From d81280607e90ae510949b717782d197375224c76 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 19 Sep 2020 06:33:17 +0700 Subject: [PATCH] [extractor/common] Relax interaction count extraction in _json_ld --- youtube_dl/extractor/common.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index c9b8b6337..021945a89 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -68,6 +68,7 @@ from ..utils import ( sanitized_Request, sanitize_filename, str_or_none, + str_to_int, strip_or_none, unescapeHTML, unified_strdate, @@ -1248,7 +1249,10 @@ class InfoExtractor(object): interaction_type = is_e.get('interactionType') if not isinstance(interaction_type, compat_str): continue - interaction_count = int_or_none(is_e.get('userInteractionCount')) + # For interaction count some sites provide string instead of + # an integer (as per spec) with non digit characters (e.g. ",") + # so extracting count with more relaxed str_to_int + interaction_count = str_to_int(is_e.get('userInteractionCount')) if interaction_count is None: continue count_kind = INTERACTION_TYPE_MAP.get(interaction_type.split('/')[-1])