[eitb] Improve, make more robust and extract f4m formats (Closes #7328)

youtube_dl/extractor/eitb.py

@@ -4,14 +4,15 @@
 from .common import InfoExtractor
 from ..compat import compat_urllib_request
 from ..utils import (
+    float_or_none,
     int_or_none,
-    unified_strdate,
+    parse_iso8601,
 )
 
 
 class EitbIE(InfoExtractor):
     IE_NAME = 'eitb.tv'
-    _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/\d+/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?eitb\.tv/(?:eu/bideoa|es/video)/[^/]+/\d+/(?P<id>\d+)'
 
     _TEST = {
         'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/4104995148001/4090227752001/lasa-y-zabala-30-anos/',
@@ -20,43 +21,71 @@ class EitbIE(InfoExtractor):
             'id': '4090227752001',
             'ext': 'mp4',
             'title': '60 minutos (Lasa y Zabala, 30 años)',
-            'description': '',
+            'description': 'Programa de reportajes de actualidad.',
-            'duration': 3996760,
+            'duration': 3996.76,
+            'timestamp': 1381789200,
             'upload_date': '20131014',
+            'tags': list,
         },
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json('http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id, video_id)['web_media'][0]
+
+        video = self._download_json(
+            'http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id,
+            video_id, 'Downloading video JSON')
+
+        media = video['web_media'][0]
 
         formats = []
-        for rendition in video_data['RENDITIONS']:
+        for rendition in media['RENDITIONS']:
+            video_url = rendition.get('PMD_URL')
+            if not video_url:
+                continue
+            tbr = float_or_none(rendition.get('ENCODING_RATE'), 1000)
+            format_id = 'http'
+            if tbr:
+                format_id += '-%d' % int(tbr)
             formats.append({
                 'url': rendition['PMD_URL'],
+                'format_id': format_id,
                 'width': int_or_none(rendition.get('FRAME_WIDTH')),
                 'height': int_or_none(rendition.get('FRAME_HEIGHT')),
-                'tbr': int_or_none(rendition.get('ENCODING_RATE')),
+                'tbr': tbr,
             })
 
-        # TODO: parse f4m manifest
+        hls_url = media.get('HLS_SURL')
-        request = compat_urllib_request.Request(
+        if hls_url:
-            'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/',
+            request = compat_urllib_request.Request(
-            headers={'Referer': url})
+                'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/',
-        token_data = self._download_json(request, video_id, fatal=False)
+                headers={'Referer': url})
-        if token_data:
+            token_data = self._download_json(
-            m3u8_formats = self._extract_m3u8_formats('%s?hdnts=%s' % (video_data['HLS_SURL'], token_data['token']), video_id, m3u8_id='hls', fatal=False)
+                request, video_id, 'Downloading auth token', fatal=False)
-            if m3u8_formats:
+            if token_data:
-                formats.extend(m3u8_formats)
+                token = token_data.get('token')
+                if token:
+                    m3u8_formats = self._extract_m3u8_formats(
+                        '%s?hdnts=%s' % (hls_url, token), video_id, m3u8_id='hls', fatal=False)
+                    if m3u8_formats:
+                        formats.extend(m3u8_formats)
+
+        hds_url = media.get('HDS_SURL').replace('euskalsvod', 'euskalvod')
+        if hds_url:
+            f4m_formats = self._extract_f4m_formats(
+                '%s?hdcore=3.7.0' % hds_url, video_id, f4m_id='hds', fatal=False)
+            if f4m_formats:
+                formats.extend(f4m_formats)
 
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': video_data['NAME_ES'],
+            'title': media.get('NAME_ES') or media.get('name') or media['NAME_EU'],
-            'description': video_data.get('SHORT_DESC_ES'),
+            'description': media.get('SHORT_DESC_ES') or video.get('desc_group') or media.get('SHORT_DESC_EU'),
-            'thumbnail': video_data.get('STILL_URL'),
+            'thumbnail': media.get('STILL_URL') or media.get('THUMBNAIL_URL'),
-            'duration': int_or_none(video_data.get('LENGTH')),
+            'duration': float_or_none(media.get('LENGTH'), 1000),
-            'upload_date': unified_strdate(video_data.get('BROADCST_DATE')),
+            'timestamp': parse_iso8601(media.get('BROADCST_DATE'), ' '),
+            'tags': media.get('TAGS'),
             'formats': formats,
         }