heckyel/yt-local
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'modular-data-extract'

Browse Source 
Commits in this branch are prefixed with "Extraction:"
This branch refactors data extraction. All such functionality has been moved to the yt_data_extract module.
Responses from requests are given to the module and it parses them into a consistent, more useful format.
The dependency on youtube-dl has also been dropped and this functionality has been built from scratch for these reasons:
(1) I've noticed youtube-dl breaks more often than invidious (which uses watch page extraction built from scratch) in response to changes from Youtube, so I'm hoping what I wrote will also be less brittle.
(2) Such breakage is inconvenient because I have to manually merge the fixes since I had to make changes to youtube-dl to make it do things such as extracting related videos.
(3) I have no control over error handling and request pooling with youtube-dl, since it does all the requests (these would require intrusive changes I don't want to maintain).
(4) I will now be able to finally display the number of comments and whether comments are disabled without making additional requests.
This commit is contained in:
James Taylor 2019-12-19 21:33:54 -08:00
commit b4406df9cf
61 changed files with 1753 additions and 32293 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
youtube/__init__.py
View File

@ -23,3 +23,10 @@ def inject_theme_preference():
'theme_path': '/youtube.com/static/' + theme_names[settings.theme] + '.css',
}
@yt_app.template_filter('commatize')
def commatize(num):
if num is None:
return ''
if isinstance(num, str):
num = int(num)
return '{:,}'.format(num)

136
youtube/channel.py
View File

@ -137,132 +137,13 @@ def get_channel_search_json(channel_id, query, page):
return polymer_json
def extract_info(polymer_json, tab):
response = polymer_json[1]['response']
try:
microformat = response['microformat']['microformatDataRenderer']
# channel doesn't exist or was terminated
# example terminated channel: https://www.youtube.com/channel/UCnKJeK_r90jDdIuzHXC0Org
except KeyError:
if 'alerts' in response and len(response['alerts']) > 0:
result = ''
for alert in response['alerts']:
result += alert['alertRenderer']['text']['simpleText'] + '\n'
flask.abort(200, result)
elif 'errors' in response['responseContext']:
for error in response['responseContext']['errors']['error']:
if error['code'] == 'INVALID_VALUE' and error['location'] == 'browse_id':
flask.abort(404, 'This channel does not exist')
raise
info = {}
info['current_tab'] = tab
# stuff from microformat (info given by youtube for every page on channel)
info['short_description'] = microformat['description']
info['channel_name'] = microformat['title']
info['avatar'] = microformat['thumbnail']['thumbnails'][0]['url']
channel_url = microformat['urlCanonical'].rstrip('/')
channel_id = channel_url[channel_url.rfind('/')+1:]
info['channel_id'] = channel_id
info['channel_url'] = 'https://www.youtube.com/channel/' + channel_id
info['items'] = []
# empty channel
if 'contents' not in response and 'continuationContents' not in response:
return info
# find the tab with content
# example channel where tabs do not have definite index: https://www.youtube.com/channel/UC4gQ8i3FD7YbhOgqUkeQEJg
# TODO: maybe use the 'selected' attribute for this?
if 'continuationContents' not in response:
tab_renderer = None
tab_content = None
for tab_json in response['contents']['twoColumnBrowseResultsRenderer']['tabs']:
try:
tab_renderer = tab_json['tabRenderer']
except KeyError:
tab_renderer = tab_json['expandableTabRenderer']
try:
tab_content = tab_renderer['content']
break
except KeyError:
pass
else: # didn't break
raise Exception("No tabs found with content")
assert tab == tab_renderer['title'].lower()
# extract tab-specific info
if tab in ('videos', 'playlists', 'search'): # find the list of items
if 'continuationContents' in response:
try:
items = response['continuationContents']['gridContinuation']['items']
except KeyError:
items = response['continuationContents']['sectionListContinuation']['contents'] # for search
else:
contents = tab_content['sectionListRenderer']['contents']
if 'itemSectionRenderer' in contents[0]:
item_section = contents[0]['itemSectionRenderer']['contents'][0]
try:
items = item_section['gridRenderer']['items']
except KeyError:
if "messageRenderer" in item_section:
items = []
else:
raise Exception('gridRenderer missing but messageRenderer not found')
else:
items = contents # for search
# TODO: Fix this URL prefixing shit
additional_info = {'author': info['channel_name'], 'author_url': '/channel/' + channel_id}
info['items'] = [yt_data_extract.renderer_info(renderer, additional_info) for renderer in items]
elif tab == 'about':
channel_metadata = tab_content['sectionListRenderer']['contents'][0]['itemSectionRenderer']['contents'][0]['channelAboutFullMetadataRenderer']
info['links'] = []
for link_json in channel_metadata.get('primaryLinks', ()):
url = link_json['navigationEndpoint']['urlEndpoint']['url']
if url.startswith('/redirect'): # youtube puts these on external links to do tracking
query_string = url[url.find('?')+1: ]
url = urllib.parse.parse_qs(query_string)['q'][0]
text = yt_data_extract.get_plain_text(link_json['title'])
info['links'].append( (text, url) )
info['stats'] = []
for stat_name in ('subscriberCountText', 'joinedDateText', 'viewCountText', 'country'):
try:
stat = channel_metadata[stat_name]
except KeyError:
continue
info['stats'].append(yt_data_extract.get_plain_text(stat))
if 'description' in channel_metadata:
info['description'] = yt_data_extract.get_text(channel_metadata['description'])
else:
info['description'] = ''
else:
raise NotImplementedError('Unknown or unsupported channel tab: ' + tab)
return info
def post_process_channel_info(info):
info['avatar'] = '/' + info['avatar']
info['channel_url'] = '/' + info['channel_url']
info['avatar'] = util.prefix_url(info['avatar'])
info['channel_url'] = util.prefix_url(info['channel_url'])
for item in info['items']:
yt_data_extract.prefix_urls(item)
yt_data_extract.add_extra_html_info(item)
util.prefix_urls(item)
util.add_extra_html_info(item)
@ -304,7 +185,9 @@ def get_channel_page(channel_id, tab='videos'):
flask.abort(404, 'Unknown channel tab: ' + tab)
info = extract_info(json.loads(polymer_json), tab)
info = yt_data_extract.extract_channel_info(json.loads(polymer_json), tab)
if info['error']:
return flask.render_template('error.html', error_message = info['error'])
post_process_channel_info(info)
if tab in ('videos', 'search'):
info['number_of_videos'] = number_of_videos
@ -344,7 +227,10 @@ def get_channel_page_general_url(base_url, tab, request):
flask.abort(404, 'Unknown channel tab: ' + tab)
info = extract_info(json.loads(polymer_json), tab)
info = yt_data_extract.extract_channel_info(json.loads(polymer_json), tab)
if info['error']:
return flask.render_template('error.html', error_message = info['error'])
post_process_channel_info(info)
if tab in ('videos', 'search'):
info['number_of_videos'] = 1000

126
youtube/comments.py
View File

@ -48,24 +48,6 @@ def comment_replies_ctoken(video_id, comment_id, max_results=500):
result = proto.nested(2, proto.string(2, video_id)) + proto.uint(3,6) + proto.nested(6, params)
return base64.urlsafe_b64encode(result).decode('ascii')
def ctoken_metadata(ctoken):
result = dict()
params = proto.parse(proto.b64_to_bytes(ctoken))
result['video_id'] = proto.parse(params[2])[2].decode('ascii')
offset_information = proto.parse(params[6])
result['offset'] = offset_information.get(5, 0)
result['is_replies'] = False
if (3 in offset_information) and (2 in proto.parse(offset_information[3])):
result['is_replies'] = True
result['sort'] = None
else:
try:
result['sort'] = proto.parse(offset_information[4])[6]
except KeyError:
result['sort'] = 0
return result
mobile_headers = {
@ -91,7 +73,9 @@ def request_comments(ctoken, replies=False):
print("got <!DOCTYPE>, retrying")
continue
break
return content
polymer_json = json.loads(util.uppercase_escape(content.decode('utf-8')))
return polymer_json
def single_comment_ctoken(video_id, comment_id):
@ -102,112 +86,40 @@ def single_comment_ctoken(video_id, comment_id):
def parse_comments_polymer(content):
try:
video_title = ''
content = json.loads(util.uppercase_escape(content.decode('utf-8')))
url = content[1]['url']
ctoken = urllib.parse.parse_qs(url[url.find('?')+1:])['ctoken'][0]
metadata = ctoken_metadata(ctoken)
try:
comments_raw = content[1]['response']['continuationContents']['commentSectionContinuation']['items']
except KeyError:
comments_raw = content[1]['response']['continuationContents']['commentRepliesContinuation']['contents']
ctoken = util.default_multi_get(content, 1, 'response', 'continuationContents', 'commentSectionContinuation', 'continuations', 0, 'nextContinuationData', 'continuation', default='')
comments = []
for comment_json in comments_raw:
number_of_replies = 0
try:
comment_thread = comment_json['commentThreadRenderer']
except KeyError:
comment_renderer = comment_json['commentRenderer']
else:
if 'commentTargetTitle' in comment_thread:
video_title = comment_thread['commentTargetTitle']['runs'][0]['text']
if 'replies' in comment_thread:
view_replies_text = yt_data_extract.get_plain_text(comment_thread['replies']['commentRepliesRenderer']['moreText'])
view_replies_text = view_replies_text.replace(',', '')
match = re.search(r'(\d+)', view_replies_text)
if match is None:
number_of_replies = 1
else:
number_of_replies = int(match.group(1))
comment_renderer = comment_thread['comment']['commentRenderer']
comment = {
'author_id': comment_renderer.get('authorId', ''),
'author_avatar': comment_renderer['authorThumbnail']['thumbnails'][0]['url'],
'likes': comment_renderer['likeCount'],
'published': yt_data_extract.get_plain_text(comment_renderer['publishedTimeText']),
'text': comment_renderer['contentText'].get('runs', ''),
'number_of_replies': number_of_replies,
'comment_id': comment_renderer['commentId'],
}
if 'authorText' in comment_renderer: # deleted channels have no name or channel link
comment['author'] = yt_data_extract.get_plain_text(comment_renderer['authorText'])
comment['author_url'] = comment_renderer['authorEndpoint']['commandMetadata']['webCommandMetadata']['url']
comment['author_channel_id'] = comment_renderer['authorEndpoint']['browseEndpoint']['browseId']
else:
comment['author'] = ''
comment['author_url'] = ''
comment['author_channel_id'] = ''
comments.append(comment)
except Exception as e:
print('Error parsing comments: ' + str(e))
comments = ()
ctoken = ''
return {
'ctoken': ctoken,
'comments': comments,
'video_title': video_title,
'video_id': metadata['video_id'],
'offset': metadata['offset'],
'is_replies': metadata['is_replies'],
'sort': metadata['sort'],
}
def post_process_comments_info(comments_info):
for comment in comments_info['comments']:
comment['author_url'] = util.URL_ORIGIN + comment['author_url']
comment['author_avatar'] = '/' + comment['author_avatar']
comment['permalink'] = util.URL_ORIGIN + '/watch?v=' + comments_info['video_id'] + '&lc=' + comment['comment_id']
comment['permalink'] = util.URL_ORIGIN + '/watch?v=' + comments_info['video_id'] + '&lc=' + comment['id']
if comment['author_channel_id'] in accounts.accounts:
if comment['author_id'] in accounts.accounts:
comment['delete_url'] = (util.URL_ORIGIN + '/delete_comment?video_id='
+ comments_info['video_id']
+ '&channel_id='+ comment['author_channel_id']
+ '&author_id=' + comment['author_id']
+ '&comment_id=' + comment['comment_id'])
+ '&channel_id='+ comment['author_id']
+ '&comment_id=' + comment['id'])
num_replies = comment['number_of_replies']
if num_replies == 0:
comment['replies_url'] = util.URL_ORIGIN + '/post_comment?parent_id=' + comment['comment_id'] + "&video_id=" + comments_info['video_id']
reply_count = comment['reply_count']
if reply_count == 0:
comment['replies_url'] = util.URL_ORIGIN + '/post_comment?parent_id=' + comment['id'] + "&video_id=" + comments_info['video_id']
else:
comment['replies_url'] = util.URL_ORIGIN + '/comments?parent_id=' + comment['comment_id'] + "&video_id=" + comments_info['video_id']
comment['replies_url'] = util.URL_ORIGIN + '/comments?parent_id=' + comment['id'] + "&video_id=" + comments_info['video_id']
if num_replies == 0:
if reply_count == 0:
comment['view_replies_text'] = 'Reply'
elif num_replies == 1:
elif reply_count == 1:
comment['view_replies_text'] = '1 reply'
else:
comment['view_replies_text'] = str(num_replies) + ' replies'
comment['view_replies_text'] = str(reply_count) + ' replies'
if comment['likes'] == 1:
if comment['like_count'] == 1:
comment['likes_text'] = '1 like'
else:
comment['likes_text'] = str(comment['likes']) + ' likes'
comment['likes_text'] = str(comment['like_count']) + ' likes'
comments_info['include_avatars'] = settings.enable_comment_avatars
if comments_info['ctoken'] != '':
if comments_info['ctoken']:
comments_info['more_comments_url'] = util.URL_ORIGIN + '/comments?ctoken=' + comments_info['ctoken']
comments_info['page_number'] = page_number = str(int(comments_info['offset']/20) + 1)
@ -222,7 +134,7 @@ def post_process_comments_info(comments_info):
def video_comments(video_id, sort=0, offset=0, lc='', secret_key=''):
if settings.comments_mode:
comments_info = parse_comments_polymer(request_comments(make_comment_ctoken(video_id, sort, offset, lc, secret_key)))
comments_info = yt_data_extract.extract_comments_info(request_comments(make_comment_ctoken(video_id, sort, offset, lc, secret_key)))
post_process_comments_info(comments_info)
post_comment_url = util.URL_ORIGIN + "/post_comment?video_id=" + video_id
@ -247,7 +159,7 @@ def get_comments_page():
ctoken = comment_replies_ctoken(video_id, parent_id)
replies = True
comments_info = parse_comments_polymer(request_comments(ctoken, replies))
comments_info = yt_data_extract.extract_comments_info(request_comments(ctoken, replies))
post_process_comments_info(comments_info)
if not replies:

2
youtube/local_playlist.py
View File

@ -57,7 +57,7 @@ def get_local_playlist_videos(name, offset=0, amount=50):
info['thumbnail'] = util.get_thumbnail_url(info['id'])
missing_thumbnails.append(info['id'])
info['type'] = 'video'
yt_data_extract.add_extra_html_info(info)
util.add_extra_html_info(info)
videos.append(info)
except json.decoder.JSONDecodeError:
if not video_json.strip() == '':

31
youtube/playlist.py
View File

@ -90,27 +90,28 @@ def get_playlist_page():
gevent.joinall(tasks)
first_page_json, this_page_json = tasks[0].value, tasks[1].value
try: # first page
video_list = this_page_json['response']['contents']['singleColumnBrowseResultsRenderer']['tabs'][0]['tabRenderer']['content']['sectionListRenderer']['contents'][0]['itemSectionRenderer']['contents'][0]['playlistVideoListRenderer']['contents']
except KeyError: # other pages
video_list = this_page_json['response']['continuationContents']['playlistVideoListContinuation']['contents']
info = yt_data_extract.extract_playlist_info(this_page_json)
if info['error']:
return flask.render_template('error.html', error_message = info['error'])
parsed_video_list = [yt_data_extract.parse_info_prepare_for_html(video_json) for video_json in video_list]
if page != '1':
info['metadata'] = yt_data_extract.extract_playlist_metadata(first_page_json)
util.prefix_urls(info['metadata'])
for item in info.get('items', ()):
util.prefix_urls(item)
util.add_extra_html_info(item)
if 'id' in item:
item['thumbnail'] = '/https://i.ytimg.com/vi/' + item['id'] + '/default.jpg'
metadata = yt_data_extract.renderer_info(first_page_json['response']['header'])
yt_data_extract.prefix_urls(metadata)
if 'description' not in metadata:
metadata['description'] = ''
video_count = int(metadata['size'].replace(',', ''))
metadata['size'] += ' videos'
video_count = yt_data_extract.deep_get(info, 'metadata', 'video_count')
if video_count is None:
video_count = 40
return flask.render_template('playlist.html',
video_list = parsed_video_list,
video_list = info.get('items', []),
num_pages = math.ceil(video_count/20),
parameters_dictionary = request.args,
**metadata
**info['metadata']
).encode('utf-8')

8
youtube/post_comment.py
View File

@ -70,7 +70,7 @@ def _post_comment_reply(text, video_id, parent_comment_id, session_token, cookie
print("Comment posting code: " + code)
return code
def _delete_comment(video_id, comment_id, author_id, session_token, cookiejar):
def _delete_comment(video_id, comment_id, session_token, cookiejar):
headers = {
'User-Agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 10_3_1 like Mac OS X) AppleWebKit/603.1.30 (KHTML, like Gecko) Version/10.0 Mobile/14E304 Safari/602.1',
'Accept': '*/*',
@ -79,7 +79,7 @@ def _delete_comment(video_id, comment_id, author_id, session_token, cookiejar):
'X-YouTube-Client-Version': '2.20180823',
'Content-Type': 'application/x-www-form-urlencoded',
}
action = proto.uint(1,6) + proto.string(3, comment_id) + proto.string(5, video_id) + proto.string(9, author_id)
action = proto.uint(1,6) + proto.string(3, comment_id) + proto.string(5, video_id)
action = proto.percent_b64encode(action).decode('ascii')
sej = json.dumps({"clickTrackingParams":"AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=","commandMetadata":{"webCommandMetadata":{"url":"/service_ajax","sendPost":True}},"performCommentActionEndpoint":{"action":action}})
@ -115,7 +115,7 @@ def delete_comment():
cookiejar = accounts.account_cookiejar(request.values['channel_id'])
token = get_session_token(video_id, cookiejar)
code = _delete_comment(video_id, request.values['comment_id'], request.values['author_id'], token, cookiejar)
code = _delete_comment(video_id, request.values['comment_id'], token, cookiejar)
if code == "SUCCESS":
return flask.redirect(util.URL_ORIGIN + '/comment_delete_success', 303)
@ -147,7 +147,7 @@ def post_comment():
@yt_app.route('/delete_comment', methods=['GET'])
def get_delete_comment_page():
parameters = [(parameter_name, request.args[parameter_name]) for parameter_name in ('video_id', 'channel_id', 'author_id', 'comment_id')]
parameters = [(parameter_name, request.args[parameter_name]) for parameter_name in ('video_id', 'channel_id', 'comment_id')]
return flask.render_template('delete_comment.html', parameters = parameters)

60
youtube/search.py
View File

@ -5,7 +5,6 @@ import settings
import json
import urllib
import base64
from math import ceil
import mimetypes
from flask import request
import flask
@ -74,59 +73,34 @@ def get_search_page():
filters['time'] = int(request.args.get("time", "0"))
filters['type'] = int(request.args.get("type", "0"))
filters['duration'] = int(request.args.get("duration", "0"))
info = get_search_json(query, page, autocorrect, sort, filters)
polymer_json = get_search_json(query, page, autocorrect, sort, filters)
estimated_results = int(info[1]['response']['estimatedResults'])
estimated_pages = ceil(estimated_results/20)
search_info = yt_data_extract.extract_search_info(polymer_json)
if search_info['error']:
return flask.render_template('error.html', error_message = search_info['error'])
# almost always is the first "section", but if there's an advertisement for a google product like Stadia or Home in the search results, then that becomes the first "section" and the search results are in the second. So just join all of them for resiliency
results = []
for section in info[1]['response']['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents']:
results += section['itemSectionRenderer']['contents']
for extract_item_info in search_info['items']:
util.prefix_urls(extract_item_info)
util.add_extra_html_info(extract_item_info)
parsed_results = []
corrections = {'type': None}
for renderer in results:
type = list(renderer.keys())[0]
if type == 'shelfRenderer':
continue
if type == 'didYouMeanRenderer':
renderer = renderer[type]
corrections = search_info['corrections']
if corrections['type'] == 'did_you_mean':
corrected_query_string = request.args.to_dict(flat=False)
corrected_query_string['query'] = [renderer['correctedQueryEndpoint']['searchEndpoint']['query']]
corrected_query_url = util.URL_ORIGIN + '/search?' + urllib.parse.urlencode(corrected_query_string, doseq=True)
corrections = {
'type': 'did_you_mean',
'corrected_query': yt_data_extract.format_text_runs(renderer['correctedQuery']['runs']),
'corrected_query_url': corrected_query_url,
}
continue
if type == 'showingResultsForRenderer':
renderer = renderer[type]
corrected_query_string['query'] = [corrections['corrected_query']]
corrections['corrected_query_url'] = util.URL_ORIGIN + '/search?' + urllib.parse.urlencode(corrected_query_string, doseq=True)
elif corrections['type'] == 'showing_results_for':
no_autocorrect_query_string = request.args.to_dict(flat=False)
no_autocorrect_query_string['autocorrect'] = ['0']
no_autocorrect_query_url = util.URL_ORIGIN + '/search?' + urllib.parse.urlencode(no_autocorrect_query_string, doseq=True)
corrections = {
'type': 'showing_results_for',
'corrected_query': yt_data_extract.format_text_runs(renderer['correctedQuery']['runs']),
'original_query_url': no_autocorrect_query_url,
'original_query': renderer['originalQuery']['simpleText'],
}
continue
info = yt_data_extract.parse_info_prepare_for_html(renderer)
if info['type'] != 'unsupported':
parsed_results.append(info)
corrections['original_query_url'] = no_autocorrect_query_url
return flask.render_template('search.html',
header_playlist_names = local_playlist.get_playlist_names(),
query = query,
estimated_results = estimated_results,
estimated_pages = estimated_pages,
corrections = corrections,
results = parsed_results,
estimated_results = search_info['estimated_results'],
estimated_pages = search_info['estimated_pages'],
corrections = search_info['corrections'],
results = search_info['items'],
parameters_dictionary = request.args,
)

17
youtube/subscriptions.py
View File

@ -172,7 +172,7 @@ def _get_videos(cursor, number_per_page, offset, tag = None):
'id': db_video[0],
'title': db_video[1],
'duration': db_video[2],
'published': exact_timestamp(db_video[3]) if db_video[4] else posix_to_dumbed_down(db_video[3]),
'time_published': exact_timestamp(db_video[3]) if db_video[4] else posix_to_dumbed_down(db_video[3]),
'author': db_video[5],
})
@ -455,10 +455,17 @@ def _get_upstream_videos(channel_id):
print('Failed to read atoma feed for ' + channel_status_name)
traceback.print_exc()
videos = channel.extract_info(json.loads(channel_tab), 'videos')['items']
channel_info = yt_data_extract.extract_channel_info(json.loads(channel_tab), 'videos')
if channel_info['error']:
print('Error checking channel ' + channel_status_name + ': ' + channel_info['error'])
return
videos = channel_info['items']
for i, video_item in enumerate(videos):
if 'description' not in video_item:
if not video_item.get('description'):
video_item['description'] = ''
else:
video_item['description'] = ''.join(run.get('text', '') for run in video_item['description'])
if video_item['id'] in times_published:
video_item['time_published'] = times_published[video_item['id']]
@ -466,7 +473,7 @@ def _get_upstream_videos(channel_id):
else:
video_item['is_time_published_exact'] = False
try:
video_item['time_published'] = youtube_timestamp_to_posix(video_item['published']) - i # subtract a few seconds off the videos so they will be in the right order
video_item['time_published'] = youtube_timestamp_to_posix(video_item['time_published']) - i # subtract a few seconds off the videos so they will be in the right order
except KeyError:
print(video_item)
@ -759,7 +766,7 @@ def get_subscriptions_page():
video['thumbnail'] = util.URL_ORIGIN + '/data/subscription_thumbnails/' + video['id'] + '.jpg'
video['type'] = 'video'
video['item_size'] = 'small'
yt_data_extract.add_extra_html_info(video)
util.add_extra_html_info(video)
tags = _get_all_tags(cursor)

4
youtube/templates/comments.html
View File

@ -12,11 +12,11 @@
<a class="author" href="{{ comment['author_url'] }}" title="{{ comment['author'] }}">{{ comment['author'] }}</a>
</address>
<a class="permalink" href="{{ comment['permalink'] }}" title="permalink">
<time datetime="">{{ comment['published'] }}</time>
<time datetime="">{{ comment['time_published'] }}</time>
</a>
<span class="text">{{ common_elements.text_runs(comment['text']) }}</span>
<span class="likes">{{ comment['likes_text'] if comment['likes'] else ''}}</span>
<span class="likes">{{ comment['likes_text'] if comment['like_count'] else ''}}</span>
<div class="bottom-row">
<a href="{{ comment['replies_url'] }}" class="replies">{{ comment['view_replies_text'] }}</a>
{% if 'delete_url' is in comment %}

36
youtube/templates/common_elements.html
View File

@ -9,42 +9,45 @@
{{ text_run["text"] }}
{%- endif -%}
{%- endfor -%}
{%- else -%}
{%- elif runs -%}
{{ runs }}
{%- endif -%}
{% endmacro %}
{% macro item(info, description=false, horizontal=true, include_author=true) %}
{% macro item(info, description=false, horizontal=true, include_author=true, include_badges=true) %}
<div class="item-box {{ info['type'] + '-item-box' }} {{'horizontal-item-box' if horizontal else 'vertical-item-box'}} {{'has-description' if description else 'no-description'}}">
{% if info['error'] %}
{{ info['error'] }}
{% else %}
<div class="item {{ info['type'] + '-item' }}">
<a class="thumbnail-box" href="{{ info['url'] }}" title="{{ info['title'] }}">
<img class="thumbnail-img" src="{{ info['thumbnail'] }}">
{% if info['type'] != 'channel' %}
<div class="thumbnail-info">
<span>{{ info['size'] if info['type'] == 'playlist' else info['duration'] }}</span>
<span>{{ (info['video_count']|string + ' videos') if info['type'] == 'playlist' else info['duration'] }}</span>
</div>
{% endif %}
</a>
<div class="title"><a class="title" href="{{ info['url'] }}" title="{{ info['title'] }}">{{ info['title'] }}</a></div>
<ul class="stats {{'vertical-stats' if horizontal and not description and include_author else 'horizontal-stats'}}">
{% if info['type'] == 'channel' %}
<li><span>{{ info['subscriber_count'] }} subscribers</span></li>
<li><span>{{ info['size'] }} videos</span></li>
{% else %}
{% if include_author %}
{% if 'author_url' is in(info) %}
<li><address title="{{ info['author'] }}">By <a href="{{ info['author_url'] }}">{{ info['author'] }}</a></address></li>
{% if info.get('author_url') %}
<address title="{{ info['author'] }}">By <a href="{{ info['author_url'] }}">{{ info['author'] }}</a></address>
{% else %}
<li><address title="{{ info['author'] }}"><b>{{ info['author'] }}</b></address></li>
<address title="{{ info['author'] }}"><b>{{ info['author'] }}</b></address>
{% endif %}
{% endif %}
{% if 'views' is in(info) %}
<li><span class="views">{{ info['views'] }}</span></li>
<ul class="stats {{'horizontal-stats' if horizontal else 'vertical-stats'}}">
{% if info['type'] == 'channel' %}
<li><span>{{ info['approx_subscriber_count'] }} subscribers</span></li>
<li><span>{{ info['video_count'] }} videos</span></li>
{% else %}
{% if info.get('approx_view_count') %}
<li><span class="views">{{ info['approx_view_count'] }} views</span></li>
{% endif %}
{% if 'published' is in(info) %}
<li><time>{{ info['published'] }}</time></li>
{% if info.get('time_published') %}
<li><time>{{ info['time_published'] }}</time></li>
{% endif %}
{% endif %}
</ul>
@ -52,11 +55,14 @@
{% if description %}
<span class="description">{{ text_runs(info.get('description', '')) }}</span>
{% endif %}
{% if include_badges %}
<span class="badges">{{ info['badges']|join(' | ') }}</span>
{% endif %}
</div>
{% if info['type'] == 'video' %}
<input class="item-checkbox" type="checkbox" name="video_info_list" value="{{ info['video_info'] }}" form="playlist-edit">
{% endif %}
{% endif %}
</div>
{% endmacro %}

5
youtube/templates/playlist.html
View File

@ -54,8 +54,9 @@
<h2 class="playlist-title">{{ title }}</h2>
<a class="playlist-author" href="{{ author_url }}">{{ author }}</a>
<div class="playlist-stats">
<div>{{ views }}</div>
<div>{{ size }}</div>
<div>{{ video_count|commatize }} videos</div>
<div>{{ view_count|commatize }} views</div>
<div>Last updated {{ time_published }}</div>
</div>
<div class="playlist-description">{{ common_elements.text_runs(description) }}</div>
</div>

6
youtube/templates/search.html
View File

@ -29,10 +29,10 @@
<div id="result-info">
<div id="number-of-results">Approximately {{ '{:,}'.format(estimated_results) }} results ({{ '{:,}'.format(estimated_pages) }} pages)</div>
{% if corrections['type'] == 'showing_results_for' %}
<div>Showing results for <a>{{ corrections['corrected_query']|safe }}</a></div>
<div>Search instead for <a href="{{ corrections['original_query_url'] }}">{{ corrections['original_query'] }}</a></div>
<div>Showing results for <a>{{ common_elements.text_runs(corrections['corrected_query_text']) }}</a></div>
<div>Search instead for <a href="{{ corrections['original_query_url'] }}">{{ corrections['original_query_text'] }}</a></div>
{% elif corrections['type'] == 'did_you_mean' %}
<div>Did you mean <a href="{{ corrections['corrected_query_url'] }}">{{ corrections['corrected_query']|safe }}</a></div>
<div>Did you mean <a href="{{ corrections['corrected_query_url'] }}">{{ common_elements.text_runs(corrections['corrected_query_text']) }}</a></div>
{% endif %}
</div>
<div class="item-list">

85
youtube/templates/watch.html
View File

@ -14,6 +14,19 @@
text-decoration: underline;
}
.playability-error{
height: 360px;
width: 640px;
grid-column: 2;
background-color: var(--video-background-color);
text-align:center;
}
.playability-error span{
position: relative;
top: 50%;
transform: translate(-50%, -50%);
}
{% if theater_mode %}
video{
grid-column: 1 / span 5;
@ -61,11 +74,20 @@
grid-column: 1 / span 2;
min-width: 0;
}
.video-info > .is-unlisted{
background-color: var(--interface-color);
.video-info > .labels{
justify-self:start;
padding-left:2px;
padding-right:2px;
list-style: none;
padding: 0px;
margin: 5px 0px;
}
.video-info > .labels:empty{
margin: 0px;
}
.labels > li{
display: inline;
margin-right:5px;
background-color: var(--interface-color);
padding: 2px 5px
}
.video-info > address{
grid-column: 1;
@ -143,9 +165,13 @@
.related-videos-inner{
padding-top: 10px;
display: grid;
grid-auto-rows: 94px;
grid-auto-rows: 90px;
grid-row-gap: 10px;
}
.thumbnail-box{ /* overides rule in shared.css */
height: 90px !important;
width: 120px !important;
}
/* Put related vids below videos when window is too small */
/* 1100px instead of 1080 because W3C is full of idiots who include scrollbar width */
@ -187,13 +213,25 @@
.format-ext{
width: 60px;
}
.format-res{
width:90px;
.format-video-quality{
width: 140px;
}
.format-audio-quality{
width: 120px;
}
.format-file-size{
width: 80px;
}
.format-codecs{
width: 120px;
}
{% endblock style %}
{% block main %}
<video controls autofocus>
{% if playability_error %}
<div class="playability-error"><span>{{ 'Error: ' + playability_error }}</span></div>
{% else %}
<video controls autofocus class="video">
{% for video_source in video_sources %}
<source src="{{ video_source['src'] }}" type="{{ video_source['type'] }}">
{% endfor %}
@ -207,18 +245,27 @@
{% endfor %}
</video>
{% endif %}
<div class="video-info">
<h2 class="title">{{ title }}</h2>
{% if unlisted %}
<span class="is-unlisted">Unlisted</span>
{% endif %}
<ul class="labels">
{%- if unlisted -%}
<li class="is-unlisted">Unlisted</li>
{%- endif -%}
{%- if age_restricted -%}
<li class="age-restricted">Age-restricted</li>
{%- endif -%}
{%- if limited_state -%}
<li>Limited state</li>
{%- endif -%}
</ul>
<address>Uploaded by <a href="{{ uploader_channel_url }}">{{ uploader }}</a></address>
<span class="views">{{ views }} views</span>
<span class="views">{{ view_count }} views</span>
<time datetime="$upload_date">Published on {{ upload_date }}</time>
<span class="likes-dislikes">{{ likes }} likes {{ dislikes }} dislikes</span>
<time datetime="$upload_date">Published on {{ time_published }}</time>
<span class="likes-dislikes">{{ like_count }} likes {{ dislike_count }} dislikes</span>
<details class="download-dropdown">
<summary class="download-dropdown-label">Download</summary>
<ul class="download-dropdown-content">
@ -227,8 +274,10 @@
<a class="download-link" href="{{ format['url'] }}">
<ol class="format-attributes">
<li class="format-ext">{{ format['ext'] }}</li>
<li class="format-res">{{ format['resolution'] }}</li>
<li class="format-note">{{ format['note'] }}</li>
<li class="format-video-quality">{{ format['video_quality'] }}</li>
<li class="format-audio-quality">{{ format['audio_quality'] }}</li>
<li class="format-file-size">{{ format['file_size'] }}</li>
<li class="format-codecs">{{ format['codecs'] }}</li>
</ol>
</a>
</li>
@ -238,7 +287,7 @@
<input class="checkbox" name="video_info_list" value="{{ video_info }}" form="playlist-edit" type="checkbox">
<span class="description">{{ description }}</span>
<span class="description">{{ common_elements.text_runs(description) }}</span>
<div class="music-list">
{% if music_list.__len__() != 0 %}
<hr>
@ -266,7 +315,7 @@
<summary>Related Videos</summary>
<nav class="related-videos-inner">
{% for info in related %}
{{ common_elements.item(info) }}
{{ common_elements.item(info, include_badges=false) }}
{% endfor %}
</nav>
</details>

62
youtube/util.py
View File

@ -1,4 +1,5 @@
import settings
from youtube import yt_data_extract
import socks, sockshandler
import gzip
import brotli
@ -6,6 +7,7 @@ import urllib.parse
import re
import time
import os
import json
import gevent
import gevent.queue
import gevent.lock
@ -176,7 +178,7 @@ def fetch_url(url, headers=(), timeout=15, report_text=None, data=None, cookieja
return content, response
return content
mobile_user_agent = 'Mozilla/5.0 (iPhone; CPU iPhone OS 10_3_1 like Mac OS X) AppleWebKit/603.1.30 (KHTML, like Gecko) Version/10.0 Mobile/14E304 Safari/602.1'
mobile_user_agent = 'Mozilla/5.0 (Linux; Android 7.0; Redmi Note 4 Build/NRD90M) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Mobile Safari/537.36'
mobile_ua = (('User-Agent', mobile_user_agent),)
desktop_user_agent = 'Mozilla/5.0 (Windows NT 6.1; rv:52.0) Gecko/20100101 Firefox/52.0'
desktop_ua = (('User-Agent', desktop_user_agent),)
@ -277,15 +279,6 @@ def video_id(url):
url_parts = urllib.parse.urlparse(url)
return urllib.parse.parse_qs(url_parts.query)['v'][0]
def default_multi_get(object, *keys, default):
''' Like dict.get(), but for nested dictionaries/sequences, supporting keys or indices. Last argument is the default value to use in case of any IndexErrors or KeyErrors '''
try:
for key in keys:
object = object[key]
return object
except (IndexError, KeyError):
return default
# default, sddefault, mqdefault, hqdefault, hq720
def get_thumbnail_url(video_id):
@ -317,3 +310,52 @@ def uppercase_escape(s):
return re.sub(
r'\\U([0-9a-fA-F]{8})',
lambda m: chr(int(m.group(1), base=16)), s)
def prefix_url(url):
if url is None:
return None
url = url.lstrip('/') # some urls have // before them, which has a special meaning
return '/' + url
def left_remove(string, substring):
'''removes substring from the start of string, if present'''
if string.startswith(substring):
return string[len(substring):]
return string
def prefix_urls(item):
try:
item['thumbnail'] = prefix_url(item['thumbnail'])
except KeyError:
pass
try:
item['author_url'] = prefix_url(item['author_url'])
except KeyError:
pass
def add_extra_html_info(item):
if item['type'] == 'video':
item['url'] = (URL_ORIGIN + '/watch?v=' + item['id']) if item.get('id') else None
video_info = {}
for key in ('id', 'title', 'author', 'duration'):
try:
video_info[key] = item[key]
except KeyError:
video_info[key] = ''
item['video_info'] = json.dumps(video_info)
elif item['type'] == 'playlist':
item['url'] = (URL_ORIGIN + '/playlist?list=' + item['id']) if item.get('id') else None
elif item['type'] == 'channel':
item['url'] = (URL_ORIGIN + "/channel/" + item['id']) if item.get('id') else None
def parse_info_prepare_for_html(renderer, additional_info={}):
item = yt_data_extract.extract_item_info(renderer, additional_info)
prefix_urls(item)
add_extra_html_info(item)
return item

