]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/crunchyroll.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / crunchyroll.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import json
6 import xml.etree.ElementTree as etree
7 import zlib
8
9 from hashlib import sha1
10 from math import pow, sqrt, floor
11 from .common import InfoExtractor
12 from .vrv import VRVIE
13 from ..compat import (
14 compat_b64decode,
15 compat_etree_fromstring,
16 compat_urllib_parse_urlencode,
17 compat_urllib_request,
18 compat_urlparse,
19 )
20 from ..utils import (
21 ExtractorError,
22 bytes_to_intlist,
23 extract_attributes,
24 float_or_none,
25 intlist_to_bytes,
26 int_or_none,
27 lowercase_escape,
28 remove_end,
29 sanitized_Request,
30 unified_strdate,
31 urlencode_postdata,
32 xpath_text,
33 )
34 from ..aes import (
35 aes_cbc_decrypt,
36 )
37
38
39 class CrunchyrollBaseIE(InfoExtractor):
40 _LOGIN_URL = 'https://www.crunchyroll.com/login'
41 _LOGIN_FORM = 'login_form'
42 _NETRC_MACHINE = 'crunchyroll'
43
44 def _call_rpc_api(self, method, video_id, note=None, data=None):
45 data = data or {}
46 data['req'] = 'RpcApi' + method
47 data = compat_urllib_parse_urlencode(data).encode('utf-8')
48 return self._download_xml(
49 'https://www.crunchyroll.com/xml/',
50 video_id, note, fatal=False, data=data, headers={
51 'Content-Type': 'application/x-www-form-urlencoded',
52 })
53
54 def _login(self):
55 username, password = self._get_login_info()
56 if username is None:
57 return
58
59 self._download_webpage(
60 'https://www.crunchyroll.com/?a=formhandler',
61 None, 'Logging in', 'Wrong login info',
62 data=urlencode_postdata({
63 'formname': 'RpcApiUser_Login',
64 'next_url': 'https://www.crunchyroll.com/acct/membership',
65 'name': username,
66 'password': password,
67 }))
68
69 '''
70 login_page = self._download_webpage(
71 self._LOGIN_URL, None, 'Downloading login page')
72
73 def is_logged(webpage):
74 return '<title>Redirecting' in webpage
75
76 # Already logged in
77 if is_logged(login_page):
78 return
79
80 login_form_str = self._search_regex(
81 r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM,
82 login_page, 'login form', group='form')
83
84 post_url = extract_attributes(login_form_str).get('action')
85 if not post_url:
86 post_url = self._LOGIN_URL
87 elif not post_url.startswith('http'):
88 post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
89
90 login_form = self._form_hidden_inputs(self._LOGIN_FORM, login_page)
91
92 login_form.update({
93 'login_form[name]': username,
94 'login_form[password]': password,
95 })
96
97 response = self._download_webpage(
98 post_url, None, 'Logging in', 'Wrong login info',
99 data=urlencode_postdata(login_form),
100 headers={'Content-Type': 'application/x-www-form-urlencoded'})
101
102 # Successful login
103 if is_logged(response):
104 return
105
106 error = self._html_search_regex(
107 '(?s)<ul[^>]+class=["\']messages["\'][^>]*>(.+?)</ul>',
108 response, 'error message', default=None)
109 if error:
110 raise ExtractorError('Unable to login: %s' % error, expected=True)
111
112 raise ExtractorError('Unable to log in')
113 '''
114
115 def _real_initialize(self):
116 self._login()
117
118 def _download_webpage(self, url_or_request, *args, **kwargs):
119 request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
120 else sanitized_Request(url_or_request))
121 # Accept-Language must be set explicitly to accept any language to avoid issues
122 # similar to https://github.com/rg3/youtube-dl/issues/6797.
123 # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction
124 # should be imposed or not (from what I can see it just takes the first language
125 # ignoring the priority and requires it to correspond the IP). By the way this causes
126 # Crunchyroll to not work in georestriction cases in some browsers that don't place
127 # the locale lang first in header. However allowing any language seems to workaround the issue.
