]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/orf.py
[README.md] Make download URLs consistent
[youtube-dl] / youtube_dl / extractor / orf.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import json
5 import re
6 import calendar
7 import datetime
8
9 from .common import InfoExtractor
10 from ..utils import (
11     HEADRequest,
12     unified_strdate,
13     ExtractorError,
14     strip_jsonp,
15     int_or_none,
16     float_or_none,
17     determine_ext,
18     remove_end,
19 )
20
21
22 class ORFTVthekIE(InfoExtractor):
23     IE_NAME = 'orf:tvthek'
24     IE_DESC = 'ORF TVthek'
25     _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics?/.+?|program/[^/]+)/(?P<id>\d+)'
26
27     _TESTS = [{
28         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
29         'playlist': [{
30             'md5': '2942210346ed779588f428a92db88712',
31             'info_dict': {
32                 'id': '8896777',
33                 'ext': 'mp4',
34                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
35                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
36                 'duration': 2668,
37                 'upload_date': '20141208',
38             },
39         }],
40         'skip': 'Blocked outside of Austria / Germany',
41     }, {
42         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
43         'info_dict': {
44             'id': '7982259',
45             'ext': 'mp4',
46             'title': 'Best of Ingrid Thurnher',
47             'upload_date': '20140527',
48             'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
49         },
50         'params': {
51             'skip_download': True,  # rtsp downloads
52         },
53         '_skip': 'Blocked outside of Austria / Germany',
54     }]
55
56     def _real_extract(self, url):
57         playlist_id = self._match_id(url)
58         webpage = self._download_webpage(url, playlist_id)
59
60         data_json = self._search_regex(
61             r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
62         all_data = json.loads(data_json)
63
64         def get_segments(all_data):
65             for data in all_data:
66                 if data['name'] in (
67                         'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
68                         'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
69                     return data['values']['segments']
70
71         sdata = get_segments(all_data)
72         if not sdata:
73             raise ExtractorError('Unable to extract segments')
74
75         def quality_to_int(s):
76             m = re.search('([0-9]+)', s)
77             if m is None:
78                 return -1
79             return int(m.group(1))
80
81         entries = []
82         for sd in sdata:
83             video_id = sd['id']
84             formats = [{
85                 'preference': -10 if fd['delivery'] == 'hls' else None,
86                 'format_id': '%s-%s-%s' % (
87                     fd['delivery'], fd['quality'], fd['quality_string']),
88                 'url': fd['src'],
89                 'protocol': fd['protocol'],
90                 'quality': quality_to_int(fd['quality']),
91             } for fd in sd['playlist_item_array']['sources']]
92
93             # Check for geoblocking.
94             # There is a property is_geoprotection, but that's always false
95             geo_str = sd.get('geoprotection_string')
96             if geo_str:
97                 try:
98                     http_url = next(
99                         f['url']
100                         for f in formats
101                         if re.match(r'^https?://.*\.mp4$', f['url']))
102                 except StopIteration:
103                     pass
104                 else:
105                     req = HEADRequest(http_url)
106                     self._request_webpage(
107                         req, video_id,
108                         note='Testing for geoblocking',
109                         errnote=((
110                             'This video seems to be blocked outside of %s. '
111                             'You may want to try the streaming-* formats.')
112                             % geo_str),
113                         fatal=False)
114
115             self._check_formats(formats, video_id)
116             self._sort_formats(formats)
117
118             upload_date = unified_strdate(sd['created_date'])
119             entries.append({
120                 '_type': 'video',
121                 'id': video_id,
122                 'title': sd['header'],
123                 'formats': formats,
124                 'description': sd.get('description'),
125                 'duration': int(sd['duration_in_seconds']),
126                 'upload_date': upload_date,
127                 'thumbnail': sd.get('image_full_url'),
128             })
129
130         return {
131             '_type': 'playlist',
132             'entries': entries,
133             'id': playlist_id,
134         }
135
136
137 class ORFOE1IE(InfoExtractor):
138     IE_NAME = 'orf:oe1'
139     IE_DESC = 'Radio Österreich 1'
140     _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole.*?#\?track_id=)(?P<id>[0-9]+)'
141
142     # Audios on ORF radio are only available for 7 days, so we can't add tests.
