youtube_dl/extractor/crunchyroll.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6 import base64
   7 import zlib
   8 import xml.etree.ElementTree
   9
  10 from hashlib import sha1
  11 from math import pow, sqrt, floor
  12 from .common import InfoExtractor
  13 from ..compat import (
  14     compat_urllib_parse,
  15     compat_urllib_parse_unquote,
  16     compat_urllib_request,
  17     compat_urlparse,
  18 )
  19 from ..utils import (
  20     ExtractorError,
  21     bytes_to_intlist,
  22     intlist_to_bytes,
  23     int_or_none,
  24     remove_end,
  25     unified_strdate,
  26     urlencode_postdata,
  27     xpath_text,
  28 )
  29 from ..aes import (
  30     aes_cbc_decrypt,
  31 )
  32
  33
  34 class CrunchyrollIE(InfoExtractor):
  35     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
  36     _NETRC_MACHINE = 'crunchyroll'
  37     _TESTS = [{
  38         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
  39         'info_dict': {
  40             'id': '645513',
  41             'ext': 'flv',
  42             'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
  43             'description': 'md5:2d17137920c64f2f49981a7797d275ef',
  44             'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
  45             'uploader': 'Yomiuri Telecasting Corporation (YTV)',
  46             'upload_date': '20131013',
  47             'url': 're:(?!.*&amp)',
  48         },
  49         'params': {
  50             # rtmp
  51             'skip_download': True,
  52         },
  53     }, {
  54         'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
  55         'info_dict': {
  56             'id': '589804',
  57             'ext': 'flv',
  58             'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
  59             'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
  60             'thumbnail': 're:^https?://.*\.jpg$',
  61             'uploader': 'Danny Choo Network',
  62             'upload_date': '20120213',
  63         },
  64         'params': {
  65             # rtmp
  66             'skip_download': True,
  67         },
  68
  69     }, {
  70         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
  71         'only_matching': True,
  72     }]
  73
  74     _FORMAT_IDS = {
  75         '360': ('60', '106'),
  76         '480': ('61', '106'),
  77         '720': ('62', '106'),
  78         '1080': ('80', '108'),
  79     }
  80
  81     def _login(self):
  82         (username, password) = self._get_login_info()
  83         if username is None:
  84             return
  85         self.report_login()
  86         login_url = 'https://www.crunchyroll.com/?a=formhandler'
  87         data = urlencode_postdata({
  88             'formname': 'RpcApiUser_Login',
  89             'name': username,
  90             'password': password,
  91         })
  92         login_request = compat_urllib_request.Request(login_url, data)
  93         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  94         self._download_webpage(login_request, None, False, 'Wrong login info')
  95
  96     def _real_initialize(self):
  97         self._login()
  98
  99     def _decrypt_subtitles(self, data, iv, id):
 100         data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
 101         iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))
 102         id = int(id)
 103
 104         def obfuscate_key_aux(count, modulo, start):
 105             output = list(start)
 106             for _ in range(count):
 107                 output.append(output[-1] + output[-2])
 108             # cut off start values
 109             output = output[2:]
 110             output = list(map(lambda x: x % modulo + 33, output))
 111             return output
 112
 113         def obfuscate_key(key):
 114             num1 = int(floor(pow(2, 25) * sqrt(6.9)))
 115             num2 = (num1 ^ key) << 5
 116             num3 = key ^ num1
 117             num4 = num3 ^ (num3 >> 3) ^ num2
 118             prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
 119             shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
 120             # Extend 160 Bit hash to 256 Bit
 121             return shaHash + [0] * 12
 122
 123         key = obfuscate_key(id)
 124
 125         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
 126         return zlib.decompress(decrypted_data)
 127
 128     def _convert_subtitles_to_srt(self, sub_root):
 129         output = ''
 130
 131         for i, event in enumerate(sub_root.findall('./events/event'), 1):
 132             start = event.attrib['start'].replace('.', ',')
 133             end = event.attrib['end'].replace('.', ',')
 134             text = event.attrib['text'].replace('\\N', '\n')
 135             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
 136         return output
 137
 138     def _convert_subtitles_to_ass(self, sub_root):
 139         output = ''
 140
 141         def ass_bool(strvalue):
 142             assvalue = '0'
 143             if strvalue == '1':
 144                 assvalue = '-1'
 145             return assvalue
 146