128 request.add_header('Accept-Language', '*')
129 return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs)
130
131 @staticmethod
132 def _add_skip_wall(url):
133 parsed_url = compat_urlparse.urlparse(url)
134 qs = compat_urlparse.parse_qs(parsed_url.query)
135 # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message:
136 # > This content may be inappropriate for some people.
137 # > Are you sure you want to continue?
138 # since it's not disabled by default in crunchyroll account's settings.
139 # See https://github.com/rg3/youtube-dl/issues/7202.
140 qs['skip_wall'] = ['1']
141 return compat_urlparse.urlunparse(
142 parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
143
144
145 class CrunchyrollIE(CrunchyrollBaseIE, VRVIE):
146 IE_NAME = 'crunchyroll'
147 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
148 _TESTS = [{
149 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
150 'info_dict': {
151 'id': '645513',
152 'ext': 'mp4',
153 'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
154 'description': 'md5:2d17137920c64f2f49981a7797d275ef',
155 'thumbnail': r're:^https?://.*\.jpg$',
156 'uploader': 'Yomiuri Telecasting Corporation (YTV)',
157 'upload_date': '20131013',
158 'url': 're:(?!.*&amp)',
159 },
160 'params': {
161 # rtmp
162 'skip_download': True,
163 },
164 }, {
165 'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
166 'info_dict': {
167 'id': '589804',
168 'ext': 'flv',
169 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
170 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
171 'thumbnail': r're:^https?://.*\.jpg$',
172 'uploader': 'Danny Choo Network',
173 'upload_date': '20120213',
174 },
175 'params': {
176 # rtmp
177 'skip_download': True,
178 },
179 'skip': 'Video gone',
180 }, {
181 'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
182 'info_dict': {
183 'id': '702409',
184 'ext': 'mp4',
185 'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
186 'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
187 'thumbnail': r're:^https?://.*\.jpg$',
188 'uploader': 'TV TOKYO',
189 'upload_date': '20160508',
190 },
191 'params': {
192 # m3u8 download
193 'skip_download': True,
194 },
195 }, {
196 'url': 'http://www.crunchyroll.com/konosuba-gods-blessing-on-this-wonderful-world/episode-1-give-me-deliverance-from-this-judicial-injustice-727589',
197 'info_dict': {
198 'id': '727589',
199 'ext': 'mp4',
200 'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance From This Judicial Injustice!",
201 'description': 'md5:cbcf05e528124b0f3a0a419fc805ea7d',
202 'thumbnail': r're:^https?://.*\.jpg$',
203 'uploader': 'Kadokawa Pictures Inc.',
204 'upload_date': '20170118',
205 'series': "KONOSUBA -God's blessing on this wonderful world!",
206 'season': "KONOSUBA -God's blessing on this wonderful world! 2",
207 'season_number': 2,
208 'episode': 'Give Me Deliverance From This Judicial Injustice!',
209 'episode_number': 1,
210 },
211 'params': {
212 # m3u8 download
213 'skip_download': True,
214 },
215 }, {
216 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
217 'only_matching': True,
218 }, {
219 # geo-restricted (US), 18+ maturity wall, non-premium available
220 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617',
221 'only_matching': True,
222 }, {
223 # A description with double quotes
224 'url': 'http://www.crunchyroll.com/11eyes/episode-1-piros-jszaka-red-night-535080',
225 'info_dict': {
226 'id': '535080',
227 'ext': 'mp4',
228 'title': '11eyes Episode 1 – Red Night ~ Piros éjszaka',
229 'description': 'Kakeru and Yuka are thrown into an alternate nightmarish world they call "Red Night".',
230 'uploader': 'Marvelous AQL Inc.',
231 'upload_date': '20091021',
232 },
233 'params': {
234 # Just test metadata extraction
235 'skip_download': True,
236 },
237 }, {
238 # make sure we can extract an uploader name that's not a link
239 'url': 'http://www.crunchyroll.com/hakuoki-reimeiroku/episode-1-dawn-of-the-divine-warriors-606899',
