2 from __future__ import unicode_literals
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,
36 class CrunchyrollBaseIE(InfoExtractor):
37 _LOGIN_URL = 'https://www.crunchyroll.com/login'
38 _LOGIN_FORM = 'login_form'
39 _NETRC_MACHINE = 'crunchyroll'
42 (username, password) = self._get_login_info()
46 login_page = self._download_webpage(
47 self._LOGIN_URL, None, 'Downloading login page')
49 def is_logged(webpage):
50 return '<title>Redirecting' in webpage
53 if is_logged(login_page):
56 login_form_str = self._search_regex(
57 r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM,
58 login_page, 'login form', group='form')
60 post_url = extract_attributes(login_form_str).get('action')
62 post_url = self._LOGIN_URL
63 elif not post_url.startswith('http'):
64 post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
66 login_form = self._form_hidden_inputs(self._LOGIN_FORM, login_page)
69 'login_form[name]': username,
70 'login_form[password]': password,
73 response = self._download_webpage(
74 post_url, None, 'Logging in', 'Wrong login info',
75 data=urlencode_postdata(login_form),
76 headers={'Content-Type': 'application/x-www-form-urlencoded'})
79 if is_logged(response):
82 error = self._html_search_regex(
83 '(?s)<ul[^>]+class=["\']messages["\'][^>]*>(.+?)</ul>',
84 response, 'error message', default=None)
86 raise ExtractorError('Unable to login: %s' % error, expected=True)
88 raise ExtractorError('Unable to log in')
90 def _real_initialize(self):
93 def _download_webpage(self, url_or_request, *args, **kwargs):
94 request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
95 else sanitized_Request(url_or_request))
96 # Accept-Language must be set explicitly to accept any language to avoid issues
97 # similar to https://github.com/rg3/youtube-dl/issues/6797.
98 # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction
99 # should be imposed or not (from what I can see it just takes the first language
100 # ignoring the priority and requires it to correspond the IP). By the way this causes
101 # Crunchyroll to not work in georestriction cases in some browsers that don't place
102 # the locale lang first in header. However allowing any language seems to workaround the issue.
103 request.add_header('Accept-Language', '*')
104 return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs)
107 def _add_skip_wall(url):
108 parsed_url = compat_urlparse.urlparse(url)
109 qs = compat_urlparse.parse_qs(parsed_url.query)
110 # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message:
111 # > This content may be inappropriate for some people.
112 # > Are you sure you want to continue?
113 # since it's not disabled by default in crunchyroll account's settings.
114 # See https://github.com/rg3/youtube-dl/issues/7202.
115 qs['skip_wall'] = ['1']
116 return compat_urlparse.urlunparse(
117 parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
120 class CrunchyrollIE(CrunchyrollBaseIE):
121 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
123 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
127 'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
128 'description': 'md5:2d17137920c64f2f49981a7797d275ef',
129 'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
130 'uploader': 'Yomiuri Telecasting Corporation (YTV)',
131 'upload_date': '20131013',
132 'url': 're:(?!.*&)',
136 'skip_download': True,
139 'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
143 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
144 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
145 'thumbnail': 're:^https?://.*\.jpg$',
146 'uploader': 'Danny Choo Network',
147 'upload_date': '20120213',
151 'skip_download': True,
154 'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
158 'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
159 'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
160 'thumbnail': 're:^https?://.*\.jpg$',
161 'uploader': 'TV TOKYO',
162 'upload_date': '20160508',
166 'skip_download': True,
169 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
170 'only_matching': True,
172 # geo-restricted (US), 18+ maturity wall, non-premium available
173 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
174 'only_matching': True,
178 '360': ('60', '106'),
179 '480': ('61', '106'),
180 '720': ('62', '106'),
181 '1080': ('80', '108'),
184 def _decrypt_subtitles(self, data, iv, id):
185 data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
186 iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))
189 def obfuscate_key_aux(count, modulo, start):
191 for _ in range(count):
192 output.append(output[-1] + output[-2])
193 # cut off start values
195 output = list(map(lambda x: x % modulo + 33, output))
198 def obfuscate_key(key):
