fix: use YouTube-provided thumbnail URLs instead of hardcoded hq720.jpg
All checks were successful
git-sync-with-mirror / git-sync (push) Successful in 15s
CI / test (push) Successful in 58s

Videos without hq720.jpg thumbnails caused mass 404 errors.
Now preserves the actual thumbnail URL from YouTube's API response,
falls back to hqdefault.jpg only when no thumbnail is provided.
Also picks highest quality thumbnail from API (thumbnails[-1])
and adds progressive fallback for subscription/download functions.
This commit is contained in:
2026-03-27 19:22:12 -05:00
parent f629565e77
commit 56ecd6cb1b
10 changed files with 81 additions and 61 deletions

View File

@@ -99,7 +99,6 @@ def proxy_site(env, start_response, video=False):
if response.status >= 400: if response.status >= 400:
print('Error: YouTube returned "%d %s" while routing %s' % ( print('Error: YouTube returned "%d %s" while routing %s' % (
response.status, response.reason, url.split('?')[0])) response.status, response.reason, url.split('?')[0]))
total_received = 0 total_received = 0
retry = False retry = False
while True: while True:

View File

@@ -406,12 +406,12 @@ def post_process_channel_info(info):
info['avatar'] = util.prefix_url(info['avatar']) info['avatar'] = util.prefix_url(info['avatar'])
info['channel_url'] = util.prefix_url(info['channel_url']) info['channel_url'] = util.prefix_url(info['channel_url'])
for item in info['items']: for item in info['items']:
# For playlists, use first_video_id for thumbnail, not playlist id # Only set thumbnail if YouTube didn't provide one
if not item.get('thumbnail'):
if item.get('type') == 'playlist' and item.get('first_video_id'): if item.get('type') == 'playlist' and item.get('first_video_id'):
item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['first_video_id']) item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['first_video_id'])
elif item.get('type') == 'video': elif item.get('type') == 'video' and item.get('id'):
item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id']) item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['id'])
# For channels and other types, keep existing thumbnail
util.prefix_urls(item) util.prefix_urls(item)
util.add_extra_html_info(item) util.add_extra_html_info(item)
if info['current_tab'] == 'about': if info['current_tab'] == 'about':

View File

@@ -150,7 +150,7 @@ def post_process_comments_info(comments_info):
util.URL_ORIGIN, '/watch?v=', comments_info['video_id']) util.URL_ORIGIN, '/watch?v=', comments_info['video_id'])
comments_info['video_thumbnail'] = concat_or_none( comments_info['video_thumbnail'] = concat_or_none(
settings.img_prefix, 'https://i.ytimg.com/vi/', settings.img_prefix, 'https://i.ytimg.com/vi/',
comments_info['video_id'], '/hq720.jpg' comments_info['video_id'], '/hqdefault.jpg'
) )

View File

@@ -106,8 +106,8 @@ def get_playlist_page():
for item in info.get('items', ()): for item in info.get('items', ()):
util.prefix_urls(item) util.prefix_urls(item)
util.add_extra_html_info(item) util.add_extra_html_info(item)
if 'id' in item: if 'id' in item and not item.get('thumbnail'):
item['thumbnail'] = f"{settings.img_prefix}https://i.ytimg.com/vi/{item['id']}/hq720.jpg" item['thumbnail'] = f"{settings.img_prefix}https://i.ytimg.com/vi/{item['id']}/hqdefault.jpg"
item['url'] += '&list=' + playlist_id item['url'] += '&list=' + playlist_id
if item['index']: if item['index']:

