Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/crunchyroll.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6 import base64
   7 import zlib
   8
   9 from hashlib import sha1
  10 from math import pow, sqrt, floor
  11 from .common import InfoExtractor
  12 from ..compat import (
  13     compat_etree_fromstring,
  14     compat_urllib_parse_urlencode,
  15     compat_urllib_request,
  16     compat_urlparse,
  17 )
  18 from ..utils import (
  19     ExtractorError,
  20     bytes_to_intlist,
  21     intlist_to_bytes,
  22     int_or_none,
  23     lowercase_escape,
  24     remove_end,
  25     sanitized_Request,
  26     unified_strdate,
  27     urlencode_postdata,
  28     xpath_text,
  29     extract_attributes,
  30 )
  31 from ..aes import (
  32     aes_cbc_decrypt,
  33 )
  34
  35
  36 class CrunchyrollBaseIE(InfoExtractor):
  37     _LOGIN_URL = 'https://www.crunchyroll.com/login'
  38     _LOGIN_FORM = 'login_form'
  39     _NETRC_MACHINE = 'crunchyroll'
  40
  41     def _call_rpc_api(self, method, video_id, note=None, data=None):
  42         data = data or {}
  43         data['req'] = 'RpcApi' + method
  44         data = compat_urllib_parse_urlencode(data).encode('utf-8')
  45         return self._download_xml(
  46             'http://www.crunchyroll.com/xml/',
  47             video_id, note, fatal=False, data=data, headers={
  48                 'Content-Type': 'application/x-www-form-urlencoded',
  49             })
  50
  51     def _login(self):
  52         (username, password) = self._get_login_info()
  53         if username is None:
  54             return
  55
  56         self._download_webpage(
  57             'https://www.crunchyroll.com/?a=formhandler',
  58             None, 'Logging in', 'Wrong login info',
  59             data=urlencode_postdata({
  60                 'formname': 'RpcApiUser_Login',
  61                 'next_url': 'https://www.crunchyroll.com/acct/membership',
  62                 'name': username,
  63                 'password': password,
  64             }))
  65
  66         '''
  67         login_page = self._download_webpage(
  68             self._LOGIN_URL, None, 'Downloading login page')
  69
  70         def is_logged(webpage):
  71             return '<title>Redirecting' in webpage
  72
  73         # Already logged in
  74         if is_logged(login_page):
  75             return
  76
  77         login_form_str = self._search_regex(
  78             r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM,
  79             login_page, 'login form', group='form')
  80
  81         post_url = extract_attributes(login_form_str).get('action')
  82         if not post_url:
  83             post_url = self._LOGIN_URL
  84         elif not post_url.startswith('http'):
  85             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  86
  87         login_form = self._form_hidden_inputs(self._LOGIN_FORM, login_page)
  88
  89         login_form.update({
  90             'login_form[name]': username,
  91             'login_form[password]': password,
  92         })
  93
  94         response = self._download_webpage(
  95             post_url, None, 'Logging in', 'Wrong login info',
  96             data=urlencode_postdata(login_form),
  97             headers={'Content-Type': 'application/x-www-form-urlencoded'})
  98
  99         # Successful login
 100         if is_logged(response):
 101             return
 102
 103         error = self._html_search_regex(
 104             '(?s)<ul[^>]+class=["\']messages["\'][^>]*>(.+?)</ul>',
 105             response, 'error message', default=None)
 106         if error:
 107             raise ExtractorError('Unable to login: %s' % error, expected=True)
 108
 109         raise ExtractorError('Unable to log in')
 110         '''
 111
 112     def _real_initialize(self):
 113         self._login()
 114
 115     def _download_webpage(self, url_or_request, *args, **kwargs):
 116         request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
 117                    else sanitized_Request(url_or_request))
 118         # Accept-Language must be set explicitly to accept any language to avoid issues
 119         # similar to https://github.com/rg3/youtube-dl/issues/6797.
 120         # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction
 121         # should be imposed or not (from what I can see it just takes the first language
 122         # ignoring the priority and requires it to correspond the IP). By the way this causes
 123         # Crunchyroll to not work in georestriction cases in some browsers that don't place
 124         # the locale lang first in header. However allowing any language seems to workaround the issue.
 125         request.add_header('Accept-Language', '*')
 126         return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs)
 127
 128     @staticmethod
 129     def _add_skip_wall(url):
 130         parsed_url = compat_urlparse.urlparse(url)
 131         qs = compat_urlparse.parse_qs(parsed_url.query)
 132         # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message:
 133         # > This content may be inappropriate for some people.
 134         # > Are you sure you want to continue?
