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