]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/orf.py
Merge tag 'upstream/2017.03.07'
[youtubedl] / youtube_dl / extractor / orf.py
index cfca2a06352287038ff367e0f83fa67bd4cee782..1e2c54e68c3eb16b0ee8e9afb7b50b07ae429207 100644 (file)
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
-import xml.etree.ElementTree
-import json
+import calendar
+import datetime
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    compat_urlparse,
-    ExtractorError,
-    find_xpath_attr,
+    HEADRequest,
+    unified_strdate,
+    strip_jsonp,
+    int_or_none,
+    float_or_none,
+    determine_ext,
+    remove_end,
+    unescapeHTML,
 )
 
-class ORFIE(InfoExtractor):
-    _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
+
+class ORFTVthekIE(InfoExtractor):
+    IE_NAME = 'orf:tvthek'
+    IE_DESC = 'ORF TVthek'
+    _VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
+        'playlist': [{
+            'md5': '2942210346ed779588f428a92db88712',
+            'info_dict': {
+                'id': '8896777',
+                'ext': 'mp4',
+                'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
+                'description': 'md5:c1272f0245537812d4e36419c207b67d',
+                'duration': 2668,
+                'upload_date': '20141208',
+            },
+        }],
+        'skip': 'Blocked outside of Austria / Germany',
+    }, {
+        'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
+        'info_dict': {
+            'id': '7982259',
+            'ext': 'mp4',
+            'title': 'Best of Ingrid Thurnher',
+            'upload_date': '20140527',
+            '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".',
+        },
+        'params': {
+            'skip_download': True,  # rtsp downloads
+        },
+        '_skip': 'Blocked outside of Austria / Germany',
+    }, {
+        'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
+        'skip_download': True,
+    }, {
+        'url': 'http://tvthek.orf.at/profile/Universum/35429',
+        'skip_download': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        playlist_id = mobj.group('id')
+        playlist_id = self._match_id(url)
         webpage = self._download_webpage(url, playlist_id)
 
