def watch_page_related_video_info(item):
result = {key: item[key] for key in ('id', 'title', 'author')}
result['duration'] = util.seconds_to_timestamp(item['length_seconds'])
try:
result['views'] = item['short_view_count_text']
except KeyError:
result['views'] = ''
result['thumbnail'] = util.get_thumbnail_url(item['id'])
return result
def get_watch_page(video_id=None):
video_id = request.args.get('v') or video_id
if not video_id:
return flask.render_template('error.html',
error_message='Missing video id'), 404
if len(video_id) < 11:
return flask.render_template(
'error.html',
error_message='Incomplete video id (too short): ' + video_id), 404
lc = request.args.get('lc', '')
playlist_id = request.args.get('list')
index = request.args.get('index')
use_invidious = bool(int(request.args.get('use_invidious', '1')))
tasks = (gevent.spawn(comments.video_comments,
video_id,
int(settings.default_comment_sorting),
lc=lc),
gevent.spawn(extract_info,
video_id,
use_invidious,
playlist_id=playlist_id,
index=index))
gevent.joinall(tasks)
util.check_gevent_exceptions(tasks[1])
comments_info, info = tasks[0].value, tasks[1].value
if info['error']:
return flask.render_template('error.html', error_message=info['error'])
video_info = {
"duration": util.seconds_to_timestamp(info["duration"] or 0),
"id": info['id'],
"title": info['title'],
"author": info['author'],
}
for item in info['related_videos']:
util.prefix_urls(item)
util.add_extra_html_info(item)
if info['playlist']:
playlist_id = info['playlist']['id']
for item in info['playlist']['items']:
util.prefix_urls(item)
util.add_extra_html_info(item)
if playlist_id:
item['url'] += '&list=' + playlist_id
if item['index']:
item['url'] += '&index=' + str(item['index'])
info['playlist']['author_url'] = util.prefix_url(
info['playlist']['author_url'])
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']):
if format['acodec'] and format['vcodec']:
codecs_string = format['acodec'] + ', ' + format['vcodec']
else:
codecs_string = format['acodec'] or format['vcodec'] or '?'
download_formats.append({
'url': format['url'],
'ext': format['ext'] or '?',
'audio_quality': audio_quality_string(format),
'video_quality': video_quality_string(format),
'file_size': format_bytes(format['file_size']),
'codecs': codecs_string,
})
video_sources = get_video_sources(info)
video_height = yt_data_extract.deep_get(video_sources,
0,
'height',
default=360)
video_width = yt_data_extract.deep_get(video_sources,
0,
'width',
default=640)
# 1 second per pixel, or the actual video width
theater_video_target_width = max(640, info['duration'] or 0, video_width)
# Check for false determination of disabled comments, which comes from
# the watch page. But if we got comments in the separate request for those,
# then the determination is wrong.
if info['comments_disabled'] and len(comments_info['comments']) != 0:
info['comments_disabled'] = False
print('Warning: False determination that comments are disabled')
print('Comment count:', info['comment_count'])
info['comment_count'] = None # hack to make it obvious there's a bug
return flask.render_template(
'watch.html',
header_playlist_names=local_playlist.get_playlist_names(),
uploader_channel_url=('/' + info['author_url'])
if info['author_url'] else '',
time_published=info['time_published'],
view_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("view_count", None)),
like_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("like_count", None)),
dislike_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get(
"dislike_count", None)),
download_formats=download_formats,
video_info=json.dumps(video_info),
video_sources=video_sources,
hls_formats=info['hls_formats'],
subtitle_sources=get_subtitle_sources(info),
related=info['related_videos'],
playlist=info['playlist'],
music_list=info['music_list'],
music_attributes=get_ordered_music_list_attributes(info['music_list']),
comments_info=comments_info,
comment_count=info['comment_count'],
