]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/crunchyroll.py
[qqmusic] Add support for playlists
[youtube-dl] / 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.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
50         'only_matching': True,
51     }]
52
53     _FORMAT_IDS = {
54         '360': ('60', '106'),
55         '480': ('61', '106'),
56         '720': ('62', '106'),
57         '1080': ('80', '108'),
58     }
59
60     def _login(self):
61         (username, password) = self._get_login_info()
62         if username is None:
63             return
64         self.report_login()
65         login_url = 'https://www.crunchyroll.com/?a=formhandler'
66         data = urlencode_postdata({
67             'formname': 'RpcApiUser_Login',
68             'name': username,
69             'password': password,
70         })
71         login_request = compat_urllib_request.Request(login_url, data)
72         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
73         self._download_webpage(login_request, None, False, 'Wrong login info')
74
75     def _real_initialize(self):
76         self._login()
77
78     def _decrypt_subtitles(self, data, iv, id):
79         data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
80         iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))
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
104         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
105         return zlib.decompress(decrypted_data)
106
107     def _convert_subtitles_to_srt(self, sub_root):
108         output = ''
109
110         for i, event in enumerate(sub_root.findall('./events/event'), 1):
111             start = event.attrib['start'].replace('.', ',')
112             end = event.attrib['end'].replace('.', ',')
113             text = event.attrib['text'].replace('\\N', '\n')
114             output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
115         return output
116
117     def _convert_subtitles_to_ass(self, sub_root):
118         output = ''
119
120         def ass_bool(strvalue):
121             assvalue = '0'
122             if strvalue == '1':
123                 assvalue = '-1'
124             return assvalue
125
126         output = '[Script Info]\n'
127         output += 'Title: %s\n' % sub_root.attrib["title"]
128         output += 'ScriptType: v4.00+\n'
129         output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"]
130         output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"]
131         output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"]
132         output += """ScaledBorderAndShadow: yes
133
134 [V4+ Styles]
135 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
136 """
137         for style in sub_root.findall('./styles/style'):
138             output += 'Style: ' + style.attrib["name"]
139             output += ',' + style.attrib["font_name"]
140             output += ',' + style.attrib["font_size"]
141             output += ',' + style.attrib["primary_colour"]
142             output += ',' + style.attrib["secondary_colour"]
143             output += ',' + style.attrib["outline_colour"]
144             output += ',' + style.attrib["back_colour"]
145             output += ',' + ass_bool(style.attrib["bold"])
146             output += ',' + ass_bool(style.attrib["italic"])
147             output += ',' + ass_bool(style.attrib["underline"])
148             output += ',' + ass_bool(style.attrib["strikeout"])
149             output += ',' + style.attrib["scale_x"]
150             output += ',' + style.attrib["scale_y"]
151             output += ',' + style.attrib["spacing"]
152             output += ',' + style.attrib["angle"]
153             output += ',' + style.attrib["border_style"]
154             output += ',' + style.attrib["outline"]
155             output += ',' + style.attrib["shadow"]
156             output += ',' + style.attrib["alignment"]
157             output += ',' + style.attrib["margin_l"]
158             output += ',' + style.attrib["margin_r"]
159             output += ',' + style.attrib["margin_v"]
160             output += ',' + style.attrib["encoding"]
161             output += '\n'
162
163         output += """
164 [Events]
165 Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
166 """
167         for event in sub_root.findall('./events/event'):
168             output += 'Dialogue: 0'
169             output += ',' + event.attrib["start"]
170             output += ',' + event.attrib["end"]
171             output += ',' + event.attrib["style"]
172             output += ',' + event.attrib["name"]
173             output += ',' + event.attrib["margin_l"]
174             output += ',' + event.attrib["margin_r"]
175             output += ',' + event.attrib["margin_v"]
176             output += ',' + event.attrib["effect"]
177             output += ',' + event.attrib["text"]
178             output += '\n'
179
180         return output
181
182     def _extract_subtitles(self, subtitle):
183         sub_root = xml.etree.ElementTree.fromstring(subtitle)
184         return [{
185             'ext': 'srt',
186             'data': self._convert_subtitles_to_srt(sub_root),
187         }, {
188             'ext': 'ass',
189             'data': self._convert_subtitles_to_ass(sub_root),
190         }]
191
192     def _get_subtitles(self, video_id, webpage):
193         subtitles = {}
194         for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
195             sub_page = self._download_webpage(
196                 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id,
197                 video_id, note='Downloading subtitles for ' + sub_name)
198             id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
199             iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
200             data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
201             if not id or not iv or not data:
202                 continue
203             subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
204             lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
205             if not lang_code:
206                 continue
207             subtitles[lang_code] = self._extract_subtitles(subtitle)
208         return subtitles
209
210     def _real_extract(self, url):
211         mobj = re.match(self._VALID_URL, url)
212         video_id = mobj.group('video_id')
213
214         if mobj.group('prefix') == 'm':
215             mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
216             webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
217         else:
218             webpage_url = 'http://www.' + mobj.group('url')
219
220         webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
221         note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
222         if note_m:
223             raise ExtractorError(note_m)
224
225         mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
226         if mobj:
227             msg = json.loads(mobj.group('msg'))
228             if msg.get('type') == 'error':
229                 raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
230
231         video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
232         video_title = re.sub(r' {2,}', ' ', video_title)
233         video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
234         if not video_description:
235             video_description = None
236         video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
237         if video_upload_date:
238             video_upload_date = unified_strdate(video_upload_date)
239         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
240
241         playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
242         playerdata_req = compat_urllib_request.Request(playerdata_url)
243         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
244         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
245         playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
246
247         stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
248         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
249
250         formats = []
251         for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
252             stream_quality, stream_format = self._FORMAT_IDS[fmt]
253             video_format = fmt + 'p'
254             streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
255             # urlencode doesn't work!
256             streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality=' + stream_quality + '&media%5Fid=' + stream_id + '&video%5Fformat=' + stream_format
257             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
258             streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
259             streamdata = self._download_xml(
260                 streamdata_req, video_id,
261                 note='Downloading media info for %s' % video_format)
262             video_url = streamdata.find('./host').text
263             video_play_path = streamdata.find('./file').text
264             formats.append({
265                 'url': video_url,
266                 'play_path': video_play_path,
267                 'ext': 'flv',
268                 'format': video_format,
269                 'format_id': video_format,
270             })
271
272         subtitles = self.extract_subtitles(video_id, webpage)
273
274         return {
275             'id': video_id,
276             'title': video_title,
277             'description': video_description,
278             'thumbnail': video_thumbnail,
279             'uploader': video_uploader,
280             'upload_date': video_upload_date,
281             'subtitles': subtitles,
282             'formats': formats,
283         }
284
285
286 class CrunchyrollShowPlaylistIE(InfoExtractor):
287     IE_NAME = "crunchyroll:playlist"
288     _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\-]+))/?$'
289
290     _TESTS = [{
291         'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
292         'info_dict': {
293             'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
294             'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi'
295         },
296         'playlist_count': 13,
297     }]
298
299     def _real_extract(self, url):
300         show_id = self._match_id(url)
301
302         webpage = self._download_webpage(url, show_id)
303         title = self._html_search_regex(
304             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
305             webpage, 'title')
306         episode_paths = re.findall(
307             r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
308             webpage)
309         entries = [
310             self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
311             for ep in episode_paths
312         ]
313         entries.reverse()
314
315         return {
316             '_type': 'playlist',
317             'id': show_id,
318             'title': title,
319             'entries': entries,
320         }