View File

@@ -121,11 +121,12 @@ window.addEventListener('DOMContentLoaded', function() {
* Priority: hq720.jpg -> sddefault.jpg -> hqdefault.jpg -> mqdefault.jpg -> default.jpg * Priority: hq720.jpg -> sddefault.jpg -> hqdefault.jpg -> mqdefault.jpg -> default.jpg
*/ */
function thumbnail_fallback(img) { function thumbnail_fallback(img) {
const src = img.src || img.dataset.src; // Once src is set (image was loaded or attempted), always work with src
const src = img.src;
if (!src) return; if (!src) return;
// Handle YouTube video thumbnails // Handle YouTube video thumbnails
if (src.includes('/i.ytimg.com/')) { if (src.includes('/i.ytimg.com/') || src.includes('/i.ytimg.com%2F')) {
// Extract video ID from URL // Extract video ID from URL
const match = src.match(/\/vi\/([^/]+)/); const match = src.match(/\/vi\/([^/]+)/);
if (!match) return; if (!match) return;
@@ -138,36 +139,32 @@ function thumbnail_fallback(img) {
'hq720.jpg', 'hq720.jpg',
'sddefault.jpg', 'sddefault.jpg',
'hqdefault.jpg', 'hqdefault.jpg',
'mqdefault.jpg',
'default.jpg'
]; ];
// Find current quality and try next fallback // Find current quality and try next fallback
for (let i = 0; i < fallbacks.length; i++) { for (let i = 0; i < fallbacks.length; i++) {
if (src.includes(fallbacks[i])) { if (src.includes(fallbacks[i])) {
// Try next quality
if (i < fallbacks.length - 1) { if (i < fallbacks.length - 1) {
const newSrc = imgPrefix + 'https://i.ytimg.com/vi/' + videoId + '/' + fallbacks[i + 1]; img.src = imgPrefix + 'https://i.ytimg.com/vi/' + videoId + '/' + fallbacks[i + 1];
if (img.dataset.src) {
img.dataset.src = newSrc;
} else { } else {
img.src = newSrc; // Last fallback failed, stop retrying
} img.onerror = null;
} }
break; return;
} }
} }
// Unknown quality format, stop retrying
img.onerror = null;
} }
// Handle YouTube channel avatars (ggpht.com) // Handle YouTube channel avatars (ggpht.com)
else if (src.includes('ggpht.com') || src.includes('yt3.ggpht.com')) { else if (src.includes('ggpht.com') || src.includes('yt3.ggpht.com')) {
// Try to increase avatar size (s88 -> s240)
const newSrc = src.replace(/=s\d+-c-k/, '=s240-c-k-c0x00ffffff-no-rj'); const newSrc = src.replace(/=s\d+-c-k/, '=s240-c-k-c0x00ffffff-no-rj');
if (newSrc !== src) { if (newSrc !== src) {
if (img.dataset.src) {
img.dataset.src = newSrc;
} else {
img.src = newSrc; img.src = newSrc;
} else {
img.onerror = null;
} }
} } else {
img.onerror = null;
} }
} }

View File

@@ -1089,12 +1089,26 @@ def serve_subscription_thumbnail(thumbnail):
f.close() f.close()
return flask.Response(image, mimetype='image/jpeg') return flask.Response(image, mimetype='image/jpeg')
url = f"https://i.ytimg.com/vi/{video_id}/hq720.jpg" image = None
for quality in ('hq720.jpg', 'sddefault.jpg', 'hqdefault.jpg'):
url = f"https://i.ytimg.com/vi/{video_id}/{quality}"
try: try:
image = util.fetch_url(url, report_text="Saved thumbnail: " + video_id) image = util.fetch_url(url, report_text="Saved thumbnail: " + video_id)
break
except util.FetchError as e:
if '404' in str(e):
continue
print("Failed to download thumbnail for " + video_id + ": " + str(e))
flask.abort(500)
except urllib.error.HTTPError as e: except urllib.error.HTTPError as e:
if e.code == 404:
continue
print("Failed to download thumbnail for " + video_id + ": " + str(e)) print("Failed to download thumbnail for " + video_id + ": " + str(e))
flask.abort(e.code) flask.abort(e.code)
if image is None:
flask.abort(404)
try: try:
f = open(thumbnail_path, 'wb') f = open(thumbnail_path, 'wb')
except FileNotFoundError: except FileNotFoundError:

View File

@@ -542,11 +542,19 @@ class RateLimitedQueue(gevent.queue.Queue):
def download_thumbnail(save_directory, video_id): def download_thumbnail(save_directory, video_id):
url = f"https://i.ytimg.com/vi/{video_id}/hq720.jpg"
save_location = os.path.join(save_directory, video_id + ".jpg") save_location = os.path.join(save_directory, video_id + ".jpg")
for quality in ('hq720.jpg', 'sddefault.jpg', 'hqdefault.jpg'):
url = f"https://i.ytimg.com/vi/{video_id}/{quality}"
try: try:
thumbnail = fetch_url(url, report_text="Saved thumbnail: " + video_id) thumbnail = fetch_url(url, report_text="Saved thumbnail: " + video_id)
except FetchError as e:
if '404' in str(e):
continue
print("Failed to download thumbnail for " + video_id + ": " + str(e))
return False
except urllib.error.HTTPError as e: except urllib.error.HTTPError as e:
if e.code == 404:
continue
print("Failed to download thumbnail for " + video_id + ": " + str(e)) print("Failed to download thumbnail for " + video_id + ": " + str(e))
return False return False
try: try:
@@ -557,6 +565,8 @@ def download_thumbnail(save_directory, video_id):
f.write(thumbnail) f.write(thumbnail)
f.close() f.close()
return True return True
print("No thumbnail available for " + video_id)
return False
def download_thumbnails(save_directory, ids): def download_thumbnails(save_directory, ids):

