app/common/innertube.py

   1 # functions that deal with parsing data from youtube's internal API ("innertube")
   2
   3 from urllib.parse import parse_qs, urlparse
   4
   5 def findall(obj, key):
   6     """
   7     given a list of dicts, where one dict contains a given key, return said key.
   8     """
   9     if obj is None: return []
  10     return [ obj[key] for obj in obj if key in obj.keys() ]
  11 def listget(obj, index, fallback=None):
  12     if obj is None: return fallback
  13     return next(iter(obj[index:]), fallback)
  14 flatten = lambda l: [item for sublist in l for item in sublist] # https://stackoverflow.com/a/952952
  15 first = lambda l: next(iter(l),{})
  16 listfind = lambda obj,key: first(findall(obj,key))
  17
  18 def prepare_searchresults(yt_results):
  19     contents = listfind(yt_results, 'response') \
  20         .get('contents',{})\
  21         .get('twoColumnSearchResultsRenderer',{})\
  22         .get('primaryContents',{})\
  23         .get('sectionListRenderer',{})\
  24         .get('contents',[])
  25     contents = flatten([c.get('contents',[]) for c in findall(contents, 'itemSectionRenderer')])
  26
  27     return parse_result_items(contents)
  28
  29 def prepare_infocards(metadata):
  30     cards = metadata.get('cards',{}).get('cardCollectionRenderer',{}).get('cards',[])
  31     return list(filter(None, map(parse_infocard, cards)))
  32
  33 def prepare_endcards(metadata):
  34     endsc = metadata.get('endscreen',{}).get('endscreenRenderer',{}).get('elements',[])
  35     return list(filter(None, map(parse_endcard, endsc)))
  36
  37 def prepare_channel(result, channel_id):
  38     response = listfind(result,'response')
  39
  40     if 'alerts' in response: # possibly got an error back
  41         from flask import current_app
  42         current_app.logger.error([(alert['alertRenderer']['type'],alert['alertRenderer']['text']['simpleText']) for alert in response['alerts']])
  43         return None,None,[],[],False
  44
  45     meta1 = response.get('metadata',{}).get('channelMetadataRenderer',{})
  46     meta2 = response.get('microformat',{}).get('microformatDataRenderer',{})
  47     title = meta1.get('title', meta2.get('title'))
  48     descr = meta1.get('description', meta2.get('description')) # meta2.description is capped at 160chars
  49     thumb = mkthumbs(meta2.get('thumbnail',meta1.get('avatar',{})).get('thumbnails',{})) # .avatar ~ 900px
  50
  51     contents = response.get('continuationContents')
  52     if not contents: # overran end of list
  53         return title, descr, thumb, [], False
  54
  55     unparsed = contents.get('gridContinuation',{}).get('items') or \
  56             contents.get('sectionListContinuation',{}).get('contents') or []
  57     items, extra = parse_channel_items(unparsed, channel_id, title)
  58     has_more = 'continuations' in (contents.get('gridContinuation') or
  59             contents.get('sectionListContinuation') or {})
  60
  61     return title, descr, thumb, items, has_more
  62
  63 def prepare_playlist(result):
  64     contents = listfind(result,'response')['continuationContents']
  65     unparsed = contents['playlistVideoListContinuation'].get('contents',[])
  66     has_more = 'continuations' in contents.get('playlistVideoListContinuation')
  67
  68     return list(filter(None, map(parse_playlist, unparsed))), has_more
  69
  70 def mkthumbs(thumbs):
  71     output = {str(e['height']): e['url'] for e in thumbs}
  72     largest=next(iter(sorted(output.keys(),reverse=True,key=int)),None)
  73     return {**output, 'largest': largest}
  74
  75 def clean_url(url):
  76     # externals URLs are redirected through youtube.com/redirect, but we
  77     # may encounter internal URLs, too
  78     return parse_qs(urlparse(url).query).get('q',[url])[0]
  79
  80 def toInt(s, fallback=0):
  81     if s is None:
  82         return fallback
  83     try:
  84         return int(''.join(filter(str.isdigit, s)))
  85     except ValueError:
  86         return fallback
  87
  88 # Remove left-/rightmost word from string:
  89 delL = lambda s: s.partition(' ')[2]
  90
  91 def age(s):
  92     if s is None: # missing from autogen'd music, some livestreams
  93         return None
  94     # Some livestreams have "Streamed 7 hours ago"
  95     s = s.replace("Streamed ","")
  96     # Now, everything should be in the form "1 year ago"
  97     value, unit, _ = s.split(" ")
  98     suffix = dict(
  99         month='mn',
 100         months='mn',
 101     ).get(unit, unit[0]) # first letter otherwise (e.g. year(s) => y)
 102
 103     return f"{value}{suffix}"
 104
 105 def log_unknown_card(data):
 106     import json
 107     try:
 108         from flask import request
 109         source = request.url
 110     except: source = "unknown"
 111     with open("/tmp/innertube.err", "a") as f:
 112         f.write(f"\n/***** {source} *****/\n")
 113         json.dump(data, f, indent=2)
 114
 115 def parse_result_items(items):
 116     # TODO: use .get() for most non-essential attributes
 117     """
 118     parses youtube search response into an easier to use format.
