[xhamster] Extract formats from xplayer settings and extract filesizes (closes #28114)

test/test_youtube_signature.py

@@ -19,55 +19,46 @@ from youtube_dl.compat import compat_str, compat_urlretrieve
 _TESTS = [
     (
         'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
-        'js',
         86,
         '>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
-        'js',
         85,
         '3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
-        'js',
         90,
         ']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js',
-        'js',
         84,
         'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=',
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js',
-        'js',
         '2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA',
         'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2',
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflBb0OQx.js',
-        'js',
         84,
         '123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQ0STUVWXYZ!"#$%&\'()*+,@./:;<=>'
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl9FYC6l.js',
-        'js',
         83,
         '123456789abcdefghijklmnopqr0tuvwxyzABCDETGHIJKLMNOPQRS>UVWXYZ!"#$%&\'()*+,-./:;<=F'
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflCGk6yw/html5player.js',
-        'js',
         '4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288',
         '82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B'
     ),
     (
         'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js',
-        'js',
         '312AA52209E3623129A412D56A40F11CB0AF14AE.3EE09501CB14E3BCDC3B2AE808BF3F1D14E7FBF12',
         '112AA5220913623229A412D56A40F11CB0AF14AE.3EE0950FCB14EEBCDC3B2AE808BF331D14E7FBF3',
     )
@@ -78,6 +69,10 @@ class TestPlayerInfo(unittest.TestCase):
     def test_youtube_extract_player_info(self):
         PLAYER_URLS = (
             ('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/en_US/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/fr_FR/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'),
             # obsolete
             ('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'),
             ('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'),
@@ -100,13 +95,13 @@ class TestSignature(unittest.TestCase):
             os.mkdir(self.TESTDATA_DIR)
 
 
-def make_tfunc(url, stype, sig_input, expected_sig):
+def make_tfunc(url, sig_input, expected_sig):
     m = re.match(r'.*-([a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.[a-z]+$', url)
     assert m, '%r should follow URL format' % url
     test_id = m.group(1)
 
     def test_func(self):
-        basename = 'player-%s.%s' % (test_id, stype)
+        basename = 'player-%s.js' % test_id
         fn = os.path.join(self.TESTDATA_DIR, basename)
 
         if not os.path.exists(fn):
@@ -114,22 +109,16 @@ def make_tfunc(url, stype, sig_input, expected_sig):
 
         ydl = FakeYDL()
         ie = YoutubeIE(ydl)
-        if stype == 'js':
-            with io.open(fn, encoding='utf-8') as testf:
-                jscode = testf.read()
-            func = ie._parse_sig_js(jscode)
-        else:
-            assert stype == 'swf'
-            with open(fn, 'rb') as testf:
-                swfcode = testf.read()
-            func = ie._parse_sig_swf(swfcode)
+        with io.open(fn, encoding='utf-8') as testf:
+            jscode = testf.read()
+        func = ie._parse_sig_js(jscode)
         src_sig = (
             compat_str(string.printable[:sig_input])
             if isinstance(sig_input, int) else sig_input)
         got_sig = func(src_sig)
         self.assertEqual(got_sig, expected_sig)
 
-    test_func.__name__ = str('test_signature_' + stype + '_' + test_id)
+    test_func.__name__ = str('test_signature_js_' + test_id)
     setattr(TestSignature, test_func.__name__, test_func)
 
 

youtube_dl/extractor/xhamster.py

@@ -11,11 +11,14 @@ from ..utils import (
     dict_get,
     extract_attributes,
     ExtractorError,
+    float_or_none,
     int_or_none,
     parse_duration,
+    str_or_none,
     try_get,
     unified_strdate,
     url_or_none,
+    urljoin,
 )
 