199 num1 = int(floor(pow(2, 25) * sqrt(6.9)))
200 num2 = (num1 ^ key) << 5
202 num4 = num3 ^ (num3 >> 3) ^ num2
203 prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
204 shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
205 # Extend 160 Bit hash to 256 Bit
206 return shaHash + [0] * 12
208 key = obfuscate_key(id)
210 decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
211 return zlib.decompress(decrypted_data)
213 def _convert_subtitles_to_srt(self, sub_root):
216 for i, event in enumerate(sub_root.findall('./events/event'), 1):
217 start = event.attrib['start'].replace('.', ',')
218 end = event.attrib['end'].replace('.', ',')
219 text = event.attrib['text'].replace('\\N', '\n')
220 output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
223 def _convert_subtitles_to_ass(self, sub_root):
226 def ass_bool(strvalue):
232 output = '[Script Info]\n'
233 output += 'Title: %s\n' % sub_root.attrib['title']
234 output += 'ScriptType: v4.00+\n'
235 output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
236 output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
237 output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']
238 output += """ScaledBorderAndShadow: yes
241 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
243 for style in sub_root.findall('./styles/style'):
244 output += 'Style: ' + style.attrib['name']
245 output += ',' + style.attrib['font_name']
246 output += ',' + style.attrib['font_size']
247 output += ',' + style.attrib['primary_colour']
248 output += ',' + style.attrib['secondary_colour']
249 output += ',' + style.attrib['outline_colour']
250 output += ',' + style.attrib['back_colour']
251 output += ',' + ass_bool(style.attrib['bold'])
252 output += ',' + ass_bool(style.attrib['italic'])
253 output += ',' + ass_bool(style.attrib['underline'])
254 output += ',' + ass_bool(style.attrib['strikeout'])
255 output += ',' + style.attrib['scale_x']
256 output += ',' + style.attrib['scale_y']
257 output += ',' + style.attrib['spacing']
258 output += ',' + style.attrib['angle']
259 output += ',' + style.attrib['border_style']
260 output += ',' + style.attrib['outline']
261 output += ',' + style.attrib['shadow']
262 output += ',' + style.attrib['alignment']
263 output += ',' + style.attrib['margin_l']
264 output += ',' + style.attrib['margin_r']
265 output += ',' + style.attrib['margin_v']
266 output += ',' + style.attrib['encoding']
271 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
273 for event in sub_root.findall('./events/event'):
274 output += 'Dialogue: 0'
275 output += ',' + event.attrib['start']
276 output += ',' + event.attrib['end']
277 output += ',' + event.attrib['style']
278 output += ',' + event.attrib['name']
279 output += ',' + event.attrib['margin_l']
280 output += ',' + event.attrib['margin_r']
281 output += ',' + event.attrib['margin_v']
282 output += ',' + event.attrib['effect']
283 output += ',' + event.attrib['text']
288 def _extract_subtitles(self, subtitle):
289 sub_root = compat_etree_fromstring(subtitle)
292 'data': self._convert_subtitles_to_srt(sub_root),
295 'data': self._convert_subtitles_to_ass(sub_root),
298 def _get_subtitles(self, video_id, webpage):
300 for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage):
301 sub_page = self._download_webpage(
302 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id,
303 video_id, note='Downloading subtitles for ' + sub_name)
304 id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
305 iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
306 data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
307 if not id or not iv or not data:
309 subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
310 lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
313 subtitles[lang_code] = self._extract_subtitles(subtitle)
316 def _real_extract(self, url):
317 mobj = re.match(self._VALID_URL, url)
318 video_id = mobj.group('video_id')
320 if mobj.group('prefix') == 'm':
321 mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
322 webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
324 webpage_url = 'http://www.' + mobj.group('url')
326 webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
327 note_m = self._html_search_regex(
328 r'<div class="showmedia-trailer-notice">(.+?)</div>',
329 webpage, 'trailer-notice', default='')
331 raise ExtractorError(note_m)
333 mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
335 msg = json.loads(mobj.group('msg'))
336 if msg.get('type') == 'error':
337 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
339 if 'To view this, please log in to verify you are 18 or older.' in webpage:
340 self.raise_login_required()
342 video_title = self._html_search_regex(