 119     """
 120     results = []
 121     extras = []
 122     for item in items:
 123         key = next(iter(item.keys()), None)
 124         content = item[key]
 125         if key == 'videoRenderer':
 126             is_live = listfind(content.get('badges',[]), 'metadataBadgeRenderer').get('style') == 'BADGE_STYLE_TYPE_LIVE_NOW'
 127             results.append({'type': 'VIDEO', 'content': {
 128                 'video_id': content['videoId'],
 129                 'title': content['title']['runs'][0]['text'],
 130                 'author': content['longBylineText']['runs'][0]['text'] or \
 131                           content['shortBylineText']['runs'][0]['text'],
 132                 'channel_id': content['ownerText']['runs'][0] \
 133                     ['navigationEndpoint']['browseEndpoint']['browseId'],
 134                 'length': content.get('lengthText',{}).get('simpleText') \
 135                     if not is_live else 'LIVE', # "44:07", "1:41:50"
 136                 'views': toInt(content.get('viewCountText',{}).get('simpleText') or # "123,456 views"
 137                     listget(content.get('viewCountText',{}).get('runs'),0,{}).get('text')), # "1,234 watching"
 138                 'published': age(content.get('publishedTimeText',{}).get('simpleText')),
 139             }})
 140         elif key == 'playlistRenderer':
 141             results.append({'type': 'PLAYLIST', 'content': {
 142                 'playlist_id': content['navigationEndpoint']['watchEndpoint']['playlistId'],
 143                 'video_id': content['navigationEndpoint']['watchEndpoint']['videoId'],
 144                 'title': content['title']['simpleText'],
 145                 'author': content['longBylineText']['runs'][0]['text'] or
 146                      content['shortBylineText']['runs'][0]['text'],
 147                 'channel_id': content['longBylineText']['runs'][0]['navigationEndpoint']['browseEndpoint']['browseId'], # OR .shortBylineText
 148                 'n_videos': toInt(content['videoCount']),
 149             }})
 150         elif key == 'radioRenderer': # "Mix" playlists
 151             results.append({'type': 'PLAYLIST', 'content': {
 152                 'playlist_id': content['playlistId'],
 153                 'video_id': content['navigationEndpoint']['watchEndpoint']['videoId'],
 154                 'title': content['title']['simpleText'],
 155                 'author': content['longBylineText']['simpleText'] or \
 156                     content['shortBylineText']['simpleText'] , # always "YouTube"
 157                 'channel_id': None,
 158                 'n_videos': content['videoCountShortText']['runs'][0]['text'] or \
 159                     content['videoCountText']['runs'][0]['text'],
 160                     # videoCountShortText: "50+"; videoCountText: "50+ videos"
 161             }})
 162         elif key == 'channelRenderer':
 163             results.append({'type': 'CHANNEL', 'content': {
 164                 'channel_id': content['channelId'],
 165                 'title': content['title']['simpleText'],
 166                 'icons': mkthumbs(content['thumbnail']['thumbnails']),
 167                 'subscribers': content.get('subscriberCountText',{}).get('simpleText'), # "2.47K subscribers"
 168             }})
 169         elif key == 'shelfRenderer':
 170             r, e = parse_result_items(content['content']['verticalListRenderer']['items'])
 171             results.extend(r)
 172             extras.extend(e)
 173         elif key == 'movieRenderer': # movies to buy/rent
 174             pass
 175         elif key in ['carouselAdRenderer','searchPyvRenderer','promotedSparklesTextSearchRenderer']: # haha, no.
 176             pass
 177         elif key == 'horizontalCardListRenderer':
 178             # suggested searches: .cards[].searchRefinementCardRenderer.query.runs[].text
 179             pass
 180         elif key == 'emergencyOneboxRenderer': # suicide prevention hotline
 181             pass
 182         elif key == 'clarificationRenderer': # COVID-19 infos
 183             pass
 184         elif key == 'didYouMeanRenderer' or key == 'showingResultsForRenderer':
 185             extras.append({
 186                 'type': 'spelling',
 187                 'query': content['correctedQueryEndpoint']['searchEndpoint']['query'], # non-misspelled query
 188                 'autocorrected': key == 'showingResultsForRenderer',
 189             })
 190         elif key == 'backgroundPromoRenderer': # e.g. "no results"
 191             extras.append({
 192                 'type': content['icon']['iconType'],
 193                 'message': content['title']['runs'][0]['text'],
 194             })
 195         else:
 196             log_unknown_card(item)
 197     return results, extras
 198
 199 def parse_infocard(card):
 200     """
 201     parses a single infocard into a format that's easier to handle.
