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