[extractor/common] Allow angle brackets in attributes in _og_regexes (#7215)

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index b928e24beed9cd161b2fa84c9b410980268c0061..4365077f1379e49bc96145e28df4d47f58b91ecb 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -165,6 +165,7 @@ class InfoExtractor(object):
                     with the "ext" entry and one of:
                         * "data": The subtitles file contents
                         * "url": A URL pointing to the subtitles file
+                    "ext" will be calculated from URL if missing
     automatic_captions: Like 'subtitles', used by the YoutubeIE for
                     automatically generated captions
     duration:       Length of the video in seconds, as an integer.
@@ -644,8 +645,9 @@ class InfoExtractor(object):
     # Helper functions for extracting OpenGraph info
     @staticmethod
     def _og_regexes(prop):
-        content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
-        property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
+        content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
+        property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
+                       % {'prop': re.escape(prop)})
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),
@@ -870,13 +872,18 @@ class InfoExtractor(object):
         time.sleep(timeout)
 
     def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
-                             transform_source=lambda s: fix_xml_ampersands(s).strip()):
+                             transform_source=lambda s: fix_xml_ampersands(s).strip(),
+                             fatal=True):
         manifest = self._download_xml(
             manifest_url, video_id, 'Downloading f4m manifest',
             'Unable to download f4m manifest',
             # Some manifests may be malformed, e.g. prosiebensat1 generated manifests
             # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244)
-            transform_source=transform_source)
+            transform_source=transform_source,
+            fatal=fatal)
+
+        if manifest is False:
+            return manifest
 
         formats = []
         manifest_version = '1.0'
@@ -897,7 +904,10 @@ class InfoExtractor(object):
                 # may differ leading to inability to resolve the format by requested
                 # bitrate in f4m downloader
                 if determine_ext(manifest_url) == 'f4m':
-                    formats.extend(self._extract_f4m_formats(manifest_url, video_id, preference, f4m_id))
+                    f4m_formats = self._extract_f4m_formats(
+                        manifest_url, video_id, preference, f4m_id, fatal=fatal)
+                    if f4m_formats:
+                        formats.extend(f4m_formats)
                     continue
             tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
@@ -1129,8 +1139,10 @@ class InfoExtractor(object):
             src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
 
             if proto == 'm3u8' or src_ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    src_url, video_id, ext or 'mp4', m3u8_id='hls'))
+                m3u8_formats = self._extract_m3u8_formats(
+                    src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
                 continue
 
             if src_ext == 'f4m':
@@ -1142,7 +1154,9 @@ 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'))
+                f4m_formats = self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)
+                if f4m_formats:
+                    formats.extend(f4m_formats)
                 continue
 
             if src_url.startswith('http') and self._is_valid_url(src, video_id):