 202     """
 203     card = card['cardRenderer']
 204     ctype = list(card['content'].keys())[0]
 205     content = card['content'][ctype]
 206     if ctype == "pollRenderer":
 207         return {'type': "POLL", 'content': {
 208             'question': content['question']['simpleText'],
 209             'answers': [(a['text']['simpleText'],a['numVotes']) \
 210                 for a in content['choices']],
 211         }}
 212     elif ctype == "videoInfoCardContentRenderer":
 213         is_live = content.get('badge',{}).get('liveBadgeRenderer') is not None
 214         return {'type': "VIDEO", 'content': {
 215             'video_id': content['action']['watchEndpoint']['videoId'],
 216             'title': content['videoTitle']['simpleText'],
 217             'author': delL(content['channelName']['simpleText']),
 218             'length': content.get('lengthString',{}).get('simpleText') \
 219                 if not is_live else "LIVE", # "23:03"
 220             'views': toInt(content.get('viewCountText',{}).get('simpleText')),
 221                 # XXX: views sometimes "Starts: July 31, 2020 at 1:30 PM"
 222         }}
 223     elif ctype == "playlistInfoCardContentRenderer":
 224         return {'type': "PLAYLIST", 'content': {
 225             'playlist_id': content['action']['watchEndpoint']['playlistId'],
 226             'video_id': content['action']['watchEndpoint']['videoId'],
 227             'title': content['playlistTitle']['simpleText'],
 228             'author': delL(content['channelName']['simpleText']),
 229             'n_videos': toInt(content['playlistVideoCount']['simpleText']),
 230         }}
 231     elif ctype == "simpleCardContentRenderer" and \
 232             'urlEndpoint' in content['command']:
 233         return {'type': "WEBSITE", 'content': {
 234             'url': clean_url(content['command']['urlEndpoint']['url']),
 235             'domain': content['displayDomain']['simpleText'],
 236             'title': content['title']['simpleText'],
 237             # XXX: no thumbnails for infocards
 238         }}
 239     elif ctype == "collaboratorInfoCardContentRenderer":
 240         return {'type': "CHANNEL", 'content': {
 241             'channel_id': content['endpoint']['browseEndpoint']['browseId'],
 242             'title': content['channelName']['simpleText'],
 243             'icons': mkthumbs(content['channelAvatar']['thumbnails']),
 244             'subscribers': content.get('subscriberCountText',{}).get('simpleText',''), # "545K subscribers"
 245         }}
 246     else:
 247         log_unknown_card(card)
 248         return None
 249
 250 def parse_endcard(card):
 251     """
 252     parses a single endcard into a format that's easier to handle.
 253     """
 254     card = card.get('endscreenElementRenderer', card) #only sometimes nested
 255     ctype = card['style']
 256     if ctype == "CHANNEL":
 257         return {'type': ctype, 'content': {
 258             'channel_id': card['endpoint']['browseEndpoint']['browseId'],
 259             'title': card['title']['simpleText'],
 260             'icons': mkthumbs(card['image']['thumbnails']),
 261         }}
 262     elif ctype == "VIDEO":
 263         return {'type': ctype, 'content': {
 264             'video_id': card['endpoint']['watchEndpoint']['videoId'],
 265             'title': card['title']['simpleText'],
 266             'length': card['videoDuration']['simpleText'],  # '12:21'
 267             'views': toInt(card['metadata']['simpleText']),
 268             # XXX: no channel name
 269         }}
 270     elif ctype == "PLAYLIST":
 271         return {'type': ctype, 'content': {
 272             'playlist_id': card['endpoint']['watchEndpoint']['playlistId'],
 273             'video_id': card['endpoint']['watchEndpoint']['videoId'],
 274             'title': card['title']['simpleText'],
 275             'author': delL(card['metadata']['simpleText']),
 276             'n_videos': toInt(card['playlistLength']['simpleText']),
 277         }}
 278     elif ctype == "WEBSITE" or ctype == "CREATOR_MERCHANDISE":
 279         url = clean_url(card['endpoint']['urlEndpoint']['url'])
 280         return {'type': "WEBSITE", 'content': {
 281             'url': url,
 282             'domain': urlparse(url).netloc,
 283             'title': card['title']['simpleText'],
 284             'icons': mkthumbs(card['image']['thumbnails']),
 285         }}
 286     else:
 287         log_unknown_card(card)
 288         return None
 289
 290 def parse_channel_items(items, channel_id, author):
 291     result = []
 292     extra = []
 293     for item in items:
 294         key = next(iter(item.keys()), None)
 295         content = item[key]
 296         if key in ["gridVideoRenderer", "videoRenderer", "videoCardRenderer"]:
 297             # only videoCardRenderer (topic channels) has author and channel, others fall back to supplied ones.
