← Back to branch summary

~facundo/encuentro/trunk

« back to all changes in this revision

Viewing changes to external/youtube-dl/youtube_dl/extractor/orf.py

  • Committer: Facundo Batista
  • Date: 2015-12-27 11:27:15 UTC
  • mto: This revision was merged to the branch mainline in revision 274.
  • Revision ID: facundo@taniquetil.com.ar-20151227112715-ztuasdhqm26hycug
Able to download TEDx.

Show diffs side-by-side

added added

removed removed

Lines of Context:
external/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
        'playlist': [{
 
44
            'md5': '68f543909aea49d621dfc7703a11cfaf',
 
45
            'info_dict': {
 
46
                'id': '7982259',
 
47
                'ext': 'mp4',
 
48
                'title': 'Best of Ingrid Thurnher',
 
49
                'upload_date': '20140527',
 
50
                '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".',
 
51
            }
 
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._sort_formats(formats)
 
116
 
 
117
            upload_date = unified_strdate(sd['created_date'])
 
118
            entries.append({
 
119
                '_type': 'video',
 
120
                'id': video_id,
 
121
                'title': sd['header'],
 
122
                'formats': formats,
 
123
                'description': sd.get('description'),
 
124
                'duration': int(sd['duration_in_seconds']),
 
125
                'upload_date': upload_date,
 
126
                'thumbnail': sd.get('image_full_url'),
 
127
            })
 
128
 
 
129
        return {
 
130
            '_type': 'playlist',
 
131
            'entries': entries,
 
132
            'id': playlist_id,
 
133
        }
 
134
 
 
135
 
 
136
class ORFOE1IE(InfoExtractor):
 
137
    IE_NAME = 'orf:oe1'
 
138
    IE_DESC = 'Radio Österreich 1'
 
139
    _VALID_URL = r'http://oe1\.orf\.at/(?:programm/|konsole.*?#\?track_id=)(?P<id>[0-9]+)'
 
140
 
 
141
    # Audios on ORF radio are only available for 7 days, so we can't add tests.
 
142
    _TEST = {
 
143
        'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
 
144
        'only_matching': True,
 
145
    }
 
146
 
 
147
    def _real_extract(self, url):
 
148
        show_id = self._match_id(url)
 
149
        data = self._download_json(
 
150
            'http://oe1.orf.at/programm/%s/konsole' % show_id,
 
151
            show_id
 
152
        )
 
153
 
 
154
        timestamp = datetime.datetime.strptime('%s %s' % (
 
155
            data['item']['day_label'],
 
156
            data['item']['time']
 
157
        ), '%d.%m.%Y %H:%M')
 
158
        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
 
159
 
 
160
        return {
 
161
            'id': show_id,
 
162
            'title': data['item']['title'],
 
163
            'url': data['item']['url_stream'],
 
164
            'ext': 'mp3',
 
165
            'description': data['item'].get('info'),
 
166
            'timestamp': unix_timestamp
 
167
        }
 
168
 
 
169
 
 
170
class ORFFM4IE(InfoExtractor):
 
171
    IE_NAME = 'orf:fm4'
 
172
    IE_DESC = 'radio FM4'
 
173
    _VALID_URL = r'http://fm4\.orf\.at/7tage/?#(?P<date>[0-9]+)/(?P<show>\w+)'
 
174
 
 
175
    def _real_extract(self, url):
 
176
        mobj = re.match(self._VALID_URL, url)
 
177
        show_date = mobj.group('date')
 
178
        show_id = mobj.group('show')
 
179
 
 
180
        data = self._download_json(
 
181
            'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
 
182
            show_id
 
183
        )
 
184
 
 
185
        def extract_entry_dict(info, title, subtitle):
 
186
            return {
 
187
                'id': info['loopStreamId'].replace('.mp3', ''),
 
188
                'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
 
189
                'title': title,
 
190
                'description': subtitle,
 
191
                'duration': (info['end'] - info['start']) / 1000,
 
192
                'timestamp': info['start'] / 1000,
 
193
                'ext': 'mp3'
 