 135         # since it's not disabled by default in crunchyroll account's settings.
 136         # See https://github.com/rg3/youtube-dl/issues/7202.
 137         qs['skip_wall'] = ['1']
 138         return compat_urlparse.urlunparse(
 139             parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
 140
 141
 142 class CrunchyrollIE(CrunchyrollBaseIE):
 143     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
 144     _TESTS = [{
 145         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
 146         'info_dict': {
 147             'id': '645513',
 148             'ext': 'mp4',
 149             'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
 150             'description': 'md5:2d17137920c64f2f49981a7797d275ef',
 151             'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
 152             'uploader': 'Yomiuri Telecasting Corporation (YTV)',
 153             'upload_date': '20131013',
 154             'url': 're:(?!.*&amp)',
 155         },
 156         'params': {
 157             # rtmp
 158             'skip_download': True,
 159         },
 160     }, {
 161         'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
 162         'info_dict': {
 163             'id': '589804',
 164             'ext': 'flv',
 165             'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
 166             'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
 167             'thumbnail': r're:^https?://.*\.jpg$',
 168             'uploader': 'Danny Choo Network',
 169             'upload_date': '20120213',
 170         },
 171         'params': {
 172             # rtmp
 173             'skip_download': True,
 174         },
 175         'skip': 'Video gone',
 176     }, {
 177         'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
 178         'info_dict': {
 179             'id': '702409',
 180             'ext': 'mp4',
 181             'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
 182             'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
 183             'thumbnail': r're:^https?://.*\.jpg$',
 184             'uploader': 'TV TOKYO',
 185             'upload_date': '20160508',
 186         },
 187         'params': {
 188             # m3u8 download
 189             'skip_download': True,
 190         },
 191     }, {
 192         'url': 'http://www.crunchyroll.com/konosuba-gods-blessing-on-this-wonderful-world/episode-1-give-me-deliverance-from-this-judicial-injustice-727589',
 193         'info_dict': {
 194             'id': '727589',
 195             'ext': 'mp4',
 196             'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance From This Judicial Injustice!",
 197             'description': 'md5:cbcf05e528124b0f3a0a419fc805ea7d',
 198             'thumbnail': r're:^https?://.*\.jpg$',
 199             'uploader': 'Kadokawa Pictures Inc.',
 200             'upload_date': '20170118',
 201             'series': "KONOSUBA -God's blessing on this wonderful world!",
 202             'season': "KONOSUBA -God's blessing on this wonderful world! 2",
 203             'season_number': 2,
 204             'episode': 'Give Me Deliverance From This Judicial Injustice!',
 205             'episode_number': 1,
 206         },
 207         'params': {
 208             # m3u8 download
 209             'skip_download': True,
 210         },
 211     }, {
 212         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
 213         'only_matching': True,
 214     }, {
 215         # geo-restricted (US), 18+ maturity wall, non-premium available
 216         'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
 217         'only_matching': True,
 218     }, {
 219         # A description with double quotes
 220         'url': 'http://www.crunchyroll.com/11eyes/episode-1-piros-jszaka-red-night-535080',
 221         'info_dict': {
 222             'id': '535080',
 223             'ext': 'mp4',
 224             'title': '11eyes Episode 1 – Piros éjszaka - Red Night',
 225             'description': 'Kakeru and Yuka are thrown into an alternate nightmarish world they call "Red Night".',
 226             'uploader': 'Marvelous AQL Inc.',
 227             'upload_date': '20091021',
 228         },
 229         'params': {
 230             # Just test metadata extraction
 231             'skip_download': True,
 232         },
 233     }, {
 234         # make sure we can extract an uploader name that's not a link
 235         'url': 'http://www.crunchyroll.com/hakuoki-reimeiroku/episode-1-dawn-of-the-divine-warriors-606899',
 236         'info_dict': {
 237             'id': '606899',
 238             'ext': 'mp4',
 239             'title': 'Hakuoki Reimeiroku Episode 1 – Dawn of the Divine Warriors',
 240             'description': 'Ryunosuke was left to die, but Serizawa-san asked him a simple question "Do you want to live?"',
 241             'uploader': 'Geneon Entertainment',
 242             'upload_date': '20120717',
 243         },
 244         'params': {
 245             # just test metadata extraction
 246             'skip_download': True,
 247         },
 248     }, {
 249         # A video with a vastly different season name compared to the series name
 250         'url': 'http://www.crunchyroll.com/nyarko-san-another-crawling-chaos/episode-1-test-590532',
 251         'info_dict': {
 252             'id': '590532',
 253             'ext': 'mp4',
 254             'title': 'Haiyoru! Nyaruani (ONA) Episode 1 – Test',
 255             'description': 'Mahiro and Nyaruko talk about official certification.',
 256             'uploader': 'TV TOKYO',
 257             'upload_date': '20120305',