350
youtube/watch.py
View File

@ -5,49 +5,20 @@ import settings
from flask import request
import flask
from youtube_dl.YoutubeDL import YoutubeDL
from youtube_dl.extractor.youtube import YoutubeError
import json
import html
import gevent
import os
import math
import traceback
import urllib
def get_related_items(info):
results = []
for item in info['related_vids']:
if 'list' in item: # playlist:
result = watch_page_related_playlist_info(item)
else:
result = watch_page_related_video_info(item)
yt_data_extract.prefix_urls(result)
yt_data_extract.add_extra_html_info(result)
results.append(result)
return results
# json of related items retrieved directly from the watch page has different names for everything
# converts these to standard names
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'])
result['type'] = 'video'
return result
with open(os.path.join(settings.data_dir, 'decrypt_function_cache.json'), 'r') as f:
decrypt_cache = json.loads(f.read())['decrypt_cache']
except FileNotFoundError:
decrypt_cache = {}
def watch_page_related_playlist_info(item):
return {
'size': item['playlist_length'] if item['playlist_length'] != "0" else "50+",
'title': item['playlist_title'],
'id': item['list'],
'first_video_id': item['video_id'],
'thumbnail': util.get_thumbnail_url(item['video_id']),
'type': 'playlist',
}
def get_video_sources(info):
video_sources = []
@ -55,9 +26,10 @@ def get_video_sources(info):
max_resolution = 360
else:
max_resolution = settings.default_resolution
for format in info['formats']:
if format['acodec'] != 'none' and format['vcodec'] != 'none' and format['height'] <= max_resolution:
if not all(format[attr] for attr in ('height', 'width', 'ext', 'url')):
continue
if format['acodec'] and format['vcodec'] and format['height'] <= max_resolution:
video_sources.append({
'src': format['url'],
'type': 'video/' + format['ext'],
@ -71,50 +43,108 @@ def get_video_sources(info):
return video_sources
def get_subtitle_sources(info):
sources = []
default_found = False
default = None
for language, formats in info['subtitles'].items():
for format in formats:
if format['ext'] == 'vtt':
source = {
'url': '/' + format['url'],
'label': language,
'srclang': language,
# set as on by default if this is the preferred language and a default-on subtitles mode is in settings
'on': language == settings.subtitles_language and settings.subtitles_mode > 0,
def make_caption_src(info, lang, auto=False, trans_lang=None):
label = lang
if auto:
label += ' (Automatic)'
if trans_lang:
label += ' -> ' + trans_lang
return {
'url': '/' + yt_data_extract.get_caption_url(info, lang, 'vtt', auto, trans_lang),
'label': label,
'srclang': trans_lang[0:2] if trans_lang else lang[0:2],
'on': False,
}
if language == settings.subtitles_language:
default_found = True
default = source
else:
sources.append(source)
break
def lang_in(lang, sequence):
'''Tests if the language is in sequence, with e.g. en and en-US considered the same'''
if lang is None:
return False
lang = lang[0:2]
return lang in (l[0:2] for l in sequence)
# Put it at the end to avoid browser bug when there are too many languages
def lang_eq(lang1, lang2):
'''Tests if two iso 639-1 codes are equal, with en and en-US considered the same.
Just because the codes are equal does not mean the dialects are mutually intelligible, but this will have to do for now without a complex language model'''
if lang1 is None or lang2 is None:
return False
return lang1[0:2] == lang2[0:2]
def equiv_lang_in(lang, sequence):
'''Extracts a language in sequence which is equivalent to lang.
e.g. if lang is en, extracts en-GB from sequence.
Necessary because if only a specific variant like en-GB is available, can't ask Youtube for simply en. Need to get the available variant.'''
lang = lang[0:2]
for l in sequence:
if l[0:2] == lang:
return l
return None
def get_subtitle_sources(info):
'''Returns these sources, ordered from least to most intelligible:
native_video_lang (Automatic)
foreign_langs (Manual)
native_video_lang (Automatic) -> pref_lang
foreign_langs (Manual) -> pref_lang
native_video_lang (Manual) -> pref_lang
pref_lang (Automatic)
pref_lang (Manual)'''
sources = []
pref_lang = settings.subtitles_language
native_video_lang = None
if info['automatic_caption_languages']:
native_video_lang = info['automatic_caption_languages'][0]
highest_fidelity_is_manual = False
# Sources are added in very specific order outlined above
# More intelligible sources are put further down to avoid browser bug when there are too many languages
# (in firefox, it is impossible to select a language near the top of the list because it is cut off)
if default_found:
sources.append(default)
try:
formats = info['automatic_captions'][settings.subtitles_language]
except KeyError:
pass
else:
for format in formats:
if format['ext'] == 'vtt':
sources.append({
'url': '/' + format['url'],
'label': settings.subtitles_language + ' - Automatic',
'srclang': settings.subtitles_language,
# native_video_lang (Automatic)
if native_video_lang and not lang_eq(native_video_lang, pref_lang):
sources.append(make_caption_src(info, native_video_lang, auto=True))
# set as on by default if this is the preferred language and a default-on subtitles mode is in settings
'on': settings.subtitles_mode == 2 and not default_found,
# foreign_langs (Manual)
for lang in info['manual_caption_languages']:
if not lang_eq(lang, pref_lang):
sources.append(make_caption_src(info, lang))
})
if (lang_in(pref_lang, info['translation_languages'])
and not lang_in(pref_lang, info['automatic_caption_languages'])
and not lang_in(pref_lang, info['manual_caption_languages'])):
# native_video_lang (Automatic) -> pref_lang
if native_video_lang and not lang_eq(pref_lang, native_video_lang):
sources.append(make_caption_src(info, native_video_lang, auto=True, trans_lang=pref_lang))
# foreign_langs (Manual) -> pref_lang
for lang in info['manual_caption_languages']:
if not lang_eq(lang, native_video_lang) and not lang_eq(lang, pref_lang):
sources.append(make_caption_src(info, lang, trans_lang=pref_lang))
# native_video_lang (Manual) -> pref_lang
if lang_in(native_video_lang, info['manual_caption_languages']):
sources.append(make_caption_src(info, native_video_lang, trans_lang=pref_lang))
# pref_lang (Automatic)
if lang_in(pref_lang, info['automatic_caption_languages']):
sources.append(make_caption_src(info, equiv_lang_in(pref_lang, info['automatic_caption_languages']), auto=True))
# pref_lang (Manual)
if lang_in(pref_lang, info['manual_caption_languages']):
sources.append(make_caption_src(info, equiv_lang_in(pref_lang, info['manual_caption_languages'])))
highest_fidelity_is_manual = True
if sources and sources[-1]['srclang'] == pref_lang:
# set as on by default since it's manual a default-on subtitles mode is in settings
if highest_fidelity_is_manual and settings.subtitles_mode > 0:
sources[-1]['on'] = True
# set as on by default since settings indicate to set it as such even if it's not manual
elif settings.subtitles_mode == 2:
sources[-1]['on'] = True
if len(sources) == 0:
assert len(info['automatic_caption_languages']) == 0 and len(info['manual_caption_languages']) == 0
return sources
@ -134,14 +164,111 @@ def get_ordered_music_list_attributes(music_list):
return ordered_attributes
def extract_info(downloader, *args, **kwargs):
def save_decrypt_cache():
try:
return downloader.extract_info(*args, **kwargs)
except YoutubeError as e:
return str(e)
f = open(os.path.join(settings.data_dir, 'decrypt_function_cache.json'), 'w')
except FileNotFoundError:
os.makedirs(settings.data_dir)
f = open(os.path.join(settings.data_dir, 'decrypt_function_cache.json'), 'w')
f.write(json.dumps({'version': 1, 'decrypt_cache':decrypt_cache}, indent=4, sort_keys=True))
f.close()
def decrypt_signatures(info):
'''return error string, or False if no errors'''
if not yt_data_extract.requires_decryption(info):
return False
if not info['player_name']:
return 'Could not find player name'
if not info['base_js']:
return 'Failed to find base.js'
player_name = info['player_name']
if player_name in decrypt_cache:
print('Using cached decryption function for: ' + player_name)
info['decryption_function'] = decrypt_cache[player_name]
else:
base_js = util.fetch_url(info['base_js'], debug_name='base.js', report_text='Fetched player ' + player_name)
base_js = base_js.decode('utf-8')
err = yt_data_extract.extract_decryption_function(info, base_js)
if err:
return err
decrypt_cache[player_name] = info['decryption_function']
save_decrypt_cache()
err = yt_data_extract.decrypt_signatures(info)
return err
headers = (
('Accept', '*/*'),
('Accept-Language', 'en-US,en;q=0.5'),
('X-YouTube-Client-Name', '2'),
('X-YouTube-Client-Version', '2.20180830'),
) + util.mobile_ua
def extract_info(video_id):
polymer_json = util.fetch_url('https://m.youtube.com/watch?v=' + video_id + '&pbj=1&bpctr=9999999999', headers=headers, debug_name='watch').decode('utf-8')
# TODO: Decide whether this should be done in yt_data_extract.extract_watch_info
try:
polymer_json = json.loads(polymer_json)
except json.decoder.JSONDecodeError:
traceback.print_exc()
return {'error': 'Failed to parse json response'}
info = yt_data_extract.extract_watch_info(polymer_json)
# age restriction bypass
if info['age_restricted']:
print('Fetching age restriction bypass page')
data = {
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
}
url = 'https://www.youtube.com/get_video_info?' + urllib.parse.urlencode(data)
video_info_page = util.fetch_url(url, debug_name='get_video_info', report_text='Fetched age restriction bypass page').decode('utf-8')
yt_data_extract.update_with_age_restricted_info(info, video_info_page)
# signature decryption
decryption_error = decrypt_signatures(info)
if decryption_error:
decryption_error = 'Error decrypting url signatures: ' + decryption_error
info['playability_error'] = decryption_error
return info
def video_quality_string(format):
if format['vcodec']:
result =str(format['width'] or '?') + 'x' + str(format['height'] or '?')
if format['fps']:
result += ' ' + str(format['fps']) + 'fps'
return result
elif format['acodec']:
return 'audio only'
return '?'
def audio_quality_string(format):
if format['acodec']:
result = str(format['audio_bitrate'] or '?') + 'k'
if format['audio_sample_rate']:
result += ' ' + str(format['audio_sample_rate']) + ' Hz'
return result
elif format['vcodec']:
return 'video only'
return '?'
# from https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py
def format_bytes(bytes):
if bytes is None:
return 'N/A'
if type(bytes) is str:
bytes = float(bytes)
if bytes == 0.0:
exponent = 0
else:
exponent = int(math.log(bytes, 1024.0))
suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent]
converted = float(bytes) / float(1024 ** exponent)
return '%.2f%s' % (converted, suffix)
@yt_app.route('/watch')
@ -152,38 +279,26 @@ def get_watch_page():
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.spawn(extract_info, video_id)
)
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)
if info['error']:
return flask.render_template('error.html', error_message = info['error'])
video_info = {
"duration": util.seconds_to_timestamp(info["duration"]),
"duration": util.seconds_to_timestamp(info["duration"] or 0),
"id": info['id'],
"title": info['title'],
"author": info['uploader'],
"author": info['author'],
}
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 = []
for item in info['related_videos']:
util.prefix_urls(item)
util.add_extra_html_info(item)
if settings.gather_googlevideo_domains:
with open(os.path.join(settings.data_dir, 'googlevideo-domains.txt'), 'a+', encoding='utf-8') as f:
@ -195,31 +310,37 @@ def get_watch_page():
download_formats = []
for format in info['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'],
'resolution': yt_dl_downloader.format_resolution(format),
'note': yt_dl_downloader._format_note(format),
'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 = video_sources[0]['height']
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'], video_sources[0]['width'])
theater_video_target_width = max(640, info['duration'] or 0, video_width)
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)),
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,
subtitle_sources = get_subtitle_sources(info),
related = related_videos,
related = info['related_videos'],
music_list = info['music_list'],
music_attributes = get_ordered_music_list_attributes(info['music_list']),
comments_info = comments_info,
@ -232,9 +353,12 @@ def get_watch_page():
theater_video_target_width = theater_video_target_width,
title = info['title'],
uploader = info['uploader'],
uploader = info['author'],
description = info['description'],
unlisted = info['unlisted'],
limited_state = info['limited_state'],
age_restricted = info['age_restricted'],
playability_error = info['playability_error'],
)

273
youtube/yt_data_extract.py
View File

@ -1,273 +0,0 @@
from youtube import util
import html
import json
# videos (all of type str):
# id
# title
# url
# author
# author_url
# thumbnail
# description
# published
# duration
# likes
# dislikes
# views
# playlist_index
# playlists:
# id
# title
# url
# author
# author_url
# thumbnail
# description
# updated
# size
# first_video_id
def get_plain_text(node):
try:
return node['simpleText']
except KeyError:
return ''.join(text_run['text'] for text_run in node['runs'])
def format_text_runs(runs):
if isinstance(runs, str):
return runs
result = ''
for text_run in runs:
if text_run.get("bold", False):
result += "<b>" + html.escape(text_run["text"]) + "</b>"
elif text_run.get('italics', False):
result += "<i>" + html.escape(text_run["text"]) + "</i>"
else:
result += html.escape(text_run["text"])
return result
def get_url(node):
try:
return node['runs'][0]['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url']
except KeyError:
return node['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url']
def get_text(node):
if node == {}:
return ''
try:
return node['simpleText']
except KeyError:
pass
try:
return node['runs'][0]['text']
except IndexError: # empty text runs
return ''
except KeyError:
print(node)
raise
def get_formatted_text(node):
try:
return node['runs']
except KeyError:
return node['simpleText']
def get_badges(node):
badges = []
for badge_node in node:
badge = badge_node['metadataBadgeRenderer']['label']
badges.append(badge)
return badges
def get_thumbnail(node):
try:
return node['thumbnails'][0]['url'] # polymer format
except KeyError:
return node['url'] # ajax format
dispatch = {
# polymer format
'title': ('title', get_text),
'publishedTimeText': ('published', get_text),
'videoId': ('id', lambda node: node),
'descriptionSnippet': ('description', get_formatted_text),
'lengthText': ('duration', get_text),
'thumbnail': ('thumbnail', get_thumbnail),
'thumbnails': ('thumbnail', lambda node: node[0]['thumbnails'][0]['url']),
'viewCountText': ('views', get_text),
'numVideosText': ('size', lambda node: get_text(node).split(' ')[0]), # the format is "324 videos"
'videoCountText': ('size', get_text),
'playlistId': ('id', lambda node: node),
'descriptionText': ('description', get_formatted_text),
'subscriberCountText': ('subscriber_count', get_text),
'channelId': ('id', lambda node: node),
'badges': ('badges', get_badges),
# ajax format
'view_count_text': ('views', get_text),
'num_videos_text': ('size', lambda node: get_text(node).split(' ')[0]),
'owner_text': ('author', get_text),
'owner_endpoint': ('author_url', lambda node: node['url']),
'description': ('description', get_formatted_text),
'index': ('playlist_index', get_text),
'short_byline': ('author', get_text),
'length': ('duration', get_text),
'video_id': ('id', lambda node: node),
}
def ajax_info(item_json):
try:
info = {}
for key, node in item_json.items():
try:
simple_key, function = dispatch[key]
except KeyError:
continue
info[simple_key] = function(node)
return info
except KeyError:
print(item_json)
raise
def prefix_urls(item):
try:
item['thumbnail'] = '/' + item['thumbnail'].lstrip('/')
except KeyError:
pass
try:
item['author_url'] = util.URL_ORIGIN + item['author_url']
except KeyError:
pass
def add_extra_html_info(item):
if item['type'] == 'video':
item['url'] = util.URL_ORIGIN + '/watch?v=' + item['id']
video_info = {}
for key in ('id', 'title', 'author', 'duration'):
try:
video_info[key] = item[key]
except KeyError:
video_info[key] = ''
item['video_info'] = json.dumps(video_info)
elif item['type'] == 'playlist':
item['url'] = util.URL_ORIGIN + '/playlist?list=' + item['id']
elif item['type'] == 'channel':
item['url'] = util.URL_ORIGIN + "/channel/" + item['id']
def renderer_info(renderer, additional_info={}):
type = list(renderer.keys())[0]
renderer = renderer[type]
info = {}
if type == 'itemSectionRenderer':
return renderer_info(renderer['contents'][0], additional_info)
if type in ('movieRenderer', 'clarificationRenderer'):
info['type'] = 'unsupported'
return info
info.update(additional_info)
if type in ('compactVideoRenderer', 'videoRenderer', 'playlistVideoRenderer', 'gridVideoRenderer'):
info['type'] = 'video'
elif type in ('playlistRenderer', 'compactPlaylistRenderer', 'gridPlaylistRenderer',
'radioRenderer', 'compactRadioRenderer', 'gridRadioRenderer',
'showRenderer', 'compactShowRenderer', 'gridShowRenderer'):
info['type'] = 'playlist'
elif type == 'channelRenderer':
info['type'] = 'channel'
elif type == 'playlistHeaderRenderer':
info['type'] = 'playlist_metadata'
else:
info['type'] = 'unsupported'
return info
try:
if 'viewCountText' in renderer: # prefer this one as it contains all the digits
info['views'] = get_text(renderer['viewCountText'])
elif 'shortViewCountText' in renderer:
info['views'] = get_text(renderer['shortViewCountText'])
if 'ownerText' in renderer:
info['author'] = renderer['ownerText']['runs'][0]['text']
info['author_url'] = renderer['ownerText']['runs'][0]['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url']
try:
overlays = renderer['thumbnailOverlays']
except KeyError:
pass
else:
for overlay in overlays:
if 'thumbnailOverlayTimeStatusRenderer' in overlay:
info['duration'] = get_text(overlay['thumbnailOverlayTimeStatusRenderer']['text'])
# show renderers don't have videoCountText
elif 'thumbnailOverlayBottomPanelRenderer' in overlay:
info['size'] = get_text(overlay['thumbnailOverlayBottomPanelRenderer']['text'])
# show renderers don't have playlistId, have to dig into the url to get it
try:
info['id'] = renderer['navigationEndpoint']['watchEndpoint']['playlistId']
except KeyError:
pass
for key, node in renderer.items():
if key in ('longBylineText', 'shortBylineText'):
info['author'] = get_text(node)
try:
info['author_url'] = get_url(node)
except KeyError:
pass
# show renderers don't have thumbnail key at top level, dig into thumbnailRenderer
elif key == 'thumbnailRenderer' and 'showCustomThumbnailRenderer' in node:
info['thumbnail'] = node['showCustomThumbnailRenderer']['thumbnail']['thumbnails'][0]['url']
else:
try:
simple_key, function = dispatch[key]
except KeyError:
continue
info[simple_key] = function(node)
if info['type'] == 'video' and 'duration' not in info:
info['duration'] = 'Live'
return info
except KeyError:
print(renderer)
raise
def parse_info_prepare_for_html(renderer, additional_info={}):
item = renderer_info(renderer, additional_info)
prefix_urls(item)
add_extra_html_info(item)
return item

11
youtube/yt_data_extract/__init__.py Normal file
View File

@ -0,0 +1,11 @@
from .common import (get, multi_get, deep_get, multi_deep_get,
liberal_update, conservative_update, remove_redirect, normalize_url,
extract_str, extract_formatted_text, extract_int, extract_approx_int,
extract_date, extract_item_info, extract_items, extract_response)
from .everything_else import (extract_channel_info, extract_search_info,
extract_playlist_metadata, extract_playlist_info, extract_comments_info)
from .watch_extraction import (extract_watch_info, get_caption_url,
update_with_age_restricted_info, requires_decryption,
extract_decryption_function, decrypt_signatures)

415
youtube/yt_data_extract/common.py Normal file
View File

@ -0,0 +1,415 @@
import re
import urllib.parse
import collections
def get(object, key, default=None, types=()):
'''Like dict.get(), but returns default if the result doesn't match one of the types.
Also works for indexing lists.'''
try:
result = object[key]
except (TypeError, IndexError, KeyError):
return default
if not types or isinstance(result, types):
return result
else:
return default
def multi_get(object, *keys, default=None, types=()):
'''Like get, but try other keys if the first fails'''
for key in keys:
try:
result = object[key]
except (TypeError, IndexError, KeyError):
pass
else:
if not types or isinstance(result, types):
return result
else:
continue
return default
def deep_get(object, *keys, default=None, types=()):
'''Like dict.get(), but for nested dictionaries/sequences, supporting keys or indices.
Last argument is the default value to use in case of any IndexErrors or KeyErrors.
If types is given and the result doesn't match one of those types, default is returned'''
try:
for key in keys:
object = object[key]
except (TypeError, IndexError, KeyError):
return default
else:
if not types or isinstance(object, types):
return object
else:
return default
def multi_deep_get(object, *key_sequences, default=None, types=()):
'''Like deep_get, but can try different key sequences in case one fails.
Return default if all of them fail. key_sequences is a list of lists'''
for key_sequence in key_sequences:
_object = object
try:
for key in key_sequence:
_object = _object[key]
except (TypeError, IndexError, KeyError):
pass
else:
if not types or isinstance(_object, types):
return _object
else:
continue
return default
def liberal_update(obj, key, value):
'''Updates obj[key] with value as long as value is not None.
Ensures obj[key] will at least get a value of None, however'''
if (value is not None) or (key not in obj):
obj[key] = value
def conservative_update(obj, key, value):
'''Only updates obj if it doesn't have key or obj[key] is None'''
if obj.get(key) is None:
obj[key] = value
def remove_redirect(url):
if re.fullmatch(r'(((https?:)?//)?(www.)?youtube.com)?/redirect\?.*', url) is not None: # youtube puts these on external links to do tracking
query_string = url[url.find('?')+1: ]
return urllib.parse.parse_qs(query_string)['q'][0]
return url
youtube_url_re = re.compile(r'^(?:(?:(?:https?:)?//)?(?:www\.)?youtube\.com)?(/.*)$')
def normalize_url(url):
if url is None:
return None
match = youtube_url_re.fullmatch(url)
if match is None:
raise Exception()
return 'https://www.youtube.com' + match.group(1)
def _recover_urls(runs):
for run in runs:
url = deep_get(run, 'navigationEndpoint', 'urlEndpoint', 'url')
text = run.get('text', '')
# second condition is necessary because youtube makes other things into urls, such as hashtags, which we want to keep as text
if url is not None and (text.startswith('http://') or text.startswith('https://')):
url = remove_redirect(url)
run['url'] = url
run['text'] = url # youtube truncates the url text, use actual url instead
def extract_str(node, default=None, recover_urls=False):
'''default is the value returned if the extraction fails. If recover_urls is true, will attempt to fix Youtube's truncation of url text (most prominently seen in descriptions)'''
if isinstance(node, str):
return node
try:
return node['simpleText']
except (KeyError, TypeError):
pass
if isinstance(node, dict) and 'runs' in node:
if recover_urls:
_recover_urls(node['runs'])
return ''.join(text_run.get('text', '') for text_run in node['runs'])
return default
def extract_formatted_text(node):
if not node:
return []
if 'runs' in node:
_recover_urls(node['runs'])
return node['runs']
elif 'simpleText' in node:
return [{'text': node['simpleText']}]
return []
def extract_int(string, default=None):
if isinstance(string, int):
return string
if not isinstance(string, str):
string = extract_str(string)
if not string:
return default
match = re.search(r'(\d+)', string.replace(',', ''))
if match is None:
return default
try:
return int(match.group(1))
except ValueError:
return default
def extract_approx_int(string):
'''e.g. "15M" from "15M subscribers"'''
if not isinstance(string, str):
string = extract_str(string)
if not string:
return None
match = re.search(r'(\d+[KMBTkmbt])', string.replace(',', ''))
if match is None:
return None
return match.group(1)
def extract_date(date_text):
'''Input: "Mar 9, 2019". Output: "2019-3-9"'''
if date_text is None:
return None
date_text = date_text.replace(',', '').lower()
parts = date_text.split()
if len(parts) >= 3:
month, day, year = parts[-3:]
month = month_abbreviations.get(month[0:3]) # slicing in case they start writing out the full month name
if month and (re.fullmatch(r'\d\d?', day) is not None) and (re.fullmatch(r'\d{4}', year) is not None):
return year + '-' + month + '-' + day
def check_missing_keys(object, *key_sequences):
for key_sequence in key_sequences:
_object = object
try:
for key in key_sequence:
_object = _object[key]
except (KeyError, IndexError, TypeError):
return 'Could not find ' + key
return None
def extract_item_info(item, additional_info={}):
if not item:
return {'error': 'No item given'}
type = get(list(item.keys()), 0)
if not type:
return {'error': 'Could not find type'}
item = item[type]
info = {'error': None}
if type in ('itemSectionRenderer', 'compactAutoplayRenderer'):
return extract_item_info(deep_get(item, 'contents', 0), additional_info)
if type in ('movieRenderer', 'clarificationRenderer'):
info['type'] = 'unsupported'
return info
info.update(additional_info)
# type looks like e.g. 'compactVideoRenderer' or 'gridVideoRenderer'
# camelCase split, https://stackoverflow.com/a/37697078
type_parts = [s.lower() for s in re.sub(r'([A-Z][a-z]+)', r' \1', type).split()]
if len(type_parts) < 2:
info['type'] = 'unsupported'
return
primary_type = type_parts[-2]
if primary_type == 'video':
info['type'] = 'video'
elif primary_type in ('playlist', 'radio', 'show'):
info['type'] = 'playlist'
elif primary_type == 'channel':
info['type'] = 'channel'
else:
info['type'] = 'unsupported'
info['title'] = extract_str(item.get('title'))
info['author'] = extract_str(multi_get(item, 'longBylineText', 'shortBylineText', 'ownerText'))
info['author_id'] = extract_str(multi_deep_get(item,
['longBylineText', 'runs', 0, 'navigationEndpoint', 'browseEndpoint', 'browseId'],
['shortBylineText', 'runs', 0, 'navigationEndpoint', 'browseEndpoint', 'browseId'],
['ownerText', 'runs', 0, 'navigationEndpoint', 'browseEndpoint', 'browseId']
))
info['author_url'] = ('https://www.youtube.com/channel/' + info['author_id']) if info['author_id'] else None
info['description'] = extract_formatted_text(multi_get(item, 'descriptionSnippet', 'descriptionText'))
info['thumbnail'] = multi_deep_get(item,
['thumbnail', 'thumbnails', 0, 'url'], # videos
['thumbnails', 0, 'thumbnails', 0, 'url'], # playlists
['thumbnailRenderer', 'showCustomThumbnailRenderer', 'thumbnail', 'thumbnails', 0, 'url'], # shows
)
info['badges'] = []
for badge_node in multi_get(item, 'badges', 'ownerBadges', default=()):
badge = deep_get(badge_node, 'metadataBadgeRenderer', 'label')
if badge:
info['badges'].append(badge)
if primary_type in ('video', 'playlist'):
info['time_published'] = extract_str(item.get('publishedTimeText'))
if primary_type == 'video':
info['id'] = item.get('videoId')
info['view_count'] = extract_int(item.get('viewCountText'))
# dig into accessibility data to get view_count for videos marked as recommended, and to get time_published
accessibility_label = deep_get(item, 'title', 'accessibility', 'accessibilityData', 'label', default='')
timestamp = re.search(r'(\d+ \w+ ago)', accessibility_label)
if timestamp:
conservative_update(info, 'time_published', timestamp.group(1))
view_count = re.search(r'(\d+) views', accessibility_label.replace(',', ''))
if view_count:
conservative_update(info, 'view_count', int(view_count.group(1)))
if info['view_count']:
info['approx_view_count'] = '{:,}'.format(info['view_count'])
else:
info['approx_view_count'] = extract_approx_int(multi_get(item, 'shortViewCountText'))
info['duration'] = extract_str(item.get('lengthText'))
elif primary_type == 'playlist':
info['id'] = item.get('playlistId')
info['video_count'] = extract_int(item.get('videoCount'))
elif primary_type == 'channel':
info['id'] = item.get('channelId')
info['approx_subscriber_count'] = extract_approx_int(item.get('subscriberCountText'))
elif primary_type == 'show':
info['id'] = deep_get(item, 'navigationEndpoint', 'watchEndpoint', 'playlistId')
if primary_type in ('playlist', 'channel'):
conservative_update(info, 'video_count', extract_int(item.get('videoCountText')))
for overlay in item.get('thumbnailOverlays', []):
conservative_update(info, 'duration', extract_str(deep_get(
overlay, 'thumbnailOverlayTimeStatusRenderer', 'text'
)))
# show renderers don't have videoCountText
conservative_update(info, 'video_count', extract_int(deep_get(
overlay, 'thumbnailOverlayBottomPanelRenderer', 'text'
)))
return info
def extract_response(polymer_json):
'''return response, error'''
response = multi_deep_get(polymer_json, [1, 'response'], ['response'], default=None, types=dict)
if response is None:
return None, 'Failed to extract response'
else:
return response, None
list_types = {
'sectionListRenderer',
'itemSectionRenderer',
'gridRenderer',
'playlistVideoListRenderer',
}
item_types = {
'movieRenderer',
'didYouMeanRenderer',
'showingResultsForRenderer',
'videoRenderer',
'compactVideoRenderer',
'compactAutoplayRenderer',
'gridVideoRenderer',
'playlistVideoRenderer',
'playlistRenderer',
'compactPlaylistRenderer',
'gridPlaylistRenderer',
'radioRenderer',
'compactRadioRenderer',
'gridRadioRenderer',
'showRenderer',
'compactShowRenderer',
'gridShowRenderer',
'channelRenderer',
'compactChannelRenderer',
'gridChannelRenderer',
'channelAboutFullMetadataRenderer',
}
def _traverse_browse_renderer(renderer):
for tab in get(renderer, 'tabs', (), types=(list, tuple)):
tab_renderer = multi_deep_get(tab, ['tabRenderer'], ['expandableTabRenderer'], default=None, types=dict)
if tab_renderer is None:
continue
if tab_renderer.get('selected', False):
return get(tab_renderer, 'content', {}, types=(dict))
print('Could not find tab with content')
return {}
def _traverse_standard_list(renderer):
renderer_list = multi_deep_get(renderer, ['contents'], ['items'], default=(), types=(list, tuple))
continuation = deep_get(renderer, 'continuations', 0, 'nextContinuationData', 'continuation')
return renderer_list, continuation
# these renderers contain one inside them
nested_renderer_dispatch = {
'singleColumnBrowseResultsRenderer': _traverse_browse_renderer,
'twoColumnBrowseResultsRenderer': _traverse_browse_renderer,
'twoColumnSearchResultsRenderer': lambda renderer: get(renderer, 'primaryContents', {}, types=dict),
}
# these renderers contain a list of renderers inside them
nested_renderer_list_dispatch = {
'sectionListRenderer': _traverse_standard_list,
'itemSectionRenderer': _traverse_standard_list,
'gridRenderer': _traverse_standard_list,
'playlistVideoListRenderer': _traverse_standard_list,
'singleColumnWatchNextResults': lambda r: (deep_get(r, 'results', 'results', 'contents', default=[], types=(list, tuple)), None),
}
def extract_items(response, item_types=item_types):
'''return items, ctoken'''
if 'continuationContents' in response:
# always has just the one [something]Continuation key, but do this just in case they add some tracking key or something
for key, renderer_continuation in get(response, 'continuationContents', {}, types=dict).items():
if key.endswith('Continuation'): # e.g. commentSectionContinuation, playlistVideoListContinuation
items = multi_deep_get(renderer_continuation, ['contents'], ['items'], default=[], types=(list, tuple))
ctoken = deep_get(renderer_continuation, 'continuations', 0, 'nextContinuationData', 'continuation', default=None, types=str)
return items, ctoken
return [], None
elif 'contents' in response:
ctoken = None
items = []
iter_stack = collections.deque()
current_iter = iter(())
renderer = get(response, 'contents', {}, types=dict)
while True:
# mode 1: dig into the current renderer
# Will stay in mode 1 (via continue) if a new renderer is found inside this one
# Otherwise, after finding that it is an item renderer,
# contains a list, or contains nothing,
# falls through into mode 2 to get a new renderer
if len(renderer) != 0:
key, value = list(renderer.items())[0]
# has a list in it, add it to the iter stack
if key in nested_renderer_list_dispatch:
renderer_list, continuation = nested_renderer_list_dispatch[key](value)
if renderer_list:
iter_stack.append(current_iter)
current_iter = iter(renderer_list)
if continuation:
ctoken = continuation
# new renderer nested inside this one
elif key in nested_renderer_dispatch:
renderer = nested_renderer_dispatch[key](value)
continue # back to mode 1
# the renderer is an item
elif key in item_types:
items.append(renderer)
# mode 2: get a new renderer by iterating.
# goes up the stack for an iterator if one has been exhausted
while current_iter is not None:
try:
renderer = current_iter.__next__()
break
except StopIteration:
try:
current_iter = iter_stack.pop() # go back up the stack
except IndexError:
return items, ctoken
else:
return [], None

273
youtube/yt_data_extract/everything_else.py Normal file
View File

@ -0,0 +1,273 @@
from .common import (get, multi_get, deep_get, multi_deep_get,
liberal_update, conservative_update, remove_redirect, normalize_url,
extract_str, extract_formatted_text, extract_int, extract_approx_int,
extract_date, check_missing_keys, extract_item_info, extract_items,
extract_response)
from youtube import proto
import re
import urllib
from math import ceil
def extract_channel_info(polymer_json, tab):
response, err = extract_response(polymer_json)
if err:
return {'error': err}
try:
microformat = response['microformat']['microformatDataRenderer']
# channel doesn't exist or was terminated
# example terminated channel: https://www.youtube.com/channel/UCnKJeK_r90jDdIuzHXC0Org
except KeyError:
if 'alerts' in response and len(response['alerts']) > 0:
return {'error': ' '.join(alert['alertRenderer']['text']['simpleText'] for alert in response['alerts']) }
elif 'errors' in response['responseContext']:
for error in response['responseContext']['errors']['error']:
if error['code'] == 'INVALID_VALUE' and error['location'] == 'browse_id':
return {'error': 'This channel does not exist'}
return {'error': 'Failure getting microformat'}
info = {'error': None}
info['current_tab'] = tab
# stuff from microformat (info given by youtube for every page on channel)
info['short_description'] = microformat['description']
info['channel_name'] = microformat['title']
info['avatar'] = microformat['thumbnail']['thumbnails'][0]['url']
channel_url = microformat['urlCanonical'].rstrip('/')
channel_id = channel_url[channel_url.rfind('/')+1:]
info['channel_id'] = channel_id
info['channel_url'] = 'https://www.youtube.com/channel/' + channel_id
info['items'] = []
# empty channel
if 'contents' not in response and 'continuationContents' not in response:
return info
items, _ = extract_items(response)
if tab in ('videos', 'playlists', 'search'):
additional_info = {'author': info['channel_name'], 'author_url': 'https://www.youtube.com/channel/' + channel_id}
info['items'] = [extract_item_info(renderer, additional_info) for renderer in items]
elif tab == 'about':
for item in items:
try:
channel_metadata = item['channelAboutFullMetadataRenderer']
break
except KeyError:
pass
else:
info['error'] = 'Could not find channelAboutFullMetadataRenderer'
return info
info['links'] = []
for link_json in channel_metadata.get('primaryLinks', ()):
url = remove_redirect(link_json['navigationEndpoint']['urlEndpoint']['url'])
text = extract_str(link_json['title'])
info['links'].append( (text, url) )
info['stats'] = []
for stat_name in ('subscriberCountText', 'joinedDateText', 'viewCountText', 'country'):
try:
stat = channel_metadata[stat_name]
except KeyError:
continue
info['stats'].append(extract_str(stat))
if 'description' in channel_metadata:
info['description'] = extract_str(channel_metadata['description'])
else:
info['description'] = ''
else:
raise NotImplementedError('Unknown or unsupported channel tab: ' + tab)
return info
def extract_search_info(polymer_json):
response, err = extract_response(polymer_json)
if err:
return {'error': err}
info = {'error': None}
info['estimated_results'] = int(response['estimatedResults'])
info['estimated_pages'] = ceil(info['estimated_results']/20)
results, _ = extract_items(response)
info['items'] = []
info['corrections'] = {'type': None}
for renderer in results:
type = list(renderer.keys())[0]
if type == 'shelfRenderer':
continue
if type == 'didYouMeanRenderer':
renderer = renderer[type]
info['corrections'] = {
'type': 'did_you_mean',
'corrected_query': renderer['correctedQueryEndpoint']['searchEndpoint']['query'],
'corrected_query_text': renderer['correctedQuery']['runs'],
}
continue
if type == 'showingResultsForRenderer':
renderer = renderer[type]
info['corrections'] = {
'type': 'showing_results_for',
'corrected_query_text': renderer['correctedQuery']['runs'],
'original_query_text': renderer['originalQuery']['simpleText'],
}
continue
i_info = extract_item_info(renderer)
if i_info.get('type') != 'unsupported':
info['items'].append(i_info)
return info
def extract_playlist_metadata(polymer_json):
response, err = extract_response(polymer_json)
if err:
return {'error': err}
metadata = {'error': None}
header = deep_get(response, 'header', 'playlistHeaderRenderer', default={})
metadata['title'] = extract_str(header.get('title'))
metadata['first_video_id'] = deep_get(header, 'playEndpoint', 'watchEndpoint', 'videoId')
first_id = re.search(r'([a-z_\-]{11})', deep_get(header,
'thumbnail', 'thumbnails', 0, 'url', default=''))
if first_id:
conservative_update(metadata, 'first_video_id', first_id.group(1))
if metadata['first_video_id'] is None:
metadata['thumbnail'] = None
else:
metadata['thumbnail'] = 'https://i.ytimg.com/vi/' + metadata['first_video_id'] + '/mqdefault.jpg'
metadata['video_count'] = extract_int(header.get('numVideosText'))
metadata['description'] = extract_str(header.get('descriptionText'), default='')
metadata['author'] = extract_str(header.get('ownerText'))
metadata['author_id'] = multi_deep_get(header,
['ownerText', 'runs', 0, 'navigationEndpoint', 'browseEndpoint', 'browseId'],
['ownerEndpoint', 'browseEndpoint', 'browseId'])
if metadata['author_id']:
metadata['author_url'] = 'https://www.youtube.com/channel/' + metadata['author_id']
else:
metadata['author_url'] = None
metadata['view_count'] = extract_int(header.get('viewCountText'))
metadata['like_count'] = extract_int(header.get('likesCountWithoutLikeText'))
for stat in header.get('stats', ()):
text = extract_str(stat)
if 'videos' in text:
conservative_update(metadata, 'video_count', extract_int(text))
elif 'views' in text:
conservative_update(metadata, 'view_count', extract_int(text))
elif 'updated' in text:
metadata['time_published'] = extract_date(text)
return metadata
def extract_playlist_info(polymer_json):
response, err = extract_response(polymer_json)
if err:
return {'error': err}
info = {'error': None}
first_page = 'continuationContents' not in response
video_list, _ = extract_items(response)
info['items'] = [extract_item_info(renderer) for renderer in video_list]
if first_page:
info['metadata'] = extract_playlist_metadata(polymer_json)
return info
def _ctoken_metadata(ctoken):
result = dict()
params = proto.parse(proto.b64_to_bytes(ctoken))
result['video_id'] = proto.parse(params[2])[2].decode('ascii')
offset_information = proto.parse(params[6])
result['offset'] = offset_information.get(5, 0)
result['is_replies'] = False
if (3 in offset_information) and (2 in proto.parse(offset_information[3])):
result['is_replies'] = True
result['sort'] = None
else:
try:
result['sort'] = proto.parse(offset_information[4])[6]
except KeyError:
result['sort'] = 0
return result
def extract_comments_info(polymer_json):
response, err = extract_response(polymer_json)
if err:
return {'error': err}
info = {'error': None}
url = multi_deep_get(polymer_json, [1, 'url'], ['url'])
if url:
ctoken = urllib.parse.parse_qs(url[url.find('?')+1:])['ctoken'][0]
metadata = _ctoken_metadata(ctoken)
else:
metadata = {}
info['video_id'] = metadata.get('video_id')
info['offset'] = metadata.get('offset')
info['is_replies'] = metadata.get('is_replies')
info['sort'] = metadata.get('sort')
info['video_title'] = None
comments, ctoken = extract_items(response)
info['comments'] = []
info['ctoken'] = ctoken
for comment in comments:
comment_info = {}
if 'commentThreadRenderer' in comment: # top level comments
conservative_update(info, 'is_replies', False)
comment_thread = comment['commentThreadRenderer']
info['video_title'] = extract_str(comment_thread.get('commentTargetTitle'))
if 'replies' not in comment_thread:
comment_info['reply_count'] = 0
else:
comment_info['reply_count'] = extract_int(deep_get(comment_thread,
'replies', 'commentRepliesRenderer', 'moreText'
), default=1) # With 1 reply, the text reads "View reply"
comment_renderer = deep_get(comment_thread, 'comment', 'commentRenderer', default={})
elif 'commentRenderer' in comment: # replies
comment_info['reply_count'] = 0 # replyCount, below, not present for replies even if the reply has further replies to it
conservative_update(info, 'is_replies', True)
comment_renderer = comment['commentRenderer']
else:
comment_renderer = {}
# These 3 are sometimes absent, likely because the channel was deleted
comment_info['author'] = extract_str(comment_renderer.get('authorText'))
comment_info['author_url'] = deep_get(comment_renderer,
'authorEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')
comment_info['author_id'] = deep_get(comment_renderer,
'authorEndpoint', 'browseEndpoint', 'browseId')
comment_info['author_avatar'] = deep_get(comment_renderer,
'authorThumbnail', 'thumbnails', 0, 'url')
comment_info['id'] = comment_renderer.get('commentId')
comment_info['text'] = extract_formatted_text(comment_renderer.get('contentText'))
comment_info['time_published'] = extract_str(comment_renderer.get('publishedTimeText'))
comment_info['like_count'] = comment_renderer.get('likeCount')
liberal_update(comment_info, 'reply_count', comment_renderer.get('replyCount'))
info['comments'].append(comment_info)
return info