 147         output = '[Script Info]\n'
 148         output += 'Title: %s\n' % sub_root.attrib["title"]
 149         output += 'ScriptType: v4.00+\n'
 150         output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
 151         output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
 152         output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
 153         output += """ScaledBorderAndShadow: yes
 154
 155 [V4+ Styles]
 156 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 157 """
 158         for style in sub_root.findall('./styles/style'):
 159             output += 'Style: ' + style.attrib["name"]
 160             output += ',' + style.attrib["font_name"]
 161             output += ',' + style.attrib["font_size"]
 162             output += ',' + style.attrib["primary_colour"]
 163             output += ',' + style.attrib["secondary_colour"]
 164             output += ',' + style.attrib["outline_colour"]
 165             output += ',' + style.attrib["back_colour"]
 166             output += ',' + ass_bool(style.attrib["bold"])
 167             output += ',' + ass_bool(style.attrib["italic"])
 168             output += ',' + ass_bool(style.attrib["underline"])
 169             output += ',' + ass_bool(style.attrib["strikeout"])
 170             output += ',' + style.attrib["scale_x"]
 171             output += ',' + style.attrib["scale_y"]
 172             output += ',' + style.attrib["spacing"]
 173             output += ',' + style.attrib["angle"]
 174             output += ',' + style.attrib["border_style"]
 175             output += ',' + style.attrib["outline"]
 176             output += ',' + style.attrib["shadow"]
 177             output += ',' + style.attrib["alignment"]
 178             output += ',' + style.attrib["margin_l"]
 179             output += ',' + style.attrib["margin_r"]
 180             output += ',' + style.attrib["margin_v"]
 181             output += ',' + style.attrib["encoding"]
 182             output += '\n'
 183
 184         output += """
 185 [Events]
 186 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 187 """
 188         for event in sub_root.findall('./events/event'):
 189             output += 'Dialogue: 0'
 190             output += ',' + event.attrib["start"]
 191             output += ',' + event.attrib["end"]
 192             output += ',' + event.attrib["style"]
 193             output += ',' + event.attrib["name"]
 194             output += ',' + event.attrib["margin_l"]
 195             output += ',' + event.attrib["margin_r"]
 196             output += ',' + event.attrib["margin_v"]
 197             output += ',' + event.attrib["effect"]
 198             output += ',' + event.attrib["text"]
 199             output += '\n'
 200
 201         return output
 202
 203     def _extract_subtitles(self, subtitle):
 204         sub_root = xml.etree.ElementTree.fromstring(subtitle)
 205         return [{
 206             'ext': 'srt',
 207             'data': self._convert_subtitles_to_srt(sub_root),
 208         }, {
 209             'ext': 'ass',
 210             'data': self._convert_subtitles_to_ass(sub_root),
 211         }]
 212
 213     def _get_subtitles(self, video_id, webpage):
 214         subtitles = {}
 215         for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
 216             sub_page = self._download_webpage(
 217                 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id,
 218                 video_id, note='Downloading subtitles for ' + sub_name)
 219             id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
 220             iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
 221             data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
 222             if not id or not iv or not data:
 223                 continue
 224             subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
 225             lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
 226             if not lang_code:
 227                 continue
 228             subtitles[lang_code] = self._extract_subtitles(subtitle)
 229         return subtitles
 230
 231     def _real_extract(self, url):
 232         mobj = re.match(self._VALID_URL, url)
 233         video_id = mobj.group('video_id')
 234
 235         if mobj.group('prefix') == 'm':
 236             mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
 237             webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
 238         else:
 239             webpage_url = 'http://www.' + mobj.group('url')
 240
 241         webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
 242         note_m = self._html_search_regex(
 243             r'<div class="showmedia-trailer-notice">(.+?)</div>',
 244             webpage, 'trailer-notice', default='')
 245         if note_m:
 246             raise ExtractorError(note_m)
 247
 248         mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
 249         if mobj:
 250             msg = json.loads(mobj.group('msg'))
 251             if msg.get('type') == 'error':
 252                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
 253