143     _TEST = {
144         'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
145         'only_matching': True,
146     }
147
148     def _real_extract(self, url):
149         show_id = self._match_id(url)
150         data = self._download_json(
151             'http://oe1.orf.at/programm/%s/konsole' % show_id,
152             show_id
153         )
154
155         timestamp = datetime.datetime.strptime('%s %s' % (
156             data['item']['day_label'],
157             data['item']['time']
158         ), '%d.%m.%Y %H:%M')
159         unix_timestamp = calendar.timegm(timestamp.utctimetuple())
160
161         return {
162             'id': show_id,
163             'title': data['item']['title'],
164             'url': data['item']['url_stream'],
165             'ext': 'mp3',
166             'description': data['item'].get('info'),
167             'timestamp': unix_timestamp
168         }
169
170
171 class ORFFM4IE(InfoExtractor):
172     IE_NAME = 'orf:fm4'
173     IE_DESC = 'radio FM4'
174     _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
175
176     _TEST = {
177         'url': 'http://fm4.orf.at/player/20160110/IS/',
178         'md5': '01e736e8f1cef7e13246e880a59ad298',
179         'info_dict': {
180             'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
181             'ext': 'mp3',
182             'title': 'Im Sumpf',
183             'description': 'md5:384c543f866c4e422a55f66a62d669cd',
184             'duration': 7173,
185             'timestamp': 1452456073,
186             'upload_date': '20160110',
187         },
188         'skip': 'Live streams on FM4 got deleted soon',
189     }
190
191     def _real_extract(self, url):
192         mobj = re.match(self._VALID_URL, url)
193         show_date = mobj.group('date')
194         show_id = mobj.group('show')
195
196         data = self._download_json(
197             'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
198             show_id
199         )
200
201         def extract_entry_dict(info, title, subtitle):
202             return {
203                 'id': info['loopStreamId'].replace('.mp3', ''),
204                 'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
205                 'title': title,
206                 'description': subtitle,
207                 'duration': (info['end'] - info['start']) / 1000,
208                 'timestamp': info['start'] / 1000,
209                 'ext': 'mp3'
210             }
211
212         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
213
214         return {
215             '_type': 'playlist',
216             'id': show_id,
217             'title': data['title'],
218             'description': data['subtitle'],
219             'entries': entries
220         }
221
222
223 class ORFIPTVIE(InfoExtractor):
224     IE_NAME = 'orf:iptv'
225     IE_DESC = 'iptv.ORF.at'
226     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
227
228     _TEST = {
229         'url': 'http://iptv.orf.at/stories/2275236/',
230         'md5': 'c8b22af4718a4b4af58342529453e3e5',
231         'info_dict': {
232             'id': '350612',
233             'ext': 'flv',
234             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
235             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
236             'duration': 68.197,
237             'thumbnail': 're:^https?://.*\.jpg$',
238             'upload_date': '20150425',
239         },
240     }
241
242     def _real_extract(self, url):
243         story_id = self._match_id(url)
244
245         webpage = self._download_webpage(
246             'http://iptv.orf.at/stories/%s' % story_id, story_id)
247
248         video_id = self._search_regex(
249             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
250
251         data = self._download_json(
252             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
253             video_id)[0]
254
255         duration = float_or_none(data['duration'], 1000)
256
257         video = data['sources']['default']
258         load_balancer_url = video['loadBalancerUrl']
259         abr = int_or_none(video.get('audioBitrate'))
260         vbr = int_or_none(video.get('bitrate'))
261         fps = int_or_none(video.get('videoFps'))
262         width = int_or_none(video.get('videoWidth'))
263         height = int_or_none(video.get('videoHeight'))
264         thumbnail = video.get('preview')
265
266         rendition = self._download_json(
267             load_balancer_url, video_id, transform_source=strip_jsonp)
268
269         f = {
270             'abr': abr,
271             'vbr': vbr,
272             'fps': fps,
273             'width': width,
274             'height': height,
275         }
276
277         formats = []
278         for format_id, format_url in rendition['redirect'].items():
279             if format_id == 'rtmp':
280                 ff = f.copy()
281                 ff.update({
282                     'url': format_url,
283                     'format_id': format_id,
284                 })
285                 formats.append(ff)
286             elif determine_ext(format_url) == 'f4m':
287                 formats.extend(self._extract_f4m_formats(
288                     format_url, video_id, f4m_id=format_id))
289             elif determine_ext(format_url) == 'm3u8':
290                 formats.extend(self._extract_m3u8_formats(
291                     format_url, video_id, 'mp4', m3u8_id=format_id))
292             else:
293                 continue
294         self._sort_formats(formats)
295
296         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
297         description = self._og_search_description(webpage)
298         upload_date = unified_strdate(self._html_search_meta(
299             'dc.date', webpage, 'upload date'))
300
301         return {
302             'id': video_id,
303             'title': title,
304             'description': description,
305             'duration': duration,
306             'thumbnail': thumbnail,
307             'upload_date': upload_date,
308             'formats': formats,
309         }