240 'info_dict': {
241 'id': '606899',
242 'ext': 'mp4',
243 'title': 'Hakuoki Reimeiroku Episode 1 – Dawn of the Divine Warriors',
244 'description': 'Ryunosuke was left to die, but Serizawa-san asked him a simple question "Do you want to live?"',
245 'uploader': 'Geneon Entertainment',
246 'upload_date': '20120717',
247 },
248 'params': {
249 # just test metadata extraction
250 'skip_download': True,
251 },
252 }, {
253 # A video with a vastly different season name compared to the series name
254 'url': 'http://www.crunchyroll.com/nyarko-san-another-crawling-chaos/episode-1-test-590532',
255 'info_dict': {
256 'id': '590532',
257 'ext': 'mp4',
258 'title': 'Haiyoru! Nyaruani (ONA) Episode 1 – Test',
259 'description': 'Mahiro and Nyaruko talk about official certification.',
260 'uploader': 'TV TOKYO',
261 'upload_date': '20120305',
262 'series': 'Nyarko-san: Another Crawling Chaos',
263 'season': 'Haiyoru! Nyaruani (ONA)',
264 },
265 'params': {
266 # Just test metadata extraction
267 'skip_download': True,
268 },
269 }, {
270 'url': 'http://www.crunchyroll.com/media-723735',
271 'only_matching': True,
272 }]
273
274 _FORMAT_IDS = {
275 '360': ('60', '106'),
276 '480': ('61', '106'),
277 '720': ('62', '106'),
278 '1080': ('80', '108'),
279 }
280
281 def _decrypt_subtitles(self, data, iv, id):
282 data = bytes_to_intlist(compat_b64decode(data))
283 iv = bytes_to_intlist(compat_b64decode(iv))
284 id = int(id)
285
286 def obfuscate_key_aux(count, modulo, start):
287 output = list(start)
288 for _ in range(count):
289 output.append(output[-1] + output[-2])
290 # cut off start values
291 output = output[2:]
292 output = list(map(lambda x: x % modulo + 33, output))
293 return output
294
295 def obfuscate_key(key):
296 num1 = int(floor(pow(2, 25) * sqrt(6.9)))
297 num2 = (num1 ^ key) << 5
298 num3 = key ^ num1
299 num4 = num3 ^ (num3 >> 3) ^ num2
300 prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
301 shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
302 # Extend 160 Bit hash to 256 Bit
303 return shaHash + [0] * 12
304
305 key = obfuscate_key(id)
306
307 decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
308 return zlib.decompress(decrypted_data)
309
310 def _convert_subtitles_to_srt(self, sub_root):
311 output = ''
312
313 for i, event in enumerate(sub_root.findall('./events/event'), 1):
314 start = event.attrib['start'].replace('.', ',')
315 end = event.attrib['end'].replace('.', ',')
316 text = event.attrib['text'].replace('\\N', '\n')
317 output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
318 return output
319
320 def _convert_subtitles_to_ass(self, sub_root):
321 output = ''
322
323 def ass_bool(strvalue):
324 assvalue = '0'
325 if strvalue == '1':
326 assvalue = '-1'
327 return assvalue
328
329 output = '[Script Info]\n'
330 output += 'Title: %s\n' % sub_root.attrib['title']
331 output += 'ScriptType: v4.00+\n'
332 output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style']
333 output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x']
334 output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y']
335 output += """
336 [V4+ Styles]
337 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
338 """
339 for style in sub_root.findall('./styles/style'):
340 output += 'Style: ' + style.attrib['name']
341 output += ',' + style.attrib['font_name']
342 output += ',' + style.attrib['font_size']
343 output += ',' + style.attrib['primary_colour']
344 output += ',' + style.attrib['secondary_colour']
345 output += ',' + style.attrib['outline_colour']
346 output += ',' + style.attrib['back_colour']
347 output += ',' + ass_bool(style.attrib['bold'])
348 output += ',' + ass_bool(style.attrib['italic'])
349 output += ',' + ass_bool(style.attrib['underline'])
350 output += ',' + ass_bool(style.attrib['strikeout'])
351 output += ',' + style.attrib['scale_x']
352 output += ',' + style.attrib['scale_y']