545
youtube/yt_data_extract/watch_extraction.py Normal file
View File

@ -0,0 +1,545 @@
from .common import (get, multi_get, deep_get, multi_deep_get,
liberal_update, conservative_update, remove_redirect, normalize_url,
extract_str, extract_formatted_text, extract_int, extract_approx_int,
extract_date, check_missing_keys, extract_item_info, extract_items,
extract_response)
import json
import urllib.parse
import traceback
import re
# from https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/extractor/youtube.py
_formats = {
'5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'audio_bitrate': 64, 'vcodec': 'h263'},
'6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'audio_bitrate': 64, 'vcodec': 'h263'},
'13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
'17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'audio_bitrate': 24, 'vcodec': 'mp4v'},
'18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'audio_bitrate': 96, 'vcodec': 'h264'},
'22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'audio_bitrate': 192, 'vcodec': 'h264'},
'34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
# itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), audio_bitrate varies as well
'36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
'37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'audio_bitrate': 192, 'vcodec': 'h264'},
'38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'audio_bitrate': 192, 'vcodec': 'h264'},
'43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'audio_bitrate': 128, 'vcodec': 'vp8'},
'44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'audio_bitrate': 128, 'vcodec': 'vp8'},
'45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'audio_bitrate': 192, 'vcodec': 'vp8'},
'46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'audio_bitrate': 192, 'vcodec': 'vp8'},
'59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
# 3D videos
'82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'audio_bitrate': 192, 'vcodec': 'h264'},
'85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'audio_bitrate': 192, 'vcodec': 'h264'},
'100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'audio_bitrate': 128, 'vcodec': 'vp8'},
'101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'audio_bitrate': 192, 'vcodec': 'vp8'},
'102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'audio_bitrate': 192, 'vcodec': 'vp8'},
# Apple HTTP Live Streaming
'91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 48, 'vcodec': 'h264'},
'92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 48, 'vcodec': 'h264'},
'93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 128, 'vcodec': 'h264'},
'95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 256, 'vcodec': 'h264'},
'96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 256, 'vcodec': 'h264'},
'132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 48, 'vcodec': 'h264'},
'151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'audio_bitrate': 24, 'vcodec': 'h264'},
# DASH mp4 video
'133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
'134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
'135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
'137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
'138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
'160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
'212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
'298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
'299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
'266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
# Dash mp4 audio
'139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'audio_bitrate': 48, 'container': 'm4a_dash'},
'140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'audio_bitrate': 128, 'container': 'm4a_dash'},
'141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'audio_bitrate': 256, 'container': 'm4a_dash'},
'256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
'258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
'325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
'328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
# Dash webm
'167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
'278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
'242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
# itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
'272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
'303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
'308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
'313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
'315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
# Dash webm audio
'171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'audio_bitrate': 128},
'172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'audio_bitrate': 256},
# Dash webm audio with opus inside
'249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'audio_bitrate': 50},
'250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'audio_bitrate': 70},
'251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'audio_bitrate': 160},
# RTMP (unnamed)
'_rtmp': {'protocol': 'rtmp'},
# av01 video only formats sometimes served with "unknown" codecs
'394': {'vcodec': 'av01.0.05M.08'},
'395': {'vcodec': 'av01.0.05M.08'},
'396': {'vcodec': 'av01.0.05M.08'},
'397': {'vcodec': 'av01.0.05M.08'},
}
def _extract_metadata_row_info(video_renderer_info):
# extract category and music list
info = {
'category': None,
'music_list': [],
}
current_song = {}
for row in deep_get(video_renderer_info, 'metadataRowContainer', 'metadataRowContainerRenderer', 'rows', default=[]):
row_title = extract_str(deep_get(row, 'metadataRowRenderer', 'title'), default='')
row_content = extract_str(deep_get(row, 'metadataRowRenderer', 'contents', 0))
if row_title == 'Category':
info['category'] = row_content
elif row_title in ('Song', 'Music'):
if current_song:
info['music_list'].append(current_song)
current_song = {'title': row_content}
elif row_title == 'Artist':
current_song['artist'] = row_content
elif row_title == 'Album':
current_song['album'] = row_content
elif row_title == 'Writers':
current_song['writers'] = row_content
elif row_title.startswith('Licensed'):
current_song['licensor'] = row_content
if current_song:
info['music_list'].append(current_song)
return info
def _extract_watch_info_mobile(top_level):
info = {}
microformat = deep_get(top_level, 'playerResponse', 'microformat', 'playerMicroformatRenderer', default={})
family_safe = microformat.get('isFamilySafe')
if family_safe is None:
info['age_restricted'] = None
else:
info['age_restricted'] = not family_safe
info['allowed_countries'] = microformat.get('availableCountries', [])
info['time_published'] = microformat.get('publishDate')
response = top_level.get('response', {})
# video info from metadata renderers
items, _ = extract_items(response, item_types={'slimVideoMetadataRenderer'})
if items:
video_info = items[0]['slimVideoMetadataRenderer']
else:
print('Failed to extract video metadata')
video_info = {}
info.update(_extract_metadata_row_info(video_info))
info['description'] = extract_str(video_info.get('description'), recover_urls=True)
info['view_count'] = extract_int(extract_str(video_info.get('expandedSubtitle')))
info['author'] = extract_str(deep_get(video_info, 'owner', 'slimOwnerRenderer', 'title'))
info['author_id'] = deep_get(video_info, 'owner', 'slimOwnerRenderer', 'navigationEndpoint', 'browseEndpoint', 'browseId')
info['title'] = extract_str(video_info.get('title'))
info['live'] = 'watching' in extract_str(video_info.get('expandedSubtitle'), default='')
info['unlisted'] = False
for badge in video_info.get('badges', []):
if deep_get(badge, 'metadataBadgeRenderer', 'label') == 'Unlisted':
info['unlisted'] = True
info['like_count'] = None
info['dislike_count'] = None
if not info['time_published']:
info['time_published'] = extract_date(extract_str(video_info.get('dateText', None)))
for button in video_info.get('buttons', ()):
button_renderer = button.get('slimMetadataToggleButtonRenderer', {})
# all the digits can be found in the accessibility data
count = extract_int(deep_get(button_renderer, 'button', 'toggleButtonRenderer', 'defaultText', 'accessibility', 'accessibilityData', 'label'))
# this count doesn't have all the digits, it's like 53K for instance
dumb_count = extract_int(extract_str(deep_get(button_renderer, 'button', 'toggleButtonRenderer', 'defaultText')))
# the accessibility text will be "No likes" or "No dislikes" or something like that, but dumb count will be 0
if dumb_count == 0:
count = 0
if 'isLike' in button_renderer:
info['like_count'] = count
elif 'isDislike' in button_renderer:
info['dislike_count'] = count
# comment section info
items, _ = extract_items(response, item_types={'commentSectionRenderer'})
if items:
comment_info = items[0]['commentSectionRenderer']
comment_count_text = extract_str(deep_get(comment_info, 'header', 'commentSectionHeaderRenderer', 'countText'))
if comment_count_text == 'Comments': # just this with no number, means 0 comments
info['comment_count'] = 0
else:
info['comment_count'] = extract_int(comment_count_text)
info['comments_disabled'] = False
else: # no comment section present means comments are disabled
info['comment_count'] = 0
info['comments_disabled'] = True
# check for limited state
items, _ = extract_items(response, item_types={'limitedStateMessageRenderer'})
if items:
info['limited_state'] = True
else:
info['limited_state'] = False
# related videos
related, _ = extract_items(response)
info['related_videos'] = [extract_item_info(renderer) for renderer in related]
return info
month_abbreviations = {'jan':'1', 'feb':'2', 'mar':'3', 'apr':'4', 'may':'5', 'jun':'6', 'jul':'7', 'aug':'8', 'sep':'9', 'oct':'10', 'nov':'11', 'dec':'12'}
def _extract_watch_info_desktop(top_level):
info = {
'comment_count': None,
'comments_disabled': None,
'allowed_countries': None,
'limited_state': None,
}
video_info = {}
for renderer in deep_get(top_level, 'response', 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', default=()):
if renderer and list(renderer.keys())[0] in ('videoPrimaryInfoRenderer', 'videoSecondaryInfoRenderer'):
video_info.update(list(renderer.values())[0])
info.update(_extract_metadata_row_info(video_info))
info['description'] = extract_str(video_info.get('description', None), recover_urls=True)
info['time_published'] = extract_date(extract_str(video_info.get('dateText', None)))
likes_dislikes = deep_get(video_info, 'sentimentBar', 'sentimentBarRenderer', 'tooltip', default='').split('/')
if len(likes_dislikes) == 2:
info['like_count'] = extract_int(likes_dislikes[0])
info['dislike_count'] = extract_int(likes_dislikes[1])
else:
info['like_count'] = None
info['dislike_count'] = None
info['title'] = extract_str(video_info.get('title', None))
info['author'] = extract_str(deep_get(video_info, 'owner', 'videoOwnerRenderer', 'title'))
info['author_id'] = deep_get(video_info, 'owner', 'videoOwnerRenderer', 'navigationEndpoint', 'browseEndpoint', 'browseId')
info['view_count'] = extract_int(extract_str(deep_get(video_info, 'viewCount', 'videoViewCountRenderer', 'viewCount')))
related = deep_get(top_level, 'response', 'contents', 'twoColumnWatchNextResults', 'secondaryResults', 'secondaryResults', 'results', default=[])
info['related_videos'] = [extract_item_info(renderer) for renderer in related]
return info
def _extract_formats(info, player_response):
streaming_data = player_response.get('streamingData', {})
yt_formats = streaming_data.get('formats', []) + streaming_data.get('adaptiveFormats', [])
info['formats'] = []
for yt_fmt in yt_formats:
fmt = {}
fmt['ext'] = None
fmt['audio_bitrate'] = None
fmt['acodec'] = None
fmt['vcodec'] = None
fmt['width'] = yt_fmt.get('width')
fmt['height'] = yt_fmt.get('height')
fmt['file_size'] = yt_fmt.get('contentLength')
fmt['audio_sample_rate'] = yt_fmt.get('audioSampleRate')
fmt['fps'] = yt_fmt.get('fps')
cipher = dict(urllib.parse.parse_qsl(yt_fmt.get('cipher', '')))
if cipher:
fmt['url'] = cipher.get('url')
else:
fmt['url'] = yt_fmt.get('url')
fmt['s'] = cipher.get('s')
fmt['sp'] = cipher.get('sp')
fmt.update(_formats.get(str(yt_fmt.get('itag')), {}))
info['formats'].append(fmt)
def _extract_playability_error(info, player_response, error_prefix=''):
if info['formats']:
info['playability_status'] = None
info['playability_error'] = None
return
playability_status = deep_get(player_response, 'playabilityStatus', 'status', default=None)
info['playability_status'] = playability_status
playability_reason = extract_str(multi_deep_get(player_response,
['playabilityStatus', 'reason'],
['playabilityStatus', 'errorScreen', 'playerErrorMessageRenderer', 'reason'],
default='Could not find playability error')
)
if playability_status not in (None, 'OK'):
info['playability_error'] = error_prefix + playability_reason
else:
info['playability_error'] = error_prefix + 'Unknown playability error'
SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
def extract_watch_info(polymer_json):
info = {'playability_error': None, 'error': None}
if isinstance(polymer_json, dict):
top_level = polymer_json
elif isinstance(polymer_json, (list, tuple)):
top_level = {}
for page_part in polymer_json:
if not isinstance(page_part, dict):
return {'error': 'Invalid page part'}
top_level.update(page_part)
else:
return {'error': 'Invalid top level polymer data'}
error = check_missing_keys(top_level,
['player', 'args'],
['player', 'assets', 'js'],
['playerResponse'],
)
if error:
info['playability_error'] = error
player_args = deep_get(top_level, 'player', 'args', default={})
player_response = json.loads(player_args['player_response']) if 'player_response' in player_args else {}
# captions
info['automatic_caption_languages'] = []
info['manual_caption_languages'] = []
info['_manual_caption_language_names'] = {} # language name written in that language, needed in some cases to create the url
info['translation_languages'] = []
captions_info = player_response.get('captions', {})
info['_captions_base_url'] = normalize_url(deep_get(captions_info, 'playerCaptionsRenderer', 'baseUrl'))
for caption_track in deep_get(captions_info, 'playerCaptionsTracklistRenderer', 'captionTracks', default=()):
lang_code = caption_track.get('languageCode')
if not lang_code:
continue
if caption_track.get('kind') == 'asr':
info['automatic_caption_languages'].append(lang_code)
else:
info['manual_caption_languages'].append(lang_code)
base_url = caption_track.get('baseUrl', '')
lang_name = deep_get(urllib.parse.parse_qs(urllib.parse.urlparse(base_url).query), 'name', 0)
if lang_name:
info['_manual_caption_language_names'][lang_code] = lang_name
for translation_lang_info in deep_get(captions_info, 'playerCaptionsTracklistRenderer', 'translationLanguages', default=()):
lang_code = translation_lang_info.get('languageCode')
if lang_code:
info['translation_languages'].append(lang_code)
if translation_lang_info.get('isTranslatable') == False:
print('WARNING: Found non-translatable caption language')
# formats
_extract_formats(info, player_response)
# playability errors
_extract_playability_error(info, player_response)
# check age-restriction
info['age_restricted'] = (info['playability_status'] == 'LOGIN_REQUIRED' and info['playability_error'] and ' age' in info['playability_error'])
# base_js (for decryption of signatures)
info['base_js'] = deep_get(top_level, 'player', 'assets', 'js')
if info['base_js']:
info['base_js'] = normalize_url(info['base_js'])
info['player_name'] = get(info['base_js'].split('/'), -2)
else:
info['player_name'] = None
# extract stuff from visible parts of page
mobile = 'singleColumnWatchNextResults' in deep_get(top_level, 'response', 'contents', default={})
if mobile:
info.update(_extract_watch_info_mobile(top_level))
else:
info.update(_extract_watch_info_desktop(top_level))
# stuff from videoDetails. Use liberal_update to prioritize info from videoDetails over existing info
vd = deep_get(top_level, 'playerResponse', 'videoDetails', default={})
liberal_update(info, 'title', extract_str(vd.get('title')))
liberal_update(info, 'duration', extract_int(vd.get('lengthSeconds')))
liberal_update(info, 'view_count', extract_int(vd.get('viewCount')))
# videos with no description have a blank string
liberal_update(info, 'description', vd.get('shortDescription'))
liberal_update(info, 'id', vd.get('videoId'))
liberal_update(info, 'author', vd.get('author'))
liberal_update(info, 'author_id', vd.get('channelId'))
liberal_update(info, 'live', vd.get('isLiveContent'))
conservative_update(info, 'unlisted', not vd.get('isCrawlable', True)) #isCrawlable is false on limited state videos even if they aren't unlisted
liberal_update(info, 'tags', vd.get('keywords', []))
# fallback stuff from microformat
mf = deep_get(top_level, 'playerResponse', 'microformat', 'playerMicroformatRenderer', default={})
conservative_update(info, 'title', extract_str(mf.get('title')))
conservative_update(info, 'duration', extract_int(mf.get('lengthSeconds')))
# this gives the view count for limited state videos
conservative_update(info, 'view_count', extract_int(mf.get('viewCount')))
conservative_update(info, 'description', extract_str(mf.get('description'), recover_urls=True))
conservative_update(info, 'author', mf.get('ownerChannelName'))
conservative_update(info, 'author_id', mf.get('externalChannelId'))
liberal_update(info, 'unlisted', mf.get('isUnlisted'))
liberal_update(info, 'category', mf.get('category'))
liberal_update(info, 'time_published', mf.get('publishDate'))
liberal_update(info, 'time_uploaded', mf.get('uploadDate'))
# other stuff
info['author_url'] = 'https://www.youtube.com/channel/' + info['author_id'] if info['author_id'] else None
return info
def get_caption_url(info, language, format, automatic=False, translation_language=None):
'''Gets the url for captions with the given language and format. If automatic is True, get the automatic captions for that language. If translation_language is given, translate the captions from `language` to `translation_language`. If automatic is true and translation_language is given, the automatic captions will be translated.'''
url = info['_captions_base_url']
url += '&lang=' + language
url += '&fmt=' + format
if automatic:
url += '&kind=asr'
elif language in info['_manual_caption_language_names']:
url += '&name=' + urllib.parse.quote(info['_manual_caption_language_names'][language], safe='')
if translation_language:
url += '&tlang=' + translation_language
return url
def update_with_age_restricted_info(info, video_info_page):
ERROR_PREFIX = 'Error bypassing age-restriction: '
video_info = urllib.parse.parse_qs(video_info_page)
player_response = deep_get(video_info, 'player_response', 0)
if player_response is None:
info['playability_error'] = ERROR_PREFIX + 'Could not find player_response in video_info_page'
return
try:
player_response = json.loads(player_response)
except json.decoder.JSONDecodeError:
traceback.print_exc()
info['playability_error'] = ERROR_PREFIX + 'Failed to parse json response'
return
_extract_formats(info, player_response)
_extract_playability_error(info, player_response, error_prefix=ERROR_PREFIX)
def requires_decryption(info):
return ('formats' in info) and info['formats'] and info['formats'][0]['s']
# adapted from youtube-dl and invidious:
# https://github.com/omarroth/invidious/blob/master/src/invidious/helpers/signatures.cr
decrypt_function_re = re.compile(r'function\(a\)\{(a=a\.split\(""\)[^\}]+)\}')
op_with_arg_re = re.compile(r'[^\.]+\.([^\(]+)\(a,(\d+)\)')
def extract_decryption_function(info, base_js):
'''Insert decryption function into info. Return error string if not successful.
Decryption function is a list of list[2] of numbers.
It is advisable to cache the decryption function (uniquely identified by info['player_name']) so base.js (1 MB) doesn't need to be redownloaded each time'''
info['decryption_function'] = None
decrypt_function_match = decrypt_function_re.search(base_js)
if decrypt_function_match is None:
return 'Could not find decryption function in base.js'
function_body = decrypt_function_match.group(1).split(';')[1:-1]
if not function_body:
return 'Empty decryption function body'
var_name = get(function_body[0].split('.'), 0)
if var_name is None:
return 'Could not find var_name'
var_body_match = re.search(r'var ' + re.escape(var_name) + r'=\{(.*?)\};', base_js, flags=re.DOTALL)
if var_body_match is None:
return 'Could not find var_body'
operations = var_body_match.group(1).replace('\n', '').split('},')
if not operations:
return 'Did not find any definitions in var_body'
operations[-1] = operations[-1][:-1] # remove the trailing '}' since we split by '},' on the others
operation_definitions = {}
for op in operations:
colon_index = op.find(':')
opening_brace_index = op.find('{')
if colon_index == -1 or opening_brace_index == -1:
return 'Could not parse operation'
op_name = op[:colon_index]
op_body = op[opening_brace_index+1:]
if op_body == 'a.reverse()':
operation_definitions[op_name] = 0
elif op_body == 'a.splice(0,b)':
operation_definitions[op_name] = 1
elif op_body.startswith('var c=a[0]'):
operation_definitions[op_name] = 2
else:
return 'Unknown op_body: ' + op_body
decryption_function = []
for op_with_arg in function_body:
match = op_with_arg_re.fullmatch(op_with_arg)
if match is None:
return 'Could not parse operation with arg'
op_name = match.group(1)
if op_name not in operation_definitions:
return 'Unknown op_name: ' + op_name
op_argument = match.group(2)
decryption_function.append([operation_definitions[op_name], int(op_argument)])
info['decryption_function'] = decryption_function
return False
def _operation_2(a, b):
c = a[0]
a[0] = a[b % len(a)]
a[b % len(a)] = c
def decrypt_signatures(info):
'''Applies info['decryption_function'] to decrypt all the signatures. Return err.'''
if not info.get('decryption_function'):
return 'decryption_function not in info'
for format in info['formats']:
if not format['s'] or not format['sp'] or not format['url']:
print('Warning: s, sp, or url not in format')
continue
a = list(format['s'])
for op, argument in info['decryption_function']:
if op == 0:
a.reverse()
elif op == 1:
a = a[argument:]
else:
_operation_2(a, argument)
signature = ''.join(a)
format['url'] += '&' + format['sp'] + '=' + signature
return False

2392
youtube_dl/YoutubeDL.py
View File

File diff suppressed because it is too large Load Diff

481
youtube_dl/__init__.py
View File

@ -1,481 +0,0 @@
#!/usr/bin/env python
# coding: utf-8
from __future__ import unicode_literals
__license__ = 'Public Domain'
import codecs
import io
import os
import random
import sys
from .options import (
parseOpts,
)
from .compat import (
compat_getpass,
compat_shlex_split,
workaround_optparse_bug9161,
)
from .utils import (
DateRange,
decodeOption,
DEFAULT_OUTTMPL,
DownloadError,
expand_path,
match_filter_func,
MaxDownloadsReached,
preferredencoding,
read_batch_urls,
SameFileError,
setproctitle,
std_headers,
write_string,
render_table,
)
from .update import update_self
from .downloader import (
FileDownloader,
)
from .extractor import gen_extractors, list_extractors
from .extractor.adobepass import MSO_INFO
from .YoutubeDL import YoutubeDL
def _real_main(argv=None):
# Compatibility fixes for Windows
if sys.platform == 'win32':
# https://github.com/rg3/youtube-dl/issues/820
codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
workaround_optparse_bug9161()
setproctitle('youtube-dl')
parser, opts, args = parseOpts(argv)
# Set user agent
if opts.user_agent is not None:
std_headers['User-Agent'] = opts.user_agent
# Set referer
if opts.referer is not None:
std_headers['Referer'] = opts.referer
# Custom HTTP headers
if opts.headers is not None:
for h in opts.headers:
if ':' not in h:
parser.error('wrong header formatting, it should be key:value, not "%s"' % h)
key, value = h.split(':', 1)
if opts.verbose:
write_string('[debug] Adding header from command line option %s:%s\n' % (key, value))
std_headers[key] = value
# Dump user agent
if opts.dump_user_agent:
write_string(std_headers['User-Agent'] + '\n', out=sys.stdout)
sys.exit(0)
# Batch file verification
batch_urls = []
if opts.batchfile is not None:
try:
if opts.batchfile == '-':
batchfd = sys.stdin
else:
batchfd = io.open(
expand_path(opts.batchfile),
'r', encoding='utf-8', errors='ignore')
batch_urls = read_batch_urls(batchfd)
if opts.verbose:
write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
except IOError:
sys.exit('ERROR: batch file could not be read')
all_urls = batch_urls + [url.strip() for url in args] # batch_urls are already striped in read_batch_urls
_enc = preferredencoding()
all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
if opts.list_extractors:
for ie in list_extractors(opts.age_limit):
write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '') + '\n', out=sys.stdout)
matchedUrls = [url for url in all_urls if ie.suitable(url)]
for mu in matchedUrls:
write_string(' ' + mu + '\n', out=sys.stdout)
sys.exit(0)
if opts.list_extractor_descriptions:
for ie in list_extractors(opts.age_limit):
if not ie._WORKING:
continue
desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
if desc is False:
continue
if hasattr(ie, 'SEARCH_KEY'):
_SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
_COUNTS = ('', '5', '10', 'all')
desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
write_string(desc + '\n', out=sys.stdout)
sys.exit(0)
if opts.ap_list_mso:
table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]
write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout)
sys.exit(0)
# Conflicting, missing and erroneous options
if opts.usenetrc and (opts.username is not None or opts.password is not None):
parser.error('using .netrc conflicts with giving username/password')
if opts.password is not None and opts.username is None:
parser.error('account username missing\n')
if opts.ap_password is not None and opts.ap_username is None:
parser.error('TV Provider account username missing\n')
if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid):
parser.error('using output template conflicts with using title, video ID or auto number')
if opts.autonumber_size is not None:
if opts.autonumber_size <= 0:
parser.error('auto number size must be positive')
if opts.autonumber_start is not None:
if opts.autonumber_start < 0:
parser.error('auto number start must be positive or 0')
if opts.usetitle and opts.useid:
parser.error('using title conflicts with using video ID')
if opts.username is not None and opts.password is None:
opts.password = compat_getpass('Type account password and press [Return]: ')
if opts.ap_username is not None and opts.ap_password is None:
opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
if opts.ratelimit is not None:
numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
if numeric_limit is None:
parser.error('invalid rate limit specified')
opts.ratelimit = numeric_limit
if opts.min_filesize is not None:
numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
if numeric_limit is None:
parser.error('invalid min_filesize specified')
opts.min_filesize = numeric_limit
if opts.max_filesize is not None:
numeric_limit = FileDownloader.parse_bytes(opts.max_filesize)
if numeric_limit is None:
parser.error('invalid max_filesize specified')
opts.max_filesize = numeric_limit
if opts.sleep_interval is not None:
if opts.sleep_interval < 0:
parser.error('sleep interval must be positive or 0')
if opts.max_sleep_interval is not None:
if opts.max_sleep_interval < 0:
parser.error('max sleep interval must be positive or 0')
if opts.max_sleep_interval < opts.sleep_interval:
parser.error('max sleep interval must be greater than or equal to min sleep interval')
else:
opts.max_sleep_interval = opts.sleep_interval
if opts.ap_mso and opts.ap_mso not in MSO_INFO:
parser.error('Unsupported TV Provider, use --ap-list-mso to get a list of supported TV Providers')
def parse_retries(retries):
if retries in ('inf', 'infinite'):
parsed_retries = float('inf')
else:
try:
parsed_retries = int(retries)
except (TypeError, ValueError):
parser.error('invalid retry count specified')
return parsed_retries
if opts.retries is not None:
opts.retries = parse_retries(opts.retries)
if opts.fragment_retries is not None:
opts.fragment_retries = parse_retries(opts.fragment_retries)
if opts.buffersize is not None:
numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
if numeric_buffersize is None:
parser.error('invalid buffer size specified')
opts.buffersize = numeric_buffersize
if opts.http_chunk_size is not None:
numeric_chunksize = FileDownloader.parse_bytes(opts.http_chunk_size)
if not numeric_chunksize:
parser.error('invalid http chunk size specified')
opts.http_chunk_size = numeric_chunksize
if opts.playliststart <= 0:
raise ValueError('Playlist start must be positive')
if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
raise ValueError('Playlist end must be greater than playlist start')
if opts.extractaudio:
if opts.audioformat not in ['best', 'aac', 'flac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
parser.error('invalid audio format specified')
if opts.audioquality:
opts.audioquality = opts.audioquality.strip('k').strip('K')
if not opts.audioquality.isdigit():
parser.error('invalid audio quality specified')
if opts.recodevideo is not None:
if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']:
parser.error('invalid video recode format specified')
if opts.convertsubtitles is not None:
if opts.convertsubtitles not in ['srt', 'vtt', 'ass', 'lrc']:
parser.error('invalid subtitle format specified')
if opts.date is not None:
date = DateRange.day(opts.date)
else:
date = DateRange(opts.dateafter, opts.datebefore)
# Do not download videos when there are audio-only formats
if opts.extractaudio and not opts.keepvideo and opts.format is None:
opts.format = 'bestaudio/best'
# --all-sub automatically sets --write-sub if --write-auto-sub is not given
# this was the old behaviour if only --all-sub was given.
if opts.allsubtitles and not opts.writeautomaticsub:
opts.writesubtitles = True
outtmpl = ((opts.outtmpl is not None and opts.outtmpl) or
(opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s') or
(opts.format == '-1' and '%(id)s-%(format)s.%(ext)s') or
(opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s') or
(opts.usetitle and '%(title)s-%(id)s.%(ext)s') or
(opts.useid and '%(id)s.%(ext)s') or
(opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s') or
DEFAULT_OUTTMPL)
if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
parser.error('Cannot download a video and extract audio into the same'
' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
' template'.format(outtmpl))
any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
any_printing = opts.print_json
download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
# PostProcessors
postprocessors = []
if opts.metafromtitle:
postprocessors.append({
'key': 'MetadataFromTitle',
'titleformat': opts.metafromtitle
})
if opts.extractaudio:
postprocessors.append({
'key': 'FFmpegExtractAudio',
'preferredcodec': opts.audioformat,
'preferredquality': opts.audioquality,
'nopostoverwrites': opts.nopostoverwrites,
})
if opts.recodevideo:
postprocessors.append({
'key': 'FFmpegVideoConvertor',
'preferedformat': opts.recodevideo,
})
# FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
# FFmpegExtractAudioPP as containers before conversion may not support
# metadata (3gp, webm, etc.)
# And this post-processor should be placed before other metadata
# manipulating post-processors (FFmpegEmbedSubtitle) to prevent loss of
# extra metadata. By default ffmpeg preserves metadata applicable for both
# source and target containers. From this point the container won't change,
# so metadata can be added here.
if opts.addmetadata:
postprocessors.append({'key': 'FFmpegMetadata'})
if opts.convertsubtitles:
postprocessors.append({
'key': 'FFmpegSubtitlesConvertor',
'format': opts.convertsubtitles,
})
if opts.embedsubtitles:
postprocessors.append({
'key': 'FFmpegEmbedSubtitle',
})
if opts.embedthumbnail:
already_have_thumbnail = opts.writethumbnail or opts.write_all_thumbnails
postprocessors.append({
'key': 'EmbedThumbnail',
'already_have_thumbnail': already_have_thumbnail
})
if not already_have_thumbnail:
opts.writethumbnail = True
# XAttrMetadataPP should be run after post-processors that may change file
# contents
if opts.xattrs:
postprocessors.append({'key': 'XAttrMetadata'})
# Please keep ExecAfterDownload towards the bottom as it allows the user to modify the final file in any way.
# So if the user is able to remove the file before your postprocessor runs it might cause a few problems.
if opts.exec_cmd:
postprocessors.append({
'key': 'ExecAfterDownload',
'exec_cmd': opts.exec_cmd,
})
external_downloader_args = None
if opts.external_downloader_args:
external_downloader_args = compat_shlex_split(opts.external_downloader_args)
postprocessor_args = None
if opts.postprocessor_args:
postprocessor_args = compat_shlex_split(opts.postprocessor_args)
match_filter = (
None if opts.match_filter is None
else match_filter_func(opts.match_filter))
ydl_opts = {
'usenetrc': opts.usenetrc,
'username': opts.username,
'password': opts.password,
'twofactor': opts.twofactor,
'videopassword': opts.videopassword,
'ap_mso': opts.ap_mso,
'ap_username': opts.ap_username,
'ap_password': opts.ap_password,
'quiet': (opts.quiet or any_getting or any_printing),
'no_warnings': opts.no_warnings,
'forceurl': opts.geturl,
'forcetitle': opts.gettitle,
'forceid': opts.getid,
'forcethumbnail': opts.getthumbnail,
'forcedescription': opts.getdescription,
'forceduration': opts.getduration,
'forcefilename': opts.getfilename,
'forceformat': opts.getformat,
'forcejson': opts.dumpjson or opts.print_json,
'dump_single_json': opts.dump_single_json,
'simulate': opts.simulate or any_getting,
'skip_download': opts.skip_download,
'format': opts.format,
'listformats': opts.listformats,
'outtmpl': outtmpl,
'autonumber_size': opts.autonumber_size,
'autonumber_start': opts.autonumber_start,
'restrictfilenames': opts.restrictfilenames,
'ignoreerrors': opts.ignoreerrors,
'force_generic_extractor': opts.force_generic_extractor,
'ratelimit': opts.ratelimit,
'nooverwrites': opts.nooverwrites,
'retries': opts.retries,
'fragment_retries': opts.fragment_retries,
'skip_unavailable_fragments': opts.skip_unavailable_fragments,
'keep_fragments': opts.keep_fragments,
'buffersize': opts.buffersize,
'noresizebuffer': opts.noresizebuffer,
'http_chunk_size': opts.http_chunk_size,
'continuedl': opts.continue_dl,
'noprogress': opts.noprogress,
'progress_with_newline': opts.progress_with_newline,
'playliststart': opts.playliststart,
'playlistend': opts.playlistend,
'playlistreverse': opts.playlist_reverse,
'playlistrandom': opts.playlist_random,
'noplaylist': opts.noplaylist,
'logtostderr': opts.outtmpl == '-',
'consoletitle': opts.consoletitle,
'nopart': opts.nopart,
'updatetime': opts.updatetime,
'writedescription': opts.writedescription,
'writeannotations': opts.writeannotations,
'writeinfojson': opts.writeinfojson,
'writethumbnail': opts.writethumbnail,
'write_all_thumbnails': opts.write_all_thumbnails,
'writesubtitles': opts.writesubtitles,
'writeautomaticsub': opts.writeautomaticsub,
'allsubtitles': opts.allsubtitles,
'listsubtitles': opts.listsubtitles,
'subtitlesformat': opts.subtitlesformat,
'subtitleslangs': opts.subtitleslangs,
'matchtitle': decodeOption(opts.matchtitle),
'rejecttitle': decodeOption(opts.rejecttitle),
'max_downloads': opts.max_downloads,
'prefer_free_formats': opts.prefer_free_formats,
'verbose': opts.verbose,
'dump_intermediate_pages': opts.dump_intermediate_pages,
'write_pages': opts.write_pages,
'test': opts.test,
'keepvideo': opts.keepvideo,
'min_filesize': opts.min_filesize,
'max_filesize': opts.max_filesize,
'min_views': opts.min_views,
'max_views': opts.max_views,
'daterange': date,
'cachedir': opts.cachedir,
'youtube_print_sig_code': opts.youtube_print_sig_code,
'age_limit': opts.age_limit,
'download_archive': download_archive_fn,
'cookiefile': opts.cookiefile,
'nocheckcertificate': opts.no_check_certificate,
'prefer_insecure': opts.prefer_insecure,
'proxy': opts.proxy,
'socket_timeout': opts.socket_timeout,
'bidi_workaround': opts.bidi_workaround,
'debug_printtraffic': opts.debug_printtraffic,
'prefer_ffmpeg': opts.prefer_ffmpeg,
'include_ads': opts.include_ads,
'default_search': opts.default_search,
'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
'encoding': opts.encoding,
'extract_flat': opts.extract_flat,
'mark_watched': opts.mark_watched,
'merge_output_format': opts.merge_output_format,
'postprocessors': postprocessors,
'fixup': opts.fixup,
'source_address': opts.source_address,
'call_home': opts.call_home,
'sleep_interval': opts.sleep_interval,
'max_sleep_interval': opts.max_sleep_interval,
'external_downloader': opts.external_downloader,
'list_thumbnails': opts.list_thumbnails,
'playlist_items': opts.playlist_items,
'xattr_set_filesize': opts.xattr_set_filesize,
'match_filter': match_filter,
'no_color': opts.no_color,
'ffmpeg_location': opts.ffmpeg_location,
'hls_prefer_native': opts.hls_prefer_native,
'hls_use_mpegts': opts.hls_use_mpegts,
'external_downloader_args': external_downloader_args,
'postprocessor_args': postprocessor_args,
'cn_verification_proxy': opts.cn_verification_proxy,
'geo_verification_proxy': opts.geo_verification_proxy,
'config_location': opts.config_location,
'geo_bypass': opts.geo_bypass,
'geo_bypass_country': opts.geo_bypass_country,
'geo_bypass_ip_block': opts.geo_bypass_ip_block,
# just for deprecation check
'autonumber': opts.autonumber if opts.autonumber is True else None,
'usetitle': opts.usetitle if opts.usetitle is True else None,
}
with YoutubeDL(ydl_opts) as ydl:
# Update version
if opts.update_self:
update_self(ydl.to_screen, opts.verbose, ydl._opener)
# Remove cache dir
if opts.rm_cachedir:
ydl.cache.remove()
# Maybe do nothing
if (len(all_urls) < 1) and (opts.load_info_filename is None):
if opts.update_self or opts.rm_cachedir:
sys.exit()
ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
parser.error(
'You must provide at least one URL.\n'
'Type youtube-dl --help to see a list of all options.')
try:
if opts.load_info_filename is not None:
retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
else:
retcode = ydl.download(all_urls)
except MaxDownloadsReached:
ydl.to_screen('--max-download limit reached, aborting.')
retcode = 101
sys.exit(retcode)
def main(argv=None):
try:
_real_main(argv)
except DownloadError:
sys.exit(1)
except SameFileError:
sys.exit('ERROR: fixed output name but more than one file to download')
except KeyboardInterrupt:
sys.exit('\nERROR: Interrupted by user')
__all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']

19
youtube_dl/__main__.py
View File

@ -1,19 +0,0 @@
#!/usr/bin/env python
from __future__ import unicode_literals
# Execute with
# $ python youtube_dl/__main__.py (2.6+)
# $ python -m youtube_dl (2.7+)
import sys
if __package__ is None and not hasattr(sys, 'frozen'):
# direct call of __main__.py
import os.path
path = os.path.realpath(os.path.abspath(__file__))
sys.path.insert(0, os.path.dirname(os.path.dirname(path)))
import youtube_dl
if __name__ == '__main__':
youtube_dl.main()

361
youtube_dl/aes.py
View File

@ -1,361 +0,0 @@
from __future__ import unicode_literals
from math import ceil
from .compat import compat_b64decode
from .utils import bytes_to_intlist, intlist_to_bytes
BLOCK_SIZE_BYTES = 16
def aes_ctr_decrypt(data, key, counter):
"""
Decrypt with aes in counter mode
@param {int[]} data cipher
@param {int[]} key 16/24/32-Byte cipher key
@param {instance} counter Instance whose next_value function (@returns {int[]} 16-Byte block)
returns the next counter block
@returns {int[]} decrypted data
"""
expanded_key = key_expansion(key)
block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
decrypted_data = []
for i in range(block_count):
counter_block = counter.next_value()
block = data[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES]
block += [0] * (BLOCK_SIZE_BYTES - len(block))
cipher_counter_block = aes_encrypt(counter_block, expanded_key)
decrypted_data += xor(block, cipher_counter_block)
decrypted_data = decrypted_data[:len(data)]
return decrypted_data
def aes_cbc_decrypt(data, key, iv):
"""
Decrypt with aes in CBC mode
@param {int[]} data cipher
@param {int[]} key 16/24/32-Byte cipher key
@param {int[]} iv 16-Byte IV
@returns {int[]} decrypted data
"""
expanded_key = key_expansion(key)
block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
decrypted_data = []
previous_cipher_block = iv
for i in range(block_count):
block = data[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES]
block += [0] * (BLOCK_SIZE_BYTES - len(block))
decrypted_block = aes_decrypt(block, expanded_key)
decrypted_data += xor(decrypted_block, previous_cipher_block)
previous_cipher_block = block
decrypted_data = decrypted_data[:len(data)]
return decrypted_data
def aes_cbc_encrypt(data, key, iv):
"""
Encrypt with aes in CBC mode. Using PKCS#7 padding
@param {int[]} data cleartext
@param {int[]} key 16/24/32-Byte cipher key
@param {int[]} iv 16-Byte IV
@returns {int[]} encrypted data
"""
expanded_key = key_expansion(key)
block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
encrypted_data = []
previous_cipher_block = iv
for i in range(block_count):
block = data[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES]
remaining_length = BLOCK_SIZE_BYTES - len(block)
block += [remaining_length] * remaining_length
mixed_block = xor(block, previous_cipher_block)
encrypted_block = aes_encrypt(mixed_block, expanded_key)
encrypted_data += encrypted_block
previous_cipher_block = encrypted_block
return encrypted_data
def key_expansion(data):
"""
Generate key schedule
@param {int[]} data 16/24/32-Byte cipher key
@returns {int[]} 176/208/240-Byte expanded key
"""
data = data[:] # copy
rcon_iteration = 1
key_size_bytes = len(data)
expanded_key_size_bytes = (key_size_bytes // 4 + 7) * BLOCK_SIZE_BYTES
while len(data) < expanded_key_size_bytes:
temp = data[-4:]
temp = key_schedule_core(temp, rcon_iteration)
rcon_iteration += 1
data += xor(temp, data[-key_size_bytes: 4 - key_size_bytes])
for _ in range(3):
temp = data[-4:]
data += xor(temp, data[-key_size_bytes: 4 - key_size_bytes])
if key_size_bytes == 32:
temp = data[-4:]
temp = sub_bytes(temp)
data += xor(temp, data[-key_size_bytes: 4 - key_size_bytes])
for _ in range(3 if key_size_bytes == 32 else 2 if key_size_bytes == 24 else 0):
temp = data[-4:]
data += xor(temp, data[-key_size_bytes: 4 - key_size_bytes])
data = data[:expanded_key_size_bytes]
return data
def aes_encrypt(data, expanded_key):
"""
Encrypt one block with aes
@param {int[]} data 16-Byte state
@param {int[]} expanded_key 176/208/240-Byte expanded key
@returns {int[]} 16-Byte cipher
"""
rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
for i in range(1, rounds + 1):
data = sub_bytes(data)
data = shift_rows(data)
if i != rounds:
data = mix_columns(data)
data = xor(data, expanded_key[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES])
return data
def aes_decrypt(data, expanded_key):
"""
Decrypt one block with aes
@param {int[]} data 16-Byte cipher
@param {int[]} expanded_key 176/208/240-Byte expanded key
@returns {int[]} 16-Byte state
"""
rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
for i in range(rounds, 0, -1):
data = xor(data, expanded_key[i * BLOCK_SIZE_BYTES: (i + 1) * BLOCK_SIZE_BYTES])
if i != rounds:
data = mix_columns_inv(data)
data = shift_rows_inv(data)
data = sub_bytes_inv(data)
data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
return data
def aes_decrypt_text(data, password, key_size_bytes):
"""
Decrypt text
- The first 8 Bytes of decoded 'data' are the 8 high Bytes of the counter
- The cipher key is retrieved by encrypting the first 16 Byte of 'password'
with the first 'key_size_bytes' Bytes from 'password' (if necessary filled with 0's)
- Mode of operation is 'counter'
@param {str} data Base64 encoded string
@param {str,unicode} password Password (will be encoded with utf-8)
@param {int} key_size_bytes Possible values: 16 for 128-Bit, 24 for 192-Bit or 32 for 256-Bit
@returns {str} Decrypted data
"""
NONCE_LENGTH_BYTES = 8
data = bytes_to_intlist(compat_b64decode(data))
password = bytes_to_intlist(password.encode('utf-8'))
key = password[:key_size_bytes] + [0] * (key_size_bytes - len(password))
key = aes_encrypt(key[:BLOCK_SIZE_BYTES], key_expansion(key)) * (key_size_bytes // BLOCK_SIZE_BYTES)
nonce = data[:NONCE_LENGTH_BYTES]
cipher = data[NONCE_LENGTH_BYTES:]
class Counter(object):
__value = nonce + [0] * (BLOCK_SIZE_BYTES - NONCE_LENGTH_BYTES)
def next_value(self):
temp = self.__value
self.__value = inc(self.__value)
return temp
decrypted_data = aes_ctr_decrypt(cipher, key, Counter())
plaintext = intlist_to_bytes(decrypted_data)
return plaintext
RCON = (0x8d, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36)
SBOX = (0x63, 0x7C, 0x77, 0x7B, 0xF2, 0x6B, 0x6F, 0xC5, 0x30, 0x01, 0x67, 0x2B, 0xFE, 0xD7, 0xAB, 0x76,
0xCA, 0x82, 0xC9, 0x7D, 0xFA, 0x59, 0x47, 0xF0, 0xAD, 0xD4, 0xA2, 0xAF, 0x9C, 0xA4, 0x72, 0xC0,
0xB7, 0xFD, 0x93, 0x26, 0x36, 0x3F, 0xF7, 0xCC, 0x34, 0xA5, 0xE5, 0xF1, 0x71, 0xD8, 0x31, 0x15,
0x04, 0xC7, 0x23, 0xC3, 0x18, 0x96, 0x05, 0x9A, 0x07, 0x12, 0x80, 0xE2, 0xEB, 0x27, 0xB2, 0x75,
0x09, 0x83, 0x2C, 0x1A, 0x1B, 0x6E, 0x5A, 0xA0, 0x52, 0x3B, 0xD6, 0xB3, 0x29, 0xE3, 0x2F, 0x84,
0x53, 0xD1, 0x00, 0xED, 0x20, 0xFC, 0xB1, 0x5B, 0x6A, 0xCB, 0xBE, 0x39, 0x4A, 0x4C, 0x58, 0xCF,
0xD0, 0xEF, 0xAA, 0xFB, 0x43, 0x4D, 0x33, 0x85, 0x45, 0xF9, 0x02, 0x7F, 0x50, 0x3C, 0x9F, 0xA8,
0x51, 0xA3, 0x40, 0x8F, 0x92, 0x9D, 0x38, 0xF5, 0xBC, 0xB6, 0xDA, 0x21, 0x10, 0xFF, 0xF3, 0xD2,
0xCD, 0x0C, 0x13, 0xEC, 0x5F, 0x97, 0x44, 0x17, 0xC4, 0xA7, 0x7E, 0x3D, 0x64, 0x5D, 0x19, 0x73,
0x60, 0x81, 0x4F, 0xDC, 0x22, 0x2A, 0x90, 0x88, 0x46, 0xEE, 0xB8, 0x14, 0xDE, 0x5E, 0x0B, 0xDB,
0xE0, 0x32, 0x3A, 0x0A, 0x49, 0x06, 0x24, 0x5C, 0xC2, 0xD3, 0xAC, 0x62, 0x91, 0x95, 0xE4, 0x79,
0xE7, 0xC8, 0x37, 0x6D, 0x8D, 0xD5, 0x4E, 0xA9, 0x6C, 0x56, 0xF4, 0xEA, 0x65, 0x7A, 0xAE, 0x08,
0xBA, 0x78, 0x25, 0x2E, 0x1C, 0xA6, 0xB4, 0xC6, 0xE8, 0xDD, 0x74, 0x1F, 0x4B, 0xBD, 0x8B, 0x8A,
0x70, 0x3E, 0xB5, 0x66, 0x48, 0x03, 0xF6, 0x0E, 0x61, 0x35, 0x57, 0xB9, 0x86, 0xC1, 0x1D, 0x9E,
0xE1, 0xF8, 0x98, 0x11, 0x69, 0xD9, 0x8E, 0x94, 0x9B, 0x1E, 0x87, 0xE9, 0xCE, 0x55, 0x28, 0xDF,
0x8C, 0xA1, 0x89, 0x0D, 0xBF, 0xE6, 0x42, 0x68, 0x41, 0x99, 0x2D, 0x0F, 0xB0, 0x54, 0xBB, 0x16)
SBOX_INV = (0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38, 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb,
0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87, 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb,
0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d, 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e,
0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2, 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25,
0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16, 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92,
0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda, 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84,
0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a, 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06,
0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02, 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b,
0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea, 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73,
0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85, 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e,
0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89, 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b,
0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20, 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4,
0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31, 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f,
0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d, 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef,
0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0, 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61,
0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26, 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d)
MIX_COLUMN_MATRIX = ((0x2, 0x3, 0x1, 0x1),
(0x1, 0x2, 0x3, 0x1),
(0x1, 0x1, 0x2, 0x3),
(0x3, 0x1, 0x1, 0x2))
MIX_COLUMN_MATRIX_INV = ((0xE, 0xB, 0xD, 0x9),
(0x9, 0xE, 0xB, 0xD),
(0xD, 0x9, 0xE, 0xB),
(0xB, 0xD, 0x9, 0xE))
RIJNDAEL_EXP_TABLE = (0x01, 0x03, 0x05, 0x0F, 0x11, 0x33, 0x55, 0xFF, 0x1A, 0x2E, 0x72, 0x96, 0xA1, 0xF8, 0x13, 0x35,
0x5F, 0xE1, 0x38, 0x48, 0xD8, 0x73, 0x95, 0xA4, 0xF7, 0x02, 0x06, 0x0A, 0x1E, 0x22, 0x66, 0xAA,
0xE5, 0x34, 0x5C, 0xE4, 0x37, 0x59, 0xEB, 0x26, 0x6A, 0xBE, 0xD9, 0x70, 0x90, 0xAB, 0xE6, 0x31,
0x53, 0xF5, 0x04, 0x0C, 0x14, 0x3C, 0x44, 0xCC, 0x4F, 0xD1, 0x68, 0xB8, 0xD3, 0x6E, 0xB2, 0xCD,
0x4C, 0xD4, 0x67, 0xA9, 0xE0, 0x3B, 0x4D, 0xD7, 0x62, 0xA6, 0xF1, 0x08, 0x18, 0x28, 0x78, 0x88,
0x83, 0x9E, 0xB9, 0xD0, 0x6B, 0xBD, 0xDC, 0x7F, 0x81, 0x98, 0xB3, 0xCE, 0x49, 0xDB, 0x76, 0x9A,
0xB5, 0xC4, 0x57, 0xF9, 0x10, 0x30, 0x50, 0xF0, 0x0B, 0x1D, 0x27, 0x69, 0xBB, 0xD6, 0x61, 0xA3,
0xFE, 0x19, 0x2B, 0x7D, 0x87, 0x92, 0xAD, 0xEC, 0x2F, 0x71, 0x93, 0xAE, 0xE9, 0x20, 0x60, 0xA0,
0xFB, 0x16, 0x3A, 0x4E, 0xD2, 0x6D, 0xB7, 0xC2, 0x5D, 0xE7, 0x32, 0x56, 0xFA, 0x15, 0x3F, 0x41,
0xC3, 0x5E, 0xE2, 0x3D, 0x47, 0xC9, 0x40, 0xC0, 0x5B, 0xED, 0x2C, 0x74, 0x9C, 0xBF, 0xDA, 0x75,
0x9F, 0xBA, 0xD5, 0x64, 0xAC, 0xEF, 0x2A, 0x7E, 0x82, 0x9D, 0xBC, 0xDF, 0x7A, 0x8E, 0x89, 0x80,
0x9B, 0xB6, 0xC1, 0x58, 0xE8, 0x23, 0x65, 0xAF, 0xEA, 0x25, 0x6F, 0xB1, 0xC8, 0x43, 0xC5, 0x54,
0xFC, 0x1F, 0x21, 0x63, 0xA5, 0xF4, 0x07, 0x09, 0x1B, 0x2D, 0x77, 0x99, 0xB0, 0xCB, 0x46, 0xCA,
0x45, 0xCF, 0x4A, 0xDE, 0x79, 0x8B, 0x86, 0x91, 0xA8, 0xE3, 0x3E, 0x42, 0xC6, 0x51, 0xF3, 0x0E,
0x12, 0x36, 0x5A, 0xEE, 0x29, 0x7B, 0x8D, 0x8C, 0x8F, 0x8A, 0x85, 0x94, 0xA7, 0xF2, 0x0D, 0x17,
0x39, 0x4B, 0xDD, 0x7C, 0x84, 0x97, 0xA2, 0xFD, 0x1C, 0x24, 0x6C, 0xB4, 0xC7, 0x52, 0xF6, 0x01)
RIJNDAEL_LOG_TABLE = (0x00, 0x00, 0x19, 0x01, 0x32, 0x02, 0x1a, 0xc6, 0x4b, 0xc7, 0x1b, 0x68, 0x33, 0xee, 0xdf, 0x03,
0x64, 0x04, 0xe0, 0x0e, 0x34, 0x8d, 0x81, 0xef, 0x4c, 0x71, 0x08, 0xc8, 0xf8, 0x69, 0x1c, 0xc1,
0x7d, 0xc2, 0x1d, 0xb5, 0xf9, 0xb9, 0x27, 0x6a, 0x4d, 0xe4, 0xa6, 0x72, 0x9a, 0xc9, 0x09, 0x78,
0x65, 0x2f, 0x8a, 0x05, 0x21, 0x0f, 0xe1, 0x24, 0x12, 0xf0, 0x82, 0x45, 0x35, 0x93, 0xda, 0x8e,
0x96, 0x8f, 0xdb, 0xbd, 0x36, 0xd0, 0xce, 0x94, 0x13, 0x5c, 0xd2, 0xf1, 0x40, 0x46, 0x83, 0x38,
0x66, 0xdd, 0xfd, 0x30, 0xbf, 0x06, 0x8b, 0x62, 0xb3, 0x25, 0xe2, 0x98, 0x22, 0x88, 0x91, 0x10,
0x7e, 0x6e, 0x48, 0xc3, 0xa3, 0xb6, 0x1e, 0x42, 0x3a, 0x6b, 0x28, 0x54, 0xfa, 0x85, 0x3d, 0xba,
0x2b, 0x79, 0x0a, 0x15, 0x9b, 0x9f, 0x5e, 0xca, 0x4e, 0xd4, 0xac, 0xe5, 0xf3, 0x73, 0xa7, 0x57,
0xaf, 0x58, 0xa8, 0x50, 0xf4, 0xea, 0xd6, 0x74, 0x4f, 0xae, 0xe9, 0xd5, 0xe7, 0xe6, 0xad, 0xe8,
0x2c, 0xd7, 0x75, 0x7a, 0xeb, 0x16, 0x0b, 0xf5, 0x59, 0xcb, 0x5f, 0xb0, 0x9c, 0xa9, 0x51, 0xa0,
0x7f, 0x0c, 0xf6, 0x6f, 0x17, 0xc4, 0x49, 0xec, 0xd8, 0x43, 0x1f, 0x2d, 0xa4, 0x76, 0x7b, 0xb7,
0xcc, 0xbb, 0x3e, 0x5a, 0xfb, 0x60, 0xb1, 0x86, 0x3b, 0x52, 0xa1, 0x6c, 0xaa, 0x55, 0x29, 0x9d,
0x97, 0xb2, 0x87, 0x90, 0x61, 0xbe, 0xdc, 0xfc, 0xbc, 0x95, 0xcf, 0xcd, 0x37, 0x3f, 0x5b, 0xd1,
0x53, 0x39, 0x84, 0x3c, 0x41, 0xa2, 0x6d, 0x47, 0x14, 0x2a, 0x9e, 0x5d, 0x56, 0xf2, 0xd3, 0xab,
0x44, 0x11, 0x92, 0xd9, 0x23, 0x20, 0x2e, 0x89, 0xb4, 0x7c, 0xb8, 0x26, 0x77, 0x99, 0xe3, 0xa5,
0x67, 0x4a, 0xed, 0xde, 0xc5, 0x31, 0xfe, 0x18, 0x0d, 0x63, 0x8c, 0x80, 0xc0, 0xf7, 0x70, 0x07)
def sub_bytes(data):
return [SBOX[x] for x in data]
def sub_bytes_inv(data):
return [SBOX_INV[x] for x in data]
def rotate(data):
return data[1:] + [data[0]]
def key_schedule_core(data, rcon_iteration):
data = rotate(data)
data = sub_bytes(data)
data[0] = data[0] ^ RCON[rcon_iteration]
return data
def xor(data1, data2):
return [x ^ y for x, y in zip(data1, data2)]
def rijndael_mul(a, b):
if(a == 0 or b == 0):
return 0
return RIJNDAEL_EXP_TABLE[(RIJNDAEL_LOG_TABLE[a] + RIJNDAEL_LOG_TABLE[b]) % 0xFF]
def mix_column(data, matrix):
data_mixed = []
for row in range(4):
mixed = 0
for column in range(4):
# xor is (+) and (-)
mixed ^= rijndael_mul(data[column], matrix[row][column])
data_mixed.append(mixed)
return data_mixed
def mix_columns(data, matrix=MIX_COLUMN_MATRIX):
data_mixed = []
for i in range(4):
column = data[i * 4: (i + 1) * 4]
data_mixed += mix_column(column, matrix)
return data_mixed
def mix_columns_inv(data):
return mix_columns(data, MIX_COLUMN_MATRIX_INV)
def shift_rows(data):
data_shifted = []
for column in range(4):
for row in range(4):
data_shifted.append(data[((column + row) & 0b11) * 4 + row])
return data_shifted
def shift_rows_inv(data):
data_shifted = []
for column in range(4):
for row in range(4):
data_shifted.append(data[((column - row) & 0b11) * 4 + row])
return data_shifted
def inc(data):
data = data[:] # copy
for i in range(len(data) - 1, -1, -1):
if data[i] == 255:
data[i] = 0
else:
data[i] = data[i] + 1
break
return data
__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_cbc_decrypt', 'aes_decrypt_text']

