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