X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcommon.py;h=b8bbaf81a22aab883b1740f25edd91b78473916c;hb=d0de6a287ac64b52630e5b21c3db52a4312675fd;hp=394f34372ea414bfe9d6e44b16a465d54c455b0d;hpb=d391b7e23d3d6c2af03c6329b4bf059ec095f33d;p=youtube-dl

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 394f34372..b8bbaf81a 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -52,6 +52,7 @@ from ..utils import (
     GeoUtils,
     int_or_none,
     js_to_json,
+    JSON_LD_RE,
     mimetype2ext,
     orderedSet,
     parse_codecs,
@@ -1149,8 +1150,7 @@ class InfoExtractor(object):
 
     def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
         json_ld = self._search_regex(
-            r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
-            html, 'JSON-LD', group='json_ld', **kwargs)
+            JSON_LD_RE, html, 'JSON-LD', group='json_ld', **kwargs)
         default = kwargs.get('default', NO_DEFAULT)
         if not json_ld:
             return default if default is not NO_DEFAULT else {}
@@ -1859,9 +1859,7 @@ class InfoExtractor(object):
                         'height': height,
                     })
                 formats.extend(m3u8_formats)
-                continue
-
-            if src_ext == 'f4m':
+            elif src_ext == 'f4m':
                 f4m_url = src_url
                 if not f4m_params:
                     f4m_params = {
@@ -1871,9 +1869,13 @@ class InfoExtractor(object):
                 f4m_url += '&' if '?' in f4m_url else '?'
                 f4m_url += compat_urllib_parse_urlencode(f4m_params)
                 formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
-                continue
-
-            if src_url.startswith('http') and self._is_valid_url(src, video_id):
+            elif src_ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    src_url, video_id, mpd_id='dash', fatal=False))
+            elif re.search(r'\.ism/[Mm]anifest', src_url):
+                formats.extend(self._extract_ism_formats(
+                    src_url, video_id, ism_id='mss', fatal=False))
+            elif src_url.startswith('http') and self._is_valid_url(src, video_id):
                 http_count += 1
                 formats.append({
                     'url': src_url,
@@ -1884,7 +1886,6 @@ class InfoExtractor(object):
                     'width': width,
                     'height': height,
                 })
-                continue
 
         return formats
 
@@ -2106,7 +2107,21 @@ class InfoExtractor(object):
                         representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
 
                         def prepare_template(template_name, identifiers):
-                            t = representation_ms_info[template_name]
+                            tmpl = representation_ms_info[template_name]
+                            # First of, % characters outside $...$ templates
+                            # must be escaped by doubling for proper processing
+                            # by % operator string formatting used further (see
+                            # https://github.com/rg3/youtube-dl/issues/16867).
+                            t = ''
+                            in_template = False
+                            for c in tmpl:
+                                t += c
+                                if c == '$':
+                                    in_template = not in_template
+                                elif c == '%' and not in_template:
+                                    t += c
+                            # Next, $...$ templates are translated to their
+                            # %(...) counterparts to be used with % operator
                             t = t.replace('$RepresentationID$', representation_id)
                             t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t)
                             t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t)
@@ -2437,6 +2452,8 @@ class InfoExtractor(object):
                         media_info['subtitles'].setdefault(lang, []).append({
                             'url': absolute_url(src),
                         })
+            for f in media_info['formats']:
+                f.setdefault('http_headers', {})['Referer'] = base_url
             if media_info['formats'] or media_info['subtitles']:
                 entries.append(media_info)
         return entries