353 output += ',' + style.attrib['spacing']
354 output += ',' + style.attrib['angle']
355 output += ',' + style.attrib['border_style']
356 output += ',' + style.attrib['outline']
357 output += ',' + style.attrib['shadow']
358 output += ',' + style.attrib['alignment']
359 output += ',' + style.attrib['margin_l']
360 output += ',' + style.attrib['margin_r']
361 output += ',' + style.attrib['margin_v']
362 output += ',' + style.attrib['encoding']
363 output += '\n'
364
365 output += """
366 [Events]
367 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
368 """
369 for event in sub_root.findall('./events/event'):
370 output += 'Dialogue: 0'
371 output += ',' + event.attrib['start']
372 output += ',' + event.attrib['end']
373 output += ',' + event.attrib['style']
374 output += ',' + event.attrib['name']
375 output += ',' + event.attrib['margin_l']
376 output += ',' + event.attrib['margin_r']
377 output += ',' + event.attrib['margin_v']
378 output += ',' + event.attrib['effect']
379 output += ',' + event.attrib['text']
380 output += '\n'
381
382 return output
383
384 def _extract_subtitles(self, subtitle):
385 sub_root = compat_etree_fromstring(subtitle)
386 return [{
387 'ext': 'srt',
388 'data': self._convert_subtitles_to_srt(sub_root),
389 }, {
390 'ext': 'ass',
391 'data': self._convert_subtitles_to_ass(sub_root),
392 }]
393
394 def _get_subtitles(self, video_id, webpage):
395 subtitles = {}
396 for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage):
397 sub_doc = self._call_rpc_api(
398 'Subtitle_GetXml', video_id,
399 'Downloading subtitles for ' + sub_name, data={
400 'subtitle_script_id': sub_id,
401 })
402 if not isinstance(sub_doc, etree.Element):
403 continue
404 sid = sub_doc.get('id')
405 iv = xpath_text(sub_doc, 'iv', 'subtitle iv')
406 data = xpath_text(sub_doc, 'data', 'subtitle data')
407 if not sid or not iv or not data:
408 continue
409 subtitle = self._decrypt_subtitles(data, iv, sid).decode('utf-8')
410 lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
411 if not lang_code:
412 continue
413 subtitles[lang_code] = self._extract_subtitles(subtitle)
414 return subtitles
415
416 def _real_extract(self, url):
417 mobj = re.match(self._VALID_URL, url)
418 video_id = mobj.group('video_id')
419
420 if mobj.group('prefix') == 'm':
421 mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
422 webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
423 else:
424 webpage_url = 'http://www.' + mobj.group('url')
425
426 webpage = self._download_webpage(
427 self._add_skip_wall(webpage_url), video_id,
428 headers=self.geo_verification_headers())
429 note_m = self._html_search_regex(
430 r'<div class="showmedia-trailer-notice">(.+?)</div>',
431 webpage, 'trailer-notice', default='')
432 if note_m:
433 raise ExtractorError(note_m)
434
435 mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
436 if mobj:
437 msg = json.loads(mobj.group('msg'))
438 if msg.get('type') == 'error':
439 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
440
441 if 'To view this, please log in to verify you are 18 or older.' in webpage:
442 self.raise_login_required()
443
444 media = self._parse_json(self._search_regex(
445 r'vilos\.config\.media\s*=\s*({.+?});',
446 webpage, 'vilos media', default='{}'), video_id)
447 media_metadata = media.get('metadata') or {}
448
449 language = self._search_regex(
450 r'(?:vilos\.config\.player\.language|LOCALE)\s*=\s*(["\'])(?P<lang>(?:(?!\1).)+)\1',
451 webpage, 'language', default=None, group='lang')
452
453 video_title = self._html_search_regex(
454 r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
455 webpage, 'video_title')
456 video_title = re.sub(r' {2,}', ' ', video_title)
457 video_description = (self._parse_json(self._html_search_regex(
458 r'<script[^>]*>\s*.+?\[media_id=%s\].+?({.+?"description"\s*:.+?})\);' % video_id,
459 webpage, 'description', default='{}'), video_id) or media_metadata).get('description')
460 if video_description:
461 video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
462 video_upload_date = self._html_search_regex(
463 [r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
464 webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
465 if video_upload_date:
466 video_upload_date = unified_strdate(video_upload_date)
467 video_uploader = self._html_search_regex(
468 # try looking for both an uploader that's a link and one that's not
469 [r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', r'<div>\s*Publisher:\s*<span>\s*(.+?)\s*</span>\s*</div>'],
470 webpage, 'video_uploader', fatal=False)
471
472 formats = []
473 for stream in media.get('streams', []):
474 audio_lang = stream.get('audio_lang')
475 hardsub_lang = stream.get('hardsub_lang')
476 vrv_formats = self._extract_vrv_formats(
477 stream.get('url'), video_id, stream.get('format'),
478 audio_lang, hardsub_lang)
479 for f in vrv_formats:
480 if not hardsub_lang:
481 f['preference'] = 1
482 language_preference = 0
483 if audio_lang == language:
484 language_preference += 1
485 if hardsub_lang == language:
486 language_preference += 1
487 if language_preference:
488 f['language_preference'] = language_preference
489 formats.extend(vrv_formats)
490 if not formats:
491 available_fmts = []
492 for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
493 attrs = extract_attributes(a)
494 href = attrs.get('href')
495 if href and '/freetrial' in href:
496 continue
497 available_fmts.append(fmt)
498 if not available_fmts:
499 for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
500 available_fmts = re.findall(p, webpage)
501 if available_fmts:
502 break
503 if not available_fmts:
504 available_fmts = self._FORMAT_IDS.keys()
505 video_encode_ids = []
506
507 for fmt in available_fmts:
508 stream_quality, stream_format = self._FORMAT_IDS[fmt]
509 video_format = fmt + 'p'
510 stream_infos = []
511 streamdata = self._call_rpc_api(
512 'VideoPlayer_GetStandardConfig', video_id,
513 'Downloading media info for %s' % video_format, data={
514 'media_id': video_id,
515 'video_format': stream_format,
516 'video_quality': stream_quality,
517 'current_page': url,
518 })
519 if isinstance(streamdata, etree.Element):
520 stream_info = streamdata.find('./{default}preload/stream_info')
521 if stream_info is not None:
522 stream_infos.append(stream_info)
523 stream_info = self._call_rpc_api(
524 'VideoEncode_GetStreamInfo', video_id,
525 'Downloading stream info for %s' % video_format, data={
526 'media_id': video_id,
527 'video_format': stream_format,
528 'video_encode_quality': stream_quality,
529 })
530 if isinstance(stream_info, etree.Element):
531 stream_infos.append(stream_info)
532 for stream_info in stream_infos:
533 video_encode_id = xpath_text(stream_info, './video_encode_id')
534 if video_encode_id in video_encode_ids:
535 continue
536 video_encode_ids.append(video_encode_id)
537
538 video_file = xpath_text(stream_info, './file')
539 if not video_file:
540 continue
541 if video_file.startswith('http'):
542 formats.extend(self._extract_m3u8_formats(
543 video_file, video_id, 'mp4', entry_protocol='m3u8_native',
544 m3u8_id='hls', fatal=False))
545 continue
546
547 video_url = xpath_text(stream_info, './host')
548 if not video_url:
549 continue
550 metadata = stream_info.find('./metadata')
551 format_info = {
552 'format': video_format,
553 'height': int_or_none(xpath_text(metadata, './height')),
554 'width': int_or_none(xpath_text(metadata, './width')),
555 }
556
557 if '.fplive.net/' in video_url:
558 video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
559 parsed_video_url = compat_urlparse.urlparse(video_url)
560 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
561 netloc='v.lvlt.crcdn.net',
562 path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
563 if self._is_valid_url(direct_video_url, video_id, video_format):
564 format_info.update({
565 'format_id': 'http-' + video_format,
