Merge remote-tracking branch 'Boris-de/wdrmaus_fix#8562'

[youtube-dl] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index a148f78ce3e60abb429d3cddfa76bc2ad8cff37f..02dbef91335a7b7cd40caac1be55ca908107c7a8 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -22,7 +22,6 @@ from ..utils import (
     xpath_with_ns,
     mimetype2ext,
     find_xpath_attr,
-    update_url_query,
 )
 
 default_ns = 'http://www.w3.org/2005/SMIL21/Language'
@@ -31,12 +30,10 @@ _x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 
 class ThePlatformBaseIE(OnceIE):
     def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
-        smil_url = update_url_query(smil_url, {'format': 'SMIL'})
-        meta = self._download_xml(smil_url, video_id, note=note)
-        error_element = find_xpath_attr(
-            meta, _x('.//smil:ref'), 'src',
-            'http://link.theplatform.com/s/errorFiles/Unavailable.mp4')
-        if error_element is not None:
+        meta = self._download_xml(smil_url, video_id, note=note, query={'format': 'SMIL'})
+        error_element = find_xpath_attr(meta, _x('.//smil:ref'), 'src')
+        if error_element is not None and error_element.attrib['src'].startswith(
+                'http://link.theplatform.com/s/errorFiles/Unavailable.'):
             raise ExtractorError(error_element.attrib['abstract'], expected=True)
 
         smil_formats = self._parse_smil_formats(
@@ -53,8 +50,6 @@ class ThePlatformBaseIE(OnceIE):
             else:
                 formats.append(_format)
 
-        self._sort_formats(formats)
-
         subtitles = self._parse_smil_subtitles(meta, default_ns)
 
         return formats, subtitles
@@ -79,13 +74,15 @@ class ThePlatformBaseIE(OnceIE):
             'description': info['description'],
             'thumbnail': info['defaultThumbnailUrl'],
             'duration': int_or_none(info.get('duration'), 1000),
+            'timestamp': int_or_none(info.get('pubDate'), 1000) or None,
+            'uploader': info.get('billingCode'),
         }
 
 
 class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
-           (?:(?P<media>(?:(?:[^/]+/)+select/)?media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
+           (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TESTS = [{
@@ -97,6 +94,9 @@ class ThePlatformIE(ThePlatformBaseIE):
             'title': 'Blackberry\'s big, bold Z30',
             'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
             'duration': 247,
+            'timestamp': 1383239700,
+            'upload_date': '20131031',
+            'uploader': 'CBSI-NEW',
         },
         'params': {
             # rtmp download
@@ -110,6 +110,9 @@ class ThePlatformIE(ThePlatformBaseIE):
             'ext': 'flv',
             'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
             'title': 'Tesla Model S: A second step towards a cleaner motoring future',
+            'timestamp': 1426176191,
+            'upload_date': '20150312',
+            'uploader': 'CBSI-NEW',
         },
         'params': {
             # rtmp download
@@ -122,6 +125,7 @@ class ThePlatformIE(ThePlatformBaseIE):
             'ext': 'mp4',
             'description': 'md5:644ad9188d655b742f942bf2e06b002d',
             'title': 'HIGHLIGHTS: USA bag first ever series Cup win',
+            'uploader': 'EGSM',
         }
     }, {
         'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7',
@@ -138,6 +142,7 @@ class ThePlatformIE(ThePlatformBaseIE):
             'thumbnail': 're:^https?://.*\.jpg$',
             'timestamp': 1435752600,
             'upload_date': '20150701',
+            'uploader': 'NBCU-NEWS',
         },
     }, {
         # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1
@@ -146,6 +151,22 @@ class ThePlatformIE(ThePlatformBaseIE):
         'only_matching': True,
     }]
 
+    @classmethod
+    def _extract_urls(cls, webpage):
+        m = re.search(
+            r'''(?x)
+                    <meta\s+
+                        property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+
+                        content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2
+            ''', webpage)
+        if m:
+            return [m.group('url')]
+
+        matches = re.findall(
+            r'<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
+        if matches:
+            return list(zip(*matches))[1]
+
     @staticmethod
     def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
         flags = '10' if include_qs else '00'
@@ -154,11 +175,11 @@ class ThePlatformIE(ThePlatformBaseIE):
         def str_to_hex(str):
             return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
 
-        def hex_to_str(hex):
-            return binascii.a2b_hex(hex)
+        def hex_to_bytes(hex):
+            return binascii.a2b_hex(hex.encode('ascii'))
 
-        relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0]
-        clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
+        relative_path = re.match(r'https?://link.theplatform.com/s/([^?]+)', url).group(1)
+        clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
         checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
         sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
         return '%s&sig=%s' % (url, sig)
@@ -173,10 +194,10 @@ class ThePlatformIE(ThePlatformBaseIE):
         if not provider_id:
             provider_id = 'dJ5BDC'
 
-        path = provider_id
+        path = provider_id + '/'
         if mobj.group('media'):
-            path += '/media'
-        path += '/' + video_id
+            path += mobj.group('media')
+        path += video_id
 
         qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
         if 'guid' in qs_dict:
@@ -234,6 +255,7 @@ class ThePlatformIE(ThePlatformBaseIE):
             smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
 
         formats, subtitles = self._extract_theplatform_smil(smil_url, video_id)
+        self._sort_formats(formats)
 
         ret = self.get_metadata(path, video_id)
         combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles)
@@ -263,6 +285,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
             'timestamp': 1391824260,
             'duration': 467.0,
             'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
+            'uploader': 'NBCU-NEWS',
         },
     }