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