96
youtube_dl/cache.py
View File

@ -1,96 +0,0 @@
from __future__ import unicode_literals
import errno
import io
import json
import os
import re
import shutil
import traceback
from .compat import compat_getenv
from .utils import (
expand_path,
write_json_file,
)
class Cache(object):
def __init__(self, ydl):
self._ydl = ydl
def _get_root_dir(self):
res = self._ydl.params.get('cachedir')
if res is None:
cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
res = os.path.join(cache_root, 'youtube-dl')
return expand_path(res)
def _get_cache_fn(self, section, key, dtype):
assert re.match(r'^[a-zA-Z0-9_.-]+$', section), \
'invalid section %r' % section
assert re.match(r'^[a-zA-Z0-9_.-]+$', key), 'invalid key %r' % key
return os.path.join(
self._get_root_dir(), section, '%s.%s' % (key, dtype))
@property
def enabled(self):
return self._ydl.params.get('cachedir') is not False
def store(self, section, key, data, dtype='json'):
assert dtype in ('json',)
if not self.enabled:
return
fn = self._get_cache_fn(section, key, dtype)
try:
try:
os.makedirs(os.path.dirname(fn))
except OSError as ose:
if ose.errno != errno.EEXIST:
raise
write_json_file(data, fn)
except Exception:
tb = traceback.format_exc()
self._ydl.report_warning(
'Writing cache to %r failed: %s' % (fn, tb))
def load(self, section, key, dtype='json', default=None):
assert dtype in ('json',)
if not self.enabled:
return default
cache_fn = self._get_cache_fn(section, key, dtype)
try:
try:
with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
return json.load(cachef)
except ValueError:
try:
file_size = os.path.getsize(cache_fn)
except (OSError, IOError) as oe:
file_size = str(oe)
self._ydl.report_warning(
'Cache retrieval from %s failed (%s)' % (cache_fn, file_size))
except IOError:
pass # No cache available
return default
def remove(self):
if not self.enabled:
self._ydl.to_screen('Cache is disabled (Did you combine --no-cache-dir and --rm-cache-dir?)')
return
cachedir = self._get_root_dir()
if not any((term in cachedir) for term in ('cache', 'tmp')):
raise Exception('Not removing directory %s - this does not look like a cache dir' % cachedir)
self._ydl.to_screen(
'Removing cache dir %s .' % cachedir, skip_eol=True)
if os.path.exists(cachedir):
self._ydl.to_screen('.', skip_eol=True)
shutil.rmtree(cachedir)
self._ydl.to_screen('.')

3016
youtube_dl/compat.py
View File

File diff suppressed because it is too large Load Diff

61
youtube_dl/downloader/__init__.py
View File

@ -1,61 +0,0 @@
from __future__ import unicode_literals
from .common import FileDownloader
from .f4m import F4mFD
from .hls import HlsFD
from .http import HttpFD
from .rtmp import RtmpFD
from .dash import DashSegmentsFD
from .rtsp import RtspFD
from .ism import IsmFD
from .external import (
get_external_downloader,
FFmpegFD,
)
from ..utils import (
determine_protocol,
)
PROTOCOL_MAP = {
'rtmp': RtmpFD,
'm3u8_native': HlsFD,
'm3u8': FFmpegFD,
'mms': RtspFD,
'rtsp': RtspFD,
'f4m': F4mFD,
'http_dash_segments': DashSegmentsFD,
'ism': IsmFD,
}
def get_suitable_downloader(info_dict, params={}):
"""Get the downloader class that can handle the info dict."""
protocol = determine_protocol(info_dict)
info_dict['protocol'] = protocol
# if (info_dict.get('start_time') or info_dict.get('end_time')) and not info_dict.get('requested_formats') and FFmpegFD.can_download(info_dict):
# return FFmpegFD
external_downloader = params.get('external_downloader')
if external_downloader is not None:
ed = get_external_downloader(external_downloader)
if ed.can_download(info_dict):
return ed
if protocol.startswith('m3u8') and info_dict.get('is_live'):
return FFmpegFD
if protocol == 'm3u8' and params.get('hls_prefer_native') is True:
return HlsFD
if protocol == 'm3u8_native' and params.get('hls_prefer_native') is False:
return FFmpegFD
return PROTOCOL_MAP.get(protocol, HttpFD)
__all__ = [
'get_suitable_downloader',
'FileDownloader',
]

389
youtube_dl/downloader/common.py
View File

@ -1,389 +0,0 @@
from __future__ import division, unicode_literals
import os
import re
import sys
import time
import random
from ..compat import compat_os_name
from ..utils import (
decodeArgument,
encodeFilename,
error_to_compat_str,
format_bytes,
shell_quote,
timeconvert,
)
class FileDownloader(object):
"""File Downloader class.
File downloader objects are the ones responsible of downloading the
actual video file and writing it to disk.
File downloaders accept a lot of parameters. In order not to saturate
the object constructor with arguments, it receives a dictionary of
options instead.
Available options:
verbose: Print additional info to stdout.
quiet: Do not print messages to stdout.
ratelimit: Download speed limit, in bytes/sec.
retries: Number of times to retry for HTTP error 5xx
buffersize: Size of download buffer in bytes.
noresizebuffer: Do not automatically resize the download buffer.
continuedl: Try to continue downloads if possible.
noprogress: Do not print the progress bar.
logtostderr: Log messages to stderr instead of stdout.
consoletitle: Display progress in console window's titlebar.
nopart: Do not use temporary .part files.
updatetime: Use the Last-modified header to set output file timestamps.
test: Download only first bytes to test the downloader.
min_filesize: Skip files smaller than this size
max_filesize: Skip files larger than this size
xattr_set_filesize: Set ytdl.filesize user xattribute with expected size.
external_downloader_args: A list of additional command-line arguments for the
external downloader.
hls_use_mpegts: Use the mpegts container for HLS videos.
http_chunk_size: Size of a chunk for chunk-based HTTP downloading. May be
useful for bypassing bandwidth throttling imposed by
a webserver (experimental)
Subclasses of this one must re-define the real_download method.
"""
_TEST_FILE_SIZE = 10241
params = None
def __init__(self, ydl, params):
"""Create a FileDownloader object with the given options."""
self.ydl = ydl
self._progress_hooks = []
self.params = params
self.add_progress_hook(self.report_progress)
@staticmethod
def format_seconds(seconds):
(mins, secs) = divmod(seconds, 60)
(hours, mins) = divmod(mins, 60)
if hours > 99:
return '--:--:--'
if hours == 0:
return '%02d:%02d' % (mins, secs)
else:
return '%02d:%02d:%02d' % (hours, mins, secs)
@staticmethod
def calc_percent(byte_counter, data_len):
if data_len is None:
return None
return float(byte_counter) / float(data_len) * 100.0
@staticmethod
def format_percent(percent):
if percent is None:
return '---.-%'
return '%6s' % ('%3.1f%%' % percent)
@staticmethod
def calc_eta(start, now, total, current):
if total is None:
return None
if now is None:
now = time.time()
dif = now - start
if current == 0 or dif < 0.001: # One millisecond
return None
rate = float(current) / dif
return int((float(total) - float(current)) / rate)
@staticmethod
def format_eta(eta):
if eta is None:
return '--:--'
return FileDownloader.format_seconds(eta)
@staticmethod
def calc_speed(start, now, bytes):
dif = now - start
if bytes == 0 or dif < 0.001: # One millisecond
return None
return float(bytes) / dif
@staticmethod
def format_speed(speed):
if speed is None:
return '%10s' % '---b/s'
return '%10s' % ('%s/s' % format_bytes(speed))
@staticmethod
def format_retries(retries):
return 'inf' if retries == float('inf') else '%.0f' % retries
@staticmethod
def best_block_size(elapsed_time, bytes):
new_min = max(bytes / 2.0, 1.0)
new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
if elapsed_time < 0.001:
return int(new_max)
rate = bytes / elapsed_time
if rate > new_max:
return int(new_max)
if rate < new_min:
return int(new_min)
return int(rate)
@staticmethod
def parse_bytes(bytestr):
"""Parse a string indicating a byte quantity into an integer."""
matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
if matchobj is None:
return None
number = float(matchobj.group(1))
multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
return int(round(number * multiplier))
def to_screen(self, *args, **kargs):
self.ydl.to_screen(*args, **kargs)
def to_stderr(self, message):
self.ydl.to_screen(message)
def to_console_title(self, message):
self.ydl.to_console_title(message)
def trouble(self, *args, **kargs):
self.ydl.trouble(*args, **kargs)
def report_warning(self, *args, **kargs):
self.ydl.report_warning(*args, **kargs)
def report_error(self, *args, **kargs):
self.ydl.report_error(*args, **kargs)
def slow_down(self, start_time, now, byte_counter):
"""Sleep if the download speed is over the rate limit."""
rate_limit = self.params.get('ratelimit')
if rate_limit is None or byte_counter == 0:
return
if now is None:
now = time.time()
elapsed = now - start_time
if elapsed <= 0.0:
return
speed = float(byte_counter) / elapsed
if speed > rate_limit:
time.sleep(max((byte_counter // rate_limit) - elapsed, 0))
def temp_name(self, filename):
"""Returns a temporary filename for the given filename."""
if self.params.get('nopart', False) or filename == '-' or \
(os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
return filename
return filename + '.part'
def undo_temp_name(self, filename):
if filename.endswith('.part'):
return filename[:-len('.part')]
return filename
def ytdl_filename(self, filename):
return filename + '.ytdl'
def try_rename(self, old_filename, new_filename):
try:
if old_filename == new_filename:
return
os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
except (IOError, OSError) as err:
self.report_error('unable to rename file: %s' % error_to_compat_str(err))
def try_utime(self, filename, last_modified_hdr):
"""Try to set the last-modified time of the given file."""
if last_modified_hdr is None:
return
if not os.path.isfile(encodeFilename(filename)):
return
timestr = last_modified_hdr
if timestr is None:
return
filetime = timeconvert(timestr)
if filetime is None:
return filetime
# Ignore obviously invalid dates
if filetime == 0:
return
try:
os.utime(filename, (time.time(), filetime))
except Exception:
pass
return filetime
def report_destination(self, filename):
"""Report destination filename."""
self.to_screen('[download] Destination: ' + filename)
def _report_progress_status(self, msg, is_last_line=False):
fullmsg = '[download] ' + msg
if self.params.get('progress_with_newline', False):
self.to_screen(fullmsg)
else:
if compat_os_name == 'nt':
prev_len = getattr(self, '_report_progress_prev_line_length',
0)
if prev_len > len(fullmsg):
fullmsg += ' ' * (prev_len - len(fullmsg))
self._report_progress_prev_line_length = len(fullmsg)
clear_line = '\r'
else:
clear_line = ('\r\x1b[K' if sys.stderr.isatty() else '\r')
self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
self.to_console_title('youtube-dl ' + msg)
def report_progress(self, s):
if s['status'] == 'finished':
if self.params.get('noprogress', False):
self.to_screen('[download] Download completed')
else:
msg_template = '100%%'
if s.get('total_bytes') is not None:
s['_total_bytes_str'] = format_bytes(s['total_bytes'])
msg_template += ' of %(_total_bytes_str)s'
if s.get('elapsed') is not None:
s['_elapsed_str'] = self.format_seconds(s['elapsed'])
msg_template += ' in %(_elapsed_str)s'
self._report_progress_status(
msg_template % s, is_last_line=True)
if self.params.get('noprogress'):
return
if s['status'] != 'downloading':
return
if s.get('eta') is not None:
s['_eta_str'] = self.format_eta(s['eta'])
else:
s['_eta_str'] = 'Unknown ETA'
if s.get('total_bytes') and s.get('downloaded_bytes') is not None:
s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes'])
elif s.get('total_bytes_estimate') and s.get('downloaded_bytes') is not None:
s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes_estimate'])
else:
if s.get('downloaded_bytes') == 0:
s['_percent_str'] = self.format_percent(0)
else:
s['_percent_str'] = 'Unknown %'
if s.get('speed') is not None:
s['_speed_str'] = self.format_speed(s['speed'])
else:
s['_speed_str'] = 'Unknown speed'
if s.get('total_bytes') is not None:
s['_total_bytes_str'] = format_bytes(s['total_bytes'])
msg_template = '%(_percent_str)s of %(_total_bytes_str)s at %(_speed_str)s ETA %(_eta_str)s'
elif s.get('total_bytes_estimate') is not None:
s['_total_bytes_estimate_str'] = format_bytes(s['total_bytes_estimate'])
msg_template = '%(_percent_str)s of ~%(_total_bytes_estimate_str)s at %(_speed_str)s ETA %(_eta_str)s'
else:
if s.get('downloaded_bytes') is not None:
s['_downloaded_bytes_str'] = format_bytes(s['downloaded_bytes'])
if s.get('elapsed'):
s['_elapsed_str'] = self.format_seconds(s['elapsed'])
msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s (%(_elapsed_str)s)'
else:
msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s'
else:
msg_template = '%(_percent_str)s % at %(_speed_str)s ETA %(_eta_str)s'
self._report_progress_status(msg_template % s)
def report_resuming_byte(self, resume_len):
"""Report attempt to resume at given byte."""
self.to_screen('[download] Resuming download at byte %s' % resume_len)
def report_retry(self, err, count, retries):
"""Report retry in case of HTTP error 5xx"""
self.to_screen(
'[download] Got server HTTP error: %s. Retrying (attempt %d of %s)...'
% (error_to_compat_str(err), count, self.format_retries(retries)))
def report_file_already_downloaded(self, file_name):
"""Report file has already been fully downloaded."""
try:
self.to_screen('[download] %s has already been downloaded' % file_name)
except UnicodeEncodeError:
self.to_screen('[download] The file has already been downloaded')
def report_unable_to_resume(self):
"""Report it was impossible to resume download."""
self.to_screen('[download] Unable to resume')
def download(self, filename, info_dict):
"""Download to a filename using the info from info_dict
Return True on success and False otherwise
"""
nooverwrites_and_exists = (
self.params.get('nooverwrites', False) and
os.path.exists(encodeFilename(filename))
)
if not hasattr(filename, 'write'):
continuedl_and_exists = (
self.params.get('continuedl', True) and
os.path.isfile(encodeFilename(filename)) and
not self.params.get('nopart', False)
)
# Check file already present
if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists):
self.report_file_already_downloaded(filename)
self._hook_progress({
'filename': filename,
'status': 'finished',
'total_bytes': os.path.getsize(encodeFilename(filename)),
})
return True
min_sleep_interval = self.params.get('sleep_interval')
if min_sleep_interval:
max_sleep_interval = self.params.get('max_sleep_interval', min_sleep_interval)
sleep_interval = random.uniform(min_sleep_interval, max_sleep_interval)
self.to_screen(
'[download] Sleeping %s seconds...' % (
int(sleep_interval) if sleep_interval.is_integer()
else '%.2f' % sleep_interval))
time.sleep(sleep_interval)
return self.real_download(filename, info_dict)
def real_download(self, filename, info_dict):
"""Real download process. Redefine in subclasses."""
raise NotImplementedError('This method must be implemented by subclasses')
def _hook_progress(self, status):
for ph in self._progress_hooks:
ph(status)
def add_progress_hook(self, ph):
# See YoutubeDl.py (search for progress_hooks) for a description of
# this interface
self._progress_hooks.append(ph)
def _debug_cmd(self, args, exe=None):
if not self.params.get('verbose', False):
return
str_args = [decodeArgument(a) for a in args]
if exe is None:
exe = os.path.basename(str_args[0])
self.to_screen('[debug] %s command line: %s' % (
exe, shell_quote(str_args)))

80
youtube_dl/downloader/dash.py
View File

@ -1,80 +0,0 @@
from __future__ import unicode_literals
from .fragment import FragmentFD
from ..compat import compat_urllib_error
from ..utils import (
DownloadError,
urljoin,
)
class DashSegmentsFD(FragmentFD):
"""
Download segments in a DASH manifest
"""
FD_NAME = 'dashsegments'
def real_download(self, filename, info_dict):
fragment_base_url = info_dict.get('fragment_base_url')
fragments = info_dict['fragments'][:1] if self.params.get(
'test', False) else info_dict['fragments']
ctx = {
'filename': filename,
'total_frags': len(fragments),
}
self._prepare_and_start_frag_download(ctx)
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
frag_index = 0
for i, fragment in enumerate(fragments):
frag_index += 1
if frag_index <= ctx['fragment_index']:
continue
# In DASH, the first segment contains necessary headers to
# generate a valid MP4 file, so always abort for the first segment
fatal = i == 0 or not skip_unavailable_fragments
count = 0
while count <= fragment_retries:
try:
fragment_url = fragment.get('url')
if not fragment_url:
assert fragment_base_url
fragment_url = urljoin(fragment_base_url, fragment['path'])
success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
if not success:
return False
self._append_fragment(ctx, frag_content)
break
except compat_urllib_error.HTTPError as err:
# YouTube may often return 404 HTTP error for a fragment causing the
# whole download to fail. However if the same fragment is immediately
# retried with the same request data this usually succeeds (1-2 attemps
# is usually enough) thus allowing to download the whole file successfully.
# To be future-proof we will retry all fragments that fail with any
# HTTP error.
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
except DownloadError:
# Don't retry fragment if error occurred during HTTP downloading
# itself since it has own retry settings
if not fatal:
self.report_skip_fragment(frag_index)
break
raise
if count > fragment_retries:
if not fatal:
self.report_skip_fragment(frag_index)
continue
self.report_error('giving up after %s fragment retries' % fragment_retries)
return False
self._finish_frag_download(ctx)
return True

354
youtube_dl/downloader/external.py
View File

@ -1,354 +0,0 @@
from __future__ import unicode_literals
import os.path
import re
import subprocess
import sys
import time
from .common import FileDownloader
from ..compat import (
compat_setenv,
compat_str,
)
from ..postprocessor.ffmpeg import FFmpegPostProcessor, EXT_TO_OUT_FORMATS
from ..utils import (
cli_option,
cli_valueless_option,
cli_bool_option,
cli_configuration_args,
encodeFilename,
encodeArgument,
handle_youtubedl_headers,
check_executable,
is_outdated_version,
)
class ExternalFD(FileDownloader):
def real_download(self, filename, info_dict):
self.report_destination(filename)
tmpfilename = self.temp_name(filename)
try:
started = time.time()
retval = self._call_downloader(tmpfilename, info_dict)
except KeyboardInterrupt:
if not info_dict.get('is_live'):
raise
# Live stream downloading cancellation should be considered as
# correct and expected termination thus all postprocessing
# should take place
retval = 0
self.to_screen('[%s] Interrupted by user' % self.get_basename())
if retval == 0:
status = {
'filename': filename,
'status': 'finished',
'elapsed': time.time() - started,
}
if filename != '-':
fsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('\r[%s] Downloaded %s bytes' % (self.get_basename(), fsize))
self.try_rename(tmpfilename, filename)
status.update({
'downloaded_bytes': fsize,
'total_bytes': fsize,
})
self._hook_progress(status)
return True
else:
self.to_stderr('\n')
self.report_error('%s exited with code %d' % (
self.get_basename(), retval))
return False
@classmethod
def get_basename(cls):
return cls.__name__[:-2].lower()
@property
def exe(self):
return self.params.get('external_downloader')
@classmethod
def available(cls):
return check_executable(cls.get_basename(), [cls.AVAILABLE_OPT])
@classmethod
def supports(cls, info_dict):
return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps')
@classmethod
def can_download(cls, info_dict):
return cls.available() and cls.supports(info_dict)
def _option(self, command_option, param):
return cli_option(self.params, command_option, param)
def _bool_option(self, command_option, param, true_value='true', false_value='false', separator=None):
return cli_bool_option(self.params, command_option, param, true_value, false_value, separator)
def _valueless_option(self, command_option, param, expected_value=True):
return cli_valueless_option(self.params, command_option, param, expected_value)
def _configuration_args(self, default=[]):
return cli_configuration_args(self.params, 'external_downloader_args', default)
def _call_downloader(self, tmpfilename, info_dict):
""" Either overwrite this or implement _make_cmd """
cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
self._debug_cmd(cmd)
p = subprocess.Popen(
cmd, stderr=subprocess.PIPE)
_, stderr = p.communicate()
if p.returncode != 0:
self.to_stderr(stderr.decode('utf-8', 'replace'))
return p.returncode
class CurlFD(ExternalFD):
AVAILABLE_OPT = '-V'
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '--location', '-o', tmpfilename]
for key, val in info_dict['http_headers'].items():
cmd += ['--header', '%s: %s' % (key, val)]
cmd += self._bool_option('--continue-at', 'continuedl', '-', '0')
cmd += self._valueless_option('--silent', 'noprogress')
cmd += self._valueless_option('--verbose', 'verbose')
cmd += self._option('--limit-rate', 'ratelimit')
cmd += self._option('--retry', 'retries')
cmd += self._option('--max-filesize', 'max_filesize')
cmd += self._option('--interface', 'source_address')
cmd += self._option('--proxy', 'proxy')
cmd += self._valueless_option('--insecure', 'nocheckcertificate')
cmd += self._configuration_args()
cmd += ['--', info_dict['url']]
return cmd
def _call_downloader(self, tmpfilename, info_dict):
cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
self._debug_cmd(cmd)
# curl writes the progress to stderr so don't capture it.
p = subprocess.Popen(cmd)
p.communicate()
return p.returncode
class AxelFD(ExternalFD):
AVAILABLE_OPT = '-V'
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '-o', tmpfilename]
for key, val in info_dict['http_headers'].items():
cmd += ['-H', '%s: %s' % (key, val)]
cmd += self._configuration_args()
cmd += ['--', info_dict['url']]
return cmd
class WgetFD(ExternalFD):
AVAILABLE_OPT = '--version'
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies']
for key, val in info_dict['http_headers'].items():
cmd += ['--header', '%s: %s' % (key, val)]
cmd += self._option('--bind-address', 'source_address')
cmd += self._option('--proxy', 'proxy')
cmd += self._valueless_option('--no-check-certificate', 'nocheckcertificate')
cmd += self._configuration_args()
cmd += ['--', info_dict['url']]
return cmd
class Aria2cFD(ExternalFD):
AVAILABLE_OPT = '-v'
def _make_cmd(self, tmpfilename, info_dict):
cmd = [self.exe, '-c']
cmd += self._configuration_args([
'--min-split-size', '1M', '--max-connection-per-server', '4'])
dn = os.path.dirname(tmpfilename)
if dn:
cmd += ['--dir', dn]
cmd += ['--out', os.path.basename(tmpfilename)]
for key, val in info_dict['http_headers'].items():
cmd += ['--header', '%s: %s' % (key, val)]
cmd += self._option('--interface', 'source_address')
cmd += self._option('--all-proxy', 'proxy')
cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=')
cmd += ['--', info_dict['url']]
return cmd
class HttpieFD(ExternalFD):
@classmethod
def available(cls):
return check_executable('http', ['--version'])
def _make_cmd(self, tmpfilename, info_dict):
cmd = ['http', '--download', '--output', tmpfilename, info_dict['url']]
for key, val in info_dict['http_headers'].items():
cmd += ['%s:%s' % (key, val)]
return cmd
class FFmpegFD(ExternalFD):
@classmethod
def supports(cls, info_dict):
return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps', 'm3u8', 'rtsp', 'rtmp', 'mms')
@classmethod
def available(cls):
return FFmpegPostProcessor().available
def _call_downloader(self, tmpfilename, info_dict):
url = info_dict['url']
ffpp = FFmpegPostProcessor(downloader=self)
if not ffpp.available:
self.report_error('m3u8 download detected but ffmpeg or avconv could not be found. Please install one.')
return False
ffpp.check_version()
args = [ffpp.executable, '-y']
for log_level in ('quiet', 'verbose'):
if self.params.get(log_level, False):
args += ['-loglevel', log_level]
break
seekable = info_dict.get('_seekable')
if seekable is not None:
# setting -seekable prevents ffmpeg from guessing if the server
# supports seeking(by adding the header `Range: bytes=0-`), which
# can cause problems in some cases
# https://github.com/rg3/youtube-dl/issues/11800#issuecomment-275037127
# http://trac.ffmpeg.org/ticket/6125#comment:10
args += ['-seekable', '1' if seekable else '0']
args += self._configuration_args()
# start_time = info_dict.get('start_time') or 0
# if start_time:
# args += ['-ss', compat_str(start_time)]
# end_time = info_dict.get('end_time')
# if end_time:
# args += ['-t', compat_str(end_time - start_time)]
if info_dict['http_headers'] and re.match(r'^https?://', url):
# Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv:
# [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header.
headers = handle_youtubedl_headers(info_dict['http_headers'])
args += [
'-headers',
''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())]
env = None
proxy = self.params.get('proxy')
if proxy:
if not re.match(r'^[\da-zA-Z]+://', proxy):
proxy = 'http://%s' % proxy
if proxy.startswith('socks'):
self.report_warning(
'%s does not support SOCKS proxies. Downloading is likely to fail. '
'Consider adding --hls-prefer-native to your command.' % self.get_basename())
# Since December 2015 ffmpeg supports -http_proxy option (see
# http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd)
# We could switch to the following code if we are able to detect version properly
# args += ['-http_proxy', proxy]
env = os.environ.copy()
compat_setenv('HTTP_PROXY', proxy, env=env)
compat_setenv('http_proxy', proxy, env=env)
protocol = info_dict.get('protocol')
if protocol == 'rtmp':
player_url = info_dict.get('player_url')
page_url = info_dict.get('page_url')
app = info_dict.get('app')
play_path = info_dict.get('play_path')
tc_url = info_dict.get('tc_url')
flash_version = info_dict.get('flash_version')
live = info_dict.get('rtmp_live', False)
if player_url is not None:
args += ['-rtmp_swfverify', player_url]
if page_url is not None:
args += ['-rtmp_pageurl', page_url]
if app is not None:
args += ['-rtmp_app', app]
if play_path is not None:
args += ['-rtmp_playpath', play_path]
if tc_url is not None:
args += ['-rtmp_tcurl', tc_url]
if flash_version is not None:
args += ['-rtmp_flashver', flash_version]
if live:
args += ['-rtmp_live', 'live']
args += ['-i', url, '-c', 'copy']
if self.params.get('test', False):
args += ['-fs', compat_str(self._TEST_FILE_SIZE)]
if protocol in ('m3u8', 'm3u8_native'):
if self.params.get('hls_use_mpegts', False) or tmpfilename == '-':
args += ['-f', 'mpegts']
else:
args += ['-f', 'mp4']
if (ffpp.basename == 'ffmpeg' and is_outdated_version(ffpp._versions['ffmpeg'], '3.2', False)) and (not info_dict.get('acodec') or info_dict['acodec'].split('.')[0] in ('aac', 'mp4a')):
args += ['-bsf:a', 'aac_adtstoasc']
elif protocol == 'rtmp':
args += ['-f', 'flv']
else:
args += ['-f', EXT_TO_OUT_FORMATS.get(info_dict['ext'], info_dict['ext'])]
args = [encodeArgument(opt) for opt in args]
args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True))
self._debug_cmd(args)
proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env)
try:
retval = proc.wait()
except KeyboardInterrupt:
# subprocces.run would send the SIGKILL signal to ffmpeg and the
# mp4 file couldn't be played, but if we ask ffmpeg to quit it
# produces a file that is playable (this is mostly useful for live
# streams). Note that Windows is not affected and produces playable
# files (see https://github.com/rg3/youtube-dl/issues/8300).
if sys.platform != 'win32':
proc.communicate(b'q')
raise
return retval
class AVconvFD(FFmpegFD):
pass
_BY_NAME = dict(
(klass.get_basename(), klass)
for name, klass in globals().items()
if name.endswith('FD') and name != 'ExternalFD'
)
def list_external_downloaders():
return sorted(_BY_NAME.keys())
def get_external_downloader(external_downloader):
""" Given the name of the executable, see whether we support the given
downloader . """
# Drop .exe extension on Windows
bn = os.path.splitext(os.path.basename(external_downloader))[0]
return _BY_NAME[bn]

438
youtube_dl/downloader/f4m.py
View File

@ -1,438 +0,0 @@
from __future__ import division, unicode_literals
import io
import itertools
import time
from .fragment import FragmentFD
from ..compat import (
compat_b64decode,
compat_etree_fromstring,
compat_urlparse,
compat_urllib_error,
compat_urllib_parse_urlparse,
compat_struct_pack,
compat_struct_unpack,
)
from ..utils import (
fix_xml_ampersands,
xpath_text,
)
class DataTruncatedError(Exception):
pass
class FlvReader(io.BytesIO):
"""
Reader for Flv files
The file format is documented in https://www.adobe.com/devnet/f4v.html
"""
def read_bytes(self, n):
data = self.read(n)
if len(data) < n:
raise DataTruncatedError(
'FlvReader error: need %d bytes while only %d bytes got' % (
n, len(data)))
return data
# Utility functions for reading numbers and strings
def read_unsigned_long_long(self):
return compat_struct_unpack('!Q', self.read_bytes(8))[0]
def read_unsigned_int(self):
return compat_struct_unpack('!I', self.read_bytes(4))[0]
def read_unsigned_char(self):
return compat_struct_unpack('!B', self.read_bytes(1))[0]
def read_string(self):
res = b''
while True:
char = self.read_bytes(1)
if char == b'\x00':
break
res += char
return res
def read_box_info(self):
"""
Read a box and return the info as a tuple: (box_size, box_type, box_data)
"""
real_size = size = self.read_unsigned_int()
box_type = self.read_bytes(4)
header_end = 8
if size == 1:
real_size = self.read_unsigned_long_long()
header_end = 16
return real_size, box_type, self.read_bytes(real_size - header_end)
def read_asrt(self):
# version
self.read_unsigned_char()
# flags
self.read_bytes(3)
quality_entry_count = self.read_unsigned_char()
# QualityEntryCount
for i in range(quality_entry_count):
self.read_string()
segment_run_count = self.read_unsigned_int()
segments = []
for i in range(segment_run_count):
first_segment = self.read_unsigned_int()
fragments_per_segment = self.read_unsigned_int()
segments.append((first_segment, fragments_per_segment))
return {
'segment_run': segments,
}
def read_afrt(self):
# version
self.read_unsigned_char()
# flags
self.read_bytes(3)
# time scale
self.read_unsigned_int()
quality_entry_count = self.read_unsigned_char()
# QualitySegmentUrlModifiers
for i in range(quality_entry_count):
self.read_string()
fragments_count = self.read_unsigned_int()
fragments = []
for i in range(fragments_count):
first = self.read_unsigned_int()
first_ts = self.read_unsigned_long_long()
duration = self.read_unsigned_int()
if duration == 0:
discontinuity_indicator = self.read_unsigned_char()
else:
discontinuity_indicator = None
fragments.append({
'first': first,
'ts': first_ts,
'duration': duration,
'discontinuity_indicator': discontinuity_indicator,
})
return {
'fragments': fragments,
}
def read_abst(self):
# version
self.read_unsigned_char()
# flags
self.read_bytes(3)
self.read_unsigned_int() # BootstrapinfoVersion
# Profile,Live,Update,Reserved
flags = self.read_unsigned_char()
live = flags & 0x20 != 0
# time scale
self.read_unsigned_int()
# CurrentMediaTime
self.read_unsigned_long_long()
# SmpteTimeCodeOffset
self.read_unsigned_long_long()
self.read_string() # MovieIdentifier
server_count = self.read_unsigned_char()
# ServerEntryTable
for i in range(server_count):
self.read_string()
quality_count = self.read_unsigned_char()
# QualityEntryTable
for i in range(quality_count):
self.read_string()
# DrmData
self.read_string()
# MetaData
self.read_string()
segments_count = self.read_unsigned_char()
segments = []
for i in range(segments_count):
box_size, box_type, box_data = self.read_box_info()
assert box_type == b'asrt'
segment = FlvReader(box_data).read_asrt()
segments.append(segment)
fragments_run_count = self.read_unsigned_char()
fragments = []
for i in range(fragments_run_count):
box_size, box_type, box_data = self.read_box_info()
assert box_type == b'afrt'
fragments.append(FlvReader(box_data).read_afrt())
return {
'segments': segments,
'fragments': fragments,
'live': live,
}
def read_bootstrap_info(self):
total_size, box_type, box_data = self.read_box_info()
assert box_type == b'abst'
return FlvReader(box_data).read_abst()
def read_bootstrap_info(bootstrap_bytes):
return FlvReader(bootstrap_bytes).read_bootstrap_info()
def build_fragments_list(boot_info):
""" Return a list of (segment, fragment) for each fragment in the video """
res = []
segment_run_table = boot_info['segments'][0]
fragment_run_entry_table = boot_info['fragments'][0]['fragments']
first_frag_number = fragment_run_entry_table[0]['first']
fragments_counter = itertools.count(first_frag_number)
for segment, fragments_count in segment_run_table['segment_run']:
# In some live HDS streams (for example Rai), `fragments_count` is
# abnormal and causing out-of-memory errors. It's OK to change the
# number of fragments for live streams as they are updated periodically
if fragments_count == 4294967295 and boot_info['live']:
fragments_count = 2
for _ in range(fragments_count):
res.append((segment, next(fragments_counter)))
if boot_info['live']:
res = res[-2:]
return res
def write_unsigned_int(stream, val):
stream.write(compat_struct_pack('!I', val))
def write_unsigned_int_24(stream, val):
stream.write(compat_struct_pack('!I', val)[1:])
def write_flv_header(stream):
"""Writes the FLV header to stream"""
# FLV header
stream.write(b'FLV\x01')
stream.write(b'\x05')
stream.write(b'\x00\x00\x00\x09')
stream.write(b'\x00\x00\x00\x00')
def write_metadata_tag(stream, metadata):
"""Writes optional metadata tag to stream"""
SCRIPT_TAG = b'\x12'
FLV_TAG_HEADER_LEN = 11
if metadata:
stream.write(SCRIPT_TAG)
write_unsigned_int_24(stream, len(metadata))
stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
stream.write(metadata)
write_unsigned_int(stream, FLV_TAG_HEADER_LEN + len(metadata))
def remove_encrypted_media(media):
return list(filter(lambda e: 'drmAdditionalHeaderId' not in e.attrib and
'drmAdditionalHeaderSetId' not in e.attrib,
media))
def _add_ns(prop, ver=1):
return '{http://ns.adobe.com/f4m/%d.0}%s' % (ver, prop)
def get_base_url(manifest):
base_url = xpath_text(
manifest, [_add_ns('baseURL'), _add_ns('baseURL', 2)],
'base URL', default=None)
if base_url:
base_url = base_url.strip()
return base_url
class F4mFD(FragmentFD):
"""
A downloader for f4m manifests or AdobeHDS.
"""
FD_NAME = 'f4m'
def _get_unencrypted_media(self, doc):
media = doc.findall(_add_ns('media'))
if not media:
self.report_error('No media found')
for e in (doc.findall(_add_ns('drmAdditionalHeader')) +
doc.findall(_add_ns('drmAdditionalHeaderSet'))):
# If id attribute is missing it's valid for all media nodes
# without drmAdditionalHeaderId or drmAdditionalHeaderSetId attribute
if 'id' not in e.attrib:
self.report_error('Missing ID in f4m DRM')
media = remove_encrypted_media(media)
if not media:
self.report_error('Unsupported DRM')
return media
def _get_bootstrap_from_url(self, bootstrap_url):
bootstrap = self.ydl.urlopen(bootstrap_url).read()
return read_bootstrap_info(bootstrap)
def _update_live_fragments(self, bootstrap_url, latest_fragment):
fragments_list = []
retries = 30
while (not fragments_list) and (retries > 0):
boot_info = self._get_bootstrap_from_url(bootstrap_url)
fragments_list = build_fragments_list(boot_info)
fragments_list = [f for f in fragments_list if f[1] > latest_fragment]
if not fragments_list:
# Retry after a while
time.sleep(5.0)
retries -= 1
if not fragments_list:
self.report_error('Failed to update fragments')
return fragments_list
def _parse_bootstrap_node(self, node, base_url):
# Sometimes non empty inline bootstrap info can be specified along
# with bootstrap url attribute (e.g. dummy inline bootstrap info
# contains whitespace characters in [1]). We will prefer bootstrap
# url over inline bootstrap info when present.
# 1. http://live-1-1.rutube.ru/stream/1024/HDS/SD/C2NKsS85HQNckgn5HdEmOQ/1454167650/S-s604419906/move/four/dirs/upper/1024-576p.f4m
bootstrap_url = node.get('url')
if bootstrap_url:
bootstrap_url = compat_urlparse.urljoin(
base_url, bootstrap_url)
boot_info = self._get_bootstrap_from_url(bootstrap_url)
else:
bootstrap_url = None
bootstrap = compat_b64decode(node.text)
boot_info = read_bootstrap_info(bootstrap)
return boot_info, bootstrap_url
def real_download(self, filename, info_dict):
man_url = info_dict['url']
requested_bitrate = info_dict.get('tbr')
self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME)
urlh = self.ydl.urlopen(self._prepare_url(info_dict, man_url))
man_url = urlh.geturl()
# Some manifests may be malformed, e.g. prosiebensat1 generated manifests
# (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244
# and https://github.com/rg3/youtube-dl/issues/7823)
manifest = fix_xml_ampersands(urlh.read().decode('utf-8', 'ignore')).strip()
doc = compat_etree_fromstring(manifest)
formats = [(int(f.attrib.get('bitrate', -1)), f)
for f in self._get_unencrypted_media(doc)]
if requested_bitrate is None or len(formats) == 1:
# get the best format
formats = sorted(formats, key=lambda f: f[0])
rate, media = formats[-1]
else:
rate, media = list(filter(
lambda f: int(f[0]) == requested_bitrate, formats))[0]
# Prefer baseURL for relative URLs as per 11.2 of F4M 3.0 spec.
man_base_url = get_base_url(doc) or man_url
base_url = compat_urlparse.urljoin(man_base_url, media.attrib['url'])
bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
boot_info, bootstrap_url = self._parse_bootstrap_node(
bootstrap_node, man_base_url)
live = boot_info['live']
metadata_node = media.find(_add_ns('metadata'))
if metadata_node is not None:
metadata = compat_b64decode(metadata_node.text)
else:
metadata = None
fragments_list = build_fragments_list(boot_info)
test = self.params.get('test', False)
if test:
# We only download the first fragment
fragments_list = fragments_list[:1]
total_frags = len(fragments_list)
# For some akamai manifests we'll need to add a query to the fragment url
akamai_pv = xpath_text(doc, _add_ns('pv-2.0'))
ctx = {
'filename': filename,
'total_frags': total_frags,
'live': live,
}
self._prepare_frag_download(ctx)
dest_stream = ctx['dest_stream']
if ctx['complete_frags_downloaded_bytes'] == 0:
write_flv_header(dest_stream)
if not live:
write_metadata_tag(dest_stream, metadata)
base_url_parsed = compat_urllib_parse_urlparse(base_url)
self._start_frag_download(ctx)
frag_index = 0
while fragments_list:
seg_i, frag_i = fragments_list.pop(0)
frag_index += 1
if frag_index <= ctx['fragment_index']:
continue
name = 'Seg%d-Frag%d' % (seg_i, frag_i)
query = []
if base_url_parsed.query:
query.append(base_url_parsed.query)
if akamai_pv:
query.append(akamai_pv.strip(';'))
if info_dict.get('extra_param_to_segment_url'):
query.append(info_dict['extra_param_to_segment_url'])
url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query))
try:
success, down_data = self._download_fragment(ctx, url_parsed.geturl(), info_dict)
if not success:
return False
reader = FlvReader(down_data)
while True:
try:
_, box_type, box_data = reader.read_box_info()
except DataTruncatedError:
if test:
# In tests, segments may be truncated, and thus
# FlvReader may not be able to parse the whole
# chunk. If so, write the segment as is
# See https://github.com/rg3/youtube-dl/issues/9214
dest_stream.write(down_data)
break
raise
if box_type == b'mdat':
self._append_fragment(ctx, box_data)
break
except (compat_urllib_error.HTTPError, ) as err:
if live and (err.code == 404 or err.code == 410):
# We didn't keep up with the live window. Continue
# with the next available fragment.
msg = 'Fragment %d unavailable' % frag_i
self.report_warning(msg)
fragments_list = []
else:
raise
if not fragments_list and not test and live and bootstrap_url:
fragments_list = self._update_live_fragments(bootstrap_url, frag_i)
total_frags += len(fragments_list)
if fragments_list and (fragments_list[0][1] > frag_i + 1):
msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1))
self.report_warning(msg)
self._finish_frag_download(ctx)
return True

