2 from __future__ import unicode_literals
8 import xml.etree.ElementTree
10 from hashlib import sha1
11 from math import pow, sqrt, floor
12 from .subtitles import SubtitlesInfoExtractor
16 compat_urllib_request,
27 from .common import InfoExtractor
30 class CrunchyrollIE(SubtitlesInfoExtractor):
31 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
33 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
34 #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
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',
46 'skip_download': True,
54 '1080': ('80', '108'),
58 (username, password) = self._get_login_info()
62 login_url = 'https://www.crunchyroll.com/?a=formhandler'
63 data = urlencode_postdata({
64 'formname': 'RpcApiUser_Login',
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')
73 def _real_initialize(self):
77 def _decrypt_subtitles(self, data, iv, id):
78 data = bytes_to_intlist(data)
79 iv = bytes_to_intlist(iv)
82 def obfuscate_key_aux(count, modulo, start):
84 for _ in range(count):
85 output.append(output[-1] + output[-2])
86 # cut off start values
88 output = list(map(lambda x: x % modulo + 33, output))
91 def obfuscate_key(key):
92 num1 = int(floor(pow(2, 25) * sqrt(6.9)))
93 num2 = (num1 ^ key) << 5
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
101 key = obfuscate_key(id)
104 def next_value(self):
106 self.__value = inc(self.__value)
108 decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
109 return zlib.decompress(decrypted_data)
111 def _convert_subtitles_to_srt(self, subtitles):
113 for i, (start, end, text) in enumerate(re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles), 1):
114 start = start.replace('.', ',')
115 end = end.replace('.', ',')
116 text = clean_html(text)
117 text = text.replace('\\N', '\n')
120 output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
123 def _convert_subtitles_to_ass(self, subtitles):
126 def ass_bool(strvalue):
132 sub_root = xml.etree.ElementTree.fromstring(subtitles)
136 output = '[Script Info]\n'
137 output += 'Title: %s\n' % sub_root.attrib["title"]
138 output += 'ScriptType: v4.00+\n'
139 output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
140 output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
141 output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
142 output += """ScaledBorderAndShadow: yes
145 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
147 for style in sub_root.findall('./styles/style'):
148 output += 'Style: ' + style.attrib["name"]
149 output += ',' + style.attrib["font_name"]
150 output += ',' + style.attrib["font_size"]
151 output += ',' + style.attrib["primary_colour"]
152 output += ',' + style.attrib["secondary_colour"]
153 output += ',' + style.attrib["outline_colour"]
154 output += ',' + style.attrib["back_colour"]
155 output += ',' + ass_bool(style.attrib["bold"])
156 output += ',' + ass_bool(style.attrib["italic"])
157 output += ',' + ass_bool(style.attrib["underline"])
158 output += ',' + ass_bool(style.attrib["strikeout"])
159 output += ',' + style.attrib["scale_x"]
160 output += ',' + style.attrib["scale_y"]
161 output += ',' + style.attrib["spacing"]
162 output += ',' + style.attrib["angle"]
163 output += ',' + style.attrib["border_style"]
164 output += ',' + style.attrib["outline"]
165 output += ',' + style.attrib["shadow"]
166 output += ',' + style.attrib["alignment"]
167 output += ',' + style.attrib["margin_l"]
168 output += ',' + style.attrib["margin_r"]
169 output += ',' + style.attrib["margin_v"]
170 output += ',' + style.attrib["encoding"]
175 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
177 for event in sub_root.findall('./events/event'):
178 output += 'Dialogue: 0'
179 output += ',' + event.attrib["start"]
180 output += ',' + event.attrib["end"]
181 output += ',' + event.attrib["style"]
182 output += ',' + event.attrib["name"]
183 output += ',' + event.attrib["margin_l"]
184 output += ',' + event.attrib["margin_r"]
185 output += ',' + event.attrib["margin_v"]
186 output += ',' + event.attrib["effect"]
187 output += ',' + event.attrib["text"]
192 def _real_extract(self,url):
193 mobj = re.match(self._VALID_URL, url)
194 video_id = mobj.group('video_id')
196 if mobj.group('prefix') == 'm':
197 mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
198 webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
200 webpage_url = 'http://www.' + mobj.group('url')
202 webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
203 note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
205 raise ExtractorError(note_m)
207 mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
209 msg = json.loads(mobj.group('msg'))
210 if msg.get('type') == 'error':
211 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
213 video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
214 video_title = re.sub(r' {2,}', ' ', video_title)
215 video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
216 if not video_description:
217 video_description = None
218 video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
219 if video_upload_date:
220 video_upload_date = unified_strdate(video_upload_date)
221 video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
223 playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
224 playerdata_req = compat_urllib_request.Request(playerdata_url)
225 playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
226 playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
227 playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
229 stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
230 video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
233 for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
234 stream_quality, stream_format = self._FORMAT_IDS[fmt]
235 video_format = fmt+'p'
236 streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
237 # urlencode doesn't work!
238 streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
239 streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
240 streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
241 streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format)
242 video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url')
243 video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path')
246 'play_path': video_play_path,
248 'format': video_format,
249 'format_id': video_format,
253 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
254 for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
255 sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
256 video_id, note='Downloading subtitles for '+sub_name)
257 id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
258 iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
259 data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
260 if not id or not iv or not data:
263 iv = base64.b64decode(iv)
264 data = base64.b64decode(data)
266 subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
267 lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
270 if sub_format == 'ass':
271 subtitles[lang_code] = self._convert_subtitles_to_ass(subtitle)
273 subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
275 if self._downloader.params.get('listsubtitles', False):
276 self._list_available_subtitles(video_id, subtitles)
281 'title': video_title,
282 'description': video_description,
283 'thumbnail': video_thumbnail,
284 'uploader': video_uploader,
285 'upload_date': video_upload_date,
286 'subtitles': subtitles,
291 class CrunchyrollShowPlaylistIE(InfoExtractor):
292 IE_NAME = "crunchyroll:playlist"
293 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<show>[\w\-]+))/?$'
294 _TITLE_EXTR = r'<span\s+itemprop="name">\s*(?P<showtitle>[\w\s]+)'
297 'url' : 'http://www.crunchyroll.com/attack-on-titan',
299 'title' : 'Attack on Titan'
301 'playlist_count' : 15
304 def _extract_title_entries(self,id,webpage):
305 _EPISODE_ID_EXTR = r'id="showview_videos_media_(?P<vidid>\d+)".*?href="/{0}/(?P<vidurl>[\w\-]+-(?P=vidid))"'.format(id)
306 title = self._html_search_regex(self._TITLE_EXTR,webpage,"title",flags=re.UNICODE|re.MULTILINE)
307 episode_urls = [self.url_result('http://www.crunchyroll.com/{0}/{1}'.format(id, showmatch[1])) for
308 showmatch in re.findall(_EPISODE_ID_EXTR, webpage,re.UNICODE|re.MULTILINE|re.DOTALL)]
309 return title, episode_urls
312 def _real_extract(self, url):
313 url_match = re.match(self._VALID_URL,url)
314 show_id = url_match.group('show')
315 webpage = self._download_webpage(url,show_id)
316 (title,entries) = self._extract_title_entries(show_id,webpage)
318 '_type' : 'playlist',