-        flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
-        flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
-        flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
-        playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
-        playlist = json.loads(playlist_json)
-
-        videos = []
-        ns = '{http://tempuri.org/XMLSchema.xsd}'
-        xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
-        webpage_description = self._og_search_description(webpage)
-        for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
-            # Get best quality url
-            rtmp_url = None
-            for q in ['Q6A', 'Q4A', 'Q1A']:
-                video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
-                if video_url is not None:
-                    rtmp_url = video_url.text
-                    break
-            if rtmp_url is None:
-                raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
-            description = self._html_search_regex(
-                r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
-                u'description', default=webpage_description, flags=re.DOTALL)
-            videos.append({
+        data_jsb = self._parse_json(
+            self._search_regex(
+                r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
+                webpage, 'playlist', group='json'),
+            playlist_id, transform_source=unescapeHTML)['playlist']['videos']
+
+        def quality_to_int(s):
+            m = re.search('([0-9]+)', s)
+            if m is None:
+                return -1
+            return int(m.group(1))
+
+        entries = []
+        for sd in data_jsb:
+            video_id, title = sd.get('id'), sd.get('title')
+            if not video_id or not title:
+                continue
+            video_id = compat_str(video_id)
+            formats = [{
+                'preference': -10 if fd['delivery'] == 'hls' else None,
+                'format_id': '%s-%s-%s' % (
+                    fd['delivery'], fd['quality'], fd['quality_string']),
+                'url': fd['src'],
+                'protocol': fd['protocol'],
+                'quality': quality_to_int(fd['quality']),
+            } for fd in sd['sources']]
+
+            # Check for geoblocking.
+            # There is a property is_geoprotection, but that's always false
+            geo_str = sd.get('geoprotection_string')
+            if geo_str:
+                try:
+                    http_url = next(
+                        f['url']
+                        for f in formats
+                        if re.match(r'^https?://.*\.mp4$', f['url']))
+                except StopIteration:
+                    pass
+                else:
+                    req = HEADRequest(http_url)
+                    self._request_webpage(
+                        req, video_id,
+                        note='Testing for geoblocking',
+                        errnote=((
+                            'This video seems to be blocked outside of %s. '
+                            'You may want to try the streaming-* formats.')
+                            % geo_str),
+                        fatal=False)
+
+            self._check_formats(formats, video_id)
+            self._sort_formats(formats)
+
+            subtitles = {}
+            for sub in sd.get('subtitles', []):
+                sub_src = sub.get('src')
+                if not sub_src:
+                    continue
+                subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
+                    'url': sub_src,
+                })
+
+            upload_date = unified_strdate(sd.get('created_date'))
+            entries.append({
                 '_type': 'video',
-                'id': info['id'],
-                'title': info['title'],
-                'url': rtmp_url,
-                'ext': 'flv',
-                'description': description,
+                'id': video_id,
+                'title': title,
+                'formats': formats,
+                'subtitles': subtitles,
+                'description': sd.get('description'),
+                'duration': int_or_none(sd.get('duration_in_seconds')),
+                'upload_date': upload_date,
+                'thumbnail': sd.get('image_full_url'),
+            })
+
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'id': playlist_id,
+        }
+
+
+class ORFOE1IE(InfoExtractor):
+    IE_NAME = 'orf:oe1'
+    IE_DESC = 'Radio Österreich 1'
+    _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole\?.*?\btrack_id=)(?P<id>[0-9]+)'
+
+    # Audios on ORF radio are only available for 7 days, so we can't add tests.
+    _TESTS = [{
+        'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
+        'only_matching': True,
+    }, {
+        'url': 'http://oe1.orf.at/konsole?show=ondemand&track_id=443608&load_day=/programm/konsole/tag/20160726',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        show_id = self._match_id(url)
+        data = self._download_json(
+            'http://oe1.orf.at/programm/%s/konsole' % show_id,
+            show_id
+        )
+
+        timestamp = datetime.datetime.strptime('%s %s' % (
+            data['item']['day_label'],
+            data['item']['time']
+        ), '%d.%m.%Y %H:%M')
+        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
+
+        return {
+            'id': show_id,
+            'title': data['item']['title'],
+            'url': data['item']['url_stream'],
+            'ext': 'mp3',
+            'description': data['item'].get('info'),
+            'timestamp': unix_timestamp
+        }
+
+
+class ORFFM4IE(InfoExtractor):
+    IE_NAME = 'orf:fm4'
+    IE_DESC = 'radio FM4'
+    _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
+
+    _TEST = {
+        'url': 'http://fm4.orf.at/player/20160110/IS/',
+        'md5': '01e736e8f1cef7e13246e880a59ad298',
+        'info_dict': {
+            'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
+            'ext': 'mp3',
+            'title': 'Im Sumpf',
+            'description': 'md5:384c543f866c4e422a55f66a62d669cd',
+            'duration': 7173,
+            'timestamp': 1452456073,
+            'upload_date': '20160110',
+        },
+        'skip': 'Live streams on FM4 got deleted soon',
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        show_date = mobj.group('date')
+        show_id = mobj.group('show')
+
+        data = self._download_json(
+            'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
+            show_id
+        )
+
+        def extract_entry_dict(info, title, subtitle):
+            return {
+                'id': info['loopStreamId'].replace('.mp3', ''),
+                'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
+                'title': title,
+                'description': subtitle,
+                'duration': (info['end'] - info['start']) / 1000,
+                'timestamp': info['start'] / 1000,
+                'ext': 'mp3'
+            }
+
+        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
+
+        return {
+            '_type': 'playlist',
+            'id': show_id,
+            'title': data['title'],
+            'description': data['subtitle'],
+            'entries': entries
+        }
+
+
+class ORFIPTVIE(InfoExtractor):
+    IE_NAME = 'orf:iptv'
+    IE_DESC = 'iptv.ORF.at'
+    _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://iptv.orf.at/stories/2275236/',
+        'md5': 'c8b22af4718a4b4af58342529453e3e5',
+        'info_dict': {
+            'id': '350612',
+            'ext': 'flv',
+            'title': 'Weitere Evakuierungen um Vulkan Calbuco',
+            'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
+            'duration': 68.197,
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'upload_date': '20150425',
+        },
+    }
+
+    def _real_extract(self, url):
+        story_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://iptv.orf.at/stories/%s' % story_id, story_id)
+
+        video_id = self._search_regex(
+            r'data-video(?:id)?="(\d+)"', webpage, 'video id')
+
+        data = self._download_json(
+            'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
+            video_id)[0]
+
+        duration = float_or_none(data['duration'], 1000)
+
+        video = data['sources']['default']
+        load_balancer_url = video['loadBalancerUrl']
+        abr = int_or_none(video.get('audioBitrate'))
+        vbr = int_or_none(video.get('bitrate'))
+        fps = int_or_none(video.get('videoFps'))
+        width = int_or_none(video.get('videoWidth'))
+        height = int_or_none(video.get('videoHeight'))
+        thumbnail = video.get('preview')
+
+        rendition = self._download_json(
+            load_balancer_url, video_id, transform_source=strip_jsonp)
+
+        f = {
+            'abr': abr,
+            'vbr': vbr,
+            'fps': fps,
+            'width': width,
+            'height': height,
+        }
+
+        formats = []
+        for format_id, format_url in rendition['redirect'].items():
+            if format_id == 'rtmp':
+                ff = f.copy()
+                ff.update({
+                    'url': format_url,
+                    'format_id': format_id,
                 })
+                formats.append(ff)
+            elif determine_ext(format_url) == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    format_url, video_id, f4m_id=format_id))
+            elif determine_ext(format_url) == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', m3u8_id=format_id))
+            else:
+                continue
+        self._sort_formats(formats)
+
+        title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
+        description = self._og_search_description(webpage)
+        upload_date = unified_strdate(self._html_search_meta(
+            'dc.date', webpage, 'upload date'))
 
-        return videos
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'formats': formats,
+        }