comments_disabled=info['comments_disabled'],
theater_mode=settings.theater_mode,
related_videos_mode=settings.related_videos_mode,
comments_mode=settings.comments_mode,
video_height=video_height,
theater_video_target_width=theater_video_target_width,
title=info['title'],
uploader=info['author'],
description=info['description'],
unlisted=info['unlisted'],
limited_state=info['limited_state'],
age_restricted=info['age_restricted'],
live=info['live'],
playability_error=info['playability_error'],
allowed_countries=info['allowed_countries'],
ip_address=info['ip_address'] if settings.route_tor else None,
invidious_used=info['invidious_used'],
invidious_reload_button=info['invidious_reload_button'],
video_url=util.URL_ORIGIN + '/watch?v=' + video_id,
)
def get_watch_page(video_id=None):
video_id = request.args.get('v') or video_id
if not video_id:
return flask.render_template('error.html',
error_message='Missing video id'), 404
if len(video_id) < 11:
return flask.render_template(
'error.html',
error_message='Incomplete video id (too short): ' + video_id), 404
time_start_str = request.args.get('t', '0s')
time_start = 0
if re.fullmatch(r'(\d+(h|m|s))+', time_start_str):
for match in re.finditer(r'(\d+)(h|m|s)', time_start_str):
time_start += int(match.group(1)) * time_table[match.group(2)]
elif re.fullmatch(r'\d+', time_start_str):
time_start = int(time_start_str)
lc = request.args.get('lc', '')
playlist_id = request.args.get('list')
index = request.args.get('index')
use_invidious = bool(int(request.args.get('use_invidious', '1')))
tasks = (gevent.spawn(comments.video_comments,
video_id,
int(settings.default_comment_sorting),
lc=lc),
gevent.spawn(extract_info,
video_id,
use_invidious,
playlist_id=playlist_id,
index=index))
gevent.joinall(tasks)
util.check_gevent_exceptions(tasks[1])
comments_info, info = tasks[0].value, tasks[1].value
if info['error']:
return flask.render_template('error.html', error_message=info['error'])
video_info = {
"duration": util.seconds_to_timestamp(info["duration"] or 0),
"id": info['id'],
"title": info['title'],
"author": info['author'],
}
# prefix urls, and other post-processing not handled by yt_data_extract
for item in info['related_videos']:
util.prefix_urls(item)
util.add_extra_html_info(item)
if info['playlist']:
playlist_id = info['playlist']['id']
for item in info['playlist']['items']:
util.prefix_urls(item)
util.add_extra_html_info(item)
if playlist_id:
item['url'] += '&list=' + playlist_id
if item['index']:
item['url'] += '&index=' + str(item['index'])
info['playlist']['author_url'] = util.prefix_url(
info['playlist']['author_url'])
# Don't prefix hls_formats for now because the urls inside the manifest
# would need to be prefixed as well.
for fmt in info['formats']:
fmt['url'] = util.prefix_url(fmt['url'])
# Add video title to end of url path so it has a filename other than just
# "videoplayback" when downloaded
title = urllib.parse.quote(util.to_valid_filename(info['title'] or ''))
for fmt in info['formats']:
filename = title
ext = fmt.get('ext')
if ext:
filename += '.' + ext
fmt['url'] = fmt['url'].replace('/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']):
if format['acodec'] and format['vcodec']:
codecs_string = format['acodec'] + ', ' + format['vcodec']
else:
codecs_string = format['acodec'] or format['vcodec'] or '?'
download_formats.append({
'url': format['url'],
'ext': format['ext'] or '?',
'audio_quality': audio_quality_string(format),
'video_quality': video_quality_string(format),
'file_size': format_bytes(format['file_size']),
'codecs': codecs_string,
})
video_sources = get_video_sources(info, tor_bypass=info['tor_bypass_used'])
video_height = yt_data_extract.deep_get(video_sources,
0,
'height',
default=360)
video_width = yt_data_extract.deep_get(video_sources,
0,
'width',
default=640)
# 1 second per pixel, or the actual video width
theater_video_target_width = max(640, info['duration'] or 0, video_width)