 258             'series': 'Nyarko-san: Another Crawling Chaos',
 259             'season': 'Haiyoru! Nyaruani (ONA)',
 260         },
 261         'params': {
 262             # Just test metadata extraction
 263             'skip_download': True,
 264         },
 265     }]
 266
 267     _FORMAT_IDS = {
 268         '360': ('60', '106'),
 269         '480': ('61', '106'),
 270         '720': ('62', '106'),
 271         '1080': ('80', '108'),
 272     }
 273
 274     def _decrypt_subtitles(self, data, iv, id):
 275         data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
 276         iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))
 277         id = int(id)
 278
 279         def obfuscate_key_aux(count, modulo, start):
 280             output = list(start)
 281             for _ in range(count):
 282                 output.append(output[-1] + output[-2])
 283             # cut off start values
 284             output = output[2:]
 285             output = list(map(lambda x: x % modulo + 33, output))
 286             return output
 287
 288         def obfuscate_key(key):
 289             num1 = int(floor(pow(2, 25) * sqrt(6.9)))
 290             num2 = (num1 ^ key) << 5
 291             num3 = key ^ num1
 292             num4 = num3 ^ (num3 >> 3) ^ num2
 293             prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
 294             shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
 295             # Extend 160 Bit hash to 256 Bit
 296             return shaHash + [0] * 12
 297
 298         key = obfuscate_key(id)
 299
 300         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
 301         return zlib.decompress(decrypted_data)
 302
 303     def _convert_subtitles_to_srt(self, sub_root):
 304         output = ''
 305
 306         for i, event in enumerate(sub_root.findall('./events/event'), 1):
 307             start = event.attrib['start'].replace('.', ',')
 308             end = event.attrib['end'].replace('.', ',')
 309             text = event.attrib['text'].replace('\\N', '\n')
 310             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
 311         return output
 312
 313     def _convert_subtitles_to_ass(self, sub_root):
 314         output = ''
 315
 316         def ass_bool(strvalue):
 317             assvalue = '0'
 318             if strvalue == '1':
 319                 assvalue = '-1'
 320             return assvalue
 321
 322         output = '[Script Info]\n'
 323         output += 'Title: %s\n' % sub_root.attrib['title']
 324         output += 'ScriptType: v4.00+\n'
 325         output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
 326         output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
 327         output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']
 328         output += """
 329 [V4+ Styles]
 330 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 331 """
 332         for style in sub_root.findall('./styles/style'):
 333             output += 'Style: ' + style.attrib['name']
 334             output += ',' + style.attrib['font_name']
 335             output += ',' + style.attrib['font_size']
 336             output += ',' + style.attrib['primary_colour']
 337             output += ',' + style.attrib['secondary_colour']
 338             output += ',' + style.attrib['outline_colour']
 339             output += ',' + style.attrib['back_colour']
 340             output += ',' + ass_bool(style.attrib['bold'])
 341             output += ',' + ass_bool(style.attrib['italic'])
 342             output += ',' + ass_bool(style.attrib['underline'])
 343             output += ',' + ass_bool(style.attrib['strikeout'])
 344             output += ',' + style.attrib['scale_x']
 345             output += ',' + style.attrib['scale_y']
 346             output += ',' + style.attrib['spacing']
 347             output += ',' + style.attrib['angle']
 348             output += ',' + style.attrib['border_style']
 349             output += ',' + style.attrib['outline']
 350             output += ',' + style.attrib['shadow']
 351             output += ',' + style.attrib['alignment']
 352             output += ',' + style.attrib['margin_l']
 353             output += ',' + style.attrib['margin_r']
 354             output += ',' + style.attrib['margin_v']
 355             output += ',' + style.attrib['encoding']
 356             output += '\n'
 357
 358         output += """
 359 [Events]
 360 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 361 """
 362         for event in sub_root.findall('./events/event'):
 363             output += 'Dialogue: 0'
 364             output += ',' + event.attrib['start']
 365             output += ',' + event.attrib['end']
 366             output += ',' + event.attrib['style']
 367             output += ',' + event.attrib['name']
 368             output += ',' + event.attrib['margin_l']
 369             output += ',' + event.attrib['margin_r']
 370             output += ',' + event.attrib['margin_v']
 371             output += ',' + event.attrib['effect']
 372             output += ',' + event.attrib['text']
 373             output += '\n'
 374
 375         return output
 376
 377     def _extract_subtitles(self, subtitle):
 378         sub_root = compat_etree_fromstring(subtitle)
 379         return [{
 380             'ext': 'srt',
 381             'data': self._convert_subtitles_to_srt(sub_root),
 382         }, {
 383             'ext': 'ass',
 384             'data': self._convert_subtitles_to_ass(sub_root),
 385         }]
 386
 387     def _get_subtitles(self, video_id, webpage):
 388         subtitles = {}
 389         for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage):