268
youtube_dl/downloader/fragment.py
View File

@ -1,268 +0,0 @@
from __future__ import division, unicode_literals
import os
import time
import json
from .common import FileDownloader
from .http import HttpFD
from ..utils import (
error_to_compat_str,
encodeFilename,
sanitize_open,
sanitized_Request,
)
class HttpQuietDownloader(HttpFD):
def to_screen(self, *args, **kargs):
pass
class FragmentFD(FileDownloader):
"""
A base file downloader class for fragmented media (e.g. f4m/m3u8 manifests).
Available options:
fragment_retries: Number of times to retry a fragment for HTTP error (DASH
and hlsnative only)
skip_unavailable_fragments:
Skip unavailable fragments (DASH and hlsnative only)
keep_fragments: Keep downloaded fragments on disk after downloading is
finished
For each incomplete fragment download youtube-dl keeps on disk a special
bookkeeping file with download state and metadata (in future such files will
be used for any incomplete download handled by youtube-dl). This file is
used to properly handle resuming, check download file consistency and detect
potential errors. The file has a .ytdl extension and represents a standard
JSON file of the following format:
extractor:
Dictionary of extractor related data. TBD.
downloader:
Dictionary of downloader related data. May contain following data:
current_fragment:
Dictionary with current (being downloaded) fragment data:
index: 0-based index of current fragment among all fragments
fragment_count:
Total count of fragments
This feature is experimental and file format may change in future.
"""
def report_retry_fragment(self, err, frag_index, count, retries):
self.to_screen(
'[download] Got server HTTP error: %s. Retrying fragment %d (attempt %d of %s)...'
% (error_to_compat_str(err), frag_index, count, self.format_retries(retries)))
def report_skip_fragment(self, frag_index):
self.to_screen('[download] Skipping fragment %d...' % frag_index)
def _prepare_url(self, info_dict, url):
headers = info_dict.get('http_headers')
return sanitized_Request(url, None, headers) if headers else url
def _prepare_and_start_frag_download(self, ctx):
self._prepare_frag_download(ctx)
self._start_frag_download(ctx)
@staticmethod
def __do_ytdl_file(ctx):
return not ctx['live'] and not ctx['tmpfilename'] == '-'
def _read_ytdl_file(self, ctx):
assert 'ytdl_corrupt' not in ctx
stream, _ = sanitize_open(self.ytdl_filename(ctx['filename']), 'r')
try:
ctx['fragment_index'] = json.loads(stream.read())['downloader']['current_fragment']['index']
except Exception:
ctx['ytdl_corrupt'] = True
finally:
stream.close()
def _write_ytdl_file(self, ctx):
frag_index_stream, _ = sanitize_open(self.ytdl_filename(ctx['filename']), 'w')
downloader = {
'current_fragment': {
'index': ctx['fragment_index'],
},
}
if ctx.get('fragment_count') is not None:
downloader['fragment_count'] = ctx['fragment_count']
frag_index_stream.write(json.dumps({'downloader': downloader}))
frag_index_stream.close()
def _download_fragment(self, ctx, frag_url, info_dict, headers=None):
fragment_filename = '%s-Frag%d' % (ctx['tmpfilename'], ctx['fragment_index'])
success = ctx['dl'].download(fragment_filename, {
'url': frag_url,
'http_headers': headers or info_dict.get('http_headers'),
})
if not success:
return False, None
down, frag_sanitized = sanitize_open(fragment_filename, 'rb')
ctx['fragment_filename_sanitized'] = frag_sanitized
frag_content = down.read()
down.close()
return True, frag_content
def _append_fragment(self, ctx, frag_content):
try:
ctx['dest_stream'].write(frag_content)
ctx['dest_stream'].flush()
finally:
if self.__do_ytdl_file(ctx):
self._write_ytdl_file(ctx)
if not self.params.get('keep_fragments', False):
os.remove(encodeFilename(ctx['fragment_filename_sanitized']))
del ctx['fragment_filename_sanitized']
def _prepare_frag_download(self, ctx):
if 'live' not in ctx:
ctx['live'] = False
if not ctx['live']:
total_frags_str = '%d' % ctx['total_frags']
ad_frags = ctx.get('ad_frags', 0)
if ad_frags:
total_frags_str += ' (not including %d ad)' % ad_frags
else:
total_frags_str = 'unknown (live)'
self.to_screen(
'[%s] Total fragments: %s' % (self.FD_NAME, total_frags_str))
self.report_destination(ctx['filename'])
dl = HttpQuietDownloader(
self.ydl,
{
'continuedl': True,
'quiet': True,
'noprogress': True,
'ratelimit': self.params.get('ratelimit'),
'retries': self.params.get('retries', 0),
'nopart': self.params.get('nopart', False),
'test': self.params.get('test', False),
}
)
tmpfilename = self.temp_name(ctx['filename'])
open_mode = 'wb'
resume_len = 0
# Establish possible resume length
if os.path.isfile(encodeFilename(tmpfilename)):
open_mode = 'ab'
resume_len = os.path.getsize(encodeFilename(tmpfilename))
# Should be initialized before ytdl file check
ctx.update({
'tmpfilename': tmpfilename,
'fragment_index': 0,
})
if self.__do_ytdl_file(ctx):
if os.path.isfile(encodeFilename(self.ytdl_filename(ctx['filename']))):
self._read_ytdl_file(ctx)
is_corrupt = ctx.get('ytdl_corrupt') is True
is_inconsistent = ctx['fragment_index'] > 0 and resume_len == 0
if is_corrupt or is_inconsistent:
message = (
'.ytdl file is corrupt' if is_corrupt else
'Inconsistent state of incomplete fragment download')
self.report_warning(
'%s. Restarting from the beginning...' % message)
ctx['fragment_index'] = resume_len = 0
if 'ytdl_corrupt' in ctx:
del ctx['ytdl_corrupt']
self._write_ytdl_file(ctx)
else:
self._write_ytdl_file(ctx)
assert ctx['fragment_index'] == 0
dest_stream, tmpfilename = sanitize_open(tmpfilename, open_mode)
ctx.update({
'dl': dl,
'dest_stream': dest_stream,
'tmpfilename': tmpfilename,
# Total complete fragments downloaded so far in bytes
'complete_frags_downloaded_bytes': resume_len,
})
def _start_frag_download(self, ctx):
total_frags = ctx['total_frags']
# This dict stores the download progress, it's updated by the progress
# hook
state = {
'status': 'downloading',
'downloaded_bytes': ctx['complete_frags_downloaded_bytes'],
'fragment_index': ctx['fragment_index'],
'fragment_count': total_frags,
'filename': ctx['filename'],
'tmpfilename': ctx['tmpfilename'],
}
start = time.time()
ctx.update({
'started': start,
# Amount of fragment's bytes downloaded by the time of the previous
# frag progress hook invocation
'prev_frag_downloaded_bytes': 0,
})
def frag_progress_hook(s):
if s['status'] not in ('downloading', 'finished'):
return
time_now = time.time()
state['elapsed'] = time_now - start
frag_total_bytes = s.get('total_bytes') or 0
if not ctx['live']:
estimated_size = (
(ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) /
(state['fragment_index'] + 1) * total_frags)
state['total_bytes_estimate'] = estimated_size
if s['status'] == 'finished':
state['fragment_index'] += 1
ctx['fragment_index'] = state['fragment_index']
state['downloaded_bytes'] += frag_total_bytes - ctx['prev_frag_downloaded_bytes']
ctx['complete_frags_downloaded_bytes'] = state['downloaded_bytes']
ctx['prev_frag_downloaded_bytes'] = 0
else:
frag_downloaded_bytes = s['downloaded_bytes']
state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes']
if not ctx['live']:
state['eta'] = self.calc_eta(
start, time_now, estimated_size,
state['downloaded_bytes'])
state['speed'] = s.get('speed') or ctx.get('speed')
ctx['speed'] = state['speed']
ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes
self._hook_progress(state)
ctx['dl'].add_progress_hook(frag_progress_hook)
return start
def _finish_frag_download(self, ctx):
ctx['dest_stream'].close()
if self.__do_ytdl_file(ctx):
ytdl_filename = encodeFilename(self.ytdl_filename(ctx['filename']))
if os.path.isfile(ytdl_filename):
os.remove(ytdl_filename)
elapsed = time.time() - ctx['started']
if ctx['tmpfilename'] == '-':
downloaded_bytes = ctx['complete_frags_downloaded_bytes']
else:
self.try_rename(ctx['tmpfilename'], ctx['filename'])
downloaded_bytes = os.path.getsize(encodeFilename(ctx['filename']))
self._hook_progress({
'downloaded_bytes': downloaded_bytes,
'total_bytes': downloaded_bytes,
'filename': ctx['filename'],
'status': 'finished',
'elapsed': elapsed,
})

204
youtube_dl/downloader/hls.py
View File

@ -1,204 +0,0 @@
from __future__ import unicode_literals
import re
import binascii
try:
from Crypto.Cipher import AES
can_decrypt_frag = True
except ImportError:
can_decrypt_frag = False
from .fragment import FragmentFD
from .external import FFmpegFD
from ..compat import (
compat_urllib_error,
compat_urlparse,
compat_struct_pack,
)
from ..utils import (
parse_m3u8_attributes,
update_url_query,
)
class HlsFD(FragmentFD):
""" A limited implementation that does not require ffmpeg """
FD_NAME = 'hlsnative'
@staticmethod
def can_download(manifest, info_dict):
UNSUPPORTED_FEATURES = (
r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)', # encrypted streams [1]
# r'#EXT-X-BYTERANGE', # playlists composed of byte ranges of media files [2]
# Live streams heuristic does not always work (e.g. geo restricted to Germany
# http://hls-geo.daserste.de/i/videoportal/Film/c_620000/622873/format,716451,716457,716450,716458,716459,.mp4.csmil/index_4_av.m3u8?null=0)
# r'#EXT-X-MEDIA-SEQUENCE:(?!0$)', # live streams [3]
# This heuristic also is not correct since segments may not be appended as well.
# Twitch vods of finished streams have EXT-X-PLAYLIST-TYPE:EVENT despite
# no segments will definitely be appended to the end of the playlist.
# r'#EXT-X-PLAYLIST-TYPE:EVENT', # media segments may be appended to the end of
# # event media playlists [4]
# 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.4
# 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2
# 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
# 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
)
check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
is_aes128_enc = '#EXT-X-KEY:METHOD=AES-128' in manifest
check_results.append(can_decrypt_frag or not is_aes128_enc)
check_results.append(not (is_aes128_enc and r'#EXT-X-BYTERANGE' in manifest))
check_results.append(not info_dict.get('is_live'))
return all(check_results)
def real_download(self, filename, info_dict):
man_url = info_dict['url']
self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME)
urlh = self.ydl.urlopen(self._prepare_url(info_dict, man_url))
man_url = urlh.geturl()
s = urlh.read().decode('utf-8', 'ignore')
if not self.can_download(s, info_dict):
if info_dict.get('extra_param_to_segment_url'):
self.report_error('pycrypto not found. Please install it.')
return False
self.report_warning(
'hlsnative has detected features it does not support, '
'extraction will be delegated to ffmpeg')
fd = FFmpegFD(self.ydl, self.params)
for ph in self._progress_hooks:
fd.add_progress_hook(ph)
return fd.real_download(filename, info_dict)
def is_ad_fragment(s):
return (s.startswith('#ANVATO-SEGMENT-INFO') and 'type=ad' in s or
s.startswith('#UPLYNK-SEGMENT') and s.endswith(',ad'))
media_frags = 0
ad_frags = 0
ad_frag_next = False
for line in s.splitlines():
line = line.strip()
if not line:
continue
if line.startswith('#'):
if is_ad_fragment(line):
ad_frags += 1
ad_frag_next = True
continue
if ad_frag_next:
ad_frag_next = False
continue
media_frags += 1
ctx = {
'filename': filename,
'total_frags': media_frags,
'ad_frags': ad_frags,
}
self._prepare_and_start_frag_download(ctx)
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
test = self.params.get('test', False)
extra_query = None
extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
if extra_param_to_segment_url:
extra_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
i = 0
media_sequence = 0
decrypt_info = {'METHOD': 'NONE'}
byte_range = {}
frag_index = 0
ad_frag_next = False
for line in s.splitlines():
line = line.strip()
if line:
if not line.startswith('#'):
if ad_frag_next:
ad_frag_next = False
continue
frag_index += 1
if frag_index <= ctx['fragment_index']:
continue
frag_url = (
line
if re.match(r'^https?://', line)
else compat_urlparse.urljoin(man_url, line))
if extra_query:
frag_url = update_url_query(frag_url, extra_query)
count = 0
headers = info_dict.get('http_headers', {})
if byte_range:
headers['Range'] = 'bytes=%d-%d' % (byte_range['start'], byte_range['end'])
while count <= fragment_retries:
try:
success, frag_content = self._download_fragment(
ctx, frag_url, info_dict, headers)
if not success:
return False
break
except compat_urllib_error.HTTPError as err:
# Unavailable (possibly temporary) fragments may be served.
# First we try to retry then either skip or abort.
# See https://github.com/rg3/youtube-dl/issues/10165,
# https://github.com/rg3/youtube-dl/issues/10448).
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
if count > fragment_retries:
if skip_unavailable_fragments:
i += 1
media_sequence += 1
self.report_skip_fragment(frag_index)
continue
self.report_error(
'giving up after %s fragment retries' % fragment_retries)
return False
if decrypt_info['METHOD'] == 'AES-128':
iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
self._prepare_url(info_dict, decrypt_info['URI'])).read()
frag_content = AES.new(
decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
self._append_fragment(ctx, frag_content)
# We only download the first fragment during the test
if test:
break
i += 1
media_sequence += 1
elif line.startswith('#EXT-X-KEY'):
decrypt_url = decrypt_info.get('URI')
decrypt_info = parse_m3u8_attributes(line[11:])
if decrypt_info['METHOD'] == 'AES-128':
if 'IV' in decrypt_info:
decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:].zfill(32))
if not re.match(r'^https?://', decrypt_info['URI']):
decrypt_info['URI'] = compat_urlparse.urljoin(
man_url, decrypt_info['URI'])
if extra_query:
decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
if decrypt_url != decrypt_info['URI']:
decrypt_info['KEY'] = None
elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
media_sequence = int(line[22:])
elif line.startswith('#EXT-X-BYTERANGE'):
splitted_byte_range = line[17:].split('@')
sub_range_start = int(splitted_byte_range[1]) if len(splitted_byte_range) == 2 else byte_range['end']
byte_range = {
'start': sub_range_start,
'end': sub_range_start + int(splitted_byte_range[0]),
}
elif is_ad_fragment(line):
ad_frag_next = True
self._finish_frag_download(ctx)
return True

354
youtube_dl/downloader/http.py
View File

@ -1,354 +0,0 @@
from __future__ import unicode_literals
import errno
import os
import socket
import time
import random
import re
from .common import FileDownloader
from ..compat import (
compat_str,
compat_urllib_error,
)
from ..utils import (
ContentTooShortError,
encodeFilename,
int_or_none,
sanitize_open,
sanitized_Request,
write_xattr,
XAttrMetadataError,
XAttrUnavailableError,
)
class HttpFD(FileDownloader):
def real_download(self, filename, info_dict):
url = info_dict['url']
class DownloadContext(dict):
__getattr__ = dict.get
__setattr__ = dict.__setitem__
__delattr__ = dict.__delitem__
ctx = DownloadContext()
ctx.filename = filename
ctx.tmpfilename = self.temp_name(filename)
ctx.stream = None
# Do not include the Accept-Encoding header
headers = {'Youtubedl-no-compression': 'True'}
add_headers = info_dict.get('http_headers')
if add_headers:
headers.update(add_headers)
is_test = self.params.get('test', False)
chunk_size = self._TEST_FILE_SIZE if is_test else (
info_dict.get('downloader_options', {}).get('http_chunk_size') or
self.params.get('http_chunk_size') or 0)
ctx.open_mode = 'wb'
ctx.resume_len = 0
ctx.data_len = None
ctx.block_size = self.params.get('buffersize', 1024)
ctx.start_time = time.time()
ctx.chunk_size = None
if self.params.get('continuedl', True):
# Establish possible resume length
if os.path.isfile(encodeFilename(ctx.tmpfilename)):
ctx.resume_len = os.path.getsize(
encodeFilename(ctx.tmpfilename))
ctx.is_resume = ctx.resume_len > 0
count = 0
retries = self.params.get('retries', 0)
class SucceedDownload(Exception):
pass
class RetryDownload(Exception):
def __init__(self, source_error):
self.source_error = source_error
class NextFragment(Exception):
pass
def set_range(req, start, end):
range_header = 'bytes=%d-' % start
if end:
range_header += compat_str(end)
req.add_header('Range', range_header)
def establish_connection():
ctx.chunk_size = (random.randint(int(chunk_size * 0.95), chunk_size)
if not is_test and chunk_size else chunk_size)
if ctx.resume_len > 0:
range_start = ctx.resume_len
if ctx.is_resume:
self.report_resuming_byte(ctx.resume_len)
ctx.open_mode = 'ab'
elif ctx.chunk_size > 0:
range_start = 0
else:
range_start = None
ctx.is_resume = False
range_end = range_start + ctx.chunk_size - 1 if ctx.chunk_size else None
if range_end and ctx.data_len is not None and range_end >= ctx.data_len:
range_end = ctx.data_len - 1
has_range = range_start is not None
ctx.has_range = has_range
request = sanitized_Request(url, None, headers)
if has_range:
set_range(request, range_start, range_end)
# Establish connection
try:
ctx.data = self.ydl.urlopen(request)
# When trying to resume, Content-Range HTTP header of response has to be checked
# to match the value of requested Range HTTP header. This is due to a webservers
# that don't support resuming and serve a whole file with no Content-Range
# set in response despite of requested Range (see
# https://github.com/rg3/youtube-dl/issues/6057#issuecomment-126129799)
if has_range:
content_range = ctx.data.headers.get('Content-Range')
if content_range:
content_range_m = re.search(r'bytes (\d+)-(\d+)?(?:/(\d+))?', content_range)
# Content-Range is present and matches requested Range, resume is possible
if content_range_m:
if range_start == int(content_range_m.group(1)):
content_range_end = int_or_none(content_range_m.group(2))
content_len = int_or_none(content_range_m.group(3))
accept_content_len = (
# Non-chunked download
not ctx.chunk_size or
# Chunked download and requested piece or
# its part is promised to be served
content_range_end == range_end or
content_len < range_end)
if accept_content_len:
ctx.data_len = content_len
return
# Content-Range is either not present or invalid. Assuming remote webserver is
# trying to send the whole file, resume is not possible, so wiping the local file
# and performing entire redownload
self.report_unable_to_resume()
ctx.resume_len = 0
ctx.open_mode = 'wb'
ctx.data_len = int_or_none(ctx.data.info().get('Content-length', None))
return
except (compat_urllib_error.HTTPError, ) as err:
if err.code == 416:
# Unable to resume (requested range not satisfiable)
try:
# Open the connection again without the range header
ctx.data = self.ydl.urlopen(
sanitized_Request(url, None, headers))
content_length = ctx.data.info()['Content-Length']
except (compat_urllib_error.HTTPError, ) as err:
if err.code < 500 or err.code >= 600:
raise
else:
# Examine the reported length
if (content_length is not None and
(ctx.resume_len - 100 < int(content_length) < ctx.resume_len + 100)):
# The file had already been fully downloaded.
# Explanation to the above condition: in issue #175 it was revealed that
# YouTube sometimes adds or removes a few bytes from the end of the file,
# changing the file size slightly and causing problems for some users. So
# I decided to implement a suggested change and consider the file
# completely downloaded if the file size differs less than 100 bytes from
# the one in the hard drive.
self.report_file_already_downloaded(ctx.filename)
self.try_rename(ctx.tmpfilename, ctx.filename)
self._hook_progress({
'filename': ctx.filename,
'status': 'finished',
'downloaded_bytes': ctx.resume_len,
'total_bytes': ctx.resume_len,
})
raise SucceedDownload()
else:
# The length does not match, we start the download over
self.report_unable_to_resume()
ctx.resume_len = 0
ctx.open_mode = 'wb'
return
elif err.code < 500 or err.code >= 600:
# Unexpected HTTP error
raise
raise RetryDownload(err)
except socket.error as err:
if err.errno != errno.ECONNRESET:
# Connection reset is no problem, just retry
raise
raise RetryDownload(err)
def download():
data_len = ctx.data.info().get('Content-length', None)
# Range HTTP header may be ignored/unsupported by a webserver
# (e.g. extractor/scivee.py, extractor/bambuser.py).
# However, for a test we still would like to download just a piece of a file.
# To achieve this we limit data_len to _TEST_FILE_SIZE and manually control
# block size when downloading a file.
if is_test and (data_len is None or int(data_len) > self._TEST_FILE_SIZE):
data_len = self._TEST_FILE_SIZE
if data_len is not None:
data_len = int(data_len) + ctx.resume_len
min_data_len = self.params.get('min_filesize')
max_data_len = self.params.get('max_filesize')
if min_data_len is not None and data_len < min_data_len:
self.to_screen('\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
return False
if max_data_len is not None and data_len > max_data_len:
self.to_screen('\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
return False
byte_counter = 0 + ctx.resume_len
block_size = ctx.block_size
start = time.time()
# measure time over whole while-loop, so slow_down() and best_block_size() work together properly
now = None # needed for slow_down() in the first loop run
before = start # start measuring
def retry(e):
to_stdout = ctx.tmpfilename == '-'
if not to_stdout:
ctx.stream.close()
ctx.stream = None
ctx.resume_len = byte_counter if to_stdout else os.path.getsize(encodeFilename(ctx.tmpfilename))
raise RetryDownload(e)
while True:
try:
# Download and write
data_block = ctx.data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
# socket.timeout is a subclass of socket.error but may not have
# errno set
except socket.timeout as e:
retry(e)
except socket.error as e:
if e.errno not in (errno.ECONNRESET, errno.ETIMEDOUT):
raise
retry(e)
byte_counter += len(data_block)
# exit loop when download is finished
if len(data_block) == 0:
break
# Open destination file just in time
if ctx.stream is None:
try:
ctx.stream, ctx.tmpfilename = sanitize_open(
ctx.tmpfilename, ctx.open_mode)
assert ctx.stream is not None
ctx.filename = self.undo_temp_name(ctx.tmpfilename)
self.report_destination(ctx.filename)
except (OSError, IOError) as err:
self.report_error('unable to open for writing: %s' % str(err))
return False
if self.params.get('xattr_set_filesize', False) and data_len is not None:
try:
write_xattr(ctx.tmpfilename, 'user.ytdl.filesize', str(data_len).encode('utf-8'))
except (XAttrUnavailableError, XAttrMetadataError) as err:
self.report_error('unable to set filesize xattr: %s' % str(err))
try:
ctx.stream.write(data_block)
except (IOError, OSError) as err:
self.to_stderr('\n')
self.report_error('unable to write data: %s' % str(err))
return False
# Apply rate limit
self.slow_down(start, now, byte_counter - ctx.resume_len)
# end measuring of one loop run
now = time.time()
after = now
# Adjust block size
if not self.params.get('noresizebuffer', False):
block_size = self.best_block_size(after - before, len(data_block))
before = after
# Progress message
speed = self.calc_speed(start, now, byte_counter - ctx.resume_len)
if ctx.data_len is None:
eta = None
else:
eta = self.calc_eta(start, time.time(), ctx.data_len - ctx.resume_len, byte_counter - ctx.resume_len)
self._hook_progress({
'status': 'downloading',
'downloaded_bytes': byte_counter,
'total_bytes': ctx.data_len,
'tmpfilename': ctx.tmpfilename,
'filename': ctx.filename,
'eta': eta,
'speed': speed,
'elapsed': now - ctx.start_time,
})
if is_test and byte_counter == data_len:
break
if not is_test and ctx.chunk_size and ctx.data_len is not None and byte_counter < ctx.data_len:
ctx.resume_len = byte_counter
# ctx.block_size = block_size
raise NextFragment()
if ctx.stream is None:
self.to_stderr('\n')
self.report_error('Did not get any data blocks')
return False
if ctx.tmpfilename != '-':
ctx.stream.close()
if data_len is not None and byte_counter != data_len:
err = ContentTooShortError(byte_counter, int(data_len))
if count <= retries:
retry(err)
raise err
self.try_rename(ctx.tmpfilename, ctx.filename)
# Update file modification time
if self.params.get('updatetime', True):
info_dict['filetime'] = self.try_utime(ctx.filename, ctx.data.info().get('last-modified', None))
self._hook_progress({
'downloaded_bytes': byte_counter,
'total_bytes': byte_counter,
'filename': ctx.filename,
'status': 'finished',
'elapsed': time.time() - ctx.start_time,
})
return True
while count <= retries:
try:
establish_connection()
return download()
except RetryDownload as e:
count += 1
if count <= retries:
self.report_retry(e.source_error, count, retries)
continue
except NextFragment:
continue
except SucceedDownload:
return True
self.report_error('giving up after %s retries' % retries)
return False

259
youtube_dl/downloader/ism.py
View File

@ -1,259 +0,0 @@
from __future__ import unicode_literals
import time
import binascii
import io
from .fragment import FragmentFD
from ..compat import (
compat_Struct,
compat_urllib_error,
)
u8 = compat_Struct('>B')
u88 = compat_Struct('>Bx')
u16 = compat_Struct('>H')
u1616 = compat_Struct('>Hxx')
u32 = compat_Struct('>I')
u64 = compat_Struct('>Q')
s88 = compat_Struct('>bx')
s16 = compat_Struct('>h')
s1616 = compat_Struct('>hxx')
s32 = compat_Struct('>i')
unity_matrix = (s32.pack(0x10000) + s32.pack(0) * 3) * 2 + s32.pack(0x40000000)
TRACK_ENABLED = 0x1
TRACK_IN_MOVIE = 0x2
TRACK_IN_PREVIEW = 0x4
SELF_CONTAINED = 0x1
def box(box_type, payload):
return u32.pack(8 + len(payload)) + box_type + payload
def full_box(box_type, version, flags, payload):
return box(box_type, u8.pack(version) + u32.pack(flags)[1:] + payload)
def write_piff_header(stream, params):
track_id = params['track_id']
fourcc = params['fourcc']
duration = params['duration']
timescale = params.get('timescale', 10000000)
language = params.get('language', 'und')
height = params.get('height', 0)
width = params.get('width', 0)
is_audio = width == 0 and height == 0
creation_time = modification_time = int(time.time())
ftyp_payload = b'isml' # major brand
ftyp_payload += u32.pack(1) # minor version
ftyp_payload += b'piff' + b'iso2' # compatible brands
stream.write(box(b'ftyp', ftyp_payload)) # File Type Box
mvhd_payload = u64.pack(creation_time)
mvhd_payload += u64.pack(modification_time)
mvhd_payload += u32.pack(timescale)
mvhd_payload += u64.pack(duration)
mvhd_payload += s1616.pack(1) # rate
mvhd_payload += s88.pack(1) # volume
mvhd_payload += u16.pack(0) # reserved
mvhd_payload += u32.pack(0) * 2 # reserved
mvhd_payload += unity_matrix
mvhd_payload += u32.pack(0) * 6 # pre defined
mvhd_payload += u32.pack(0xffffffff) # next track id
moov_payload = full_box(b'mvhd', 1, 0, mvhd_payload) # Movie Header Box
tkhd_payload = u64.pack(creation_time)
tkhd_payload += u64.pack(modification_time)
tkhd_payload += u32.pack(track_id) # track id
tkhd_payload += u32.pack(0) # reserved
tkhd_payload += u64.pack(duration)
tkhd_payload += u32.pack(0) * 2 # reserved
tkhd_payload += s16.pack(0) # layer
tkhd_payload += s16.pack(0) # alternate group
tkhd_payload += s88.pack(1 if is_audio else 0) # volume
tkhd_payload += u16.pack(0) # reserved
tkhd_payload += unity_matrix
tkhd_payload += u1616.pack(width)
tkhd_payload += u1616.pack(height)
trak_payload = full_box(b'tkhd', 1, TRACK_ENABLED | TRACK_IN_MOVIE | TRACK_IN_PREVIEW, tkhd_payload) # Track Header Box
mdhd_payload = u64.pack(creation_time)
mdhd_payload += u64.pack(modification_time)
mdhd_payload += u32.pack(timescale)
mdhd_payload += u64.pack(duration)
mdhd_payload += u16.pack(((ord(language[0]) - 0x60) << 10) | ((ord(language[1]) - 0x60) << 5) | (ord(language[2]) - 0x60))
mdhd_payload += u16.pack(0) # pre defined
mdia_payload = full_box(b'mdhd', 1, 0, mdhd_payload) # Media Header Box
hdlr_payload = u32.pack(0) # pre defined
hdlr_payload += b'soun' if is_audio else b'vide' # handler type
hdlr_payload += u32.pack(0) * 3 # reserved
hdlr_payload += (b'Sound' if is_audio else b'Video') + b'Handler\0' # name
mdia_payload += full_box(b'hdlr', 0, 0, hdlr_payload) # Handler Reference Box
if is_audio:
smhd_payload = s88.pack(0) # balance
smhd_payload += u16.pack(0) # reserved
media_header_box = full_box(b'smhd', 0, 0, smhd_payload) # Sound Media Header
else:
vmhd_payload = u16.pack(0) # graphics mode
vmhd_payload += u16.pack(0) * 3 # opcolor
media_header_box = full_box(b'vmhd', 0, 1, vmhd_payload) # Video Media Header
minf_payload = media_header_box
dref_payload = u32.pack(1) # entry count
dref_payload += full_box(b'url ', 0, SELF_CONTAINED, b'') # Data Entry URL Box
dinf_payload = full_box(b'dref', 0, 0, dref_payload) # Data Reference Box
minf_payload += box(b'dinf', dinf_payload) # Data Information Box
stsd_payload = u32.pack(1) # entry count
sample_entry_payload = u8.pack(0) * 6 # reserved
sample_entry_payload += u16.pack(1) # data reference index
if is_audio:
sample_entry_payload += u32.pack(0) * 2 # reserved
sample_entry_payload += u16.pack(params.get('channels', 2))
sample_entry_payload += u16.pack(params.get('bits_per_sample', 16))
sample_entry_payload += u16.pack(0) # pre defined
sample_entry_payload += u16.pack(0) # reserved
sample_entry_payload += u1616.pack(params['sampling_rate'])
if fourcc == 'AACL':
sample_entry_box = box(b'mp4a', sample_entry_payload)
else:
sample_entry_payload += u16.pack(0) # pre defined
sample_entry_payload += u16.pack(0) # reserved
sample_entry_payload += u32.pack(0) * 3 # pre defined
sample_entry_payload += u16.pack(width)
sample_entry_payload += u16.pack(height)
sample_entry_payload += u1616.pack(0x48) # horiz resolution 72 dpi
sample_entry_payload += u1616.pack(0x48) # vert resolution 72 dpi
sample_entry_payload += u32.pack(0) # reserved
sample_entry_payload += u16.pack(1) # frame count
sample_entry_payload += u8.pack(0) * 32 # compressor name
sample_entry_payload += u16.pack(0x18) # depth
sample_entry_payload += s16.pack(-1) # pre defined
codec_private_data = binascii.unhexlify(params['codec_private_data'].encode('utf-8'))
if fourcc in ('H264', 'AVC1'):
sps, pps = codec_private_data.split(u32.pack(1))[1:]
avcc_payload = u8.pack(1) # configuration version
avcc_payload += sps[1:4] # avc profile indication + profile compatibility + avc level indication
avcc_payload += u8.pack(0xfc | (params.get('nal_unit_length_field', 4) - 1)) # complete represenation (1) + reserved (11111) + length size minus one
avcc_payload += u8.pack(1) # reserved (0) + number of sps (0000001)
avcc_payload += u16.pack(len(sps))
avcc_payload += sps
avcc_payload += u8.pack(1) # number of pps
avcc_payload += u16.pack(len(pps))
avcc_payload += pps
sample_entry_payload += box(b'avcC', avcc_payload) # AVC Decoder Configuration Record
sample_entry_box = box(b'avc1', sample_entry_payload) # AVC Simple Entry
stsd_payload += sample_entry_box
stbl_payload = full_box(b'stsd', 0, 0, stsd_payload) # Sample Description Box
stts_payload = u32.pack(0) # entry count
stbl_payload += full_box(b'stts', 0, 0, stts_payload) # Decoding Time to Sample Box
stsc_payload = u32.pack(0) # entry count
stbl_payload += full_box(b'stsc', 0, 0, stsc_payload) # Sample To Chunk Box
stco_payload = u32.pack(0) # entry count
stbl_payload += full_box(b'stco', 0, 0, stco_payload) # Chunk Offset Box
minf_payload += box(b'stbl', stbl_payload) # Sample Table Box
mdia_payload += box(b'minf', minf_payload) # Media Information Box
trak_payload += box(b'mdia', mdia_payload) # Media Box
moov_payload += box(b'trak', trak_payload) # Track Box
mehd_payload = u64.pack(duration)
mvex_payload = full_box(b'mehd', 1, 0, mehd_payload) # Movie Extends Header Box
trex_payload = u32.pack(track_id) # track id
trex_payload += u32.pack(1) # default sample description index
trex_payload += u32.pack(0) # default sample duration
trex_payload += u32.pack(0) # default sample size
trex_payload += u32.pack(0) # default sample flags
mvex_payload += full_box(b'trex', 0, 0, trex_payload) # Track Extends Box
moov_payload += box(b'mvex', mvex_payload) # Movie Extends Box
stream.write(box(b'moov', moov_payload)) # Movie Box
def extract_box_data(data, box_sequence):
data_reader = io.BytesIO(data)
while True:
box_size = u32.unpack(data_reader.read(4))[0]
box_type = data_reader.read(4)
if box_type == box_sequence[0]:
box_data = data_reader.read(box_size - 8)
if len(box_sequence) == 1:
return box_data
return extract_box_data(box_data, box_sequence[1:])
data_reader.seek(box_size - 8, 1)
class IsmFD(FragmentFD):
"""
Download segments in a ISM manifest
"""
FD_NAME = 'ism'
def real_download(self, filename, info_dict):
segments = info_dict['fragments'][:1] if self.params.get(
'test', False) else info_dict['fragments']
ctx = {
'filename': filename,
'total_frags': len(segments),
}
self._prepare_and_start_frag_download(ctx)
fragment_retries = self.params.get('fragment_retries', 0)
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
track_written = False
frag_index = 0
for i, segment in enumerate(segments):
frag_index += 1
if frag_index <= ctx['fragment_index']:
continue
count = 0
while count <= fragment_retries:
try:
success, frag_content = self._download_fragment(ctx, segment['url'], info_dict)
if not success:
return False
if not track_written:
tfhd_data = extract_box_data(frag_content, [b'moof', b'traf', b'tfhd'])
info_dict['_download_params']['track_id'] = u32.unpack(tfhd_data[4:8])[0]
write_piff_header(ctx['dest_stream'], info_dict['_download_params'])
track_written = True
self._append_fragment(ctx, frag_content)
break
except compat_urllib_error.HTTPError as err:
count += 1
if count <= fragment_retries:
self.report_retry_fragment(err, frag_index, count, fragment_retries)
if count > fragment_retries:
if skip_unavailable_fragments:
self.report_skip_fragment(frag_index)
continue
self.report_error('giving up after %s fragment retries' % fragment_retries)
return False
self._finish_frag_download(ctx)
return True

