]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/cspan.py
Merge tag 'upstream/2017.03.26'
[youtubedl] / youtube_dl / extractor / cspan.py
index 5411066846eb94b9c9295bae4f8860e07112b2d1..d4576160b4489e599e4ca7dabc1e18c9d685610f 100644 (file)
@@ -7,31 +7,33 @@ from ..utils import (
     int_or_none,
     unescapeHTML,
     find_xpath_attr,
+    smuggle_url,
+    determine_ext,
+    ExtractorError,
 )
+from .senateisvp import SenateISVPIE
+from .ustream import UstreamIE
 
 
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
+    _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
     IE_DESC = 'C-SPAN'
     _TESTS = [{
         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
-        'md5': '8e44ce11f0f725527daccc453f553eb0',
+        'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
         'info_dict': {
             'id': '315139',
-            'ext': 'mp4',
             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
-            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',
         },
+        'playlist_mincount': 2,
         'skip': 'Regularly fails on travis, for unknown reasons',
     }, {
         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
-        # For whatever reason, the served video alternates between
-        # two different ones
-        #'md5': 'dbb0f047376d457f2ab8b3929cbb2d0c',
+        # md5 is unstable
         'info_dict': {
-            'id': '340723',
+            'id': 'c4486943',
             'ext': 'mp4',
-            'title': 'International Health Care Models',
+            'title': 'CSPAN - International Health Care Models',
             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
         }
     }, {
@@ -40,51 +42,127 @@ class CSpanIE(InfoExtractor):
             'id': '342759',
             'title': 'General Motors Ignition Switch Recall',
         },
-        'playlist_duration_sum': 14855,
+        'playlist_mincount': 6,
+    }, {
+        # Video from senate.gov
+        'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
+        'info_dict': {
+            'id': 'judiciary031715',
+            'ext': 'mp4',
+            'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
+        },
+        'params': {
+            'skip_download': True,  # m3u8 downloads
+        }
+    }, {
+        # Ustream embedded video
+        'url': 'https://www.c-span.org/video/?114917-1/armed-services',
+        'info_dict': {
+            'id': '58428542',
+            'ext': 'flv',
+            'title': 'USHR07 Armed Services Committee',
+            'description': 'hsas00-2118-20150204-1000et-07\n\n\nUSHR07 Armed Services Committee',
+            'timestamp': 1423060374,
+            'upload_date': '20150204',
+            'uploader': 'HouseCommittee',
+            'uploader_id': '12987475',
+        },
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        page_id = mobj.group('id')
-        webpage = self._download_webpage(url, page_id)
-        video_id = self._search_regex(r'progid=\'?([0-9]+)\'?>', webpage, 'video id')
+        video_id = self._match_id(url)
+        video_type = None
+        webpage = self._download_webpage(url, video_id)
+
+        ustream_url = UstreamIE._extract_url(webpage)
+        if ustream_url:
+            return self.url_result(ustream_url, UstreamIE.ie_key())
 
-        description = self._html_search_regex(
-            [
-                # The full description
-                r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
-                # If the description is small enough the other div is not
-                # present, otherwise this is a stripped version
-                r'<p class=\'initial\'>(.*?)</p>'
-            ],
-            webpage, 'description', flags=re.DOTALL)
+        # We first look for clipid, because clipprog always appears before
+        patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
+        results = list(filter(None, (re.search(p, webpage) for p in patterns)))
+        if results:
+            matches = results[0]
+            video_type, video_id = matches.groups()
+            video_type = 'clip' if video_type == 'id' else 'program'
+        else:
+            m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
+            if m:
+                video_id = m.group('id')
+                video_type = 'program' if m.group('type') == 'prog' else 'clip'
+            else:
+                senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
+                if senate_isvp_url:
+                    title = self._og_search_title(webpage)
+                    surl = smuggle_url(senate_isvp_url, {'force_title': title})
+                    return self.url_result(surl, 'SenateISVP', video_id, title)
+        if video_type is None or video_id is None:
+            raise ExtractorError('unable to find video id and type')
 
-        info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
-        data = self._download_json(info_url, video_id)
+        def get_text_attr(d, attr):
+            return d.get(attr, {}).get('#text')
+
+        data = self._download_json(
+            'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
+            video_id)['video']
+        if data['@status'] != 'Success':
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
 
         doc = self._download_xml(
-            'http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
+            'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
             video_id)
 
+        description = self._html_search_meta('description', webpage)
+
         title = find_xpath_attr(doc, './/string', 'name', 'title').text
         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 
-        files = data['video']['files']
+        files = data['files']
+        capfile = get_text_attr(data, 'capfile')
 
-        entries = [{
-            'id': '%s_%d' % (video_id, partnum + 1),
-            'title': (
-                title if len(files) == 1 else
-                '%s part %d' % (title, partnum + 1)),
-            'url': unescapeHTML(f['path']['#text']),
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': int_or_none(f.get('length', {}).get('#text')),
-        } for partnum, f in enumerate(files)]
+        entries = []
+        for partnum, f in enumerate(files):
+            formats = []
+            for quality in f['qualities']:
+                formats.append({
+                    'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
+                    'url': unescapeHTML(get_text_attr(quality, 'file')),
+                    'height': int_or_none(get_text_attr(quality, 'height')),
+                    'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
+                })
+            if not formats:
+                path = unescapeHTML(get_text_attr(f, 'path'))
+                if not path:
+                    continue
+                formats = self._extract_m3u8_formats(
+                    path, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
+            self._sort_formats(formats)
+            entries.append({
+                'id': '%s_%d' % (video_id, partnum + 1),
+                'title': (
+                    title if len(files) == 1 else
+                    '%s part %d' % (title, partnum + 1)),
+                'formats': formats,
+                'description': description,
+                'thumbnail': thumbnail,
+                'duration': int_or_none(get_text_attr(f, 'length')),
+                'subtitles': {
+                    'en': [{
+                        'url': capfile,
+                        'ext': determine_ext(capfile, 'dfxp')
+                    }],
+                } if capfile else None,
+            })
 
-        return {
-            '_type': 'playlist',
-            'entries': entries,
-            'title': title,
-            'id': video_id,
-        }
+        if len(entries) == 1:
+            entry = dict(entries[0])
+            entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
+            return entry
+        else:
+            return {
+                '_type': 'playlist',
+                'entries': entries,
+                'title': title,
+                'id': 'c' + video_id if video_type == 'clip' else video_id,
+            }