 298             result.append({'type': 'VIDEO', 'content': {
 299                 'video_id': content['videoId'],
 300                 'title': content['title'].get('simpleText') or content['title'].get('runs',[{}])[0].get('text'),
 301                 'author': content.get('bylineText',{}).get('runs',[{}])[0].get('text') or author,
 302                 'channel_id': content.get('bylineText',{}).get('runs',[{}])[0] \
 303                     .get('navigationEndpoint',{}).get('browseEndpoint',{}).get('browseId') or channel_id,
 304                 'length': (content.get('lengthText',{}).get('simpleText') or # topic channel
 305                     listfind(content.get('thumbnailOverlays',[]),'thumbnailOverlayTimeStatusRenderer')
 306                         .get('text',{}).get('simpleText')),
 307                 # topic channel: .metadataText.simpleText = "22M views \u00b7 2 months ago"
 308                 'views': toInt(content.get('viewCountText',{}).get('simpleText')),
 309                 'published': age(content.get('publishedTimeText',{}).get('simpleText')),
 310             }})
 311         elif key == "gridPlaylistRenderer" or key == "playlistRenderer":
 312             result.append({'type': 'PLAYLIST', 'content': {
 313                 'playlist_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('playlistId') or content.get('playlistId'),
 314                 'video_id': content['navigationEndpoint'].get('watchEndpoint',{}).get('videoId',{}),
 315                 'title': (content['title'].get('simpleText') or # playlistRenderer
 316                     content['title']['runs'][0]['text']), # gridPlaylistRenderer
 317                 'author': author,
 318                 'channel_id': channel_id,
 319                 'n_videos': toInt(content.get('videoCount') or # playlistRenderer
 320                     content.get('videoCountShortText',{}).get('simpleText') or # grid(1)
 321                     content.get('videoCountText',{}).get('runs',[{}])[0].get('text')), # grid(2)
 322             }})
 323         elif key in ["itemSectionRenderer", "gridRenderer", "horizontalCardListRenderer"]:
 324                 newkey = {
 325                     "itemSectionRenderer": 'contents',
 326                     "gridRenderer": 'items',
 327                     "horizontalCardListRenderer": 'cards',
 328                 }.get(key)
 329                 r, e = parse_channel_items(content[newkey], channel_id, author)
 330                 result.extend(r)
 331                 extra.extend(e)
 332         elif key == "shelfRenderer":
 333                 r, e = parse_channel_items([content['content']], channel_id, author)
 334                 result.extend(r)
 335                 extra.extend(e)
 336         elif key == "messageRenderer":
 337             # e.g. {'messageRenderer': {'text': {'runs': [{'text': 'This channel has no playlists.'}]}}}
 338             pass
 339         elif key == "gameCardRenderer":
 340             pass
 341         else:
 342             log_unknown_card(item)
 343
 344     return result, extra
 345
 346 def parse_playlist(item):
 347     key = next(iter(item.keys()), None)
 348     content = item[key]
 349     if key == "playlistVideoRenderer":
 350         if not content.get('isPlayable', False):
 351             return None # private or deleted video
 352
 353         return {'type': 'VIDEO', 'content': {
 354             'video_id': content['videoId'],
 355             'title': (content['title'].get('simpleText') or # playable videos
 356                 content['title'].get('runs',[{}])[0].get('text')), # "[Private video]"
 357             'playlist_id': content['navigationEndpoint']['watchEndpoint']['playlistId'],
 358             'index': content['navigationEndpoint']['watchEndpoint'].get('index',0), #or int(content['index']['simpleText']) (absent on course intros; e.g. PL96C35uN7xGJu6skU4TBYrIWxggkZBrF5)
 359             # rest is missing from unplayable videos:
 360             'author': content.get('shortBylineText',{}).get('runs',[{}])[0].get('text'),
 361             'channel_id':content.get('shortBylineText',{}).get('runs',[{}])[0].get('navigationEndpoint',{}).get('browseEndpoint',{}).get('browseId'),
 362             'length': (content.get("lengthText",{}).get("simpleText") or # "8:51"
 363                 int(content.get("lengthSeconds", 0))), # "531"
 364             'starttime': content['navigationEndpoint']['watchEndpoint'].get('startTimeSeconds'),
 365         }}
 366     else:
 367         raise Exception(item) # XXX TODO