]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/crunchyroll.py
Prepare for upload.
[youtubedl] / 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 clean_html,
21 urlencode_postdata,
22 )
23 from ..aes import (
24 aes_cbc_decrypt,
25 inc,
26 )
27 from .common import InfoExtractor
28
29
30 class CrunchyrollIE(SubtitlesInfoExtractor):
31 _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
32 _TEST = {
33 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
34 #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
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
51 _FORMAT_IDS = {
52 '360': ('60', '106'),
53 '480': ('61', '106'),
54 '720': ('62', '106'),
55 '1080': ('80', '108'),
56 }
57
58 def _login(self):
59 (username, password) = self._get_login_info()
60 if username is None:
61 return
62 self.report_login()
63 login_url = 'https://www.crunchyroll.com/?a=formhandler'
64 data = urlencode_postdata({
65 'formname': 'RpcApiUser_Login',
66 'name': username,
67 'password': password,
68 })
69 login_request = compat_urllib_request.Request(login_url, data)
70 login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
71 self._download_webpage(login_request, None, False, 'Wrong login info')
72
73
74 def _real_initialize(self):
75 self._login()
76
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 class Counter:
104 __value = iv
105 def next_value(self):
106 temp = self.__value
107 self.__value = inc(self.__value)
108 return temp
109 decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
110 return zlib.decompress(decrypted_data)
111
112 def _convert_subtitles_to_srt(self, sub_root):
113 output = ''
114
115 for i, event in enumerate(sub_root.findall('./events/event'), 1):
116 start = event.attrib['start'].replace('.', ',')
117 end = event.attrib['end'].replace('.', ',')
118 text = event.attrib['text'].replace('\\N', '\n')
119 output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
120 return output
121
122 def _convert_subtitles_to_ass(self, sub_root):
123 output = ''
124
125 def ass_bool(strvalue):
126 assvalue = '0'
127 if strvalue == '1':
128 assvalue = '-1'
129 return assvalue
130
131 output = '[Script Info]\n'
132 output += 'Title: %s\n' % sub_root.attrib["title"]
133 output += 'ScriptType: v4.00+\n'
134 output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
135 output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
136 output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
137 output += """ScaledBorderAndShadow: yes
138
139 [V4+ Styles]
140 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
141 """
142 for style in sub_root.findall('./styles/style'):
143 output += 'Style: ' + style.attrib["name"]
144 output += ',' + style.attrib["font_name"]
145 output += ',' + style.attrib["font_size"]
146 output += ',' + style.attrib["primary_colour"]
147 output += ',' + style.attrib["secondary_colour"]
148 output += ',' + style.attrib["outline_colour"]
149 output += ',' + style.attrib["back_colour"]
150 output += ',' + ass_bool(style.attrib["bold"])
151 output += ',' + ass_bool(style.attrib["italic"])
152 output += ',' + ass_bool(style.attrib["underline"])
153 output += ',' + ass_bool(style.attrib["strikeout"])
154 output += ',' + style.attrib["scale_x"]
155 output += ',' + style.attrib["scale_y"]
156 output += ',' + style.attrib["spacing"]
157 output += ',' + style.attrib["angle"]
158 output += ',' + style.attrib["border_style"]
159 output += ',' + style.attrib["outline"]
160 output += ',' + style.attrib["shadow"]
161 output += ',' + style.attrib["alignment"]
162 output += ',' + style.attrib["margin_l"]
163 output += ',' + style.attrib["margin_r"]
164 output += ',' + style.attrib["margin_v"]
165 output += ',' + style.attrib["encoding"]
166 output += '\n'
167
168 output += """
169 [Events]
170 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
171 """
172 for event in sub_root.findall('./events/event'):
173 output += 'Dialogue: 0'
174 output += ',' + event.attrib["start"]
175 output += ',' + event.attrib["end"]
176 output += ',' + event.attrib["style"]
177 output += ',' + event.attrib["name"]
178 output += ',' + event.attrib["margin_l"]
179 output += ',' + event.attrib["margin_r"]
180 output += ',' + event.attrib["margin_v"]
181 output += ',' + event.attrib["effect"]
182 output += ',' + event.attrib["text"]
183 output += '\n'
184
185 return output
186
187 def _real_extract(self,url):
188 mobj = re.match(self._VALID_URL, url)
189 video_id = mobj.group('video_id')
190
191 if mobj.group('prefix') == 'm':
192 mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
193 webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
194 else:
195 webpage_url = 'http://www.' + mobj.group('url')
196
197 webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
198 note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
199 if note_m:
200 raise ExtractorError(note_m)
201
202 mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
203 if mobj:
204 msg = json.loads(mobj.group('msg'))
205 if msg.get('type') == 'error':
206 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
207
208 video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
209 video_title = re.sub(r' {2,}', ' ', video_title)
210 video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
211 if not video_description:
212 video_description = None
213 video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
214 if video_upload_date:
215 video_upload_date = unified_strdate(video_upload_date)
216 video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
217
218 playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
219 playerdata_req = compat_urllib_request.Request(playerdata_url)
220 playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
221 playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
222 playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
223
224 stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
225 video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
226
227 formats = []
228 for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
229 stream_quality, stream_format = self._FORMAT_IDS[fmt]
230 video_format = fmt+'p'
231 streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
232 # urlencode doesn't work!
233 streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
234 streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
235 streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
236 streamdata = self._download_xml(
237 streamdata_req, video_id,
238 note='Downloading media info for %s' % video_format)
239 video_url = streamdata.find('.//host').text
240 video_play_path = streamdata.find('.//file').text
241 formats.append({
242 'url': video_url,
243 'play_path': video_play_path,
244 'ext': 'flv',
245 'format': video_format,
246 'format_id': video_format,
247 })
248
249 subtitles = {}
250 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
251 for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
252 sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
253 video_id, note='Downloading subtitles for '+sub_name)
254 id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
255 iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
256 data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
257 if not id or not iv or not data:
258 continue
259 id = int(id)
260 iv = base64.b64decode(iv)
261 data = base64.b64decode(data)
262
263 subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
264 lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
265 if not lang_code:
266 continue
267 sub_root = xml.etree.ElementTree.fromstring(subtitle)
268 if not sub_root:
269 subtitles[lang_code] = ''
270 if sub_format == 'ass':
271 subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root)
272 else:
273 subtitles[lang_code] = self._convert_subtitles_to_srt(sub_root)
274
275 if self._downloader.params.get('listsubtitles', False):
276 self._list_available_subtitles(video_id, subtitles)
277 return
278
279 return {
280 'id': video_id,
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,
287 'formats': formats,
288 }
289
290
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<id>[\w\-]+))/?$'
294
295 _TESTS = [{
296 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
297 'info_dict': {
298 'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
299 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
300 },
301 'playlist_count': 13,
302 }]
303
304 def _real_extract(self, url):
305 show_id = self._match_id(url)
306
307 webpage = self._download_webpage(url, show_id)
308 title = self._html_search_regex(
309 r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
310 webpage, 'title')
311 episode_paths = re.findall(
312 r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
313 webpage)
314 entries = [
315 self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
316 for ep in episode_paths
317 ]
318 entries.reverse()
319
320 return {
321 '_type': 'playlist',
322 'id': show_id,
323 'title': title,
324 'entries': entries,
325 }