214
youtube_dl/downloader/rtmp.py
View File

@ -1,214 +0,0 @@
from __future__ import unicode_literals
import os
import re
import subprocess
import time
from .common import FileDownloader
from ..compat import compat_str
from ..utils import (
check_executable,
encodeFilename,
encodeArgument,
get_exe_version,
)
def rtmpdump_version():
return get_exe_version(
'rtmpdump', ['--help'], r'(?i)RTMPDump\s*v?([0-9a-zA-Z._-]+)')
class RtmpFD(FileDownloader):
def real_download(self, filename, info_dict):
def run_rtmpdump(args):
start = time.time()
resume_percent = None
resume_downloaded_data_len = None
proc = subprocess.Popen(args, stderr=subprocess.PIPE)
cursor_in_new_line = True
proc_stderr_closed = False
try:
while not proc_stderr_closed:
# read line from stderr
line = ''
while True:
char = proc.stderr.read(1)
if not char:
proc_stderr_closed = True
break
if char in [b'\r', b'\n']:
break
line += char.decode('ascii', 'replace')
if not line:
# proc_stderr_closed is True
continue
mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
if mobj:
downloaded_data_len = int(float(mobj.group(1)) * 1024)
percent = float(mobj.group(2))
if not resume_percent:
resume_percent = percent
resume_downloaded_data_len = downloaded_data_len
time_now = time.time()
eta = self.calc_eta(start, time_now, 100 - resume_percent, percent - resume_percent)
speed = self.calc_speed(start, time_now, downloaded_data_len - resume_downloaded_data_len)
data_len = None
if percent > 0:
data_len = int(downloaded_data_len * 100 / percent)
self._hook_progress({
'status': 'downloading',
'downloaded_bytes': downloaded_data_len,
'total_bytes_estimate': data_len,
'tmpfilename': tmpfilename,
'filename': filename,
'eta': eta,
'elapsed': time_now - start,
'speed': speed,
})
cursor_in_new_line = False
else:
# no percent for live streams
mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
if mobj:
downloaded_data_len = int(float(mobj.group(1)) * 1024)
time_now = time.time()
speed = self.calc_speed(start, time_now, downloaded_data_len)
self._hook_progress({
'downloaded_bytes': downloaded_data_len,
'tmpfilename': tmpfilename,
'filename': filename,
'status': 'downloading',
'elapsed': time_now - start,
'speed': speed,
})
cursor_in_new_line = False
elif self.params.get('verbose', False):
if not cursor_in_new_line:
self.to_screen('')
cursor_in_new_line = True
self.to_screen('[rtmpdump] ' + line)
finally:
proc.wait()
if not cursor_in_new_line:
self.to_screen('')
return proc.returncode
url = info_dict['url']
player_url = info_dict.get('player_url')
page_url = info_dict.get('page_url')
app = info_dict.get('app')
play_path = info_dict.get('play_path')
tc_url = info_dict.get('tc_url')
flash_version = info_dict.get('flash_version')
live = info_dict.get('rtmp_live', False)
conn = info_dict.get('rtmp_conn')
protocol = info_dict.get('rtmp_protocol')
real_time = info_dict.get('rtmp_real_time', False)
no_resume = info_dict.get('no_resume', False)
continue_dl = self.params.get('continuedl', True)
self.report_destination(filename)
tmpfilename = self.temp_name(filename)
test = self.params.get('test', False)
# Check for rtmpdump first
if not check_executable('rtmpdump', ['-h']):
self.report_error('RTMP download detected but "rtmpdump" could not be run. Please install it.')
return False
# Download using rtmpdump. rtmpdump returns exit code 2 when
# the connection was interrupted and resuming appears to be
# possible. This is part of rtmpdump's normal usage, AFAIK.
basic_args = [
'rtmpdump', '--verbose', '-r', url,
'-o', tmpfilename]
if player_url is not None:
basic_args += ['--swfVfy', player_url]
if page_url is not None:
basic_args += ['--pageUrl', page_url]
if app is not None:
basic_args += ['--app', app]
if play_path is not None:
basic_args += ['--playpath', play_path]
if tc_url is not None:
basic_args += ['--tcUrl', tc_url]
if test:
basic_args += ['--stop', '1']
if flash_version is not None:
basic_args += ['--flashVer', flash_version]
if live:
basic_args += ['--live']
if isinstance(conn, list):
for entry in conn:
basic_args += ['--conn', entry]
elif isinstance(conn, compat_str):
basic_args += ['--conn', conn]
if protocol is not None:
basic_args += ['--protocol', protocol]
if real_time:
basic_args += ['--realtime']
args = basic_args
if not no_resume and continue_dl and not live:
args += ['--resume']
if not live and continue_dl:
args += ['--skip', '1']
args = [encodeArgument(a) for a in args]
self._debug_cmd(args, exe='rtmpdump')
RD_SUCCESS = 0
RD_FAILED = 1
RD_INCOMPLETE = 2
RD_NO_CONNECT = 3
started = time.time()
try:
retval = run_rtmpdump(args)
except KeyboardInterrupt:
if not info_dict.get('is_live'):
raise
retval = RD_SUCCESS
self.to_screen('\n[rtmpdump] Interrupted by user')
if retval == RD_NO_CONNECT:
self.report_error('[rtmpdump] Could not connect to RTMP server.')
return False
while retval in (RD_INCOMPLETE, RD_FAILED) and not test and not live:
prevsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('[rtmpdump] Downloaded %s bytes' % prevsize)
time.sleep(5.0) # This seems to be needed
args = basic_args + ['--resume']
if retval == RD_FAILED:
args += ['--skip', '1']
args = [encodeArgument(a) for a in args]
retval = run_rtmpdump(args)
cursize = os.path.getsize(encodeFilename(tmpfilename))
if prevsize == cursize and retval == RD_FAILED:
break
# Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
if prevsize == cursize and retval == RD_INCOMPLETE and cursize > 1024:
self.to_screen('[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
retval = RD_SUCCESS
break
if retval == RD_SUCCESS or (test and retval == RD_INCOMPLETE):
fsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('[rtmpdump] Downloaded %s bytes' % fsize)
self.try_rename(tmpfilename, filename)
self._hook_progress({
'downloaded_bytes': fsize,
'total_bytes': fsize,
'filename': filename,
'status': 'finished',
'elapsed': time.time() - started,
})
return True
else:
self.to_stderr('\n')
self.report_error('rtmpdump exited with code %d' % retval)
return False

47
youtube_dl/downloader/rtsp.py
View File

@ -1,47 +0,0 @@
from __future__ import unicode_literals
import os
import subprocess
from .common import FileDownloader
from ..utils import (
check_executable,
encodeFilename,
)
class RtspFD(FileDownloader):
def real_download(self, filename, info_dict):
url = info_dict['url']
self.report_destination(filename)
tmpfilename = self.temp_name(filename)
if check_executable('mplayer', ['-h']):
args = [
'mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy',
'-dumpstream', '-dumpfile', tmpfilename, url]
elif check_executable('mpv', ['-h']):
args = [
'mpv', '-really-quiet', '--vo=null', '--stream-dump=' + tmpfilename, url]
else:
self.report_error('MMS or RTSP download detected but neither "mplayer" nor "mpv" could be run. Please install any.')
return False
self._debug_cmd(args)
retval = subprocess.call(args)
if retval == 0:
fsize = os.path.getsize(encodeFilename(tmpfilename))
self.to_screen('\r[%s] %s bytes' % (args[0], fsize))
self.try_rename(tmpfilename, filename)
self._hook_progress({
'downloaded_bytes': fsize,
'total_bytes': fsize,
'filename': filename,
'status': 'finished',
})
return True
else:
self.to_stderr('\n')
self.report_error('%s exited with code %d' % (args[0], retval))
return False

46
youtube_dl/extractor/__init__.py
View File

@ -1,46 +0,0 @@
from __future__ import unicode_literals
try:
from .lazy_extractors import *
from .lazy_extractors import _ALL_CLASSES
_LAZY_LOADER = True
except ImportError:
_LAZY_LOADER = False
from .extractors import *
_ALL_CLASSES = [
klass
for name, klass in globals().items()
if name.endswith('IE') and name != 'GenericIE'
]
#_ALL_CLASSES.append(GenericIE)
def gen_extractor_classes():
""" Return a list of supported extractors.
The order does matter; the first extractor matched is the one handling the URL.
"""
return _ALL_CLASSES
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
"""
return [klass() for klass in gen_extractor_classes()]
def list_extractors(age_limit):
"""
Return a list of extractors that are suitable for the given age,
sorted by extractor ID.
"""
return sorted(
filter(lambda ie: ie.is_suitable(age_limit), gen_extractors()),
key=lambda ie: ie.IE_NAME.lower())
def get_info_extractor(ie_name):
"""Returns the info extractor class with the given ie_name"""
return globals()[ie_name + 'IE']

1567
youtube_dl/extractor/adobepass.py
View File

File diff suppressed because it is too large Load Diff

2862
youtube_dl/extractor/common.py
View File

File diff suppressed because it is too large Load Diff

50
youtube_dl/extractor/commonmistakes.py
View File

@ -1,50 +0,0 @@
from __future__ import unicode_literals
import sys
from .common import InfoExtractor
from ..utils import ExtractorError
class CommonMistakesIE(InfoExtractor):
IE_DESC = False # Do not list
_VALID_URL = r'''(?x)
(?:url|URL)$
'''
_TESTS = [{
'url': 'url',
'only_matching': True,
}, {
'url': 'URL',
'only_matching': True,
}]
def _real_extract(self, url):
msg = (
'You\'ve asked youtube-dl to download the URL "%s". '
'That doesn\'t make any sense. '
'Simply remove the parameter in your command or configuration.'
) % url
if not self._downloader.params.get('verbose'):
msg += ' Add -v to the command line to see what arguments and configuration youtube-dl got.'
raise ExtractorError(msg, expected=True)
class UnicodeBOMIE(InfoExtractor):
IE_DESC = False
_VALID_URL = r'(?P<bom>\ufeff)(?P<id>.*)$'
# Disable test for python 3.2 since BOM is broken in re in this version
# (see https://github.com/rg3/youtube-dl/issues/9751)
_TESTS = [] if (3, 0) < sys.version_info <= (3, 3) else [{
'url': '\ufeffhttp://www.youtube.com/watch?v=BaW_jenozKc',
'only_matching': True,
}]
def _real_extract(self, url):
real_url = self._match_id(url)
self.report_warning(
'Your URL starts with a Byte Order Mark (BOM). '
'Removing the BOM and looking for "%s" ...' % real_url)
return self.url_result(real_url)

60
youtube_dl/extractor/commonprotocols.py
View File

@ -1,60 +0,0 @@
from __future__ import unicode_literals
from .common import InfoExtractor
from ..compat import (
compat_urlparse,
)
class RtmpIE(InfoExtractor):
IE_DESC = False # Do not list
_VALID_URL = r'(?i)rtmp[est]?://.+'
_TESTS = [{
'url': 'rtmp://cp44293.edgefcs.net/ondemand?auth=daEcTdydfdqcsb8cZcDbAaCbhamacbbawaS-bw7dBb-bWG-GqpGFqCpNCnGoyL&aifp=v001&slist=public/unsecure/audio/2c97899446428e4301471a8cb72b4b97--audio--pmg-20110908-0900a_flv_aac_med_int.mp4',
'only_matching': True,
}, {
'url': 'rtmp://edge.live.hitbox.tv/live/dimak',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._generic_id(url)
title = self._generic_title(url)
return {
'id': video_id,
'title': title,
'formats': [{
'url': url,
'ext': 'flv',
'format_id': compat_urlparse.urlparse(url).scheme,
}],
}
class MmsIE(InfoExtractor):
IE_DESC = False # Do not list
_VALID_URL = r'(?i)mms://.+'
_TEST = {
# Direct MMS link
'url': 'mms://kentro.kaist.ac.kr/200907/MilesReid(0709).wmv',
'info_dict': {
'id': 'MilesReid(0709)',
'ext': 'wmv',
'title': 'MilesReid(0709)',
},
'params': {
'skip_download': True, # rtsp downloads, requiring mplayer or mpv
},
}
def _real_extract(self, url):
video_id = self._generic_id(url)
title = self._generic_title(url)
return {
'id': video_id,
'title': title,
'url': url,
}

31
youtube_dl/extractor/extractors.py
View File

@ -1,31 +0,0 @@
# flake8: noqa
from __future__ import unicode_literals
from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
from .commonprotocols import (
MmsIE,
RtmpIE,
)
from .openload import OpenloadIE
from .youtube import (
YoutubeIE,
YoutubeChannelIE,
YoutubeFavouritesIE,
YoutubeHistoryIE,
YoutubeLiveIE,
YoutubePlaylistIE,
YoutubePlaylistsIE,
YoutubeRecommendedIE,
YoutubeSearchDateIE,
YoutubeSearchIE,
YoutubeSearchURLIE,
YoutubeShowIE,
YoutubeSubscriptionsIE,
YoutubeTruncatedIDIE,
YoutubeTruncatedURLIE,
YoutubeUserIE,
YoutubeWatchLaterIE,
)

3335
youtube_dl/extractor/generic.py
View File

File diff suppressed because it is too large Load Diff

379
youtube_dl/extractor/openload.py
View File

@ -1,379 +0,0 @@
# coding: utf-8
from __future__ import unicode_literals
import json
import os
import re
import subprocess
import tempfile
from .common import InfoExtractor
from ..compat import (
compat_urlparse,
compat_kwargs,
)
from ..utils import (
check_executable,
determine_ext,
encodeArgument,
ExtractorError,
get_element_by_id,
get_exe_version,
is_outdated_version,
std_headers,
)
def cookie_to_dict(cookie):
cookie_dict = {
'name': cookie.name,
'value': cookie.value,
}
if cookie.port_specified:
cookie_dict['port'] = cookie.port
if cookie.domain_specified:
cookie_dict['domain'] = cookie.domain
if cookie.path_specified:
cookie_dict['path'] = cookie.path
if cookie.expires is not None:
cookie_dict['expires'] = cookie.expires
if cookie.secure is not None:
cookie_dict['secure'] = cookie.secure
if cookie.discard is not None:
cookie_dict['discard'] = cookie.discard
try:
if (cookie.has_nonstandard_attr('httpOnly') or
cookie.has_nonstandard_attr('httponly') or
cookie.has_nonstandard_attr('HttpOnly')):
cookie_dict['httponly'] = True
except TypeError:
pass
return cookie_dict
def cookie_jar_to_list(cookie_jar):
return [cookie_to_dict(cookie) for cookie in cookie_jar]
class PhantomJSwrapper(object):
"""PhantomJS wrapper class
This class is experimental.
"""
_TEMPLATE = r'''
phantom.onError = function(msg, trace) {{
var msgStack = ['PHANTOM ERROR: ' + msg];
if(trace && trace.length) {{
msgStack.push('TRACE:');
trace.forEach(function(t) {{
msgStack.push(' -> ' + (t.file || t.sourceURL) + ': ' + t.line
+ (t.function ? ' (in function ' + t.function +')' : ''));
}});
}}
console.error(msgStack.join('\n'));
phantom.exit(1);
}};
var page = require('webpage').create();
var fs = require('fs');
var read = {{ mode: 'r', charset: 'utf-8' }};
var write = {{ mode: 'w', charset: 'utf-8' }};
JSON.parse(fs.read("{cookies}", read)).forEach(function(x) {{
phantom.addCookie(x);
}});
page.settings.resourceTimeout = {timeout};
page.settings.userAgent = "{ua}";
page.onLoadStarted = function() {{
page.evaluate(function() {{
delete window._phantom;
delete window.callPhantom;
}});
}};
var saveAndExit = function() {{
fs.write("{html}", page.content, write);
fs.write("{cookies}", JSON.stringify(phantom.cookies), write);
phantom.exit();
}};
page.onLoadFinished = function(status) {{
if(page.url === "") {{
page.setContent(fs.read("{html}", read), "{url}");
}}
else {{
{jscode}
}}
}};
page.open("");
'''
_TMP_FILE_NAMES = ['script', 'html', 'cookies']
@staticmethod
def _version():
return get_exe_version('phantomjs', version_re=r'([0-9.]+)')
def __init__(self, extractor, required_version=None, timeout=10000):
self._TMP_FILES = {}
self.exe = check_executable('phantomjs', ['-v'])
if not self.exe:
raise ExtractorError('PhantomJS executable not found in PATH, '
'download it from http://phantomjs.org',
expected=True)
self.extractor = extractor
if required_version:
version = self._version()
if is_outdated_version(version, required_version):
self.extractor._downloader.report_warning(
'Your copy of PhantomJS is outdated, update it to version '
'%s or newer if you encounter any errors.' % required_version)
self.options = {
'timeout': timeout,
}
for name in self._TMP_FILE_NAMES:
tmp = tempfile.NamedTemporaryFile(delete=False)
tmp.close()
self._TMP_FILES[name] = tmp
def __del__(self):
for name in self._TMP_FILE_NAMES:
try:
os.remove(self._TMP_FILES[name].name)
except (IOError, OSError, KeyError):
pass
def _save_cookies(self, url):
cookies = cookie_jar_to_list(self.extractor._downloader.cookiejar)
for cookie in cookies:
if 'path' not in cookie:
cookie['path'] = '/'
if 'domain' not in cookie:
cookie['domain'] = compat_urlparse.urlparse(url).netloc
with open(self._TMP_FILES['cookies'].name, 'wb') as f:
f.write(json.dumps(cookies).encode('utf-8'))
def _load_cookies(self):
with open(self._TMP_FILES['cookies'].name, 'rb') as f:
cookies = json.loads(f.read().decode('utf-8'))
for cookie in cookies:
if cookie['httponly'] is True:
cookie['rest'] = {'httpOnly': None}
if 'expiry' in cookie:
cookie['expire_time'] = cookie['expiry']
self.extractor._set_cookie(**compat_kwargs(cookie))
def get(self, url, html=None, video_id=None, note=None, note2='Executing JS on webpage', headers={}, jscode='saveAndExit();'):
"""
Downloads webpage (if needed) and executes JS
Params:
url: website url
html: optional, html code of website
video_id: video id
note: optional, displayed when downloading webpage
note2: optional, displayed when executing JS
headers: custom http headers
jscode: code to be executed when page is loaded
Returns tuple with:
* downloaded website (after JS execution)
* anything you print with `console.log` (but not inside `page.execute`!)
In most cases you don't need to add any `jscode`.
It is executed in `page.onLoadFinished`.
`saveAndExit();` is mandatory, use it instead of `phantom.exit()`
It is possible to wait for some element on the webpage, for example:
var check = function() {
var elementFound = page.evaluate(function() {
return document.querySelector('#b.done') !== null;
});
if(elementFound)
saveAndExit();
else
window.setTimeout(check, 500);
}
page.evaluate(function(){
document.querySelector('#a').click();
});
check();
"""
if 'saveAndExit();' not in jscode:
raise ExtractorError('`saveAndExit();` not found in `jscode`')
if not html:
html = self.extractor._download_webpage(url, video_id, note=note, headers=headers)
with open(self._TMP_FILES['html'].name, 'wb') as f:
f.write(html.encode('utf-8'))
self._save_cookies(url)
replaces = self.options
replaces['url'] = url
user_agent = headers.get('User-Agent') or std_headers['User-Agent']
replaces['ua'] = user_agent.replace('"', '\\"')
replaces['jscode'] = jscode
for x in self._TMP_FILE_NAMES:
replaces[x] = self._TMP_FILES[x].name.replace('\\', '\\\\').replace('"', '\\"')
with open(self._TMP_FILES['script'].name, 'wb') as f:
f.write(self._TEMPLATE.format(**replaces).encode('utf-8'))
if video_id is None:
self.extractor.to_screen('%s' % (note2,))
else:
self.extractor.to_screen('%s: %s' % (video_id, note2))
p = subprocess.Popen([
self.exe, '--ssl-protocol=any',
self._TMP_FILES['script'].name
], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
out, err = p.communicate()
if p.returncode != 0:
raise ExtractorError(
'Executing JS failed\n:' + encodeArgument(err))
with open(self._TMP_FILES['html'].name, 'rb') as f:
html = f.read().decode('utf-8')
self._load_cookies()
return (html, encodeArgument(out))
class OpenloadIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?(?:openload\.(?:co|io|link)|oload\.(?:tv|stream|site|xyz|win|download))/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
_TESTS = [{
'url': 'https://openload.co/f/kUEfGclsU9o',
'md5': 'bf1c059b004ebc7a256f89408e65c36e',
'info_dict': {
'id': 'kUEfGclsU9o',
'ext': 'mp4',
'title': 'skyrim_no-audio_1080.mp4',
'thumbnail': r're:^https?://.*\.jpg$',
},
}, {
'url': 'https://openload.co/embed/rjC09fkPLYs',
'info_dict': {
'id': 'rjC09fkPLYs',
'ext': 'mp4',
'title': 'movie.mp4',
'thumbnail': r're:^https?://.*\.jpg$',
'subtitles': {
'en': [{
'ext': 'vtt',
}],
},
},
'params': {
'skip_download': True, # test subtitles only
},
}, {
'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
'only_matching': True,
}, {
'url': 'https://openload.io/f/ZAn6oz-VZGE/',
'only_matching': True,
}, {
'url': 'https://openload.co/f/_-ztPaZtMhM/',
'only_matching': True,
}, {
# unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
# for title and ext
'url': 'https://openload.co/embed/Sxz5sADo82g/',
'only_matching': True,
}, {
# unavailable via https://openload.co/embed/e-Ixz9ZR5L0/ but available
# via https://openload.co/f/e-Ixz9ZR5L0/
'url': 'https://openload.co/f/e-Ixz9ZR5L0/',
'only_matching': True,
}, {
'url': 'https://oload.tv/embed/KnG-kKZdcfY/',
'only_matching': True,
}, {
'url': 'http://www.openload.link/f/KnG-kKZdcfY',
'only_matching': True,
}, {
'url': 'https://oload.stream/f/KnG-kKZdcfY',
'only_matching': True,
}, {
'url': 'https://oload.xyz/f/WwRBpzW8Wtk',
'only_matching': True,
}, {
'url': 'https://oload.win/f/kUEfGclsU9o',
'only_matching': True,
}, {
'url': 'https://oload.download/f/kUEfGclsU9o',
'only_matching': True,
}, {
# Its title has not got its extension but url has it
'url': 'https://oload.download/f/N4Otkw39VCw/Tomb.Raider.2018.HDRip.XviD.AC3-EVO.avi.mp4',
'only_matching': True,
}]
_USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)',
webpage)
def _real_extract(self, url):
video_id = self._match_id(url)
url_pattern = 'https://openload.co/%%s/%s/' % video_id
headers = {
'User-Agent': self._USER_AGENT,
}
for path in ('embed', 'f'):
page_url = url_pattern % path
last = path == 'f'
webpage = self._download_webpage(
page_url, video_id, 'Downloading %s webpage' % path,
headers=headers, fatal=last)
if not webpage:
continue
if 'File not found' in webpage or 'deleted by the owner' in webpage:
if not last:
continue
raise ExtractorError('File not found', expected=True, video_id=video_id)
break
phantom = PhantomJSwrapper(self, required_version='2.0')
webpage, _ = phantom.get(page_url, html=webpage, video_id=video_id, headers=headers)
decoded_id = (get_element_by_id('streamurl', webpage) or
get_element_by_id('streamuri', webpage) or
get_element_by_id('streamurj', webpage) or
self._search_regex(
(r'>\s*([\w-]+~\d{10,}~\d+\.\d+\.0\.0~[\w-]+)\s*<',
r'>\s*([\w~-]+~\d+\.\d+\.\d+\.\d+~[\w~-]+)',
r'>\s*([\w-]+~\d{10,}~(?:[a-f\d]+:){2}:~[\w-]+)\s*<',
r'>\s*([\w~-]+~[a-f0-9:]+~[\w~-]+)\s*<',
r'>\s*([\w~-]+~[a-f0-9:]+~[\w~-]+)'), webpage,
'stream URL'))
video_url = 'https://openload.co/stream/%s?mime=true' % decoded_id
title = self._og_search_title(webpage, default=None) or self._search_regex(
r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
'title', default=None) or self._html_search_meta(
'description', webpage, 'title', fatal=True)
entries = self._parse_html5_media_entries(page_url, webpage, video_id)
entry = entries[0] if entries else {}
subtitles = entry.get('subtitles')
info_dict = {
'id': video_id,
'title': title,
'thumbnail': entry.get('thumbnail') or self._og_search_thumbnail(webpage, default=None),
'url': video_url,
'ext': determine_ext(title, None) or determine_ext(url, 'mp4'),
'subtitles': subtitles,
'http_headers': headers,
}
return info_dict

3394
youtube_dl/extractor/youtube.py
View File

File diff suppressed because it is too large Load Diff

3325
youtube_dl/extractor/youtube_unmodified_reference.py
View File

File diff suppressed because it is too large Load Diff

262
youtube_dl/jsinterp.py
View File

@ -1,262 +0,0 @@
from __future__ import unicode_literals
import json
import operator
import re
from .utils import (
ExtractorError,
remove_quotes,
)
_OPERATORS = [
('|', operator.or_),
('^', operator.xor),
('&', operator.and_),
('>>', operator.rshift),
('<<', operator.lshift),
('-', operator.sub),
('+', operator.add),
('%', operator.mod),
('/', operator.truediv),
('*', operator.mul),
]
_ASSIGN_OPERATORS = [(op + '=', opfunc) for op, opfunc in _OPERATORS]
_ASSIGN_OPERATORS.append(('=', lambda cur, right: right))
_NAME_RE = r'[a-zA-Z_$][a-zA-Z_$0-9]*'
class JSInterpreter(object):
def __init__(self, code, objects=None):
if objects is None:
objects = {}
self.code = code
self._functions = {}
self._objects = objects
def interpret_statement(self, stmt, local_vars, allow_recursion=100):
if allow_recursion < 0:
raise ExtractorError('Recursion limit reached')
should_abort = False
stmt = stmt.lstrip()
stmt_m = re.match(r'var\s', stmt)
if stmt_m:
expr = stmt[len(stmt_m.group(0)):]
else:
return_m = re.match(r'return(?:\s+|$)', stmt)
if return_m:
expr = stmt[len(return_m.group(0)):]
should_abort = True
else:
# Try interpreting it as an expression
expr = stmt
v = self.interpret_expression(expr, local_vars, allow_recursion)
return v, should_abort
def interpret_expression(self, expr, local_vars, allow_recursion):
expr = expr.strip()
if expr == '': # Empty expression
return None
if expr.startswith('('):
parens_count = 0
for m in re.finditer(r'[()]', expr):
if m.group(0) == '(':
parens_count += 1
else:
parens_count -= 1
if parens_count == 0:
sub_expr = expr[1:m.start()]
sub_result = self.interpret_expression(
sub_expr, local_vars, allow_recursion)
remaining_expr = expr[m.end():].strip()
if not remaining_expr:
return sub_result
else:
expr = json.dumps(sub_result) + remaining_expr
break
else:
raise ExtractorError('Premature end of parens in %r' % expr)
for op, opfunc in _ASSIGN_OPERATORS:
m = re.match(r'''(?x)
(?P<out>%s)(?:\[(?P<index>[^\]]+?)\])?
\s*%s
(?P<expr>.*)$''' % (_NAME_RE, re.escape(op)), expr)
if not m:
continue
right_val = self.interpret_expression(
m.group('expr'), local_vars, allow_recursion - 1)
if m.groupdict().get('index'):
lvar = local_vars[m.group('out')]
idx = self.interpret_expression(
m.group('index'), local_vars, allow_recursion)
assert isinstance(idx, int)
cur = lvar[idx]
val = opfunc(cur, right_val)
lvar[idx] = val
return val
else:
cur = local_vars.get(m.group('out'))
val = opfunc(cur, right_val)
local_vars[m.group('out')] = val
return val
if expr.isdigit():
return int(expr)
var_m = re.match(
r'(?!if|return|true|false)(?P<name>%s)$' % _NAME_RE,
expr)
if var_m:
return local_vars[var_m.group('name')]
try:
return json.loads(expr)
except ValueError:
pass
m = re.match(
r'(?P<in>%s)\[(?P<idx>.+)\]$' % _NAME_RE, expr)
if m:
val = local_vars[m.group('in')]
idx = self.interpret_expression(
m.group('idx'), local_vars, allow_recursion - 1)
return val[idx]
m = re.match(
r'(?P<var>%s)(?:\.(?P<member>[^(]+)|\[(?P<member2>[^]]+)\])\s*(?:\(+(?P<args>[^()]*)\))?$' % _NAME_RE,
expr)
if m:
variable = m.group('var')
member = remove_quotes(m.group('member') or m.group('member2'))
arg_str = m.group('args')
if variable in local_vars:
obj = local_vars[variable]
else:
if variable not in self._objects:
self._objects[variable] = self.extract_object(variable)
obj = self._objects[variable]
if arg_str is None:
# Member access
if member == 'length':
return len(obj)
return obj[member]
assert expr.endswith(')')
# Function call
if arg_str == '':
argvals = tuple()
else:
argvals = tuple([
self.interpret_expression(v, local_vars, allow_recursion)
for v in arg_str.split(',')])
if member == 'split':
assert argvals == ('',)
return list(obj)
if member == 'join':
assert len(argvals) == 1
return argvals[0].join(obj)
if member == 'reverse':
assert len(argvals) == 0
obj.reverse()
return obj
if member == 'slice':
assert len(argvals) == 1
return obj[argvals[0]:]
if member == 'splice':
assert isinstance(obj, list)
index, howMany = argvals
res = []
for i in range(index, min(index + howMany, len(obj))):
res.append(obj.pop(index))
return res
return obj[member](argvals)
for op, opfunc in _OPERATORS:
m = re.match(r'(?P<x>.+?)%s(?P<y>.+)' % re.escape(op), expr)
if not m:
continue
x, abort = self.interpret_statement(
m.group('x'), local_vars, allow_recursion - 1)
if abort:
raise ExtractorError(
'Premature left-side return of %s in %r' % (op, expr))
y, abort = self.interpret_statement(
m.group('y'), local_vars, allow_recursion - 1)
if abort:
raise ExtractorError(
'Premature right-side return of %s in %r' % (op, expr))
return opfunc(x, y)
m = re.match(
r'^(?P<func>%s)\((?P<args>[a-zA-Z0-9_$,]*)\)$' % _NAME_RE, expr)
if m:
fname = m.group('func')
argvals = tuple([
int(v) if v.isdigit() else local_vars[v]
for v in m.group('args').split(',')]) if len(m.group('args')) > 0 else tuple()
if fname not in self._functions:
self._functions[fname] = self.extract_function(fname)
return self._functions[fname](argvals)
raise ExtractorError('Unsupported JS expression %r' % expr)
def extract_object(self, objname):
_FUNC_NAME_RE = r'''(?:[a-zA-Z$0-9]+|"[a-zA-Z$0-9]+"|'[a-zA-Z$0-9]+')'''
obj = {}
obj_m = re.search(
r'''(?x)
(?<!this\.)%s\s*=\s*{\s*
(?P<fields>(%s\s*:\s*function\s*\(.*?\)\s*{.*?}(?:,\s*)?)*)
}\s*;
''' % (re.escape(objname), _FUNC_NAME_RE),
self.code)
fields = obj_m.group('fields')
# Currently, it only supports function definitions
fields_m = re.finditer(
r'''(?x)
(?P<key>%s)\s*:\s*function\s*\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}
''' % _FUNC_NAME_RE,
fields)
for f in fields_m:
argnames = f.group('args').split(',')
obj[remove_quotes(f.group('key'))] = self.build_function(argnames, f.group('code'))
return obj
def extract_function(self, funcname):
func_m = re.search(
r'''(?x)
(?:function\s+%s|[{;,]\s*%s\s*=\s*function|var\s+%s\s*=\s*function)\s*
\((?P<args>[^)]*)\)\s*
\{(?P<code>[^}]+)\}''' % (
re.escape(funcname), re.escape(funcname), re.escape(funcname)),
self.code)
if func_m is None:
raise ExtractorError('Could not find JS function %r' % funcname)
argnames = func_m.group('args').split(',')
return self.build_function(argnames, func_m.group('code'))
def call_function(self, funcname, *args):
f = self.extract_function(funcname)
return f(args)
def build_function(self, argnames, code):
def resf(args):
local_vars = dict(zip(argnames, args))
for stmt in code.split(';'):
res, abort = self.interpret_statement(stmt, local_vars)
if abort:
break
return res
return resf

