Update version v0.3.2HEADmaster

1 files changed, 68 insertions, 66 deletions

diff --git a/youtube/watch.py b/youtube/watch.py
index 2cfece5..0274cd0 100644
--- a/youtube/watch.py
+++ b/youtube/watch.py
@@ -2,7 +2,6 @@ import youtube
 from youtube import yt_app
 from youtube import util, comments, local_playlist, yt_data_extract
 from youtube.util import time_utc_isoformat
-from youtube.util import INNERTUBE_CLIENTS
 import settings
 
 from flask import request
@@ -344,7 +343,6 @@ def _add_to_error(info, key, additional_message):
 def fetch_player_response(client, video_id):
     return util.call_youtube_api(client, 'player', {
         'videoId': video_id,
-        'params': 'CgIIAdgDAQ==',
     })
 
 
@@ -369,84 +367,94 @@ def fetch_watch_page_info(video_id, playlist_id, index):
     watch_page = watch_page.decode('utf-8')
     return yt_data_extract.extract_watch_info_from_html(watch_page)
 
+
 def extract_info(video_id, use_invidious, playlist_id=None, index=None):
-    for client in INNERTUBE_CLIENTS:
-        tasks = (
-            gevent.spawn(fetch_watch_page_info, video_id, playlist_id, index),
-            gevent.spawn(fetch_player_response, client, video_id)  # Use client from INNERTUBE_CLIENTS
-        )
-        gevent.joinall(tasks)
-        util.check_gevent_exceptions(*tasks)
-        info, player_response = tasks[0].value, tasks[1].value
+    primary_client = 'android_vr'
+    fallback_client = 'ios'
+    last_resort_client = 'tv_embedded'
+
+    tasks = (
+        # Get video metadata from here
+        gevent.spawn(fetch_watch_page_info, video_id, playlist_id, index),
+        gevent.spawn(fetch_player_response, primary_client, video_id)
+    )
+    gevent.joinall(tasks)
+    util.check_gevent_exceptions(*tasks)
 
+    info = tasks[0].value or {}
+    player_response = tasks[1].value or {}
+
+    yt_data_extract.update_with_new_urls(info, player_response)
+
+    # Fallback to 'ios' if no valid URLs are found
+    if not info.get('formats') or info.get('player_urls_missing'):
+        print(f"No URLs found in '{primary_client}', attempting with '{fallback_client}'.")
+        player_response = fetch_player_response(fallback_client, video_id) or {}
         yt_data_extract.update_with_new_urls(info, player_response)
 
-        # Age restricted video, retry
-        if info['age_restricted'] or info['player_urls_missing']:
-            if info['age_restricted']:
-                print('Age restricted video, retrying')
-            else:
-                print('Player urls missing, retrying')
-            player_response = fetch_player_response('tv_embedded', video_id)
-            yt_data_extract.update_with_new_urls(info, player_response)
+    # Final attempt with 'tv_embedded' if there are still no URLs
+    if not info.get('formats') or info.get('player_urls_missing'):
+        print(f"No URLs found in '{fallback_client}', attempting with '{last_resort_client}'")
+        player_response = fetch_player_response(last_resort_client, video_id) or {}
+        yt_data_extract.update_with_new_urls(info, player_response)
 
-        # signature decryption
+    # signature decryption
+    if info.get('formats'):
         decryption_error = decrypt_signatures(info, video_id)
         if decryption_error:
-            decryption_error = 'Error decrypting url signatures: ' + decryption_error
-            info['playability_error'] = decryption_error
+            info['playability_error'] = 'Error decrypting url signatures: ' + decryption_error
 
-        # check if urls ready (non-live format) in former livestream
-        # urls not ready if all of them have no filesize
-        if info['was_live']:
-            info['urls_ready'] = False
-            for fmt in info['formats']:
-                if fmt['file_size'] is not None:
-                    info['urls_ready'] = True
-        else:
-            info['urls_ready'] = True
+    # check if urls ready (non-live format) in former livestream
+    # urls not ready if all of them have no filesize
+    if info['was_live']:
+        info['urls_ready'] = False
+        for fmt in info['formats']:
+            if fmt['file_size'] is not None:
+                info['urls_ready'] = True
+    else:
+        info['urls_ready'] = True
 
-        # livestream urls
-        # sometimes only the livestream urls work soon after the livestream is over
-        if (info['hls_manifest_url']
-            and (info['live'] or not info['formats'] or not info['urls_ready'])
-        ):
+    # livestream urls
+    # sometimes only the livestream urls work soon after the livestream is over
+    info['hls_formats'] = []
+    if info.get('hls_manifest_url') and (info.get('live') or not info.get('formats') or not info['urls_ready']):
+        try:
             manifest = util.fetch_url(info['hls_manifest_url'],
                 debug_name='hls_manifest.m3u8',
                 report_text='Fetched hls manifest'
             ).decode('utf-8')
-
             info['hls_formats'], err = yt_data_extract.extract_hls_formats(manifest)
             if not err:
                 info['playability_error'] = None
             for fmt in info['hls_formats']:
                 fmt['video_quality'] = video_quality_string(fmt)
-        else:
+        except Exception as e:
+            print(f"Error obteniendo HLS manifest: {e}")
             info['hls_formats'] = []
 
-        # check for 403. Unnecessary for tor video routing b/c ip address is same
-        info['invidious_used'] = False
-        info['invidious_reload_button'] = False
-        info['tor_bypass_used'] = False
-        if (settings.route_tor == 1
-                and info['formats'] and info['formats'][0]['url']):
-            try:
-                response = util.head(info['formats'][0]['url'],
-                                     report_text='Checked for URL access')
-            except urllib3.exceptions.HTTPError:
-                print('Error while checking for URL access:\n')
-                traceback.print_exc()
-                return info
-
-            if response.status == 403:
-                print('Access denied (403) for video urls.')
-                print('Routing video through Tor')
-                info['tor_bypass_used'] = True
-                for fmt in info['formats']:
-                    fmt['url'] += '&use_tor=1'
-            elif 300 <= response.status < 400:
-                print('Error: exceeded max redirects while checking video URL')
-        return info
+    # check for 403. Unnecessary for tor video routing b/c ip address is same
+    info['invidious_used'] = False
+    info['invidious_reload_button'] = False
+    info['tor_bypass_used'] = False
+    if (settings.route_tor == 1
+            and info['formats'] and info['formats'][0]['url']):
+        try:
+            response = util.head(info['formats'][0]['url'],
+                                 report_text='Checked for URL access')
+        except urllib3.exceptions.HTTPError:
+            print('Error while checking for URL access:\n')
+            traceback.print_exc()
+            return info
+
+        if response.status == 403:
+            print('Access denied (403) for video urls.')
+            print('Routing video through Tor')
+            info['tor_bypass_used'] = True
+            for fmt in info['formats']:
+                fmt['url'] += '&use_tor=1'
+        elif 300 <= response.status < 400:
+            print('Error: exceeded max redirects while checking video URL')
+    return info
 
 
 def video_quality_string(format):
@@ -651,12 +659,6 @@ def get_watch_page(video_id=None):
             '/videoplayback',
             '/videoplayback/name/' + filename)
 
-    if settings.gather_googlevideo_domains:
-        with open(os.path.join(settings.data_dir, 'googlevideo-domains.txt'), 'a+', encoding='utf-8') as f:
-            url = info['formats'][0]['url']
-            subdomain = url[0:url.find(".googlevideo.com")]
-            f.write(subdomain + "\n")
-
     download_formats = []
 
     for format in (info['formats'] + info['hls_formats']):