343 r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
344 webpage, 'video_title')
345 video_title = re.sub(r' {2,}', ' ', video_title)
346 video_description = self._html_search_regex(
347 r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id,
348 webpage, 'description', default=None)
349 if video_description:
350 video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
351 video_upload_date = self._html_search_regex(
352 [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
353 webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
354 if video_upload_date:
355 video_upload_date = unified_strdate(video_upload_date)
356 video_uploader = self._html_search_regex(
357 r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
358 'video_uploader', fatal=False)
361 for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
362 attrs = extract_attributes(a)
363 href = attrs.get('href')
364 if href and '/freetrial' in href:
366 available_fmts.append(fmt)
367 if not available_fmts:
368 for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
369 available_fmts = re.findall(p, webpage)
372 video_encode_ids = []
374 for fmt in available_fmts:
375 stream_quality, stream_format = self._FORMAT_IDS[fmt]
376 video_format = fmt + 'p'
377 streamdata_req = sanitized_Request(
378 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
379 % (video_id, stream_format, stream_quality),
380 compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8'))
381 streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
382 streamdata = self._download_xml(
383 streamdata_req, video_id,
384 note='Downloading media info for %s' % video_format)
385 stream_info = streamdata.find('./{default}preload/stream_info')
386 video_encode_id = xpath_text(stream_info, './video_encode_id')
387 if video_encode_id in video_encode_ids:
389 video_encode_ids.append(video_encode_id)
391 video_file = xpath_text(stream_info, './file')
394 if video_file.startswith('http'):
395 formats.extend(self._extract_m3u8_formats(
396 video_file, video_id, 'mp4', entry_protocol='m3u8_native',
397 m3u8_id='hls', fatal=False))
400 video_url = xpath_text(stream_info, './host')
403 metadata = stream_info.find('./metadata')
405 'format': video_format,
406 'format_id': video_format,
407 'height': int_or_none(xpath_text(metadata, './height')),
408 'width': int_or_none(xpath_text(metadata, './width')),
411 if '.fplive.net/' in video_url:
412 video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
413 parsed_video_url = compat_urlparse.urlparse(video_url)
414 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
415 netloc='v.lvlt.crcdn.net',
416 path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
417 if self._is_valid_url(direct_video_url, video_id, video_format):
419 'url': direct_video_url,
421 formats.append(format_info)
426 'play_path': video_file,
429 formats.append(format_info)
430 self._sort_formats(formats)
432 metadata = self._download_xml(
433 'http://www.crunchyroll.com/xml', video_id,
434 note='Downloading media info', query={
435 'req': 'RpcApiVideoPlayer_GetMediaMetadata',
436 'media_id': video_id,
439 subtitles = self.extract_subtitles(video_id, webpage)
443 'title': video_title,
444 'description': video_description,
445 'thumbnail': xpath_text(metadata, 'episode_image_url'),
446 'uploader': video_uploader,
447 'upload_date': video_upload_date,
448 'series': xpath_text(metadata, 'series_title'),
449 'episode': xpath_text(metadata, 'episode_title'),
450 'episode_number': int_or_none(xpath_text(metadata, 'episode_number')),
451 'subtitles': subtitles,
456 class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
457 IE_NAME = 'crunchyroll:playlist'
458 _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\-]+))/?(?:\?|$)'
461 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
463 'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
464 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
466 'playlist_count': 13,
468 # geo-restricted (US), 18+ maturity wall, non-premium available
469 'url': 'http://www.crunchyroll.com/cosplay-complex-ova',
471 'id': 'cosplay-complex-ova',
472 'title': 'Cosplay Complex OVA'
475 'skip': 'Georestricted',
477 # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14
478 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1',
479 'only_matching': True,
482 def _real_extract(self, url):
483 show_id = self._match_id(url)
485 webpage = self._download_webpage(self._add_skip_wall(url), show_id)
486 title = self._html_search_regex(
487 r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
489 episode_paths = re.findall(
490 r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
493 self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
494 for ep in episode_paths