View File

@@ -628,12 +628,12 @@ def get_watch_page(video_id=None):
# prefix urls, and other post-processing not handled by yt_data_extract # prefix urls, and other post-processing not handled by yt_data_extract
for item in info['related_videos']: for item in info['related_videos']:
# For playlists, use first_video_id for thumbnail, not playlist id # Only set thumbnail if YouTube didn't provide one
if not item.get('thumbnail'):
if item.get('type') == 'playlist' and item.get('first_video_id'): if item.get('type') == 'playlist' and item.get('first_video_id'):
item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['first_video_id']) item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['first_video_id'])
elif item.get('type') == 'video': elif item.get('type') == 'video' and item.get('id'):
item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id']) item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['id'])
# For other types, keep existing thumbnail or skip
util.prefix_urls(item) util.prefix_urls(item)
util.add_extra_html_info(item) util.add_extra_html_info(item)
for song in info['music_list']: for song in info['music_list']:
@@ -641,9 +641,9 @@ def get_watch_page(video_id=None):
if info['playlist']: if info['playlist']:
playlist_id = info['playlist']['id'] playlist_id = info['playlist']['id']
for item in info['playlist']['items']: for item in info['playlist']['items']:
# Set high quality thumbnail for playlist videos # Only set thumbnail if YouTube didn't provide one
if item.get('type') == 'video' and item.get('id'): if not item.get('thumbnail') and item.get('type') == 'video' and item.get('id'):
item['thumbnail'] = "https://i.ytimg.com/vi/{}/hq720.jpg".format(item['id']) item['thumbnail'] = "https://i.ytimg.com/vi/{}/hqdefault.jpg".format(item['id'])
util.prefix_urls(item) util.prefix_urls(item)
util.add_extra_html_info(item) util.add_extra_html_info(item)
if playlist_id: if playlist_id:

View File

@@ -369,9 +369,9 @@ def extract_item_info(item, additional_info={}):
['detailedMetadataSnippets', 0, 'snippetText'], ['detailedMetadataSnippets', 0, 'snippetText'],
)) ))
info['thumbnail'] = normalize_url(multi_deep_get(item, info['thumbnail'] = normalize_url(multi_deep_get(item,
['thumbnail', 'thumbnails', 0, 'url'], # videos ['thumbnail', 'thumbnails', -1, 'url'], # videos (highest quality)
['thumbnails', 0, 'thumbnails', 0, 'url'], # playlists ['thumbnails', 0, 'thumbnails', -1, 'url'], # playlists
['thumbnailRenderer', 'showCustomThumbnailRenderer', 'thumbnail', 'thumbnails', 0, 'url'], # shows ['thumbnailRenderer', 'showCustomThumbnailRenderer', 'thumbnail', 'thumbnails', -1, 'url'], # shows
)) ))
info['badges'] = [] info['badges'] = []

View File

@@ -229,7 +229,7 @@ def extract_playlist_metadata(polymer_json):
if metadata['first_video_id'] is None: if metadata['first_video_id'] is None:
metadata['thumbnail'] = None metadata['thumbnail'] = None
else: else:
metadata['thumbnail'] = f"https://i.ytimg.com/vi/{metadata['first_video_id']}/hq720.jpg" metadata['thumbnail'] = f"https://i.ytimg.com/vi/{metadata['first_video_id']}/hqdefault.jpg"
metadata['video_count'] = extract_int(header.get('numVideosText')) metadata['video_count'] = extract_int(header.get('numVideosText'))
metadata['description'] = extract_str(header.get('descriptionText'), default='') metadata['description'] = extract_str(header.get('descriptionText'), default='')