 
     # create thumbnail file
-    data = requests.get(metadata["thumbnail_url"]).content
-    thumbnail_filename = f'{audio_stream.title}.jpg'
+    data = requests.get(yt.thumbnail_url).content
+    thumbnail_filename = f'{fix_filename(audio_stream.title)}.jpg'
     with open(thumbnail_filename, 'wb') as f:
         f.write(data)
 
     command = [
         'ffmpeg',
-        '-i', audio_stream.default_filename,
+        '-i', fix_filename(audio_stream.default_filename),
         '-i', thumbnail_filename,
         '-map', '0',
         '-map', '1',
-        '-metadata', f'title={audio_stream.title}',
-        '-metadata', f'artist={metadata["artist"]}',
-        '-metadata', f'date={metadata["publish_date"]}',
-        '-metadata', f'comment={big_num_format(metadata["views"]) + " views"}',
-        audio_stream.title + ".mp4",
-        '-y'
+        '-metadata', f'title={fix_filename(audio_stream.title)}',
+        '-metadata', f'artist={yt.author}',
+        '-metadata', f'date={yt.publish_date}',
+        '-metadata', f'comment={big_num_format(yt.views) + " views"}',
+        fix_filename(audio_stream.title) + ".mp4",
+        '-n'
     ]
     subprocess.run(command)
 
     # clean up tmp files
     os.remove(thumbnail_filename)
-    os.remove(audio_stream.default_filename)
\ No newline at end of file
+    os.remove(fix_filename(audio_stream.default_filename))
-- 
2.54.0