 390             sub_doc = self._call_rpc_api(
 391                 'Subtitle_GetXml', video_id,
 392                 'Downloading subtitles for ' + sub_name, data={
 393                     'subtitle_script_id': sub_id,
 394                 })
 395             if sub_doc is None:
 396                 continue
 397             sid = sub_doc.get('id')
 398             iv = xpath_text(sub_doc, 'iv', 'subtitle iv')
 399             data = xpath_text(sub_doc, 'data', 'subtitle data')
 400             if not sid or not iv or not data:
 401                 continue
 402             subtitle = self._decrypt_subtitles(data, iv, sid).decode('utf-8')
 403             lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
 404             if not lang_code:
 405                 continue
 406             subtitles[lang_code] = self._extract_subtitles(subtitle)
 407         return subtitles
 408
 409     def _real_extract(self, url):
 410         mobj = re.match(self._VALID_URL, url)
 411         video_id = mobj.group('video_id')
 412
 413         if mobj.group('prefix') == 'm':
 414             mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
 415             webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
 416         else:
 417             webpage_url = 'http://www.' + mobj.group('url')
 418
 419         webpage = self._download_webpage(
 420             self._add_skip_wall(webpage_url), video_id,
 421             headers=self.geo_verification_headers())
 422         note_m = self._html_search_regex(
 423             r'<div class="showmedia-trailer-notice">(.+?)</div>',
 424             webpage, 'trailer-notice', default='')
 425         if note_m:
 426             raise ExtractorError(note_m)
 427
 428         mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
 429         if mobj:
 430             msg = json.loads(mobj.group('msg'))
 431             if msg.get('type') == 'error':
 432                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
 433
 434         if 'To view this, please log in to verify you are 18 or older.' in webpage:
 435             self.raise_login_required()
 436
 437         video_title = self._html_search_regex(
 438             r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
 439             webpage, 'video_title')
 440         video_title = re.sub(r' {2,}', ' ', video_title)
 441         video_description = self._parse_json(self._html_search_regex(
 442             r'<script[^>]*>\s*.+?\[media_id=%s\].+?({.+?"description"\s*:.+?})\);' % video_id,
 443             webpage, 'description', default='{}'), video_id).get('description')
 444         if video_description:
 445             video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
 446         video_upload_date = self._html_search_regex(
 447             [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
 448             webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
 449         if video_upload_date:
 450             video_upload_date = unified_strdate(video_upload_date)
 451         video_uploader = self._html_search_regex(
 452             # try looking for both an uploader that's a link and one that's not
 453             [r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', r'<div>\s*Publisher:\s*<span>\s*(.+?)\s*</span>\s*</div>'],
 454             webpage, 'video_uploader', fatal=False)
 455
 456         available_fmts = []
 457         for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
 458             attrs = extract_attributes(a)
 459             href = attrs.get('href')
 460             if href and '/freetrial' in href:
 461                 continue
 462             available_fmts.append(fmt)
 463         if not available_fmts:
 464             for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
 465                 available_fmts = re.findall(p, webpage)
 466                 if available_fmts:
 467                     break
 468         video_encode_ids = []
 469         formats = []
 470         for fmt in available_fmts:
 471             stream_quality, stream_format = self._FORMAT_IDS[fmt]
 472             video_format = fmt + 'p'
 473             stream_infos = []
 474             streamdata = self._call_rpc_api(
 475                 'VideoPlayer_GetStandardConfig', video_id,
 476                 'Downloading media info for %s' % video_format, data={
 477                     'media_id': video_id,
 478                     'video_format': stream_format,
 479                     'video_quality': stream_quality,
 480                     'current_page': url,
 481                 })
 482             if streamdata is not None:
 483                 stream_info = streamdata.find('./{default}preload/stream_info')
 484                 if stream_info is not None:
 485                     stream_infos.append(stream_info)
 486             stream_info = self._call_rpc_api(
 487                 'VideoEncode_GetStreamInfo', video_id,
 488                 'Downloading stream info for %s' % video_format, data={
 489                     'media_id': video_id,
 490                     'video_format': stream_format,
 491                     'video_encode_quality': stream_quality,
 492                 })
 493             if stream_info is not None:
 494                 stream_infos.append(stream_info)
 495             for stream_info in stream_infos:
 496                 video_encode_id = xpath_text(stream_info, './video_encode_id')
 497                 if video_encode_id in video_encode_ids:
 498                     continue
 499                 video_encode_ids.append(video_encode_id)
 500
 501                 video_file = xpath_text(stream_info, './file')
