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 .subtitles import SubtitlesInfoExtractor
  13 from ..utils import (
  14     ExtractorError,
  15     compat_urllib_parse,
  16     compat_urllib_request,
  17     bytes_to_intlist,
  18     intlist_to_bytes,
  19     unified_strdate,
  20     urlencode_postdata,
  21 )
  22 from ..aes import (
  23     aes_cbc_decrypt,
  24     inc,
  25 )
  26 from .common import InfoExtractor
  27
  28
  29 class CrunchyrollIE(SubtitlesInfoExtractor):
  30     _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
  31     _TEST = {
  32         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
  33         #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
  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
  50     _FORMAT_IDS = {
  51         '360': ('60', '106'),
  52         '480': ('61', '106'),
  53         '720': ('62', '106'),
  54         '1080': ('80', '108'),
  55     }
  56
  57     def _login(self):
  58         (username, password) = self._get_login_info()
  59         if username is None:
  60             return
  61         self.report_login()
  62         login_url = 'https://www.crunchyroll.com/?a=formhandler'
  63         data = urlencode_postdata({
  64             'formname': 'RpcApiUser_Login',
  65             'name': username,
  66             'password': password,
  67         })
  68         login_request = compat_urllib_request.Request(login_url, data)
  69         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  70         self._download_webpage(login_request, None, False, 'Wrong login info')
  71
  72
  73     def _real_initialize(self):
  74         self._login()
  75
  76
  77     def _decrypt_subtitles(self, data, iv, id):
  78         data = bytes_to_intlist(data)
  79         iv = bytes_to_intlist(iv)
  80         id = int(id)
  81
  82         def obfuscate_key_aux(count, modulo, start):
  83             output = list(start)
  84             for _ in range(count):
  85                 output.append(output[-1] + output[-2])
  86             # cut off start values
  87             output = output[2:]
  88             output = list(map(lambda x: x % modulo + 33, output))
  89             return output
  90
  91         def obfuscate_key(key):
  92             num1 = int(floor(pow(2, 25) * sqrt(6.9)))
  93             num2 = (num1 ^ key) << 5
  94             num3 = key ^ num1
  95             num4 = num3 ^ (num3 >> 3) ^ num2
  96             prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
  97             shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
  98             # Extend 160 Bit hash to 256 Bit
  99             return shaHash + [0] * 12
 100
 101         key = obfuscate_key(id)
 102         class Counter:
 103             __value = iv
 104             def next_value(self):
 105                 temp = self.__value
 106                 self.__value = inc(self.__value)
 107                 return temp
 108         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
 109         return zlib.decompress(decrypted_data)
 110
 111     def _convert_subtitles_to_srt(self, sub_root):
 112         output = ''
 113
 114         for i, event in enumerate(sub_root.findall('./events/event'), 1):
 115             start = event.attrib['start'].replace('.', ',')
 116             end = event.attrib['end'].replace('.', ',')
 117             text = event.attrib['text'].replace('\\N', '\n')
 118             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
 119         return output
 120
 121     def _convert_subtitles_to_ass(self, sub_root):
 122         output = ''
 123
 124         def ass_bool(strvalue):
 125             assvalue = '0'
 126             if strvalue == '1':
 127                 assvalue = '-1'
 128             return assvalue
 129
 130         output = '[Script Info]\n'
 131         output += 'Title: %s\n' % sub_root.attrib["title"]
 132         output += 'ScriptType: v4.00+\n'
 133         output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
 134         output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
 135         output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
 136         output += """ScaledBorderAndShadow: yes
 137
 138 [V4+ Styles]
 139 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 140 """
 141         for style in sub_root.findall('./styles/style'):
 142             output += 'Style: ' + style.attrib["name"]
 143             output += ',' + style.attrib["font_name"]
 144             output += ',' + style.attrib["font_size"]
 145             output += ',' + style.attrib["primary_colour"]
 146             output += ',' + style.attrib["secondary_colour"]
 147             output += ',' + style.attrib["outline_colour"]
 148             output += ',' + style.attrib["back_colour"]
 149             output += ',' + ass_bool(style.attrib["bold"])
 150             output += ',' + ass_bool(style.attrib["italic"])
 151             output += ',' + ass_bool(style.attrib["underline"])
 152             output += ',' + ass_bool(style.attrib["strikeout"])
 153             output += ',' + style.attrib["scale_x"]
 154             output += ',' + style.attrib["scale_y"]
 155             output += ',' + style.attrib["spacing"]
 156             output += ',' + style.attrib["angle"]
 157             output += ',' + style.attrib["border_style"]
 158             output += ',' + style.attrib["outline"]
 159             output += ',' + style.attrib["shadow"]
 160             output += ',' + style.attrib["alignment"]
 161             output += ',' + style.attrib["margin_l"]
 162             output += ',' + style.attrib["margin_r"]
 163             output += ',' + style.attrib["margin_v"]
 164             output += ',' + style.attrib["encoding"]
 165             output += '\n'
 166
 167         output += """
 168 [Events]
 169 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 170 """
 171         for event in sub_root.findall('./events/event'):
 172             output += 'Dialogue: 0'
 173             output += ',' + event.attrib["start"]
 174             output += ',' + event.attrib["end"]
 175             output += ',' + event.attrib["style"]
 176             output += ',' + event.attrib["name"]
 177             output += ',' + event.attrib["margin_l"]
 178             output += ',' + event.attrib["margin_r"]
 179             output += ',' + event.attrib["margin_v"]
