File size: 11,806 Bytes
340cbde |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 |
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}
metadata = deep_get(response, 'metadata', 'channelMetadataRenderer',
default={})
if not metadata:
metadata = deep_get(response, 'microformat', 'microformatDataRenderer',
default={})
# channel doesn't exist or was terminated
# example terminated channel: https://www.youtube.com/channel/UCnKJeK_r90jDdIuzHXC0Org
if not metadata:
if response.get('alerts'):
error_string = ' '.join(
extract_str(deep_get(alert, 'alertRenderer', 'text'), default='')
for alert in response['alerts']
)
if not error_string:
error_string = 'Failed to extract error'
return {'error': error_string}
elif deep_get(response, 'responseContext', 'errors'):
for error in response['responseContext']['errors'].get('error', []):
if error.get('code') == 'INVALID_VALUE' and error.get('location') == 'browse_id':
return {'error': 'This channel does not exist'}
return {'error': 'Failure getting metadata'}
info = {'error': None}
info['current_tab'] = tab
info['approx_subscriber_count'] = extract_approx_int(deep_get(response,
'header', 'c4TabbedHeaderRenderer', 'subscriberCountText'))
# stuff from microformat (info given by youtube for every page on channel)
info['short_description'] = metadata.get('description')
if info['short_description'] and len(info['short_description']) > 730:
info['short_description'] = info['short_description'][0:730] + '...'
info['channel_name'] = metadata.get('title')
info['avatar'] = multi_deep_get(metadata,
['avatar', 'thumbnails', 0, 'url'],
['thumbnail', 'thumbnails', 0, 'url'],
)
channel_url = multi_get(metadata, 'urlCanonical', 'channelUrl')
if channel_url:
channel_id = get(channel_url.rstrip('/').split('/'), -1)
info['channel_id'] = channel_id
else:
info['channel_id'] = metadata.get('externalId')
if info['channel_id']:
info['channel_url'] = 'https://www.youtube.com/channel/' + channel_id
else:
info['channel_url'] = None
# get items
info['items'] = []
# empty channel
if 'contents' not in response and 'continuationContents' not in response:
return info
if tab in ('videos', 'playlists', 'search'):
items, ctoken = extract_items(response)
additional_info = {'author': info['channel_name'], 'author_url': info['channel_url']}
info['items'] = [extract_item_info(renderer, additional_info) for renderer in items]
if tab == 'search':
info['is_last_page'] = (ctoken is None)
elif tab == 'about':
items, _ = extract_items(response, item_types={'channelAboutFullMetadataRenderer'})
if not items:
info['error'] = 'Could not find channelAboutFullMetadataRenderer'
return info
channel_metadata = items[0]['channelAboutFullMetadataRenderer']
info['links'] = []
for link_json in channel_metadata.get('primaryLinks', ()):
url = remove_redirect(deep_get(link_json, 'navigationEndpoint', 'urlEndpoint', 'url'))
text = extract_str(link_json.get('title'))
info['links'].append( (text, url) )
info['date_joined'] = extract_date(channel_metadata.get('joinedDateText'))
info['view_count'] = extract_int(channel_metadata.get('viewCountText'))
info['description'] = extract_str(channel_metadata.get('description'), default='')
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,
item_types={'commentThreadRenderer', 'commentRenderer'})
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
|