diff options
Diffstat (limited to 'yt_dlp/extractor/common.py')
| -rw-r--r-- | yt_dlp/extractor/common.py | 20 |
1 files changed, 14 insertions, 6 deletions
diff --git a/yt_dlp/extractor/common.py b/yt_dlp/extractor/common.py index ebf2e3cea..9abbaf04f 100644 --- a/yt_dlp/extractor/common.py +++ b/yt_dlp/extractor/common.py @@ -163,9 +163,8 @@ class InfoExtractor(object): * filesize_approx An estimate for the number of bytes * player_url SWF Player URL (used for rtmpdump). * protocol The protocol that will be used for the actual - download, lower-case. - "http", "https", "rtsp", "rtmp", "rtmp_ffmpeg", "rtmpe", - "m3u8", "m3u8_native" or "http_dash_segments". + download, lower-case. One of "http", "https" or + one of the protocols defined in downloader.PROTOCOL_MAP * fragment_base_url Base URL for fragments. Each fragment's path value (if present) will be relative to @@ -181,6 +180,8 @@ class InfoExtractor(object): fragment_base_url * "duration" (optional, int or float) * "filesize" (optional, int) + * is_from_start Is a live format that can be downloaded + from the start. Boolean * preference Order number of this format. If this field is present and not None, the formats get sorted by this field, regardless of all other values. @@ -1451,8 +1452,13 @@ class InfoExtractor(object): }) extract_interaction_statistic(e) - for e in json_ld: - if '@context' in e: + def traverse_json_ld(json_ld, at_top_level=True): + for e in json_ld: + if at_top_level and '@context' not in e: + continue + if at_top_level and set(e.keys()) == {'@context', '@graph'}: + traverse_json_ld(variadic(e['@graph'], allowed_types=(dict,)), at_top_level=False) + break item_type = e.get('@type') if expected_type is not None and expected_type != item_type: continue @@ -1488,7 +1494,7 @@ class InfoExtractor(object): info.update({ 'timestamp': parse_iso8601(e.get('datePublished')), 'title': unescapeHTML(e.get('headline')), - 'description': unescapeHTML(e.get('articleBody')), + 'description': unescapeHTML(e.get('articleBody') or e.get('description')), }) elif item_type == 'VideoObject': extract_video_object(e) @@ -1503,6 +1509,8 @@ class InfoExtractor(object): continue else: break + traverse_json_ld(json_ld) + return dict((k, v) for k, v in info.items() if v is not None) def _search_nextjs_data(self, webpage, video_id, **kw): |