916
youtube_dl/options.py
View File

@ -1,916 +0,0 @@
from __future__ import unicode_literals
import os.path
import optparse
import re
import sys
from .downloader.external import list_external_downloaders
from .compat import (
compat_expanduser,
compat_get_terminal_size,
compat_getenv,
compat_kwargs,
compat_shlex_split,
)
from .utils import (
preferredencoding,
write_string,
)
from .version import __version__
def _hide_login_info(opts):
PRIVATE_OPTS = set(['-p', '--password', '-u', '--username', '--video-password', '--ap-password', '--ap-username'])
eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$')
def _scrub_eq(o):
m = eqre.match(o)
if m:
return m.group('key') + '=PRIVATE'
else:
return o
opts = list(map(_scrub_eq, opts))
for idx, opt in enumerate(opts):
if opt in PRIVATE_OPTS and idx + 1 < len(opts):
opts[idx + 1] = 'PRIVATE'
return opts
def parseOpts(overrideArguments=None):
def _readOptions(filename_bytes, default=[]):
try:
optionf = open(filename_bytes)
except IOError:
return default # silently skip if file is not present
try:
# FIXME: https://github.com/rg3/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56
contents = optionf.read()
if sys.version_info < (3,):
contents = contents.decode(preferredencoding())
res = compat_shlex_split(contents, comments=True)
finally:
optionf.close()
return res
def _readUserConf():
xdg_config_home = compat_getenv('XDG_CONFIG_HOME')
if xdg_config_home:
userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
if not os.path.isfile(userConfFile):
userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
else:
userConfFile = os.path.join(compat_expanduser('~'), '.config', 'youtube-dl', 'config')
if not os.path.isfile(userConfFile):
userConfFile = os.path.join(compat_expanduser('~'), '.config', 'youtube-dl.conf')
userConf = _readOptions(userConfFile, None)
if userConf is None:
appdata_dir = compat_getenv('appdata')
if appdata_dir:
userConf = _readOptions(
os.path.join(appdata_dir, 'youtube-dl', 'config'),
default=None)
if userConf is None:
userConf = _readOptions(
os.path.join(appdata_dir, 'youtube-dl', 'config.txt'),
default=None)
if userConf is None:
userConf = _readOptions(
os.path.join(compat_expanduser('~'), 'youtube-dl.conf'),
default=None)
if userConf is None:
userConf = _readOptions(
os.path.join(compat_expanduser('~'), 'youtube-dl.conf.txt'),
default=None)
if userConf is None:
userConf = []
return userConf
def _format_option_string(option):
''' ('-o', '--option') -> -o, --format METAVAR'''
opts = []
if option._short_opts:
opts.append(option._short_opts[0])
if option._long_opts:
opts.append(option._long_opts[0])
if len(opts) > 1:
opts.insert(1, ', ')
if option.takes_value():
opts.append(' %s' % option.metavar)
return ''.join(opts)
def _comma_separated_values_options_callback(option, opt_str, value, parser):
setattr(parser.values, option.dest, value.split(','))
# No need to wrap help messages if we're on a wide console
columns = compat_get_terminal_size().columns
max_width = columns if columns else 80
max_help_position = 80
fmt = optparse.IndentedHelpFormatter(width=max_width, max_help_position=max_help_position)
fmt.format_option_strings = _format_option_string
kw = {
'version': __version__,
'formatter': fmt,
'usage': '%prog [OPTIONS] URL [URL...]',
'conflict_handler': 'resolve',
}
parser = optparse.OptionParser(**compat_kwargs(kw))
general = optparse.OptionGroup(parser, 'General Options')
general.add_option(
'-h', '--help',
action='help',
help='Print this help text and exit')
general.add_option(
'-v', '--version',
action='version',
help='Print program version and exit')
general.add_option(
'-U', '--update',
action='store_true', dest='update_self',
help='Update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)')
general.add_option(
'-i', '--ignore-errors',
action='store_true', dest='ignoreerrors', default=False,
help='Continue on download errors, for example to skip unavailable videos in a playlist')
general.add_option(
'--abort-on-error',
action='store_false', dest='ignoreerrors',
help='Abort downloading of further videos (in the playlist or the command line) if an error occurs')
general.add_option(
'--dump-user-agent',
action='store_true', dest='dump_user_agent', default=False,
help='Display the current browser identification')
general.add_option(
'--list-extractors',
action='store_true', dest='list_extractors', default=False,
help='List all supported extractors')
general.add_option(
'--extractor-descriptions',
action='store_true', dest='list_extractor_descriptions', default=False,
help='Output descriptions of all supported extractors')
general.add_option(
'--force-generic-extractor',
action='store_true', dest='force_generic_extractor', default=False,
help='Force extraction to use the generic extractor')
general.add_option(
'--default-search',
dest='default_search', metavar='PREFIX',
help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for youtube-dl "large apple". Use the value "auto" to let youtube-dl guess ("auto_warning" to emit a warning when guessing). "error" just throws an error. The default value "fixup_error" repairs broken URLs, but emits an error if this is not possible instead of searching.')
general.add_option(
'--ignore-config',
action='store_true',
help='Do not read configuration files. '
'When given in the global configuration file /etc/youtube-dl.conf: '
'Do not read the user configuration in ~/.config/youtube-dl/config '
'(%APPDATA%/youtube-dl/config.txt on Windows)')
general.add_option(
'--config-location',
dest='config_location', metavar='PATH',
help='Location of the configuration file; either the path to the config or its containing directory.')
general.add_option(
'--flat-playlist',
action='store_const', dest='extract_flat', const='in_playlist',
default=False,
help='Do not extract the videos of a playlist, only list them.')
general.add_option(
'--mark-watched',
action='store_true', dest='mark_watched', default=False,
help='Mark videos watched (YouTube only)')
general.add_option(
'--no-mark-watched',
action='store_false', dest='mark_watched', default=False,
help='Do not mark videos watched (YouTube only)')
general.add_option(
'--no-color', '--no-colors',
action='store_true', dest='no_color',
default=False,
help='Do not emit color codes in output')
network = optparse.OptionGroup(parser, 'Network Options')
network.add_option(
'--proxy', dest='proxy',
default=None, metavar='URL',
help='Use the specified HTTP/HTTPS/SOCKS proxy. To enable '
'SOCKS proxy, specify a proper scheme. For example '
'socks5://127.0.0.1:1080/. Pass in an empty string (--proxy "") '
'for direct connection')
network.add_option(
'--socket-timeout',
dest='socket_timeout', type=float, default=None, metavar='SECONDS',
help='Time to wait before giving up, in seconds')
network.add_option(
'--source-address',
metavar='IP', dest='source_address', default=None,
help='Client-side IP address to bind to',
)
network.add_option(
'-4', '--force-ipv4',
action='store_const', const='0.0.0.0', dest='source_address',
help='Make all connections via IPv4',
)
network.add_option(
'-6', '--force-ipv6',
action='store_const', const='::', dest='source_address',
help='Make all connections via IPv6',
)
geo = optparse.OptionGroup(parser, 'Geo Restriction')
geo.add_option(
'--geo-verification-proxy',
dest='geo_verification_proxy', default=None, metavar='URL',
help='Use this proxy to verify the IP address for some geo-restricted sites. '
'The default proxy specified by --proxy (or none, if the option is not present) is used for the actual downloading.')
geo.add_option(
'--cn-verification-proxy',
dest='cn_verification_proxy', default=None, metavar='URL',
help=optparse.SUPPRESS_HELP)
geo.add_option(
'--geo-bypass',
action='store_true', dest='geo_bypass', default=True,
help='Bypass geographic restriction via faking X-Forwarded-For HTTP header')
geo.add_option(
'--no-geo-bypass',
action='store_false', dest='geo_bypass', default=True,
help='Do not bypass geographic restriction via faking X-Forwarded-For HTTP header')
geo.add_option(
'--geo-bypass-country', metavar='CODE',
dest='geo_bypass_country', default=None,
help='Force bypass geographic restriction with explicitly provided two-letter ISO 3166-2 country code')
geo.add_option(
'--geo-bypass-ip-block', metavar='IP_BLOCK',
dest='geo_bypass_ip_block', default=None,
help='Force bypass geographic restriction with explicitly provided IP block in CIDR notation')
selection = optparse.OptionGroup(parser, 'Video Selection')
selection.add_option(
'--playlist-start',
dest='playliststart', metavar='NUMBER', default=1, type=int,
help='Playlist video to start at (default is %default)')
selection.add_option(
'--playlist-end',
dest='playlistend', metavar='NUMBER', default=None, type=int,
help='Playlist video to end at (default is last)')
selection.add_option(
'--playlist-items',
dest='playlist_items', metavar='ITEM_SPEC', default=None,
help='Playlist video items to download. Specify indices of the videos in the playlist separated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.')
selection.add_option(
'--match-title',
dest='matchtitle', metavar='REGEX',
help='Download only matching titles (regex or caseless sub-string)')
selection.add_option(
'--reject-title',
dest='rejecttitle', metavar='REGEX',
help='Skip download for matching titles (regex or caseless sub-string)')
selection.add_option(
'--max-downloads',
dest='max_downloads', metavar='NUMBER', type=int, default=None,
help='Abort after downloading NUMBER files')
selection.add_option(
'--min-filesize',
metavar='SIZE', dest='min_filesize', default=None,
help='Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)')
selection.add_option(
'--max-filesize',
metavar='SIZE', dest='max_filesize', default=None,
help='Do not download any videos larger than SIZE (e.g. 50k or 44.6m)')
selection.add_option(
'--date',
metavar='DATE', dest='date', default=None,
help='Download only videos uploaded in this date')
selection.add_option(
'--datebefore',
metavar='DATE', dest='datebefore', default=None,
help='Download only videos uploaded on or before this date (i.e. inclusive)')
selection.add_option(
'--dateafter',
metavar='DATE', dest='dateafter', default=None,
help='Download only videos uploaded on or after this date (i.e. inclusive)')
selection.add_option(
'--min-views',
metavar='COUNT', dest='min_views', default=None, type=int,
help='Do not download any videos with less than COUNT views')
selection.add_option(
'--max-views',
metavar='COUNT', dest='max_views', default=None, type=int,
help='Do not download any videos with more than COUNT views')
selection.add_option(
'--match-filter',
metavar='FILTER', dest='match_filter', default=None,
help=(
'Generic video filter. '
'Specify any key (see the "OUTPUT TEMPLATE" for a list of available keys) to '
'match if the key is present, '
'!key to check if the key is not present, '
'key > NUMBER (like "comment_count > 12", also works with '
'>=, <, <=, !=, =) to compare against a number, '
'key = \'LITERAL\' (like "uploader = \'Mike Smith\'", also works with !=) '
'to match against a string literal '
'and & to require multiple matches. '
'Values which are not known are excluded unless you '
'put a question mark (?) after the operator. '
'For example, to only match videos that have been liked more than '
'100 times and disliked less than 50 times (or the dislike '
'functionality is not available at the given service), but who '
'also have a description, use --match-filter '
'"like_count > 100 & dislike_count <? 50 & description" .'
))
selection.add_option(
'--no-playlist',
action='store_true', dest='noplaylist', default=False,
help='Download only the video, if the URL refers to a video and a playlist.')
selection.add_option(
'--yes-playlist',
action='store_false', dest='noplaylist', default=False,
help='Download the playlist, if the URL refers to a video and a playlist.')
selection.add_option(
'--age-limit',
metavar='YEARS', dest='age_limit', default=None, type=int,
help='Download only videos suitable for the given age')
selection.add_option(
'--download-archive', metavar='FILE',
dest='download_archive',
help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.')
selection.add_option(
'--include-ads',
dest='include_ads', action='store_true',
help='Download advertisements as well (experimental)')
authentication = optparse.OptionGroup(parser, 'Authentication Options')
authentication.add_option(
'-u', '--username',
dest='username', metavar='USERNAME',
help='Login with this account ID')
authentication.add_option(
'-p', '--password',
dest='password', metavar='PASSWORD',
help='Account password. If this option is left out, youtube-dl will ask interactively.')
authentication.add_option(
'-2', '--twofactor',
dest='twofactor', metavar='TWOFACTOR',
help='Two-factor authentication code')
authentication.add_option(
'-n', '--netrc',
action='store_true', dest='usenetrc', default=False,
help='Use .netrc authentication data')
authentication.add_option(
'--video-password',
dest='videopassword', metavar='PASSWORD',
help='Video password (vimeo, smotri, youku)')
adobe_pass = optparse.OptionGroup(parser, 'Adobe Pass Options')
adobe_pass.add_option(
'--ap-mso',
dest='ap_mso', metavar='MSO',
help='Adobe Pass multiple-system operator (TV provider) identifier, use --ap-list-mso for a list of available MSOs')
adobe_pass.add_option(
'--ap-username',
dest='ap_username', metavar='USERNAME',
help='Multiple-system operator account login')
adobe_pass.add_option(
'--ap-password',
dest='ap_password', metavar='PASSWORD',
help='Multiple-system operator account password. If this option is left out, youtube-dl will ask interactively.')
adobe_pass.add_option(
'--ap-list-mso',
action='store_true', dest='ap_list_mso', default=False,
help='List all supported multiple-system operators')
video_format = optparse.OptionGroup(parser, 'Video Format Options')
video_format.add_option(
'-f', '--format',
action='store', dest='format', metavar='FORMAT', default=None,
help='Video format code, see the "FORMAT SELECTION" for all the info')
video_format.add_option(
'--all-formats',
action='store_const', dest='format', const='all',
help='Download all available video formats')
video_format.add_option(
'--prefer-free-formats',
action='store_true', dest='prefer_free_formats', default=False,
help='Prefer free video formats unless a specific one is requested')
video_format.add_option(
'-F', '--list-formats',
action='store_true', dest='listformats',
help='List all available formats of requested videos')
video_format.add_option(
'--youtube-include-dash-manifest',
action='store_true', dest='youtube_include_dash_manifest', default=True,
help=optparse.SUPPRESS_HELP)
video_format.add_option(
'--youtube-skip-dash-manifest',
action='store_false', dest='youtube_include_dash_manifest',
help='Do not download the DASH manifests and related data on YouTube videos')
video_format.add_option(
'--merge-output-format',
action='store', dest='merge_output_format', metavar='FORMAT', default=None,
help=(
'If a merge is required (e.g. bestvideo+bestaudio), '
'output to given container format. One of mkv, mp4, ogg, webm, flv. '
'Ignored if no merge is required'))
subtitles = optparse.OptionGroup(parser, 'Subtitle Options')
subtitles.add_option(
'--write-sub', '--write-srt',
action='store_true', dest='writesubtitles', default=False,
help='Write subtitle file')
subtitles.add_option(
'--write-auto-sub', '--write-automatic-sub',
action='store_true', dest='writeautomaticsub', default=False,
help='Write automatically generated subtitle file (YouTube only)')
subtitles.add_option(
'--all-subs',
action='store_true', dest='allsubtitles', default=False,
help='Download all the available subtitles of the video')
subtitles.add_option(
'--list-subs',
action='store_true', dest='listsubtitles', default=False,
help='List all available subtitles for the video')
subtitles.add_option(
'--sub-format',
action='store', dest='subtitlesformat', metavar='FORMAT', default='best',
help='Subtitle format, accepts formats preference, for example: "srt" or "ass/srt/best"')
subtitles.add_option(
'--sub-lang', '--sub-langs', '--srt-lang',
action='callback', dest='subtitleslangs', metavar='LANGS', type='str',
default=[], callback=_comma_separated_values_options_callback,
help='Languages of the subtitles to download (optional) separated by commas, use --list-subs for available language tags')
downloader = optparse.OptionGroup(parser, 'Download Options')
downloader.add_option(
'-r', '--limit-rate', '--rate-limit',
dest='ratelimit', metavar='RATE',
help='Maximum download rate in bytes per second (e.g. 50K or 4.2M)')
downloader.add_option(
'-R', '--retries',
dest='retries', metavar='RETRIES', default=10,
help='Number of retries (default is %default), or "infinite".')
downloader.add_option(
'--fragment-retries',
dest='fragment_retries', metavar='RETRIES', default=10,
help='Number of retries for a fragment (default is %default), or "infinite" (DASH, hlsnative and ISM)')
downloader.add_option(
'--skip-unavailable-fragments',
action='store_true', dest='skip_unavailable_fragments', default=True,
help='Skip unavailable fragments (DASH, hlsnative and ISM)')
downloader.add_option(
'--abort-on-unavailable-fragment',
action='store_false', dest='skip_unavailable_fragments',
help='Abort downloading when some fragment is not available')
downloader.add_option(
'--keep-fragments',
action='store_true', dest='keep_fragments', default=False,
help='Keep downloaded fragments on disk after downloading is finished; fragments are erased by default')
downloader.add_option(
'--buffer-size',
dest='buffersize', metavar='SIZE', default='1024',
help='Size of download buffer (e.g. 1024 or 16K) (default is %default)')
downloader.add_option(
'--no-resize-buffer',
action='store_true', dest='noresizebuffer', default=False,
help='Do not automatically adjust the buffer size. By default, the buffer size is automatically resized from an initial value of SIZE.')
downloader.add_option(
'--http-chunk-size',
dest='http_chunk_size', metavar='SIZE', default=None,
help='Size of a chunk for chunk-based HTTP downloading (e.g. 10485760 or 10M) (default is disabled). '
'May be useful for bypassing bandwidth throttling imposed by a webserver (experimental)')
downloader.add_option(
'--test',
action='store_true', dest='test', default=False,
help=optparse.SUPPRESS_HELP)
downloader.add_option(
'--playlist-reverse',
action='store_true',
help='Download playlist videos in reverse order')
downloader.add_option(
'--playlist-random',
action='store_true',
help='Download playlist videos in random order')
downloader.add_option(
'--xattr-set-filesize',
dest='xattr_set_filesize', action='store_true',
help='Set file xattribute ytdl.filesize with expected file size')
downloader.add_option(
'--hls-prefer-native',
dest='hls_prefer_native', action='store_true', default=None,
help='Use the native HLS downloader instead of ffmpeg')
downloader.add_option(
'--hls-prefer-ffmpeg',
dest='hls_prefer_native', action='store_false', default=None,
help='Use ffmpeg instead of the native HLS downloader')
downloader.add_option(
'--hls-use-mpegts',
dest='hls_use_mpegts', action='store_true',
help='Use the mpegts container for HLS videos, allowing to play the '
'video while downloading (some players may not be able to play it)')
downloader.add_option(
'--external-downloader',
dest='external_downloader', metavar='COMMAND',
help='Use the specified external downloader. '
'Currently supports %s' % ','.join(list_external_downloaders()))
downloader.add_option(
'--external-downloader-args',
dest='external_downloader_args', metavar='ARGS',
help='Give these arguments to the external downloader')
workarounds = optparse.OptionGroup(parser, 'Workarounds')
workarounds.add_option(
'--encoding',
dest='encoding', metavar='ENCODING',
help='Force the specified encoding (experimental)')
workarounds.add_option(
'--no-check-certificate',
action='store_true', dest='no_check_certificate', default=False,
help='Suppress HTTPS certificate validation')
workarounds.add_option(
'--prefer-insecure',
'--prefer-unsecure', action='store_true', dest='prefer_insecure',
help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)')
workarounds.add_option(
'--user-agent',
metavar='UA', dest='user_agent',
help='Specify a custom user agent')
workarounds.add_option(
'--referer',
metavar='URL', dest='referer', default=None,
help='Specify a custom referer, use if the video access is restricted to one domain',
)
workarounds.add_option(
'--add-header',
metavar='FIELD:VALUE', dest='headers', action='append',
help='Specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times',
)
workarounds.add_option(
'--bidi-workaround',
dest='bidi_workaround', action='store_true',
help='Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
workarounds.add_option(
'--sleep-interval', '--min-sleep-interval', metavar='SECONDS',
dest='sleep_interval', type=float,
help=(
'Number of seconds to sleep before each download when used alone '
'or a lower bound of a range for randomized sleep before each download '
'(minimum possible number of seconds to sleep) when used along with '
'--max-sleep-interval.'))
workarounds.add_option(
'--max-sleep-interval', metavar='SECONDS',
dest='max_sleep_interval', type=float,
help=(
'Upper bound of a range for randomized sleep before each download '
'(maximum possible number of seconds to sleep). Must only be used '
'along with --min-sleep-interval.'))
verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
verbosity.add_option(
'-q', '--quiet',
action='store_true', dest='quiet', default=False,
help='Activate quiet mode')
verbosity.add_option(
'--no-warnings',
dest='no_warnings', action='store_true', default=False,
help='Ignore warnings')
verbosity.add_option(
'-s', '--simulate',
action='store_true', dest='simulate', default=False,
help='Do not download the video and do not write anything to disk')
verbosity.add_option(
'--skip-download',
action='store_true', dest='skip_download', default=False,
help='Do not download the video')
verbosity.add_option(
'-g', '--get-url',
action='store_true', dest='geturl', default=False,
help='Simulate, quiet but print URL')
verbosity.add_option(
'-e', '--get-title',
action='store_true', dest='gettitle', default=False,
help='Simulate, quiet but print title')
verbosity.add_option(
'--get-id',
action='store_true', dest='getid', default=False,
help='Simulate, quiet but print id')
verbosity.add_option(
'--get-thumbnail',
action='store_true', dest='getthumbnail', default=False,
help='Simulate, quiet but print thumbnail URL')
verbosity.add_option(
'--get-description',
action='store_true', dest='getdescription', default=False,
help='Simulate, quiet but print video description')
verbosity.add_option(
'--get-duration',
action='store_true', dest='getduration', default=False,
help='Simulate, quiet but print video length')
verbosity.add_option(
'--get-filename',
action='store_true', dest='getfilename', default=False,
help='Simulate, quiet but print output filename')
verbosity.add_option(
'--get-format',
action='store_true', dest='getformat', default=False,
help='Simulate, quiet but print output format')
verbosity.add_option(
'-j', '--dump-json',
action='store_true', dest='dumpjson', default=False,
help='Simulate, quiet but print JSON information. See the "OUTPUT TEMPLATE" for a description of available keys.')
verbosity.add_option(
'-J', '--dump-single-json',
action='store_true', dest='dump_single_json', default=False,
help='Simulate, quiet but print JSON information for each command-line argument. If the URL refers to a playlist, dump the whole playlist information in a single line.')
verbosity.add_option(
'--print-json',
action='store_true', dest='print_json', default=False,
help='Be quiet and print the video information as JSON (video is still being downloaded).',
)
verbosity.add_option(
'--newline',
action='store_true', dest='progress_with_newline', default=False,
help='Output progress bar as new lines')
verbosity.add_option(
'--no-progress',
action='store_true', dest='noprogress', default=False,
help='Do not print progress bar')
verbosity.add_option(
'--console-title',
action='store_true', dest='consoletitle', default=False,
help='Display progress in console titlebar')
verbosity.add_option(
'-v', '--verbose',
action='store_true', dest='verbose', default=False,
help='Print various debugging information')
verbosity.add_option(
'--dump-pages', '--dump-intermediate-pages',
action='store_true', dest='dump_intermediate_pages', default=False,
help='Print downloaded pages encoded using base64 to debug problems (very verbose)')
verbosity.add_option(
'--write-pages',
action='store_true', dest='write_pages', default=False,
help='Write downloaded intermediary pages to files in the current directory to debug problems')
verbosity.add_option(
'--youtube-print-sig-code',
action='store_true', dest='youtube_print_sig_code', default=False,
help=optparse.SUPPRESS_HELP)
verbosity.add_option(
'--print-traffic', '--dump-headers',
dest='debug_printtraffic', action='store_true', default=False,
help='Display sent and read HTTP traffic')
verbosity.add_option(
'-C', '--call-home',
dest='call_home', action='store_true', default=False,
help='Contact the youtube-dl server for debugging')
verbosity.add_option(
'--no-call-home',
dest='call_home', action='store_false', default=False,
help='Do NOT contact the youtube-dl server for debugging')
filesystem = optparse.OptionGroup(parser, 'Filesystem Options')
filesystem.add_option(
'-a', '--batch-file',
dest='batchfile', metavar='FILE',
help="File containing URLs to download ('-' for stdin), one URL per line. "
"Lines starting with '#', ';' or ']' are considered as comments and ignored.")
filesystem.add_option(
'--id', default=False,
action='store_true', dest='useid', help='Use only video ID in file name')
filesystem.add_option(
'-o', '--output',
dest='outtmpl', metavar='TEMPLATE',
help=('Output filename template, see the "OUTPUT TEMPLATE" for all the info'))
filesystem.add_option(
'--autonumber-size',
dest='autonumber_size', metavar='NUMBER', type=int,
help=optparse.SUPPRESS_HELP)
filesystem.add_option(
'--autonumber-start',
dest='autonumber_start', metavar='NUMBER', default=1, type=int,
help='Specify the start value for %(autonumber)s (default is %default)')
filesystem.add_option(
'--restrict-filenames',
action='store_true', dest='restrictfilenames', default=False,
help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames')
filesystem.add_option(
'-A', '--auto-number',
action='store_true', dest='autonumber', default=False,
help=optparse.SUPPRESS_HELP)
filesystem.add_option(
'-t', '--title',
action='store_true', dest='usetitle', default=False,
help=optparse.SUPPRESS_HELP)
filesystem.add_option(
'-l', '--literal', default=False,
action='store_true', dest='usetitle',
help=optparse.SUPPRESS_HELP)
filesystem.add_option(
'-w', '--no-overwrites',
action='store_true', dest='nooverwrites', default=False,
help='Do not overwrite files')
filesystem.add_option(
'-c', '--continue',
action='store_true', dest='continue_dl', default=True,
help='Force resume of partially downloaded files. By default, youtube-dl will resume downloads if possible.')
filesystem.add_option(
'--no-continue',
action='store_false', dest='continue_dl',
help='Do not resume partially downloaded files (restart from beginning)')
filesystem.add_option(
'--no-part',
action='store_true', dest='nopart', default=False,
help='Do not use .part files - write directly into output file')
filesystem.add_option(
'--no-mtime',
action='store_false', dest='updatetime', default=True,
help='Do not use the Last-modified header to set the file modification time')
filesystem.add_option(
'--write-description',
action='store_true', dest='writedescription', default=False,
help='Write video description to a .description file')
filesystem.add_option(
'--write-info-json',
action='store_true', dest='writeinfojson', default=False,
help='Write video metadata to a .info.json file')
filesystem.add_option(
'--write-annotations',
action='store_true', dest='writeannotations', default=False,
help='Write video annotations to a .annotations.xml file')
filesystem.add_option(
'--load-info-json', '--load-info',
dest='load_info_filename', metavar='FILE',
help='JSON file containing the video information (created with the "--write-info-json" option)')
filesystem.add_option(
'--cookies',
dest='cookiefile', metavar='FILE',
help='File to read cookies from and dump cookie jar in')
filesystem.add_option(
'--cache-dir', dest='cachedir', default=None, metavar='DIR',
help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
filesystem.add_option(
'--no-cache-dir', action='store_const', const=False, dest='cachedir',
help='Disable filesystem caching')
filesystem.add_option(
'--rm-cache-dir',
action='store_true', dest='rm_cachedir',
help='Delete all filesystem cache files')
thumbnail = optparse.OptionGroup(parser, 'Thumbnail images')
thumbnail.add_option(
'--write-thumbnail',
action='store_true', dest='writethumbnail', default=False,
help='Write thumbnail image to disk')
thumbnail.add_option(
'--write-all-thumbnails',
action='store_true', dest='write_all_thumbnails', default=False,
help='Write all thumbnail image formats to disk')
thumbnail.add_option(
'--list-thumbnails',
action='store_true', dest='list_thumbnails', default=False,
help='Simulate and list all available thumbnail formats')
postproc = optparse.OptionGroup(parser, 'Post-processing Options')
postproc.add_option(
'-x', '--extract-audio',
action='store_true', dest='extractaudio', default=False,
help='Convert video files to audio-only files (requires ffmpeg or avconv and ffprobe or avprobe)')
postproc.add_option(
'--audio-format', metavar='FORMAT', dest='audioformat', default='best',
help='Specify audio format: "best", "aac", "flac", "mp3", "m4a", "opus", "vorbis", or "wav"; "%default" by default; No effect without -x')
postproc.add_option(
'--audio-quality', metavar='QUALITY',
dest='audioquality', default='5',
help='Specify ffmpeg/avconv audio quality, insert a value between 0 (better) and 9 (worse) for VBR or a specific bitrate like 128K (default %default)')
postproc.add_option(
'--recode-video',
metavar='FORMAT', dest='recodevideo', default=None,
help='Encode the video to another format if necessary (currently supported: mp4|flv|ogg|webm|mkv|avi)')
postproc.add_option(
'--postprocessor-args',
dest='postprocessor_args', metavar='ARGS',
help='Give these arguments to the postprocessor')
postproc.add_option(
'-k', '--keep-video',
action='store_true', dest='keepvideo', default=False,
help='Keep the video file on disk after the post-processing; the video is erased by default')
postproc.add_option(
'--no-post-overwrites',
action='store_true', dest='nopostoverwrites', default=False,
help='Do not overwrite post-processed files; the post-processed files are overwritten by default')
postproc.add_option(
'--embed-subs',
action='store_true', dest='embedsubtitles', default=False,
help='Embed subtitles in the video (only for mp4, webm and mkv videos)')
postproc.add_option(
'--embed-thumbnail',
action='store_true', dest='embedthumbnail', default=False,
help='Embed thumbnail in the audio as cover art')
postproc.add_option(
'--add-metadata',
action='store_true', dest='addmetadata', default=False,
help='Write metadata to the video file')
postproc.add_option(
'--metadata-from-title',
metavar='FORMAT', dest='metafromtitle',
help='Parse additional metadata like song title / artist from the video title. '
'The format syntax is the same as --output. Regular expression with '
'named capture groups may also be used. '
'The parsed parameters replace existing values. '
'Example: --metadata-from-title "%(artist)s - %(title)s" matches a title like '
'"Coldplay - Paradise". '
'Example (regex): --metadata-from-title "(?P<artist>.+?) - (?P<title>.+)"')
postproc.add_option(
'--xattrs',
action='store_true', dest='xattrs', default=False,
help='Write metadata to the video file\'s xattrs (using dublin core and xdg standards)')
postproc.add_option(
'--fixup',
metavar='POLICY', dest='fixup', default='detect_or_warn',
help='Automatically correct known faults of the file. '
'One of never (do nothing), warn (only emit a warning), '
'detect_or_warn (the default; fix file if we can, warn otherwise)')
postproc.add_option(
'--prefer-avconv',
action='store_false', dest='prefer_ffmpeg',
help='Prefer avconv over ffmpeg for running the postprocessors')
postproc.add_option(
'--prefer-ffmpeg',
action='store_true', dest='prefer_ffmpeg',
help='Prefer ffmpeg over avconv for running the postprocessors (default)')
postproc.add_option(
'--ffmpeg-location', '--avconv-location', metavar='PATH',
dest='ffmpeg_location',
help='Location of the ffmpeg/avconv binary; either the path to the binary or its containing directory.')
postproc.add_option(
'--exec',
metavar='CMD', dest='exec_cmd',
help='Execute a command on the file after downloading, similar to find\'s -exec syntax. Example: --exec \'adb push {} /sdcard/Music/ && rm {}\'')
postproc.add_option(
'--convert-subs', '--convert-subtitles',
metavar='FORMAT', dest='convertsubtitles', default=None,
help='Convert the subtitles to other format (currently supported: srt|ass|vtt|lrc)')
parser.add_option_group(general)
parser.add_option_group(network)
parser.add_option_group(geo)
parser.add_option_group(selection)
parser.add_option_group(downloader)
parser.add_option_group(filesystem)
parser.add_option_group(thumbnail)
parser.add_option_group(verbosity)
parser.add_option_group(workarounds)
parser.add_option_group(video_format)
parser.add_option_group(subtitles)
parser.add_option_group(authentication)
parser.add_option_group(adobe_pass)
parser.add_option_group(postproc)
if overrideArguments is not None:
opts, args = parser.parse_args(overrideArguments)
if opts.verbose:
write_string('[debug] Override config: ' + repr(overrideArguments) + '\n')
else:
def compat_conf(conf):
if sys.version_info < (3,):
return [a.decode(preferredencoding(), 'replace') for a in conf]
return conf
command_line_conf = compat_conf(sys.argv[1:])
opts, args = parser.parse_args(command_line_conf)
system_conf = user_conf = custom_conf = []
if '--config-location' in command_line_conf:
location = compat_expanduser(opts.config_location)
if os.path.isdir(location):
location = os.path.join(location, 'youtube-dl.conf')
if not os.path.exists(location):
parser.error('config-location %s does not exist.' % location)
custom_conf = _readOptions(location)
elif '--ignore-config' in command_line_conf:
pass
else:
system_conf = _readOptions('/etc/youtube-dl.conf')
if '--ignore-config' not in system_conf:
user_conf = _readUserConf()
argv = system_conf + user_conf + custom_conf + command_line_conf
opts, args = parser.parse_args(argv)
if opts.verbose:
for conf_label, conf in (
('System config', system_conf),
('User config', user_conf),
('Custom config', custom_conf),
('Command-line args', command_line_conf)):
write_string('[debug] %s: %s\n' % (conf_label, repr(_hide_login_info(conf))))
return parser, opts, args

40
youtube_dl/postprocessor/__init__.py
View File

@ -1,40 +0,0 @@
from __future__ import unicode_literals
from .embedthumbnail import EmbedThumbnailPP
from .ffmpeg import (
FFmpegPostProcessor,
FFmpegEmbedSubtitlePP,
FFmpegExtractAudioPP,
FFmpegFixupStretchedPP,
FFmpegFixupM3u8PP,
FFmpegFixupM4aPP,
FFmpegMergerPP,
FFmpegMetadataPP,
FFmpegVideoConvertorPP,
FFmpegSubtitlesConvertorPP,
)
from .xattrpp import XAttrMetadataPP
from .execafterdownload import ExecAfterDownloadPP
from .metadatafromtitle import MetadataFromTitlePP
def get_postprocessor(key):
return globals()[key + 'PP']
__all__ = [
'EmbedThumbnailPP',
'ExecAfterDownloadPP',
'FFmpegEmbedSubtitlePP',
'FFmpegExtractAudioPP',
'FFmpegFixupM3u8PP',
'FFmpegFixupM4aPP',
'FFmpegFixupStretchedPP',
'FFmpegMergerPP',
'FFmpegMetadataPP',
'FFmpegPostProcessor',
'FFmpegSubtitlesConvertorPP',
'FFmpegVideoConvertorPP',
'MetadataFromTitlePP',
'XAttrMetadataPP',
]

69
youtube_dl/postprocessor/common.py
View File

@ -1,69 +0,0 @@
from __future__ import unicode_literals
import os
from ..utils import (
PostProcessingError,
cli_configuration_args,
encodeFilename,
)
class PostProcessor(object):
"""Post Processor class.
PostProcessor objects can be added to downloaders with their
add_post_processor() method. When the downloader has finished a
successful download, it will take its internal chain of PostProcessors
and start calling the run() method on each one of them, first with
an initial argument and then with the returned value of the previous
PostProcessor.
The chain will be stopped if one of them ever returns None or the end
of the chain is reached.
PostProcessor objects follow a "mutual registration" process similar
to InfoExtractor objects.
Optionally PostProcessor can use a list of additional command-line arguments
with self._configuration_args.
"""
_downloader = None
def __init__(self, downloader=None):
self._downloader = downloader
def set_downloader(self, downloader):
"""Sets the downloader for this PP."""
self._downloader = downloader
def run(self, information):
"""Run the PostProcessor.
The "information" argument is a dictionary like the ones
composed by InfoExtractors. The only difference is that this
one has an extra field called "filepath" that points to the
downloaded file.
This method returns a tuple, the first element is a list of the files
that can be deleted, and the second of which is the updated
information.
In addition, this method may raise a PostProcessingError
exception if post processing fails.
"""
return [], information # by default, keep file and do nothing
def try_utime(self, path, atime, mtime, errnote='Cannot update utime of file'):
try:
os.utime(encodeFilename(path), (atime, mtime))
except Exception:
self._downloader.report_warning(errnote)
def _configuration_args(self, default=[]):
return cli_configuration_args(self._downloader.params, 'postprocessor_args', default)
class AudioConversionError(PostProcessingError):
pass

93
youtube_dl/postprocessor/embedthumbnail.py
View File

@ -1,93 +0,0 @@
# coding: utf-8
from __future__ import unicode_literals
import os
import subprocess
from .ffmpeg import FFmpegPostProcessor
from ..utils import (
check_executable,
encodeArgument,
encodeFilename,
PostProcessingError,
prepend_extension,
shell_quote
)
class EmbedThumbnailPPError(PostProcessingError):
pass
class EmbedThumbnailPP(FFmpegPostProcessor):
def __init__(self, downloader=None, already_have_thumbnail=False):
super(EmbedThumbnailPP, self).__init__(downloader)
self._already_have_thumbnail = already_have_thumbnail
def run(self, info):
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
if not info.get('thumbnails'):
self._downloader.to_screen('[embedthumbnail] There aren\'t any thumbnails to embed')
return [], info
thumbnail_filename = info['thumbnails'][-1]['filename']
if not os.path.exists(encodeFilename(thumbnail_filename)):
self._downloader.report_warning(
'Skipping embedding the thumbnail because the file is missing.')
return [], info
if info['ext'] == 'mp3':
options = [
'-c', 'copy', '-map', '0', '-map', '1',
'-metadata:s:v', 'title="Album cover"', '-metadata:s:v', 'comment="Cover (Front)"']
self._downloader.to_screen('[ffmpeg] Adding thumbnail to "%s"' % filename)
self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options)
if not self._already_have_thumbnail:
os.remove(encodeFilename(thumbnail_filename))
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
elif info['ext'] in ['m4a', 'mp4']:
if not check_executable('AtomicParsley', ['-v']):
raise EmbedThumbnailPPError('AtomicParsley was not found. Please install.')
cmd = [encodeFilename('AtomicParsley', True),
encodeFilename(filename, True),
encodeArgument('--artwork'),
encodeFilename(thumbnail_filename, True),
encodeArgument('-o'),
encodeFilename(temp_filename, True)]
self._downloader.to_screen('[atomicparsley] Adding thumbnail to "%s"' % filename)
if self._downloader.params.get('verbose', False):
self._downloader.to_screen('[debug] AtomicParsley command line: %s' % shell_quote(cmd))
p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
stdout, stderr = p.communicate()
if p.returncode != 0:
msg = stderr.decode('utf-8', 'replace').strip()
raise EmbedThumbnailPPError(msg)
if not self._already_have_thumbnail:
os.remove(encodeFilename(thumbnail_filename))
# for formats that don't support thumbnails (like 3gp) AtomicParsley
# won't create to the temporary file
if b'No changes' in stdout:
self._downloader.report_warning('The file format doesn\'t support embedding a thumbnail')
else:
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
else:
raise EmbedThumbnailPPError('Only mp3 and m4a/mp4 are supported for thumbnail embedding for now.')
return [], info

31
youtube_dl/postprocessor/execafterdownload.py
View File

@ -1,31 +0,0 @@
from __future__ import unicode_literals
import subprocess
from .common import PostProcessor
from ..compat import compat_shlex_quote
from ..utils import (
encodeArgument,
PostProcessingError,
)
class ExecAfterDownloadPP(PostProcessor):
def __init__(self, downloader, exec_cmd):
super(ExecAfterDownloadPP, self).__init__(downloader)
self.exec_cmd = exec_cmd
def run(self, information):
cmd = self.exec_cmd
if '{}' not in cmd:
cmd += ' {}'
cmd = cmd.replace('{}', compat_shlex_quote(information['filepath']))
self._downloader.to_screen('[exec] Executing command: %s' % cmd)
retCode = subprocess.call(encodeArgument(cmd), shell=True)
if retCode != 0:
raise PostProcessingError(
'Command returned error code %d' % retCode)
return [], information

613
youtube_dl/postprocessor/ffmpeg.py
View File

@ -1,613 +0,0 @@
from __future__ import unicode_literals
import io
import os
import subprocess
import time
import re
from .common import AudioConversionError, PostProcessor
from ..compat import (
compat_subprocess_get_DEVNULL,
)
from ..utils import (
encodeArgument,
encodeFilename,
get_exe_version,
is_outdated_version,
PostProcessingError,
prepend_extension,
shell_quote,
subtitles_filename,
dfxp2srt,
ISO639Utils,
replace_extension,
)
EXT_TO_OUT_FORMATS = {
'aac': 'adts',
'flac': 'flac',
'm4a': 'ipod',
'mka': 'matroska',
'mkv': 'matroska',
'mpg': 'mpeg',
'ogv': 'ogg',
'ts': 'mpegts',
'wma': 'asf',
'wmv': 'asf',
}
ACODECS = {
'mp3': 'libmp3lame',
'aac': 'aac',
'flac': 'flac',
'm4a': 'aac',
'opus': 'libopus',
'vorbis': 'libvorbis',
'wav': None,
}
class FFmpegPostProcessorError(PostProcessingError):
pass
class FFmpegPostProcessor(PostProcessor):
def __init__(self, downloader=None):
PostProcessor.__init__(self, downloader)
self._determine_executables()
def check_version(self):
if not self.available:
raise FFmpegPostProcessorError('ffmpeg or avconv not found. Please install one.')
required_version = '10-0' if self.basename == 'avconv' else '1.0'
if is_outdated_version(
self._versions[self.basename], required_version):
warning = 'Your copy of %s is outdated, update %s to version %s or newer if you encounter any errors.' % (
self.basename, self.basename, required_version)
if self._downloader:
self._downloader.report_warning(warning)
@staticmethod
def get_versions(downloader=None):
return FFmpegPostProcessor(downloader)._versions
def _determine_executables(self):
programs = ['avprobe', 'avconv', 'ffmpeg', 'ffprobe']
prefer_ffmpeg = True
self.basename = None
self.probe_basename = None
self._paths = None
self._versions = None
if self._downloader:
prefer_ffmpeg = self._downloader.params.get('prefer_ffmpeg', True)
location = self._downloader.params.get('ffmpeg_location')
if location is not None:
if not os.path.exists(location):
self._downloader.report_warning(
'ffmpeg-location %s does not exist! '
'Continuing without avconv/ffmpeg.' % (location))
self._versions = {}
return
elif not os.path.isdir(location):
basename = os.path.splitext(os.path.basename(location))[0]
if basename not in programs:
self._downloader.report_warning(
'Cannot identify executable %s, its basename should be one of %s. '
'Continuing without avconv/ffmpeg.' %
(location, ', '.join(programs)))
self._versions = {}
return None
location = os.path.dirname(os.path.abspath(location))
if basename in ('ffmpeg', 'ffprobe'):
prefer_ffmpeg = True
self._paths = dict(
(p, os.path.join(location, p)) for p in programs)
self._versions = dict(
(p, get_exe_version(self._paths[p], args=['-version']))
for p in programs)
if self._versions is None:
self._versions = dict(
(p, get_exe_version(p, args=['-version'])) for p in programs)
self._paths = dict((p, p) for p in programs)
if prefer_ffmpeg is False:
prefs = ('avconv', 'ffmpeg')
else:
prefs = ('ffmpeg', 'avconv')
for p in prefs:
if self._versions[p]:
self.basename = p
break
if prefer_ffmpeg is False:
prefs = ('avprobe', 'ffprobe')
else:
prefs = ('ffprobe', 'avprobe')
for p in prefs:
if self._versions[p]:
self.probe_basename = p
break
@property
def available(self):
return self.basename is not None
@property
def executable(self):
return self._paths[self.basename]
@property
def probe_available(self):
return self.probe_basename is not None
@property
def probe_executable(self):
return self._paths[self.probe_basename]
def get_audio_codec(self, path):
if not self.probe_available:
raise PostProcessingError('ffprobe or avprobe not found. Please install one.')
try:
cmd = [
encodeFilename(self.probe_executable, True),
encodeArgument('-show_streams'),
encodeFilename(self._ffmpeg_filename_argument(path), True)]
if self._downloader.params.get('verbose', False):
self._downloader.to_screen('[debug] %s command line: %s' % (self.basename, shell_quote(cmd)))
handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE, stdin=subprocess.PIPE)
output = handle.communicate()[0]
if handle.wait() != 0:
return None
except (IOError, OSError):
return None
audio_codec = None
for line in output.decode('ascii', 'ignore').split('\n'):
if line.startswith('codec_name='):
audio_codec = line.split('=')[1].strip()
elif line.strip() == 'codec_type=audio' and audio_codec is not None:
return audio_codec
return None
def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
self.check_version()
oldest_mtime = min(
os.stat(encodeFilename(path)).st_mtime for path in input_paths)
opts += self._configuration_args()
files_cmd = []
for path in input_paths:
files_cmd.extend([
encodeArgument('-i'),
encodeFilename(self._ffmpeg_filename_argument(path), True)
])
cmd = ([encodeFilename(self.executable, True), encodeArgument('-y')] +
files_cmd +
[encodeArgument(o) for o in opts] +
[encodeFilename(self._ffmpeg_filename_argument(out_path), True)])
if self._downloader.params.get('verbose', False):
self._downloader.to_screen('[debug] ffmpeg command line: %s' % shell_quote(cmd))
p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE)
stdout, stderr = p.communicate()
if p.returncode != 0:
stderr = stderr.decode('utf-8', 'replace')
msg = stderr.strip().split('\n')[-1]
raise FFmpegPostProcessorError(msg)
self.try_utime(out_path, oldest_mtime, oldest_mtime)
def run_ffmpeg(self, path, out_path, opts):
self.run_ffmpeg_multiple_files([path], out_path, opts)
def _ffmpeg_filename_argument(self, fn):
# Always use 'file:' because the filename may contain ':' (ffmpeg
# interprets that as a protocol) or can start with '-' (-- is broken in
# ffmpeg, see https://ffmpeg.org/trac/ffmpeg/ticket/2127 for details)
# Also leave '-' intact in order not to break streaming to stdout.
return 'file:' + fn if fn != '-' else fn
class FFmpegExtractAudioPP(FFmpegPostProcessor):
def __init__(self, downloader=None, preferredcodec=None, preferredquality=None, nopostoverwrites=False):
FFmpegPostProcessor.__init__(self, downloader)
if preferredcodec is None:
preferredcodec = 'best'
self._preferredcodec = preferredcodec
self._preferredquality = preferredquality
self._nopostoverwrites = nopostoverwrites
def run_ffmpeg(self, path, out_path, codec, more_opts):
if codec is None:
acodec_opts = []
else:
acodec_opts = ['-acodec', codec]
opts = ['-vn'] + acodec_opts + more_opts
try:
FFmpegPostProcessor.run_ffmpeg(self, path, out_path, opts)
except FFmpegPostProcessorError as err:
raise AudioConversionError(err.msg)
def run(self, information):
path = information['filepath']
filecodec = self.get_audio_codec(path)
if filecodec is None:
raise PostProcessingError('WARNING: unable to obtain file audio codec with ffprobe')
more_opts = []
if self._preferredcodec == 'best' or self._preferredcodec == filecodec or (self._preferredcodec == 'm4a' and filecodec == 'aac'):
if filecodec == 'aac' and self._preferredcodec in ['m4a', 'best']:
# Lossless, but in another container
acodec = 'copy'
extension = 'm4a'
more_opts = ['-bsf:a', 'aac_adtstoasc']
elif filecodec in ['aac', 'flac', 'mp3', 'vorbis', 'opus']:
# Lossless if possible
acodec = 'copy'
extension = filecodec
if filecodec == 'aac':
more_opts = ['-f', 'adts']
if filecodec == 'vorbis':
extension = 'ogg'
else:
# MP3 otherwise.
acodec = 'libmp3lame'
extension = 'mp3'
more_opts = []
if self._preferredquality is not None:
if int(self._preferredquality) < 10:
more_opts += ['-q:a', self._preferredquality]
else:
more_opts += ['-b:a', self._preferredquality + 'k']
else:
# We convert the audio (lossy if codec is lossy)
acodec = ACODECS[self._preferredcodec]
extension = self._preferredcodec
more_opts = []
if self._preferredquality is not None:
# The opus codec doesn't support the -aq option
if int(self._preferredquality) < 10 and extension != 'opus':
more_opts += ['-q:a', self._preferredquality]
else:
more_opts += ['-b:a', self._preferredquality + 'k']
if self._preferredcodec == 'aac':
more_opts += ['-f', 'adts']
if self._preferredcodec == 'm4a':
more_opts += ['-bsf:a', 'aac_adtstoasc']
if self._preferredcodec == 'vorbis':
extension = 'ogg'
if self._preferredcodec == 'wav':
extension = 'wav'
more_opts += ['-f', 'wav']
prefix, sep, ext = path.rpartition('.') # not os.path.splitext, since the latter does not work on unicode in all setups
new_path = prefix + sep + extension
information['filepath'] = new_path
information['ext'] = extension
# If we download foo.mp3 and convert it to... foo.mp3, then don't delete foo.mp3, silly.
if (new_path == path or
(self._nopostoverwrites and os.path.exists(encodeFilename(new_path)))):
self._downloader.to_screen('[ffmpeg] Post-process file %s exists, skipping' % new_path)
return [], information
try:
self._downloader.to_screen('[ffmpeg] Destination: ' + new_path)
self.run_ffmpeg(path, new_path, acodec, more_opts)
except AudioConversionError as e:
raise PostProcessingError(
'audio conversion failed: ' + e.msg)
except Exception:
raise PostProcessingError('error running ' + self.basename)
# Try to update the date time for extracted audio file.
if information.get('filetime') is not None:
self.try_utime(
new_path, time.time(), information['filetime'],
errnote='Cannot update utime of audio file')
return [path], information
class FFmpegVideoConvertorPP(FFmpegPostProcessor):
def __init__(self, downloader=None, preferedformat=None):
super(FFmpegVideoConvertorPP, self).__init__(downloader)
self._preferedformat = preferedformat
def run(self, information):
path = information['filepath']
if information['ext'] == self._preferedformat:
self._downloader.to_screen('[ffmpeg] Not converting video file %s - already is in target format %s' % (path, self._preferedformat))
return [], information
options = []
if self._preferedformat == 'avi':
options.extend(['-c:v', 'libxvid', '-vtag', 'XVID'])
prefix, sep, ext = path.rpartition('.')
outpath = prefix + sep + self._preferedformat
self._downloader.to_screen('[' + 'ffmpeg' + '] Converting video from %s to %s, Destination: ' % (information['ext'], self._preferedformat) + outpath)
self.run_ffmpeg(path, outpath, options)
information['filepath'] = outpath
information['format'] = self._preferedformat
information['ext'] = self._preferedformat
return [path], information
class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
def run(self, information):
if information['ext'] not in ('mp4', 'webm', 'mkv'):
self._downloader.to_screen('[ffmpeg] Subtitles can only be embedded in mp4, webm or mkv files')
return [], information
subtitles = information.get('requested_subtitles')
if not subtitles:
self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to embed')
return [], information
filename = information['filepath']
ext = information['ext']
sub_langs = []
sub_filenames = []
webm_vtt_warn = False
for lang, sub_info in subtitles.items():
sub_ext = sub_info['ext']
if ext != 'webm' or ext == 'webm' and sub_ext == 'vtt':
sub_langs.append(lang)
sub_filenames.append(subtitles_filename(filename, lang, sub_ext))
else:
if not webm_vtt_warn and ext == 'webm' and sub_ext != 'vtt':
webm_vtt_warn = True
self._downloader.to_screen('[ffmpeg] Only WebVTT subtitles can be embedded in webm files')
if not sub_langs:
return [], information
input_files = [filename] + sub_filenames
opts = [
'-map', '0',
'-c', 'copy',
# Don't copy the existing subtitles, we may be running the
# postprocessor a second time
'-map', '-0:s',
]
if information['ext'] == 'mp4':
opts += ['-c:s', 'mov_text']
for (i, lang) in enumerate(sub_langs):
opts.extend(['-map', '%d:0' % (i + 1)])
lang_code = ISO639Utils.short2long(lang)
if lang_code is not None:
opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code])
temp_filename = prepend_extension(filename, 'temp')
self._downloader.to_screen('[ffmpeg] Embedding subtitles in \'%s\'' % filename)
self.run_ffmpeg_multiple_files(input_files, temp_filename, opts)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return sub_filenames, information
class FFmpegMetadataPP(FFmpegPostProcessor):
def run(self, info):
metadata = {}
def add(meta_list, info_list=None):
if not info_list:
info_list = meta_list
if not isinstance(meta_list, (list, tuple)):
meta_list = (meta_list,)
if not isinstance(info_list, (list, tuple)):
info_list = (info_list,)
for info_f in info_list:
if info.get(info_f) is not None:
for meta_f in meta_list:
metadata[meta_f] = info[info_f]
break
add('title', ('track', 'title'))
add('date', 'upload_date')
add(('description', 'comment'), 'description')
add('purl', 'webpage_url')
add('track', 'track_number')
add('artist', ('artist', 'creator', 'uploader', 'uploader_id'))
add('genre')
add('album')
add('album_artist')
add('disc', 'disc_number')
if not metadata:
self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add')
return [], info
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
in_filenames = [filename]
options = []
if info['ext'] == 'm4a':
options.extend(['-vn', '-acodec', 'copy'])
else:
options.extend(['-c', 'copy'])
for (name, value) in metadata.items():
options.extend(['-metadata', '%s=%s' % (name, value)])
chapters = info.get('chapters', [])
if chapters:
metadata_filename = replace_extension(filename, 'meta')
with io.open(metadata_filename, 'wt', encoding='utf-8') as f:
def ffmpeg_escape(text):
return re.sub(r'(=|;|#|\\|\n)', r'\\\1', text)
metadata_file_content = ';FFMETADATA1\n'
for chapter in chapters:
metadata_file_content += '[CHAPTER]\nTIMEBASE=1/1000\n'
metadata_file_content += 'START=%d\n' % (chapter['start_time'] * 1000)
metadata_file_content += 'END=%d\n' % (chapter['end_time'] * 1000)
chapter_title = chapter.get('title')
if chapter_title:
metadata_file_content += 'title=%s\n' % ffmpeg_escape(chapter_title)
f.write(metadata_file_content)
in_filenames.append(metadata_filename)
options.extend(['-map_metadata', '1'])
self._downloader.to_screen('[ffmpeg] Adding metadata to \'%s\'' % filename)
self.run_ffmpeg_multiple_files(in_filenames, temp_filename, options)
if chapters:
os.remove(metadata_filename)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return [], info
class FFmpegMergerPP(FFmpegPostProcessor):
def run(self, info):
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
args = ['-c', 'copy', '-map', '0:v:0', '-map', '1:a:0']
self._downloader.to_screen('[ffmpeg] Merging formats into "%s"' % filename)
self.run_ffmpeg_multiple_files(info['__files_to_merge'], temp_filename, args)
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return info['__files_to_merge'], info
def can_merge(self):
# TODO: figure out merge-capable ffmpeg version
if self.basename != 'avconv':
return True
required_version = '10-0'
if is_outdated_version(
self._versions[self.basename], required_version):
warning = ('Your copy of %s is outdated and unable to properly mux separate video and audio files, '
'youtube-dl will download single file media. '
'Update %s to version %s or newer to fix this.') % (
self.basename, self.basename, required_version)
if self._downloader:
self._downloader.report_warning(warning)
return False
return True
class FFmpegFixupStretchedPP(FFmpegPostProcessor):
def run(self, info):
stretched_ratio = info.get('stretched_ratio')
if stretched_ratio is None or stretched_ratio == 1:
return [], info
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
options = ['-c', 'copy', '-aspect', '%f' % stretched_ratio]
self._downloader.to_screen('[ffmpeg] Fixing aspect ratio in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return [], info
class FFmpegFixupM4aPP(FFmpegPostProcessor):
def run(self, info):
if info.get('container') != 'm4a_dash':
return [], info
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
options = ['-c', 'copy', '-f', 'mp4']
self._downloader.to_screen('[ffmpeg] Correcting container in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return [], info
class FFmpegFixupM3u8PP(FFmpegPostProcessor):
def run(self, info):
filename = info['filepath']
if self.get_audio_codec(filename) == 'aac':
temp_filename = prepend_extension(filename, 'temp')
options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options)
os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return [], info
class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
def __init__(self, downloader=None, format=None):
super(FFmpegSubtitlesConvertorPP, self).__init__(downloader)
self.format = format
def run(self, info):
subs = info.get('requested_subtitles')
filename = info['filepath']
new_ext = self.format
new_format = new_ext
if new_format == 'vtt':
new_format = 'webvtt'
if subs is None:
self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to convert')
return [], info
self._downloader.to_screen('[ffmpeg] Converting subtitles')
sub_filenames = []
for lang, sub in subs.items():
ext = sub['ext']
if ext == new_ext:
self._downloader.to_screen(
'[ffmpeg] Subtitle file for %s is already in the requested format' % new_ext)
continue
old_file = subtitles_filename(filename, lang, ext)
sub_filenames.append(old_file)
new_file = subtitles_filename(filename, lang, new_ext)
if ext in ('dfxp', 'ttml', 'tt'):
self._downloader.report_warning(
'You have requested to convert dfxp (TTML) subtitles into another format, '
'which results in style information loss')
dfxp_file = old_file
srt_file = subtitles_filename(filename, lang, 'srt')
with open(dfxp_file, 'rb') as f:
srt_data = dfxp2srt(f.read())
with io.open(srt_file, 'wt', encoding='utf-8') as f:
f.write(srt_data)
old_file = srt_file
subs[lang] = {
'ext': 'srt',
'data': srt_data
}
if new_ext == 'srt':
continue
else:
sub_filenames.append(srt_file)
self.run_ffmpeg(old_file, new_file, ['-f', new_format])
with io.open(new_file, 'rt', encoding='utf-8') as f:
subs[lang] = {
'ext': new_ext,
'data': f.read(),
}
return sub_filenames, info

