Support BBC News (bbc.com/news)

diff --git a/docs/supportedsites.md b/docs/supportedsites.md
index 220e52b988ace1314ec866f59a8ef34562f1b583..d4ccbbd3a63dd87a20be57cf803abc4cd8861157 100644 (file)
--- a/docs/supportedsites.md
+++ b/docs/supportedsites.md
@@ -50,6 +50,7 @@
  - **Bandcamp**
  - **Bandcamp:album**
  - **bbc.co.uk**: BBC iPlayer
+ - **bbc.com**: BBC news videos
  - **BeatportPro**
  - **Beeg**
  - **BehindKink**

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 6fdaf90b2d69799091db469a9eef5810cef8542e..51d2d20e9033ef820099c15e5bf89077e6bbd9cc 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -36,6 +36,7 @@ from .baidu import BaiduVideoIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bbccouk import BBCCoUkIE
+from .bbcnews import BBCNewsIE
 from .beeg import BeegIE
 from .behindkink import BehindKinkIE
 from .beatportpro import BeatportProIE

diff --git a/youtube_dl/extractor/bbcnews.py b/youtube_dl/extractor/bbcnews.py
new file mode 100644 (file)
index 0000000..b10e30a
--- /dev/null
+++ b/youtube_dl/extractor/bbcnews.py
@@ -0,0 +1,162 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
+from ..compat import compat_HTTPError
+import re
+from .bbccouk import BBCCoUkIE
+
+class BBCNewsIE(BBCCoUkIE):
+    IE_NAME = 'bbc.com'
+    IE_DESC = 'BBC news'
+    _VALID_URL = r'https?://(?:www\.)?(?:bbc\.co\.uk|bbc\.com)/news/(?P<id>[^/]+)'
+
+    _TESTS = [{
+        'url': 'http://www.bbc.com/news/world-europe-32668511',
+        'info_dict': {
+            'id': 'world-europe-32668511',
+            'title': 'Russia stages massive WW2 parade despite Western boycott',
+        },
+        'playlist_count': 2,
+    },{
+        'url': 'http://www.bbc.com/news/business-28299555',
+        'info_dict': {
+            'id': 'business-28299555',
+            'title': 'Farnborough Airshow: Video highlights',
+        },
+        'playlist_count': 9,
+    },{
+        'url': 'http://www.bbc.com/news/world-europe-32041533',
+        'note': 'Video',
+        'info_dict': {
+            'id': 'p02mprgb',
+            'ext': 'mp4',
+            'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+            'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+            'duration': 47,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }]
+
+    def _duration_str2int(self, str):
+        if not str:
+            return None
+        ret = re.match(r'^\d+$', str)
+        if ret:
+            return int(ret.group(0))
+        ret = re.match(r'PT((?P<h>\d+)H)?((?P<m>\d+)M)?(?P<s>\d+)S$', str)
+        if ret:
+            total=int(ret.group('s'))
+            if ret.group('m'):
+                total+=(int(ret.group('m'))*60)
+            if ret.group('h'):
+                total+=(int(ret.group('h'))*3600)
+            return total
+        return None
+
+    def _download_media_selector(self, programme_id):
+        # bbc news uses http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/ not
+        # http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/
+        # Could add third urlspec arg to BBCCoUkIE._download_media_selector instead of duplicating it
+
+        try:
+            media_selection = self._download_xml(
+               'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s' % programme_id,
+                programme_id, 'Downloading media selection XML')
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
+                media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
+            else:
+                raise
+        formats = []
+        subtitles = None
+
+        for media in self._extract_medias(media_selection):
+            kind = media.get('kind')
+            if kind == 'audio':
+                formats.extend(self._extract_audio(media, programme_id))
+            elif kind == 'video':
+                formats.extend(self._extract_video(media, programme_id))
+            elif kind == 'captions':
+                subtitles = self.extract_subtitles(media, programme_id)
+
+        formats = []
+        subtitles = None
+
+        for media in self._extract_medias(media_selection):
+            kind = media.get('kind')
+            if kind == 'audio':
+                formats.extend(self._extract_audio(media, programme_id))
+            elif kind == 'video':
+                formats.extend(self._extract_video(media, programme_id))
+            elif kind == 'captions':
+                subtitles = self.extract_subtitles(media, programme_id)
+
+        return formats, subtitles
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+        webpage = self._download_webpage(url, list_id)
+
+        list_title = self._html_search_regex(r'<title>(.*?)(?:\s*-\s*BBC News)?</title>', webpage, 'list title')
+
+        pubdate = self._html_search_regex(r'"datePublished":\s*"(\d+-\d+-\d+)', webpage, 'date', default=None)
+        if pubdate:
+           pubdate = pubdate.replace('-','')
+
+        ret = []
+        # works with bbc.com/news/something-something-123456 articles
+        matches = re.findall(r"data-media-meta='({[^']+})'", webpage)
+        if not matches:
+           # stubbornly generic extractor for {json with "image":{allvideoshavethis},etc}
+           # in http://www.bbc.com/news/video_and_audio/international
+           matches = re.findall(r"({[^{}]+image\":{[^}]+}[^}]+})", webpage)
+        if not matches:
+           raise ExtractorError('No video found', expected=True)
+
+        for ent in matches:
+            jent = self._parse_json(ent,list_id)
+
+            programme_id = jent.get('externalId',None)
+            xml_url = jent.get('href', None)
+
+            title = jent['caption']
+            duration = self._duration_str2int(jent.get('duration',None))
+            description = list_title + ' - ' + jent.get('caption','')
+            thumbnail = None
+            if jent.has_key('image'):
+               thumbnail=jent['image'].get('href',None)
+
+            if programme_id:
+               formats, subtitles = self._download_media_selector(programme_id)
+            elif xml_url:
+               # Cheap fallback
+               # http://playlists.bbc.co.uk/news/(list_id)[ABC..]/playlist.sxml
+               xml = self._download_webpage(xml_url, programme_id, 'Downloading playlist.sxml for externalId (fallback)')
+               programme_id = self._search_regex(r'<mediator [^>]*identifier="(.+?)"', xml, 'playlist.sxml (externalId fallback)')
+               formats, subtitles = self._download_media_selector(programme_id)
+            else:
+               raise ExtractorError('data-media-meta entry has no externalId or href value.')
+               
+            self._sort_formats(formats)
+
+            ret.append( {
+                'id': programme_id,
+                'uploader': 'BBC News',
+                'upload_date': pubdate,
+                'title': title,
+                'description': description,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'formats': formats,
+                'subtitles': subtitles,
+            } )
+
+        if len(ret) > 0:
+           return self.playlist_result(ret, list_id, list_title)
+        raise ExtractorError('No video found', expected=True)