 180             output += ',' + event.attrib["effect"]
 181             output += ',' + event.attrib["text"]
 182             output += '\n'
 183
 184         return output
 185
 186     def _real_extract(self,url):
 187         mobj = re.match(self._VALID_URL, url)
 188         video_id = mobj.group('video_id')
 189
 190         if mobj.group('prefix') == 'm':
 191             mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
 192             webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
 193         else:
 194             webpage_url = 'http://www.' + mobj.group('url')
 195
 196         webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
 197         note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
 198         if note_m:
 199             raise ExtractorError(note_m)
 200
 201         mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
 202         if mobj:
 203             msg = json.loads(mobj.group('msg'))
 204             if msg.get('type') == 'error':
 205                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
 206
 207         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
 208         video_title = re.sub(r' {2,}', ' ', video_title)
 209         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
 210         if not video_description:
 211             video_description = None
 212         video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
 213         if video_upload_date:
 214             video_upload_date = unified_strdate(video_upload_date)
 215         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 216
 217         playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
 218         playerdata_req = compat_urllib_request.Request(playerdata_url)
 219         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
 220         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
 221         playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
 222
 223         stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
 224         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 225
 226         formats = []
 227         for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
 228             stream_quality, stream_format = self._FORMAT_IDS[fmt]
 229             video_format = fmt+'p'
 230             streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
 231             # urlencode doesn't work!
 232             streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
 233             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
 234             streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
 235             streamdata = self._download_xml(
 236                 streamdata_req, video_id,
 237                 note='Downloading media info for %s' % video_format)
 238             video_url = streamdata.find('.//host').text
 239             video_play_path = streamdata.find('.//file').text
 240             formats.append({
 241                 'url': video_url,
 242                 'play_path': video_play_path,
 243                 'ext': 'flv',
 244                 'format': video_format,
 245                 'format_id': video_format,
 246             })
 247
 248         subtitles = {}
 249         sub_format = self._downloader.params.get('subtitlesformat', 'srt')
 250         for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
 251             sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
 252                                               video_id, note='Downloading subtitles for '+sub_name)
 253             id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
 254             iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
 255             data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
 256             if not id or not iv or not data:
 257                 continue
 258             id = int(id)
 259             iv = base64.b64decode(iv)
 260             data = base64.b64decode(data)
 261
 262             subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
 263             lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
 264             if not lang_code:
 265                 continue
 266             sub_root = xml.etree.ElementTree.fromstring(subtitle)
 267             if sub_format == 'ass':
 268                 subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root)
 269             else:
 270                 subtitles[lang_code] = self._convert_subtitles_to_srt(sub_root)
 271
 272         if self._downloader.params.get('listsubtitles', False):
 273             self._list_available_subtitles(video_id, subtitles)
 274             return
 275
 276         return {
 277             'id':          video_id,
 278             'title':       video_title,
 279             'description': video_description,
 280             'thumbnail':   video_thumbnail,
 281             'uploader':    video_uploader,
 282             'upload_date': video_upload_date,
 283             'subtitles':   subtitles,
 284             'formats':     formats,
 285         }
 286
 287
 288 class CrunchyrollShowPlaylistIE(InfoExtractor):
 289     IE_NAME = "crunchyroll:playlist"
 290     _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\-]+))/?$'
 291
 292     _TESTS = [{
 293         'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 294         'info_dict': {
 295             'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
 296             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
 297         },
 298         'playlist_count': 13,
 299     }]
 300
 301     def _real_extract(self, url):
 302         show_id = self._match_id(url)
 303
 304         webpage = self._download_webpage(url, show_id)
 305         title = self._html_search_regex(
 306             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
 307             webpage, 'title')
 308         episode_paths = re.findall(
 309             r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
 310             webpage)
 311         entries = [
 312             self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
 313             for ep in episode_paths
 314         ]
 315         entries.reverse()
 316
 317         return {
 318             '_type': 'playlist',
 319             'id': show_id,
 320             'title': title,
 321             'entries': entries,
 322         }