 502                 if not video_file:
 503                     continue
 504                 if video_file.startswith('http'):
 505                     formats.extend(self._extract_m3u8_formats(
 506                         video_file, video_id, 'mp4', entry_protocol='m3u8_native',
 507                         m3u8_id='hls', fatal=False))
 508                     continue
 509
 510                 video_url = xpath_text(stream_info, './host')
 511                 if not video_url:
 512                     continue
 513                 metadata = stream_info.find('./metadata')
 514                 format_info = {
 515                     'format': video_format,
 516                     'height': int_or_none(xpath_text(metadata, './height')),
 517                     'width': int_or_none(xpath_text(metadata, './width')),
 518                 }
 519
 520                 if '.fplive.net/' in video_url:
 521                     video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
 522                     parsed_video_url = compat_urlparse.urlparse(video_url)
 523                     direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
 524                         netloc='v.lvlt.crcdn.net',
 525                         path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
 526                     if self._is_valid_url(direct_video_url, video_id, video_format):
 527                         format_info.update({
 528                             'format_id': 'http-' + video_format,
 529                             'url': direct_video_url,
 530                         })
 531                         formats.append(format_info)
 532                         continue
 533
 534                 format_info.update({
 535                     'format_id': 'rtmp-' + video_format,
 536                     'url': video_url,
 537                     'play_path': video_file,
 538                     'ext': 'flv',
 539                 })
 540                 formats.append(format_info)
 541         self._sort_formats(formats, ('height', 'width', 'tbr', 'fps'))
 542
 543         metadata = self._call_rpc_api(
 544             'VideoPlayer_GetMediaMetadata', video_id,
 545             note='Downloading media info', data={
 546                 'media_id': video_id,
 547             })
 548
 549         subtitles = self.extract_subtitles(video_id, webpage)
 550
 551         # webpage provide more accurate data than series_title from XML
 552         series = self._html_search_regex(
 553             r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
 554             webpage, 'series', fatal=False)
 555         season = xpath_text(metadata, 'series_title')
 556
 557         episode = xpath_text(metadata, 'episode_title')
 558         episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
 559
 560         season_number = int_or_none(self._search_regex(
 561             r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
 562             webpage, 'season number', default=None))
 563
 564         return {
 565             'id': video_id,
 566             'title': video_title,
 567             'description': video_description,
 568             'thumbnail': xpath_text(metadata, 'episode_image_url'),
 569             'uploader': video_uploader,
 570             'upload_date': video_upload_date,
 571             'series': series,
 572             'season': season,
 573             'season_number': season_number,
 574             'episode': episode,
 575             'episode_number': episode_number,
 576             'subtitles': subtitles,
 577             'formats': formats,
 578         }
 579
 580
 581 class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
 582     IE_NAME = 'crunchyroll:playlist'
 583     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)'
 584
 585     _TESTS = [{
 586         'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 587         'info_dict': {
 588             'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 589             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
 590         },
 591         'playlist_count': 13,
 592     }, {
 593         # geo-restricted (US), 18+ maturity wall, non-premium available
 594         'url': 'http://www.crunchyroll.com/cosplay-complex-ova',
 595         'info_dict': {
 596             'id': 'cosplay-complex-ova',
 597             'title': 'Cosplay Complex OVA'
 598         },
 599         'playlist_count': 3,
 600         'skip': 'Georestricted',
 601     }, {
 602         # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14
 603         'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1',
 604         'only_matching': True,
 605     }]
 606
 607     def _real_extract(self, url):
 608         show_id = self._match_id(url)
 609
 610         webpage = self._download_webpage(
 611             self._add_skip_wall(url), show_id,
 612             headers=self.geo_verification_headers())
 613         title = self._html_search_regex(
 614             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
 615             webpage, 'title')
 616         episode_paths = re.findall(
 617             r'(?s)<li id="showview_videos_media_(\d+)"[^>]+>.*?<a href="([^"]+)"',
 618             webpage)
 619         entries = [
 620             self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll', ep_id)
 621             for ep_id, ep in episode_paths
 622         ]
 623         entries.reverse()
 624
 625         return {
 626             '_type': 'playlist',
 627             'id': show_id,
 628             'title': title,
 629             'entries': entries,
 630         }