# Check for false determination of disabled comments, which comes from
# the watch page. But if we got comments in the separate request for those,
# then the determination is wrong.
if info['comments_disabled'] and len(comments_info['comments']) != 0:
info['comments_disabled'] = False
print('Warning: False determination that comments are disabled')
print('Comment count:', info['comment_count'])
info['comment_count'] = None # hack to make it obvious there's a bug
# captions and transcript
subtitle_sources = get_subtitle_sources(info)
other_downloads = []
for source in subtitle_sources:
best_caption_parse = urllib.parse.urlparse(source['url'].lstrip('/'))
transcript_url = (util.URL_ORIGIN + '/watch/transcript' +
best_caption_parse.path + '?' +
best_caption_parse.query)
other_downloads.append({
'label': 'Video Transcript: ' + source['label'],
'ext': 'txt',
'url': transcript_url
})
return flask.render_template(
'watch.html',
header_playlist_names=local_playlist.get_playlist_names(),
uploader_channel_url=('/' + info['author_url'])
if info['author_url'] else '',
time_published=info['time_published'],
view_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("view_count", None)),
like_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("like_count", None)),
dislike_count=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get(
"dislike_count", None)),
download_formats=download_formats,
other_downloads=other_downloads,
video_info=json.dumps(video_info),
video_sources=video_sources,
hls_formats=info['hls_formats'],
subtitle_sources=subtitle_sources,
related=info['related_videos'],
playlist=info['playlist'],
music_list=info['music_list'],
music_attributes=get_ordered_music_list_attributes(info['music_list']),
comments_info=comments_info,
comment_count=info['comment_count'],
comments_disabled=info['comments_disabled'],
video_height=video_height,
video_width=video_width,
theater_video_target_width=theater_video_target_width,
title=info['title'],
uploader=info['author'],
description=info['description'],
unlisted=info['unlisted'],
limited_state=info['limited_state'],
age_restricted=info['age_restricted'],
live=info['live'],
playability_error=info['playability_error'],
allowed_countries=info['allowed_countries'],
ip_address=info['ip_address'] if settings.route_tor else None,
invidious_used=info['invidious_used'],
invidious_reload_button=info['invidious_reload_button'],
video_url=util.URL_ORIGIN + '/watch?v=' + video_id,
time_start=time_start,
js_data={
'video_id': video_info['id'],
})
def get_watch_page():
video_id = request.args['v']
if len(video_id) < 11:
flask.abort(404)
flask.abort(
flask.Response('Incomplete video id (too short): ' + video_id))
lc = request.args.get('lc', '')
if settings.route_tor:
proxy = 'socks5://127.0.0.1:9150/'
else:
proxy = ''
yt_dl_downloader = YoutubeDL(params={
'youtube_include_dash_manifest': False,
'proxy': proxy
})
tasks = (gevent.spawn(comments.video_comments,
video_id,
int(settings.default_comment_sorting),
lc=lc),
gevent.spawn(extract_info,
yt_dl_downloader,
"https://www.youtube.com/watch?v=" + video_id,
download=False))
gevent.joinall(tasks)
comments_info, info = tasks[0].value, tasks[1].value
if isinstance(info, str): # youtube error
return flask.render_template('error.html', error_message=info)
video_info = {
"duration": util.seconds_to_timestamp(info["duration"]),
"id": info['id'],
"title": info['title'],
"author": info['uploader'],
}
upload_year = info["upload_date"][0:4]
upload_month = info["upload_date"][4:6]
upload_day = info["upload_date"][6:8]
upload_date = upload_month + "/" + upload_day + "/" + upload_year
if settings.related_videos_mode:
related_videos = get_related_items(info)
else:
related_videos = []
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']:
download_formats.append({
'url':
format['url'],
'ext':
format['ext'],
'resolution':
yt_dl_downloader.format_resolution(format),
'note':
yt_dl_downloader._format_note(format),
})
return flask.render_template(
'watch.html',
header_playlist_names=local_playlist.get_playlist_names(),
uploader_channel_url='/' + info['uploader_url'],
upload_date=upload_date,
views=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("view_count", None)),
likes=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("like_count", None)),
dislikes=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("dislike_count", None)),