566 'url': direct_video_url,
567 })
568 formats.append(format_info)
569 continue
570
571 format_info.update({
572 'format_id': 'rtmp-' + video_format,
573 'url': video_url,
574 'play_path': video_file,
575 'ext': 'flv',
576 })
577 formats.append(format_info)
578 self._sort_formats(formats, ('preference', 'language_preference', 'height', 'width', 'tbr', 'fps'))
579
580 metadata = self._call_rpc_api(
581 'VideoPlayer_GetMediaMetadata', video_id,
582 note='Downloading media info', data={
583 'media_id': video_id,
584 })
585
586 subtitles = {}
587 for subtitle in media.get('subtitles', []):
588 subtitle_url = subtitle.get('url')
589 if not subtitle_url:
590 continue
591 subtitles.setdefault(subtitle.get('language', 'enUS'), []).append({
592 'url': subtitle_url,
593 'ext': subtitle.get('format', 'ass'),
594 })
595 if not subtitles:
596 subtitles = self.extract_subtitles(video_id, webpage)
597
598 # webpage provide more accurate data than series_title from XML
599 series = self._html_search_regex(
600 r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
601 webpage, 'series', fatal=False)
602
603 season = episode = episode_number = duration = thumbnail = None
604
605 if isinstance(metadata, etree.Element):
606 season = xpath_text(metadata, 'series_title')
607 episode = xpath_text(metadata, 'episode_title')
608 episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
609 duration = float_or_none(media_metadata.get('duration'), 1000)
610 thumbnail = xpath_text(metadata, 'episode_image_url')
611
612 if not episode:
613 episode = media_metadata.get('title')
614 if not episode_number:
615 episode_number = int_or_none(media_metadata.get('episode_number'))
616 if not thumbnail:
617 thumbnail = media_metadata.get('thumbnail', {}).get('url')
618
619 season_number = int_or_none(self._search_regex(
620 r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
621 webpage, 'season number', default=None))
622
623 return {
624 'id': video_id,
625 'title': video_title,
626 'description': video_description,
627 'duration': duration,
628 'thumbnail': thumbnail,
629 'uploader': video_uploader,
630 'upload_date': video_upload_date,
631 'series': series,
632 'season': season,
633 'season_number': season_number,
634 'episode': episode,
635 'episode_number': episode_number,
636 'subtitles': subtitles,
637 'formats': formats,
638 }
639
640
641 class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
642 IE_NAME = 'crunchyroll:playlist'
643 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login|media-\d+))(?P<id>[\w\-]+))/?(?:\?|$)'
644
645 _TESTS = [{
646 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
647 'info_dict': {
648 'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
649 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
650 },
651 'playlist_count': 13,
652 }, {
653 # geo-restricted (US), 18+ maturity wall, non-premium available
654 'url': 'http://www.crunchyroll.com/cosplay-complex-ova',
655 'info_dict': {
656 'id': 'cosplay-complex-ova',
657 'title': 'Cosplay Complex OVA'
658 },
659 'playlist_count': 3,
660 'skip': 'Georestricted',
661 }, {
662 # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14
663 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1',
664 'only_matching': True,
665 }]
666
667 def _real_extract(self, url):
668 show_id = self._match_id(url)
669
670 webpage = self._download_webpage(
671 self._add_skip_wall(url), show_id,
672 headers=self.geo_verification_headers())
673 title = self._html_search_regex(
674 r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
675 webpage, 'title')
676 episode_paths = re.findall(
677 r'(?s)<li id="showview_videos_media_(\d+)"[^>]+>.*?<a href="([^"]+)"',
678 webpage)
679 entries = [
680 self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll', ep_id)
681 for ep_id, ep in episode_paths
682 ]
683 entries.reverse()
684
685 return {
686 '_type': 'playlist',
687 'id': show_id,
688 'title': title,
689 'entries': entries,
690 }