48
youtube_dl/postprocessor/metadatafromtitle.py
View File

@ -1,48 +0,0 @@
from __future__ import unicode_literals
import re
from .common import PostProcessor
class MetadataFromTitlePP(PostProcessor):
def __init__(self, downloader, titleformat):
super(MetadataFromTitlePP, self).__init__(downloader)
self._titleformat = titleformat
self._titleregex = (self.format_to_regex(titleformat)
if re.search(r'%\(\w+\)s', titleformat)
else titleformat)
def format_to_regex(self, fmt):
r"""
Converts a string like
'%(title)s - %(artist)s'
to a regex like
'(?P<title>.+)\ \-\ (?P<artist>.+)'
"""
lastpos = 0
regex = ''
# replace %(..)s with regex group and escape other string parts
for match in re.finditer(r'%\((\w+)\)s', fmt):
regex += re.escape(fmt[lastpos:match.start()])
regex += r'(?P<' + match.group(1) + '>.+)'
lastpos = match.end()
if lastpos < len(fmt):
regex += re.escape(fmt[lastpos:])
return regex
def run(self, info):
title = info['title']
match = re.match(self._titleregex, title)
if match is None:
self._downloader.to_screen(
'[fromtitle] Could not interpret title of video as "%s"'
% self._titleformat)
return [], info
for attribute, value in match.groupdict().items():
info[attribute] = value
self._downloader.to_screen(
'[fromtitle] parsed %s: %s'
% (attribute, value if value is not None else 'NA'))
return [], info

79
youtube_dl/postprocessor/xattrpp.py
View File

@ -1,79 +0,0 @@
from __future__ import unicode_literals
from .common import PostProcessor
from ..compat import compat_os_name
from ..utils import (
hyphenate_date,
write_xattr,
XAttrMetadataError,
XAttrUnavailableError,
)
class XAttrMetadataPP(PostProcessor):
#
# More info about extended attributes for media:
# http://freedesktop.org/wiki/CommonExtendedAttributes/
# http://www.freedesktop.org/wiki/PhreedomDraft/
# http://dublincore.org/documents/usageguide/elements.shtml
#
# TODO:
# * capture youtube keywords and put them in 'user.dublincore.subject' (comma-separated)
# * figure out which xattrs can be used for 'duration', 'thumbnail', 'resolution'
#
def run(self, info):
""" Set extended attributes on downloaded file (if xattr support is found). """
# Write the metadata to the file's xattrs
self._downloader.to_screen('[metadata] Writing metadata to file\'s xattrs')
filename = info['filepath']
try:
xattr_mapping = {
'user.xdg.referrer.url': 'webpage_url',
# 'user.xdg.comment': 'description',
'user.dublincore.title': 'title',
'user.dublincore.date': 'upload_date',
'user.dublincore.description': 'description',
'user.dublincore.contributor': 'uploader',
'user.dublincore.format': 'format',
}
num_written = 0
for xattrname, infoname in xattr_mapping.items():
value = info.get(infoname)
if value:
if infoname == 'upload_date':
value = hyphenate_date(value)
byte_value = value.encode('utf-8')
write_xattr(filename, xattrname, byte_value)
num_written += 1
return [], info
except XAttrUnavailableError as e:
self._downloader.report_error(str(e))
return [], info
except XAttrMetadataError as e:
if e.reason == 'NO_SPACE':
self._downloader.report_warning(
'There\'s no disk space left, disk quota exceeded or filesystem xattr limit exceeded. ' +
(('Some ' if num_written else '') + 'extended attributes are not written.').capitalize())
elif e.reason == 'VALUE_TOO_LONG':
self._downloader.report_warning(
'Unable to write extended attributes due to too long values.')
else:
msg = 'This filesystem doesn\'t support extended attributes. '
if compat_os_name == 'nt':
msg += 'You need to use NTFS.'
else:
msg += '(You may have to enable them in your /etc/fstab)'
self._downloader.report_error(msg)
return [], info

273
youtube_dl/socks.py
View File

@ -1,273 +0,0 @@
# Public Domain SOCKS proxy protocol implementation
# Adapted from https://gist.github.com/bluec0re/cafd3764412967417fd3
from __future__ import unicode_literals
# References:
# SOCKS4 protocol http://www.openssh.com/txt/socks4.protocol
# SOCKS4A protocol http://www.openssh.com/txt/socks4a.protocol
# SOCKS5 protocol https://tools.ietf.org/html/rfc1928
# SOCKS5 username/password authentication https://tools.ietf.org/html/rfc1929
import collections
import socket
from .compat import (
compat_ord,
compat_struct_pack,
compat_struct_unpack,
)
__author__ = 'Timo Schmid <coding@timoschmid.de>'
SOCKS4_VERSION = 4
SOCKS4_REPLY_VERSION = 0x00
# Excerpt from SOCKS4A protocol:
# if the client cannot resolve the destination host's domain name to find its
# IP address, it should set the first three bytes of DSTIP to NULL and the last
# byte to a non-zero value.
SOCKS4_DEFAULT_DSTIP = compat_struct_pack('!BBBB', 0, 0, 0, 0xFF)
SOCKS5_VERSION = 5
SOCKS5_USER_AUTH_VERSION = 0x01
SOCKS5_USER_AUTH_SUCCESS = 0x00
class Socks4Command(object):
CMD_CONNECT = 0x01
CMD_BIND = 0x02
class Socks5Command(Socks4Command):
CMD_UDP_ASSOCIATE = 0x03
class Socks5Auth(object):
AUTH_NONE = 0x00
AUTH_GSSAPI = 0x01
AUTH_USER_PASS = 0x02
AUTH_NO_ACCEPTABLE = 0xFF # For server response
class Socks5AddressType(object):
ATYP_IPV4 = 0x01
ATYP_DOMAINNAME = 0x03
ATYP_IPV6 = 0x04
class ProxyError(socket.error):
ERR_SUCCESS = 0x00
def __init__(self, code=None, msg=None):
if code is not None and msg is None:
msg = self.CODES.get(code) or 'unknown error'
super(ProxyError, self).__init__(code, msg)
class InvalidVersionError(ProxyError):
def __init__(self, expected_version, got_version):
msg = ('Invalid response version from server. Expected {0:02x} got '
'{1:02x}'.format(expected_version, got_version))
super(InvalidVersionError, self).__init__(0, msg)
class Socks4Error(ProxyError):
ERR_SUCCESS = 90
CODES = {
91: 'request rejected or failed',
92: 'request rejected because SOCKS server cannot connect to identd on the client',
93: 'request rejected because the client program and identd report different user-ids'
}
class Socks5Error(ProxyError):
ERR_GENERAL_FAILURE = 0x01
CODES = {
0x01: 'general SOCKS server failure',
0x02: 'connection not allowed by ruleset',
0x03: 'Network unreachable',
0x04: 'Host unreachable',
0x05: 'Connection refused',
0x06: 'TTL expired',
0x07: 'Command not supported',
0x08: 'Address type not supported',
0xFE: 'unknown username or invalid password',
0xFF: 'all offered authentication methods were rejected'
}
class ProxyType(object):
SOCKS4 = 0
SOCKS4A = 1
SOCKS5 = 2
Proxy = collections.namedtuple('Proxy', (
'type', 'host', 'port', 'username', 'password', 'remote_dns'))
class sockssocket(socket.socket):
def __init__(self, *args, **kwargs):
self._proxy = None
super(sockssocket, self).__init__(*args, **kwargs)
def setproxy(self, proxytype, addr, port, rdns=True, username=None, password=None):
assert proxytype in (ProxyType.SOCKS4, ProxyType.SOCKS4A, ProxyType.SOCKS5)
self._proxy = Proxy(proxytype, addr, port, username, password, rdns)
def recvall(self, cnt):
data = b''
while len(data) < cnt:
cur = self.recv(cnt - len(data))
if not cur:
raise EOFError('{0} bytes missing'.format(cnt - len(data)))
data += cur
return data
def _recv_bytes(self, cnt):
data = self.recvall(cnt)
return compat_struct_unpack('!{0}B'.format(cnt), data)
@staticmethod
def _len_and_data(data):
return compat_struct_pack('!B', len(data)) + data
def _check_response_version(self, expected_version, got_version):
if got_version != expected_version:
self.close()
raise InvalidVersionError(expected_version, got_version)
def _resolve_address(self, destaddr, default, use_remote_dns):
try:
return socket.inet_aton(destaddr)
except socket.error:
if use_remote_dns and self._proxy.remote_dns:
return default
else:
return socket.inet_aton(socket.gethostbyname(destaddr))
def _setup_socks4(self, address, is_4a=False):
destaddr, port = address
ipaddr = self._resolve_address(destaddr, SOCKS4_DEFAULT_DSTIP, use_remote_dns=is_4a)
packet = compat_struct_pack('!BBH', SOCKS4_VERSION, Socks4Command.CMD_CONNECT, port) + ipaddr
username = (self._proxy.username or '').encode('utf-8')
packet += username + b'\x00'
if is_4a and self._proxy.remote_dns:
packet += destaddr.encode('utf-8') + b'\x00'
self.sendall(packet)
version, resp_code, dstport, dsthost = compat_struct_unpack('!BBHI', self.recvall(8))
self._check_response_version(SOCKS4_REPLY_VERSION, version)
if resp_code != Socks4Error.ERR_SUCCESS:
self.close()
raise Socks4Error(resp_code)
return (dsthost, dstport)
def _setup_socks4a(self, address):
self._setup_socks4(address, is_4a=True)
def _socks5_auth(self):
packet = compat_struct_pack('!B', SOCKS5_VERSION)
auth_methods = [Socks5Auth.AUTH_NONE]
if self._proxy.username and self._proxy.password:
auth_methods.append(Socks5Auth.AUTH_USER_PASS)
packet += compat_struct_pack('!B', len(auth_methods))
packet += compat_struct_pack('!{0}B'.format(len(auth_methods)), *auth_methods)
self.sendall(packet)
version, method = self._recv_bytes(2)
self._check_response_version(SOCKS5_VERSION, version)
if method == Socks5Auth.AUTH_NO_ACCEPTABLE or (
method == Socks5Auth.AUTH_USER_PASS and (not self._proxy.username or not self._proxy.password)):
self.close()
raise Socks5Error(Socks5Auth.AUTH_NO_ACCEPTABLE)
if method == Socks5Auth.AUTH_USER_PASS:
username = self._proxy.username.encode('utf-8')
password = self._proxy.password.encode('utf-8')
packet = compat_struct_pack('!B', SOCKS5_USER_AUTH_VERSION)
packet += self._len_and_data(username) + self._len_and_data(password)
self.sendall(packet)
version, status = self._recv_bytes(2)
self._check_response_version(SOCKS5_USER_AUTH_VERSION, version)
if status != SOCKS5_USER_AUTH_SUCCESS:
self.close()
raise Socks5Error(Socks5Error.ERR_GENERAL_FAILURE)
def _setup_socks5(self, address):
destaddr, port = address
ipaddr = self._resolve_address(destaddr, None, use_remote_dns=True)
self._socks5_auth()
reserved = 0
packet = compat_struct_pack('!BBB', SOCKS5_VERSION, Socks5Command.CMD_CONNECT, reserved)
if ipaddr is None:
destaddr = destaddr.encode('utf-8')
packet += compat_struct_pack('!B', Socks5AddressType.ATYP_DOMAINNAME)
packet += self._len_and_data(destaddr)
else:
packet += compat_struct_pack('!B', Socks5AddressType.ATYP_IPV4) + ipaddr
packet += compat_struct_pack('!H', port)
self.sendall(packet)
version, status, reserved, atype = self._recv_bytes(4)
self._check_response_version(SOCKS5_VERSION, version)
if status != Socks5Error.ERR_SUCCESS:
self.close()
raise Socks5Error(status)
if atype == Socks5AddressType.ATYP_IPV4:
destaddr = self.recvall(4)
elif atype == Socks5AddressType.ATYP_DOMAINNAME:
alen = compat_ord(self.recv(1))
destaddr = self.recvall(alen)
elif atype == Socks5AddressType.ATYP_IPV6:
destaddr = self.recvall(16)
destport = compat_struct_unpack('!H', self.recvall(2))[0]
return (destaddr, destport)
def _make_proxy(self, connect_func, address):
if not self._proxy:
return connect_func(self, address)
result = connect_func(self, (self._proxy.host, self._proxy.port))
if result != 0 and result is not None:
return result
setup_funcs = {
ProxyType.SOCKS4: self._setup_socks4,
ProxyType.SOCKS4A: self._setup_socks4a,
ProxyType.SOCKS5: self._setup_socks5,
}
setup_funcs[self._proxy.type](address)
return result
def connect(self, address):
self._make_proxy(socket.socket.connect, address)
def connect_ex(self, address):
return self._make_proxy(socket.socket.connect_ex, address)

834
youtube_dl/swfinterp.py
View File

@ -1,834 +0,0 @@
from __future__ import unicode_literals
import collections
import io
import zlib
from .compat import (
compat_str,
compat_struct_unpack,
)
from .utils import (
ExtractorError,
)
def _extract_tags(file_contents):
if file_contents[1:3] != b'WS':
raise ExtractorError(
'Not an SWF file; header is %r' % file_contents[:3])
if file_contents[:1] == b'C':
content = zlib.decompress(file_contents[8:])
else:
raise NotImplementedError(
'Unsupported compression format %r' %
file_contents[:1])
# Determine number of bits in framesize rectangle
framesize_nbits = compat_struct_unpack('!B', content[:1])[0] >> 3
framesize_len = (5 + 4 * framesize_nbits + 7) // 8
pos = framesize_len + 2 + 2
while pos < len(content):
header16 = compat_struct_unpack('<H', content[pos:pos + 2])[0]
pos += 2
tag_code = header16 >> 6
tag_len = header16 & 0x3f
if tag_len == 0x3f:
tag_len = compat_struct_unpack('<I', content[pos:pos + 4])[0]
pos += 4
assert pos + tag_len <= len(content), \
('Tag %d ends at %d+%d - that\'s longer than the file (%d)'
% (tag_code, pos, tag_len, len(content)))
yield (tag_code, content[pos:pos + tag_len])
pos += tag_len
class _AVMClass_Object(object):
def __init__(self, avm_class):
self.avm_class = avm_class
def __repr__(self):
return '%s#%x' % (self.avm_class.name, id(self))
class _ScopeDict(dict):
def __init__(self, avm_class):
super(_ScopeDict, self).__init__()
self.avm_class = avm_class
def __repr__(self):
return '%s__Scope(%s)' % (
self.avm_class.name,
super(_ScopeDict, self).__repr__())
class _AVMClass(object):
def __init__(self, name_idx, name, static_properties=None):
self.name_idx = name_idx
self.name = name
self.method_names = {}
self.method_idxs = {}
self.methods = {}
self.method_pyfunctions = {}
self.static_properties = static_properties if static_properties else {}
self.variables = _ScopeDict(self)
self.constants = {}
def make_object(self):
return _AVMClass_Object(self)
def __repr__(self):
return '_AVMClass(%s)' % (self.name)
def register_methods(self, methods):
self.method_names.update(methods.items())
self.method_idxs.update(dict(
(idx, name)
for name, idx in methods.items()))
class _Multiname(object):
def __init__(self, kind):
self.kind = kind
def __repr__(self):
return '[MULTINAME kind: 0x%x]' % self.kind
def _read_int(reader):
res = 0
shift = 0
for _ in range(5):
buf = reader.read(1)
assert len(buf) == 1
b = compat_struct_unpack('<B', buf)[0]
res = res | ((b & 0x7f) << shift)
if b & 0x80 == 0:
break
shift += 7
return res
def _u30(reader):
res = _read_int(reader)
assert res & 0xf0000000 == 0
return res
_u32 = _read_int
def _s32(reader):
v = _read_int(reader)
if v & 0x80000000 != 0:
v = - ((v ^ 0xffffffff) + 1)
return v
def _s24(reader):
bs = reader.read(3)
assert len(bs) == 3
last_byte = b'\xff' if (ord(bs[2:3]) >= 0x80) else b'\x00'
return compat_struct_unpack('<i', bs + last_byte)[0]
def _read_string(reader):
slen = _u30(reader)
resb = reader.read(slen)
assert len(resb) == slen
return resb.decode('utf-8')
def _read_bytes(count, reader):
assert count >= 0
resb = reader.read(count)
assert len(resb) == count
return resb
def _read_byte(reader):
resb = _read_bytes(1, reader=reader)
res = compat_struct_unpack('<B', resb)[0]
return res
StringClass = _AVMClass('(no name idx)', 'String')
ByteArrayClass = _AVMClass('(no name idx)', 'ByteArray')
TimerClass = _AVMClass('(no name idx)', 'Timer')
TimerEventClass = _AVMClass('(no name idx)', 'TimerEvent', {'TIMER': 'timer'})
_builtin_classes = {
StringClass.name: StringClass,
ByteArrayClass.name: ByteArrayClass,
TimerClass.name: TimerClass,
TimerEventClass.name: TimerEventClass,
}
class _Undefined(object):
def __bool__(self):
return False
__nonzero__ = __bool__
def __hash__(self):
return 0
def __str__(self):
return 'undefined'
__repr__ = __str__
undefined = _Undefined()
class SWFInterpreter(object):
def __init__(self, file_contents):
self._patched_functions = {
(TimerClass, 'addEventListener'): lambda params: undefined,
}
code_tag = next(tag
for tag_code, tag in _extract_tags(file_contents)
if tag_code == 82)
p = code_tag.index(b'\0', 4) + 1
code_reader = io.BytesIO(code_tag[p:])
# Parse ABC (AVM2 ByteCode)
# Define a couple convenience methods
u30 = lambda *args: _u30(*args, reader=code_reader)
s32 = lambda *args: _s32(*args, reader=code_reader)
u32 = lambda *args: _u32(*args, reader=code_reader)
read_bytes = lambda *args: _read_bytes(*args, reader=code_reader)
read_byte = lambda *args: _read_byte(*args, reader=code_reader)
# minor_version + major_version
read_bytes(2 + 2)
# Constant pool
int_count = u30()
self.constant_ints = [0]
for _c in range(1, int_count):
self.constant_ints.append(s32())
self.constant_uints = [0]
uint_count = u30()
for _c in range(1, uint_count):
self.constant_uints.append(u32())
double_count = u30()
read_bytes(max(0, (double_count - 1)) * 8)
string_count = u30()
self.constant_strings = ['']
for _c in range(1, string_count):
s = _read_string(code_reader)
self.constant_strings.append(s)
namespace_count = u30()
for _c in range(1, namespace_count):
read_bytes(1) # kind
u30() # name
ns_set_count = u30()
for _c in range(1, ns_set_count):
count = u30()
for _c2 in range(count):
u30()
multiname_count = u30()
MULTINAME_SIZES = {
0x07: 2, # QName
0x0d: 2, # QNameA
0x0f: 1, # RTQName
0x10: 1, # RTQNameA
0x11: 0, # RTQNameL
0x12: 0, # RTQNameLA
0x09: 2, # Multiname
0x0e: 2, # MultinameA
0x1b: 1, # MultinameL
0x1c: 1, # MultinameLA
}
self.multinames = ['']
for _c in range(1, multiname_count):
kind = u30()
assert kind in MULTINAME_SIZES, 'Invalid multiname kind %r' % kind
if kind == 0x07:
u30() # namespace_idx
name_idx = u30()
self.multinames.append(self.constant_strings[name_idx])
elif kind == 0x09:
name_idx = u30()
u30()
self.multinames.append(self.constant_strings[name_idx])
else:
self.multinames.append(_Multiname(kind))
for _c2 in range(MULTINAME_SIZES[kind]):
u30()
# Methods
method_count = u30()
MethodInfo = collections.namedtuple(
'MethodInfo',
['NEED_ARGUMENTS', 'NEED_REST'])
method_infos = []
for method_id in range(method_count):
param_count = u30()
u30() # return type
for _ in range(param_count):
u30() # param type
u30() # name index (always 0 for youtube)
flags = read_byte()
if flags & 0x08 != 0:
# Options present
option_count = u30()
for c in range(option_count):
u30() # val
read_bytes(1) # kind
if flags & 0x80 != 0:
# Param names present
for _ in range(param_count):
u30() # param name
mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0)
method_infos.append(mi)
# Metadata
metadata_count = u30()
for _c in range(metadata_count):
u30() # name
item_count = u30()
for _c2 in range(item_count):
u30() # key
u30() # value
def parse_traits_info():
trait_name_idx = u30()
kind_full = read_byte()
kind = kind_full & 0x0f
attrs = kind_full >> 4
methods = {}
constants = None
if kind == 0x00: # Slot
u30() # Slot id
u30() # type_name_idx
vindex = u30()
if vindex != 0:
read_byte() # vkind
elif kind == 0x06: # Const
u30() # Slot id
u30() # type_name_idx
vindex = u30()
vkind = 'any'
if vindex != 0:
vkind = read_byte()
if vkind == 0x03: # Constant_Int
value = self.constant_ints[vindex]
elif vkind == 0x04: # Constant_UInt
value = self.constant_uints[vindex]
else:
return {}, None # Ignore silently for now
constants = {self.multinames[trait_name_idx]: value}
elif kind in (0x01, 0x02, 0x03): # Method / Getter / Setter
u30() # disp_id
method_idx = u30()
methods[self.multinames[trait_name_idx]] = method_idx
elif kind == 0x04: # Class
u30() # slot_id
u30() # classi
elif kind == 0x05: # Function
u30() # slot_id
function_idx = u30()
methods[function_idx] = self.multinames[trait_name_idx]
else:
raise ExtractorError('Unsupported trait kind %d' % kind)
if attrs & 0x4 != 0: # Metadata present
metadata_count = u30()
for _c3 in range(metadata_count):
u30() # metadata index
return methods, constants
# Classes
class_count = u30()
classes = []
for class_id in range(class_count):
name_idx = u30()
cname = self.multinames[name_idx]
avm_class = _AVMClass(name_idx, cname)
classes.append(avm_class)
u30() # super_name idx
flags = read_byte()
if flags & 0x08 != 0: # Protected namespace is present
u30() # protected_ns_idx
intrf_count = u30()
for _c2 in range(intrf_count):
u30()
u30() # iinit
trait_count = u30()
for _c2 in range(trait_count):
trait_methods, trait_constants = parse_traits_info()
avm_class.register_methods(trait_methods)
if trait_constants:
avm_class.constants.update(trait_constants)
assert len(classes) == class_count
self._classes_by_name = dict((c.name, c) for c in classes)
for avm_class in classes:
avm_class.cinit_idx = u30()
trait_count = u30()
for _c2 in range(trait_count):
trait_methods, trait_constants = parse_traits_info()
avm_class.register_methods(trait_methods)
if trait_constants:
avm_class.constants.update(trait_constants)
# Scripts
script_count = u30()
for _c in range(script_count):
u30() # init
trait_count = u30()
for _c2 in range(trait_count):
parse_traits_info()
# Method bodies
method_body_count = u30()
Method = collections.namedtuple('Method', ['code', 'local_count'])
self._all_methods = []
for _c in range(method_body_count):
method_idx = u30()
u30() # max_stack
local_count = u30()
u30() # init_scope_depth
u30() # max_scope_depth
code_length = u30()
code = read_bytes(code_length)
m = Method(code, local_count)
self._all_methods.append(m)
for avm_class in classes:
if method_idx in avm_class.method_idxs:
avm_class.methods[avm_class.method_idxs[method_idx]] = m
exception_count = u30()
for _c2 in range(exception_count):
u30() # from
u30() # to
u30() # target
u30() # exc_type
u30() # var_name
trait_count = u30()
for _c2 in range(trait_count):
parse_traits_info()
assert p + code_reader.tell() == len(code_tag)
def patch_function(self, avm_class, func_name, f):
self._patched_functions[(avm_class, func_name)] = f
def extract_class(self, class_name, call_cinit=True):
try:
res = self._classes_by_name[class_name]
except KeyError:
raise ExtractorError('Class %r not found' % class_name)
if call_cinit and hasattr(res, 'cinit_idx'):
res.register_methods({'$cinit': res.cinit_idx})
res.methods['$cinit'] = self._all_methods[res.cinit_idx]
cinit = self.extract_function(res, '$cinit')
cinit([])
return res
def extract_function(self, avm_class, func_name):
p = self._patched_functions.get((avm_class, func_name))
if p:
return p
if func_name in avm_class.method_pyfunctions:
return avm_class.method_pyfunctions[func_name]
if func_name in self._classes_by_name:
return self._classes_by_name[func_name].make_object()
if func_name not in avm_class.methods:
raise ExtractorError('Cannot find function %s.%s' % (
avm_class.name, func_name))
m = avm_class.methods[func_name]
def resfunc(args):
# Helper functions
coder = io.BytesIO(m.code)
s24 = lambda: _s24(coder)
u30 = lambda: _u30(coder)
registers = [avm_class.variables] + list(args) + [None] * m.local_count
stack = []
scopes = collections.deque([
self._classes_by_name, avm_class.constants, avm_class.variables])
while True:
opcode = _read_byte(coder)
if opcode == 9: # label
pass # Spec says: "Do nothing."
elif opcode == 16: # jump
offset = s24()
coder.seek(coder.tell() + offset)
elif opcode == 17: # iftrue
offset = s24()
value = stack.pop()
if value:
coder.seek(coder.tell() + offset)
elif opcode == 18: # iffalse
offset = s24()
value = stack.pop()
if not value:
coder.seek(coder.tell() + offset)
elif opcode == 19: # ifeq
offset = s24()
value2 = stack.pop()
value1 = stack.pop()
if value2 == value1:
coder.seek(coder.tell() + offset)
elif opcode == 20: # ifne
offset = s24()
value2 = stack.pop()
value1 = stack.pop()
if value2 != value1:
coder.seek(coder.tell() + offset)
elif opcode == 21: # iflt
offset = s24()
value2 = stack.pop()
value1 = stack.pop()
if value1 < value2:
coder.seek(coder.tell() + offset)
elif opcode == 32: # pushnull
stack.append(None)
elif opcode == 33: # pushundefined
stack.append(undefined)
elif opcode == 36: # pushbyte
v = _read_byte(coder)
stack.append(v)
elif opcode == 37: # pushshort
v = u30()
stack.append(v)
elif opcode == 38: # pushtrue
stack.append(True)
elif opcode == 39: # pushfalse
stack.append(False)
elif opcode == 40: # pushnan
stack.append(float('NaN'))
elif opcode == 42: # dup
value = stack[-1]
stack.append(value)
elif opcode == 44: # pushstring
idx = u30()
stack.append(self.constant_strings[idx])
elif opcode == 48: # pushscope
new_scope = stack.pop()
scopes.append(new_scope)
elif opcode == 66: # construct
arg_count = u30()
args = list(reversed(
[stack.pop() for _ in range(arg_count)]))
obj = stack.pop()
res = obj.avm_class.make_object()
stack.append(res)
elif opcode == 70: # callproperty
index = u30()
mname = self.multinames[index]
arg_count = u30()
args = list(reversed(
[stack.pop() for _ in range(arg_count)]))
obj = stack.pop()
if obj == StringClass:
if mname == 'String':
assert len(args) == 1
assert isinstance(args[0], (
int, compat_str, _Undefined))
if args[0] == undefined:
res = 'undefined'
else:
res = compat_str(args[0])
stack.append(res)
continue
else:
raise NotImplementedError(
'Function String.%s is not yet implemented'
% mname)
elif isinstance(obj, _AVMClass_Object):
func = self.extract_function(obj.avm_class, mname)
res = func(args)
stack.append(res)
continue
elif isinstance(obj, _AVMClass):
func = self.extract_function(obj, mname)
res = func(args)
stack.append(res)
continue
elif isinstance(obj, _ScopeDict):
if mname in obj.avm_class.method_names:
func = self.extract_function(obj.avm_class, mname)
res = func(args)
else:
res = obj[mname]
stack.append(res)
continue
elif isinstance(obj, compat_str):
if mname == 'split':
assert len(args) == 1
assert isinstance(args[0], compat_str)
if args[0] == '':
res = list(obj)
else:
res = obj.split(args[0])
stack.append(res)
continue
elif mname == 'charCodeAt':
assert len(args) <= 1
idx = 0 if len(args) == 0 else args[0]
assert isinstance(idx, int)
res = ord(obj[idx])
stack.append(res)
continue
elif isinstance(obj, list):
if mname == 'slice':
assert len(args) == 1
assert isinstance(args[0], int)
res = obj[args[0]:]
stack.append(res)
continue
elif mname == 'join':
assert len(args) == 1
assert isinstance(args[0], compat_str)
res = args[0].join(obj)
stack.append(res)
continue
raise NotImplementedError(
'Unsupported property %r on %r'
% (mname, obj))
elif opcode == 71: # returnvoid
res = undefined
return res
elif opcode == 72: # returnvalue
res = stack.pop()
return res
elif opcode == 73: # constructsuper
# Not yet implemented, just hope it works without it
arg_count = u30()
args = list(reversed(
[stack.pop() for _ in range(arg_count)]))
obj = stack.pop()
elif opcode == 74: # constructproperty
index = u30()
arg_count = u30()
args = list(reversed(
[stack.pop() for _ in range(arg_count)]))
obj = stack.pop()
mname = self.multinames[index]
assert isinstance(obj, _AVMClass)
# We do not actually call the constructor for now;
# we just pretend it does nothing
stack.append(obj.make_object())
elif opcode == 79: # callpropvoid
index = u30()
mname = self.multinames[index]
arg_count = u30()
args = list(reversed(
[stack.pop() for _ in range(arg_count)]))
obj = stack.pop()
if isinstance(obj, _AVMClass_Object):
func = self.extract_function(obj.avm_class, mname)
res = func(args)
assert res is undefined
continue
if isinstance(obj, _ScopeDict):
assert mname in obj.avm_class.method_names
func = self.extract_function(obj.avm_class, mname)
res = func(args)
assert res is undefined
continue
if mname == 'reverse':
assert isinstance(obj, list)
obj.reverse()
else:
raise NotImplementedError(
'Unsupported (void) property %r on %r'
% (mname, obj))
elif opcode == 86: # newarray
arg_count = u30()
arr = []
for i in range(arg_count):
arr.append(stack.pop())
arr = arr[::-1]
stack.append(arr)
elif opcode == 93: # findpropstrict
index = u30()
mname = self.multinames[index]
for s in reversed(scopes):
if mname in s:
res = s
break
else:
res = scopes[0]
if mname not in res and mname in _builtin_classes:
stack.append(_builtin_classes[mname])
else:
stack.append(res[mname])
elif opcode == 94: # findproperty
index = u30()
mname = self.multinames[index]
for s in reversed(scopes):
if mname in s:
res = s
break
else:
res = avm_class.variables
stack.append(res)
elif opcode == 96: # getlex
index = u30()
mname = self.multinames[index]
for s in reversed(scopes):
if mname in s:
scope = s
break
else:
scope = avm_class.variables
if mname in scope:
res = scope[mname]
elif mname in _builtin_classes:
res = _builtin_classes[mname]
else:
# Assume uninitialized
# TODO warn here
res = undefined
stack.append(res)
elif opcode == 97: # setproperty
index = u30()
value = stack.pop()
idx = self.multinames[index]
if isinstance(idx, _Multiname):
idx = stack.pop()
obj = stack.pop()
obj[idx] = value
elif opcode == 98: # getlocal
index = u30()
stack.append(registers[index])
elif opcode == 99: # setlocal
index = u30()
value = stack.pop()
registers[index] = value
elif opcode == 102: # getproperty
index = u30()
pname = self.multinames[index]
if pname == 'length':
obj = stack.pop()
assert isinstance(obj, (compat_str, list))
stack.append(len(obj))
elif isinstance(pname, compat_str): # Member access
obj = stack.pop()
if isinstance(obj, _AVMClass):
res = obj.static_properties[pname]
stack.append(res)
continue
assert isinstance(obj, (dict, _ScopeDict)),\
'Accessing member %r on %r' % (pname, obj)
res = obj.get(pname, undefined)
stack.append(res)
else: # Assume attribute access
idx = stack.pop()
assert isinstance(idx, int)
obj = stack.pop()
assert isinstance(obj, list)
stack.append(obj[idx])
elif opcode == 104: # initproperty
index = u30()
value = stack.pop()
idx = self.multinames[index]
if isinstance(idx, _Multiname):
idx = stack.pop()
obj = stack.pop()
obj[idx] = value
elif opcode == 115: # convert_
value = stack.pop()
intvalue = int(value)
stack.append(intvalue)
elif opcode == 128: # coerce
u30()
elif opcode == 130: # coerce_a
value = stack.pop()
# um, yes, it's any value
stack.append(value)
elif opcode == 133: # coerce_s
assert isinstance(stack[-1], (type(None), compat_str))
elif opcode == 147: # decrement
value = stack.pop()
assert isinstance(value, int)
stack.append(value - 1)
elif opcode == 149: # typeof
value = stack.pop()
return {
_Undefined: 'undefined',
compat_str: 'String',
int: 'Number',
float: 'Number',
}[type(value)]
elif opcode == 160: # add
value2 = stack.pop()
value1 = stack.pop()
res = value1 + value2
stack.append(res)
elif opcode == 161: # subtract
value2 = stack.pop()
value1 = stack.pop()
res = value1 - value2
stack.append(res)
elif opcode == 162: # multiply
value2 = stack.pop()
value1 = stack.pop()
res = value1 * value2
stack.append(res)
elif opcode == 164: # modulo
value2 = stack.pop()
value1 = stack.pop()
res = value1 % value2
stack.append(res)
elif opcode == 168: # bitand
value2 = stack.pop()
value1 = stack.pop()
assert isinstance(value1, int)
assert isinstance(value2, int)
res = value1 & value2
stack.append(res)
elif opcode == 171: # equals
value2 = stack.pop()
value1 = stack.pop()
result = value1 == value2
stack.append(result)
elif opcode == 175: # greaterequals
value2 = stack.pop()
value1 = stack.pop()
result = value1 >= value2
stack.append(result)
elif opcode == 192: # increment_i
value = stack.pop()
assert isinstance(value, int)
stack.append(value + 1)
elif opcode == 208: # getlocal_0
stack.append(registers[0])
elif opcode == 209: # getlocal_1
stack.append(registers[1])
elif opcode == 210: # getlocal_2
stack.append(registers[2])
elif opcode == 211: # getlocal_3
stack.append(registers[3])
elif opcode == 212: # setlocal_0
registers[0] = stack.pop()
elif opcode == 213: # setlocal_1
registers[1] = stack.pop()
elif opcode == 214: # setlocal_2
registers[2] = stack.pop()
elif opcode == 215: # setlocal_3
registers[3] = stack.pop()
else:
raise NotImplementedError(
'Unsupported opcode %d' % opcode)
avm_class.method_pyfunctions[func_name] = resfunc
return resfunc

187
youtube_dl/update.py
View File

@ -1,187 +0,0 @@
from __future__ import unicode_literals
import io
import json
import traceback
import hashlib
import os
import subprocess
import sys
from zipimport import zipimporter
from .utils import encode_compat_str
from .version import __version__
def rsa_verify(message, signature, key):
from hashlib import sha256
assert isinstance(message, bytes)
byte_size = (len(bin(key[0])) - 2 + 8 - 1) // 8
signature = ('%x' % pow(int(signature, 16), key[1], key[0])).encode()
signature = (byte_size * 2 - len(signature)) * b'0' + signature
asn1 = b'3031300d060960864801650304020105000420'
asn1 += sha256(message).hexdigest().encode()
if byte_size < len(asn1) // 2 + 11:
return False
expected = b'0001' + (byte_size - len(asn1) // 2 - 3) * b'ff' + b'00' + asn1
return expected == signature
def update_self(to_screen, verbose, opener):
"""Update the program file with the latest version from the repository"""
UPDATE_URL = 'https://rg3.github.io/youtube-dl/update/'
VERSION_URL = UPDATE_URL + 'LATEST_VERSION'
JSON_URL = UPDATE_URL + 'versions.json'
UPDATES_RSA_KEY = (0x9d60ee4d8f805312fdb15a62f87b95bd66177b91df176765d13514a0f1754bcd2057295c5b6f1d35daa6742c3ffc9a82d3e118861c207995a8031e151d863c9927e304576bc80692bc8e094896fcf11b66f3e29e04e3a71e9a11558558acea1840aec37fc396fb6b65dc81a1c4144e03bd1c011de62e3f1357b327d08426fe93, 65537)
if not isinstance(globals().get('__loader__'), zipimporter) and not hasattr(sys, 'frozen'):
to_screen('It looks like you installed youtube-dl with a package manager, pip, setup.py or a tarball. Please use that to update.')
return
# Check if there is a new version
try:
newversion = opener.open(VERSION_URL).read().decode('utf-8').strip()
except Exception:
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: can\'t find the current version. Please try again later.')
return
if newversion == __version__:
to_screen('youtube-dl is up-to-date (' + __version__ + ')')
return
# Download and check versions info
try:
versions_info = opener.open(JSON_URL).read().decode('utf-8')
versions_info = json.loads(versions_info)
except Exception:
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: can\'t obtain versions info. Please try again later.')
return
if 'signature' not in versions_info:
to_screen('ERROR: the versions file is not signed or corrupted. Aborting.')
return
signature = versions_info['signature']
del versions_info['signature']
if not rsa_verify(json.dumps(versions_info, sort_keys=True).encode('utf-8'), signature, UPDATES_RSA_KEY):
to_screen('ERROR: the versions file signature is invalid. Aborting.')
return
version_id = versions_info['latest']
def version_tuple(version_str):
return tuple(map(int, version_str.split('.')))
if version_tuple(__version__) >= version_tuple(version_id):
to_screen('youtube-dl is up to date (%s)' % __version__)
return
to_screen('Updating to version ' + version_id + ' ...')
version = versions_info['versions'][version_id]
print_notes(to_screen, versions_info['versions'])
# sys.executable is set to the full pathname of the exe-file for py2exe
filename = sys.executable if hasattr(sys, 'frozen') else sys.argv[0]
if not os.access(filename, os.W_OK):
to_screen('ERROR: no write permissions on %s' % filename)
return
# Py2EXE
if hasattr(sys, 'frozen'):
exe = filename
directory = os.path.dirname(exe)
if not os.access(directory, os.W_OK):
to_screen('ERROR: no write permissions on %s' % directory)
return
try:
urlh = opener.open(version['exe'][0])
newcontent = urlh.read()
urlh.close()
except (IOError, OSError):
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: unable to download latest version')
return
newcontent_hash = hashlib.sha256(newcontent).hexdigest()
if newcontent_hash != version['exe'][1]:
to_screen('ERROR: the downloaded file hash does not match. Aborting.')
return
try:
with open(exe + '.new', 'wb') as outf:
outf.write(newcontent)
except (IOError, OSError):
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: unable to write the new version')
return
try:
bat = os.path.join(directory, 'youtube-dl-updater.bat')
with io.open(bat, 'w') as batfile:
batfile.write('''
@echo off
echo Waiting for file handle to be closed ...
ping 127.0.0.1 -n 5 -w 1000 > NUL
move /Y "%s.new" "%s" > NUL
echo Updated youtube-dl to version %s.
start /b "" cmd /c del "%%~f0"&exit /b"
\n''' % (exe, exe, version_id))
subprocess.Popen([bat]) # Continues to run in the background
return # Do not show premature success messages
except (IOError, OSError):
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: unable to overwrite current version')
return
# Zip unix package
elif isinstance(globals().get('__loader__'), zipimporter):
try:
urlh = opener.open(version['bin'][0])
newcontent = urlh.read()
urlh.close()
except (IOError, OSError):
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: unable to download latest version')
return
newcontent_hash = hashlib.sha256(newcontent).hexdigest()
if newcontent_hash != version['bin'][1]:
to_screen('ERROR: the downloaded file hash does not match. Aborting.')
return
try:
with open(filename, 'wb') as outf:
outf.write(newcontent)
except (IOError, OSError):
if verbose:
to_screen(encode_compat_str(traceback.format_exc()))
to_screen('ERROR: unable to overwrite current version')
return
to_screen('Updated youtube-dl. Restart youtube-dl to use the new version.')
def get_notes(versions, fromVersion):
notes = []
for v, vdata in sorted(versions.items()):
if v > fromVersion:
notes.extend(vdata.get('notes', []))
return notes
def print_notes(to_screen, versions, fromVersion=__version__):
notes = get_notes(versions, fromVersion)
if notes:
to_screen('PLEASE NOTE:')
for note in notes:
to_screen(note)

3990
youtube_dl/utils.py
View File

File diff suppressed because it is too large Load Diff

3
youtube_dl/version.py
View File

@ -1,3 +0,0 @@
from __future__ import unicode_literals
__version__ = '2018.07.10'