download_formats=download_formats,
video_info=json.dumps(video_info),
video_sources=get_video_sources(info),
subtitle_sources=get_subtitle_sources(info),
related=related_videos,
music_list=info['music_list'],
music_attributes=get_ordered_music_list_attributes(info['music_list']),
comments_info=comments_info,
title=info['title'],
uploader=info['uploader'],
description=info['description'],
unlisted=info['unlisted'],
)
def get_watch_page(env, start_response):
video_id = env['parameters']['v'][0]
if len(video_id) < 11:
start_response('404 Not Found', [
('Content-type', 'text/plain'),
])
return b'Incomplete video id (too short): ' + video_id.encode('ascii')
start_response('200 OK', [
('Content-type', 'text/html'),
])
lc = util.default_multi_get(env['parameters'], 'lc', 0, default='')
if settings.route_tor:
proxy = 'socks5://127.0.0.1:9150/'
else:
proxy = ''
downloader = YoutubeDL(params={
'youtube_include_dash_manifest': False,
'proxy': proxy
})
tasks = (gevent.spawn(comments.video_comments,
video_id,
int(settings.default_comment_sorting),
lc=lc),
gevent.spawn(extract_info,
downloader,
"https://www.youtube.com/watch?v=" + video_id,
download=False))
gevent.joinall(tasks)
comments_html, info = tasks[0].value, tasks[1].value
#comments_html = comments.comments_html(video_id(url))
#info = YoutubeDL().extract_info(url, download=False)
#chosen_format = choose_format(info)
if isinstance(info, str): # youtube error
return html_common.yt_basic_template.substitute(
page_title="Error",
style="",
header=html_common.get_header(),
page=html.escape(info),
).encode('utf-8')
sorted_formats = sort_formats(info)
video_info = {
"duration": util.seconds_to_timestamp(info["duration"]),
"id": info['id'],
"title": info['title'],
"author": info['uploader'],
}
upload_year = info["upload_date"][0:4]
upload_month = info["upload_date"][4:6]
upload_day = info["upload_date"][6:8]
upload_date = upload_month + "/" + upload_day + "/" + upload_year
if settings.enable_related_videos:
related_videos_html = get_related_items_html(info)
else:
related_videos_html = ''
music_list = info['music_list']
if len(music_list) == 0:
music_list_html = ''
else:
# get the set of attributes which are used by atleast 1 track
# so there isn't an empty, extraneous album column which no tracks use, for example
used_attributes = set()
for track in music_list:
used_attributes = used_attributes | track.keys()
# now put them in the right order
ordered_attributes = []
for attribute in ('Artist', 'Title', 'Album'):
if attribute.lower() in used_attributes:
ordered_attributes.append(attribute)
music_list_html = '''<hr>
<table>
<caption>Music</caption>
<tr>
'''
# table headings
for attribute in ordered_attributes:
music_list_html += "<th>" + attribute + "</th>\n"
music_list_html += '''</tr>\n'''
for track in music_list:
music_list_html += '''<tr>\n'''
for attribute in ordered_attributes:
try:
value = track[attribute.lower()]
except KeyError:
music_list_html += '''<td></td>'''
else:
music_list_html += '''<td>''' + html.escape(
value) + '''</td>'''
music_list_html += '''</tr>\n'''
music_list_html += '''</table>\n'''
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_options = ''
for format in info['formats']:
download_options += download_link_template.substitute(
url=html.escape(format['url']),
ext=html.escape(format['ext']),
resolution=html.escape(downloader.format_resolution(format)),
note=html.escape(downloader._format_note(format)),
)
page = yt_watch_template.substitute(
video_title=html.escape(info["title"]),
page_title=html.escape(info["title"]),
header=html_common.get_header(),
uploader=html.escape(info["uploader"]),
uploader_channel_url='/' + info["uploader_url"],
upload_date=upload_date,
views=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("view_count", None)),
likes=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("like_count", None)),
dislikes=(lambda x: '{:,}'.format(x)
if x is not None else "")(info.get("dislike_count", None)),
download_options=download_options,
video_info=html.escape(json.dumps(video_info)),
description=html.escape(info["description"]),
video_sources=formats_html(sorted_formats) + subtitles_html(info),
related=related_videos_html,
comments=comments_html,
music_list=music_list_html,
is_unlisted='<span class="is-unlisted">Unlisted</span>'
if info['unlisted'] else '',
)
return page.encode('utf-8')