From e7e4a6e0f9166cee82c165ca69a6a3c94ddc5f45 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 28 Apr 2018 02:48:03 +0700 Subject: [PATCH] [extractor/common] Extract interaction statistic --- youtube_dl/extractor/common.py | 35 ++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index e0c3c8eb0c..a9939b0fdc 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -1020,6 +1020,40 @@ def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None): if isinstance(json_ld, dict): json_ld = [json_ld] + INTERACTION_TYPE_MAP = { + 'CommentAction': 'comment', + 'AgreeAction': 'like', + 'DisagreeAction': 'dislike', + 'LikeAction': 'like', + 'DislikeAction': 'dislike', + 'ListenAction': 'view', + 'WatchAction': 'view', + 'ViewAction': 'view', + } + + def extract_interaction_statistic(e): + interaction_statistic = e.get('interactionStatistic') + if not isinstance(interaction_statistic, list): + return + for is_e in interaction_statistic: + if not isinstance(is_e, dict): + continue + if is_e.get('@type') != 'InteractionCounter': + continue + interaction_type = is_e.get('interactionType') + if not isinstance(interaction_type, compat_str): + continue + interaction_count = int_or_none(is_e.get('userInteractionCount')) + if interaction_count is None: + continue + count_kind = INTERACTION_TYPE_MAP.get(interaction_type.split('/')[-1]) + if not count_kind: + continue + count_key = '%s_count' % count_kind + if info.get(count_key) is not None: + continue + info[count_key] = interaction_count + def extract_video_object(e): assert e['@type'] == 'VideoObject' info.update({ @@ -1035,6 +1069,7 @@ def extract_video_object(e): 'height': int_or_none(e.get('height')), 'view_count': int_or_none(e.get('interactionCount')), }) + extract_interaction_statistic(e) for e in json_ld: if isinstance(e.get('@context'), compat_str) and re.match(r'^https?://schema.org/?$', e.get('@context')):