194
            }
 
195
 
 
196
        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
 
197
 
 
198
        return {
 
199
            '_type': 'playlist',
 
200
            'id': show_id,
 
201
            'title': data['title'],
 
202
            'description': data['subtitle'],
 
203
            'entries': entries
 
204
        }
 
205
 
 
206
 
 
207
class ORFIPTVIE(InfoExtractor):
 
208
    IE_NAME = 'orf:iptv'
 
209
    IE_DESC = 'iptv.ORF.at'
 
210
    _VALID_URL = r'http://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 
211
 
 
212
    _TEST = {
 
213
        'url': 'http://iptv.orf.at/stories/2275236/',
 
214
        'md5': 'c8b22af4718a4b4af58342529453e3e5',
 
215
        'info_dict': {
 
216
            'id': '350612',
 
217
            'ext': 'flv',
 
218
            'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 
219
            'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 
220
            'duration': 68.197,
 
221
            'thumbnail': 're:^https?://.*\.jpg$',
 
222
            'upload_date': '20150425',
 
223
        },
 
224
    }
 
225
 
 
226
    def _real_extract(self, url):
 
227
        story_id = self._match_id(url)
 
228
 
 
229
        webpage = self._download_webpage(
 
230
            'http://iptv.orf.at/stories/%s' % story_id, story_id)
 
231
 
 
232
        video_id = self._search_regex(
 
233
            r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 
234
 
 
235
        data = self._download_json(
 
236
            'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 
237
            video_id)[0]
 
238
 
 
239
        duration = float_or_none(data['duration'], 1000)
 
240
 
 
241
        video = data['sources']['default']
 
242
        load_balancer_url = video['loadBalancerUrl']
 
243
        abr = int_or_none(video.get('audioBitrate'))
 
244
        vbr = int_or_none(video.get('bitrate'))
 
245
        fps = int_or_none(video.get('videoFps'))
 
246
        width = int_or_none(video.get('videoWidth'))
 
247
        height = int_or_none(video.get('videoHeight'))
 
248
        thumbnail = video.get('preview')
 
249
 
 
250
        rendition = self._download_json(
 
251
            load_balancer_url, video_id, transform_source=strip_jsonp)
 
252
 
 
253
        f = {
 
254
            'abr': abr,
 
255
            'vbr': vbr,
 
256
            'fps': fps,
 
257
            'width': width,
 
258
            'height': height,
 
259
        }
 
260
 
 
261
        formats = []
 
262
        for format_id, format_url in rendition['redirect'].items():
 
263
            if format_id == 'rtmp':
 
264
                ff = f.copy()
 
265
                ff.update({
 
266
                    'url': format_url,
 
267
                    'format_id': format_id,
 
268
                })
 
269
                formats.append(ff)
 
270
            elif determine_ext(format_url) == 'f4m':
 
271
                formats.extend(self._extract_f4m_formats(
 
272
                    format_url, video_id, f4m_id=format_id))
 
273
            elif determine_ext(format_url) == 'm3u8':
 
274
                formats.extend(self._extract_m3u8_formats(
 
275
                    format_url, video_id, 'mp4', m3u8_id=format_id))
 
276
            else:
 
277
                continue
 
278
        self._sort_formats(formats)
 
279
 
 
280
        title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 
281
        description = self._og_search_description(webpage)
 
282
        upload_date = unified_strdate(self._html_search_meta(
 
283
            'dc.date', webpage, 'upload date'))
 
284
 
 
285
        return {
 
286
            'id': video_id,
 
287
            'title': title,
 
288
            'description': description,
 
289
            'duration': duration,
 
290
            'thumbnail': thumbnail,
 
291
            'upload_date': upload_date,
 
292
            'formats': formats,
 
293
        }