 254         if 'To view this, please log in to verify you are 18 or older.' in webpage:
 255             self.raise_login_required()
 256
 257         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
 258         video_title = re.sub(r' {2,}', ' ', video_title)
 259         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
 260         if not video_description:
 261             video_description = None
 262         video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
 263         if video_upload_date:
 264             video_upload_date = unified_strdate(video_upload_date)
 265         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 266
 267         playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
 268         playerdata_req = compat_urllib_request.Request(playerdata_url)
 269         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
 270         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
 271         playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
 272
 273         stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
 274         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 275
 276         formats = []
 277         for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
 278             stream_quality, stream_format = self._FORMAT_IDS[fmt]
 279             video_format = fmt + 'p'
 280             streamdata_req = compat_urllib_request.Request(
 281                 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
 282                 % (stream_id, stream_format, stream_quality),
 283                 compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8'))
 284             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
 285             streamdata = self._download_xml(
 286                 streamdata_req, video_id,
 287                 note='Downloading media info for %s' % video_format)
 288             stream_info = streamdata.find('./{default}preload/stream_info')
 289             video_url = stream_info.find('./host').text
 290             video_play_path = stream_info.find('./file').text
 291             metadata = stream_info.find('./metadata')
 292             format_info = {
 293                 'format': video_format,
 294                 'format_id': video_format,
 295                 'height': int_or_none(xpath_text(metadata, './height')),
 296                 'width': int_or_none(xpath_text(metadata, './width')),
 297             }
 298
 299             if '.fplive.net/' in video_url:
 300                 video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
 301                 parsed_video_url = compat_urlparse.urlparse(video_url)
 302                 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
 303                     netloc='v.lvlt.crcdn.net',
 304                     path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
 305                 if self._is_valid_url(direct_video_url, video_id, video_format):
 306                     format_info.update({
 307                         'url': direct_video_url,
 308                     })
 309                     formats.append(format_info)
 310                     continue
 311
 312             format_info.update({
 313                 'url': video_url,
 314                 'play_path': video_play_path,
 315                 'ext': 'flv',
 316             })
 317             formats.append(format_info)
 318
 319         subtitles = self.extract_subtitles(video_id, webpage)
 320
 321         return {
 322             'id': video_id,
 323             'title': video_title,
 324             'description': video_description,
 325             'thumbnail': video_thumbnail,
 326             'uploader': video_uploader,
 327             'upload_date': video_upload_date,
 328             'subtitles': subtitles,
 329             'formats': formats,
 330         }
 331
 332
 333 class CrunchyrollShowPlaylistIE(InfoExtractor):
 334     IE_NAME = "crunchyroll:playlist"
 335     _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\-]+))/?$'
 336
 337     _TESTS = [{
 338         'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 339         'info_dict': {
 340             'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 341             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
 342         },
 343         'playlist_count': 13,
 344     }]
 345
 346     def _real_extract(self, url):
 347         show_id = self._match_id(url)
 348
 349         webpage = self._download_webpage(url, show_id)
 350         title = self._html_search_regex(
 351             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
 352             webpage, 'title')
 353         episode_paths = re.findall(
 354             r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
 355             webpage)
 356         entries = [
 357             self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
 358             for ep in episode_paths
 359         ]
 360         entries.reverse()
 361
 362         return {
 363             '_type': 'playlist',
 364             'id': show_id,
 365             'title': title,
 366             'entries': entries,
 367         }