 
@@ -146,36 +149,89 @@ class XHamsterIE(InfoExtractor):
             video = initials['videoModel']
             title = video['title']
             formats = []
-            for format_id, formats_dict in video['sources'].items():
+            format_urls = set()
+            format_sizes = {}
+            sources = try_get(video, lambda x: x['sources'], dict) or {}
+            for format_id, formats_dict in sources.items():
                 if not isinstance(formats_dict, dict):
                     continue
+                download_sources = try_get(sources, lambda x: x['download'], dict) or {}
+                for quality, format_dict in download_sources.items():
+                    if not isinstance(format_dict, dict):
+                        continue
+                    format_sizes[quality] = float_or_none(format_dict.get('size'))
                 for quality, format_item in formats_dict.items():
                     if format_id == 'download':
                         # Download link takes some time to be generated,
                         # skipping for now
                         continue
-                        if not isinstance(format_item, dict):
-                            continue
-                        format_url = format_item.get('link')
-                        filesize = int_or_none(
-                            format_item.get('size'), invscale=1000000)
-                    else:
-                        format_url = format_item
-                        filesize = None
+                    format_url = format_item
                     format_url = url_or_none(format_url)
-                    if not format_url:
+                    if not format_url or format_url in format_urls:
                         continue
+                    format_urls.add(format_url)
                     formats.append({
                         'format_id': '%s-%s' % (format_id, quality),
                         'url': format_url,
                         'ext': determine_ext(format_url, 'mp4'),
                         'height': get_height(quality),
-                        'filesize': filesize,
+                        'filesize': format_sizes.get(quality),
                         'http_headers': {
                             'Referer': urlh.geturl(),
                         },
                     })
-            self._sort_formats(formats)
+            xplayer_sources = try_get(
+                initials, lambda x: x['xplayerSettings']['sources'], dict)
+            if xplayer_sources:
+                hls_sources = xplayer_sources.get('hls')
+                if isinstance(hls_sources, dict):
+                    for hls_format_key in ('url', 'fallback'):
+                        hls_url = hls_sources.get(hls_format_key)
+                        if not hls_url:
+                            continue
+                        hls_url = urljoin(url, hls_url)
+                        if not hls_url or hls_url in format_urls:
+                            continue
+                        format_urls.add(hls_url)
+                        formats.extend(self._extract_m3u8_formats(
+                            hls_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                            m3u8_id='hls', fatal=False))
+                standard_sources = xplayer_sources.get('standard')
+                if isinstance(standard_sources, dict):
+                    for format_id, formats_list in standard_sources.items():
+                        if not isinstance(formats_list, list):
+                            continue
+                        for standard_format in formats_list:
+                            if not isinstance(standard_format, dict):
+                                continue
+                            for standard_format_key in ('url', 'fallback'):
+                                standard_url = standard_format.get(standard_format_key)
+                                if not standard_url:
+                                    continue
+                                standard_url = urljoin(url, standard_url)
+                                if not standard_url or standard_url in format_urls:
+                                    continue
+                                format_urls.add(standard_url)
+                                ext = determine_ext(standard_url, 'mp4')
+                                if ext == 'm3u8':
+                                    formats.extend(self._extract_m3u8_formats(
+                                        standard_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                                        m3u8_id='hls', fatal=False))
+                                    continue
+                                quality = (str_or_none(standard_format.get('quality'))
+                                           or str_or_none(standard_format.get('label'))
+                                           or '')
+                                formats.append({
+                                    'format_id': '%s-%s' % (format_id, quality),
+                                    'url': standard_url,
+                                    'ext': ext,
+                                    'height': get_height(quality),
+                                    'filesize': format_sizes.get(quality),
+                                    'http_headers': {
+                                        'Referer': standard_url,
+                                    },
+                                })
+            self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
 
             categories_list = video.get('categories')
             if isinstance(categories_list, list):

youtube_dl/extractor/youtube.py

@@ -398,7 +398,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                      (?(1).+)?                                                # if we found the ID, everything can follow
                      $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
     _PLAYER_INFO_RE = (
-        r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.js$',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
+        r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
         r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
     )
     _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
@@ -1237,6 +1238,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         funcname = self._search_regex(
             (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
              r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+             r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
+             r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
+             r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
              r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
              r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
              # Obsolete patterns