[ie/tiktok] Extract save_count (#15054)

Closes #15053
Authored by: pomtnp

README.md

@@ -1351,6 +1351,7 @@ The available fields are:
  - `repost_count` (numeric): Number of reposts of the video
  - `average_rating` (numeric): Average rating given by users, the scale used depends on the webpage
  - `comment_count` (numeric): Number of comments on the video (For some extractors, comments are only downloaded at the end, and so this field cannot be used)
+ - `save_count` (numeric): Number of times the video has been saved or bookmarked
  - `age_limit` (numeric): Age restriction for the video (years)
  - `live_status` (string): One of "not_live", "is_live", "is_upcoming", "was_live", "post_live" (was live, but VOD is not yet processed)
  - `is_live` (boolean): Whether this video is a live stream or a fixed-length video

test/helper.py

@@ -263,7 +263,7 @@ def expect_info_dict(self, got_dict, expected_dict):
         # NB: Keep in sync with the docstring of extractor/common.py
         'ie_key', 'url', 'id', 'ext', 'direct', 'display_id', 'title', 'alt_title', 'description', 'media_type',
         'uploader', 'uploader_id', 'uploader_url', 'channel', 'channel_id', 'channel_url', 'channel_is_verified',
-        'channel_follower_count', 'comment_count', 'view_count', 'concurrent_view_count',
+        'channel_follower_count', 'comment_count', 'view_count', 'concurrent_view_count', 'save_count',
         'like_count', 'dislike_count', 'repost_count', 'average_rating', 'age_limit', 'duration', 'thumbnail', 'heatmap',
         'chapters', 'chapter', 'chapter_number', 'chapter_id', 'start_time', 'end_time', 'section_start', 'section_end',
         'categories', 'tags', 'cast', 'composers', 'artists', 'album_artists', 'creators', 'genres',

yt_dlp/YoutubeDL.py

@@ -595,7 +595,7 @@ class YoutubeDL:
         'width', 'height', 'asr', 'audio_channels', 'fps',
         'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
         'timestamp', 'release_timestamp', 'available_at',
-        'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
+        'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', 'save_count',
         'average_rating', 'comment_count', 'age_limit',
         'start_time', 'end_time',
         'chapter_number', 'season_number', 'episode_number',

yt_dlp/extractor/common.py

@@ -348,6 +348,7 @@ class InfoExtractor:
     duration:       Length of the video in seconds, as an integer or float.
     view_count:     How many users have watched the video on the platform.
     concurrent_view_count: How many users are currently watching the video on the platform.
+    save_count:     Number of times the video has been saved or bookmarked
     like_count:     Number of positive ratings of the video
     dislike_count:  Number of negative ratings of the video
     repost_count:   Number of reposts of the video

yt_dlp/extractor/tiktok.py

@@ -454,6 +454,7 @@ class TikTokBaseIE(InfoExtractor):
                 'like_count': 'digg_count',
                 'repost_count': 'share_count',
                 'comment_count': 'comment_count',
+                'save_count': 'collect_count',
             }, expected_type=int_or_none),
             **author_info,
             'channel_url': format_field(author_info, 'channel_id', self._UPLOADER_URL_FORMAT, default=None),
@@ -607,6 +608,7 @@ class TikTokBaseIE(InfoExtractor):
                 'like_count': 'diggCount',
                 'repost_count': 'shareCount',
                 'comment_count': 'commentCount',
+                'save_count': 'collectCount',
             }), expected_type=int_or_none),
             'thumbnails': [
                 {
@@ -646,6 +648,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'artist': 'Ysrbeats',
             'album': 'Lehanga',
             'track': 'Lehanga',
@@ -675,6 +678,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'artists': ['Evan Todd', 'Jessica Keenan Wynn', 'Alice Lee', 'Barrett Wilbert Weed', 'Jon Eidson'],
             'track': 'Big Fun',
         },
@@ -702,6 +706,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
         },
     }, {
         # Sponsored video, only available with feed workaround
@@ -725,6 +730,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
         },
         'skip': 'This video is unavailable',
     }, {
@@ -751,6 +757,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
         },
     }, {
         # hydration JSON is sent in a <script> element
@@ -773,6 +780,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
         },
         'skip': 'This video is unavailable',
     }, {
@@ -798,6 +806,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:^https://.+\.(?:webp|jpe?g)',
         },
     }, {
@@ -824,6 +833,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:^https://.+',
             'thumbnails': 'count:3',
         },
@@ -851,6 +861,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:^https://.+\.webp',
         },
         'skip': 'Unavailable via feed API, only audio available via web',
@@ -879,6 +890,7 @@ class TikTokIE(TikTokBaseIE):
             'like_count': int,
             'comment_count': int,
             'repost_count': int,
+            'save_count': int,
             'thumbnail': r're:^https://.+\.(?:webp|jpe?g)',
         },
     }, {
@@ -1288,6 +1300,7 @@ class DouyinIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:https?://.+\.jpe?g',
         },
     }, {
@@ -1312,6 +1325,7 @@ class DouyinIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:https?://.+\.jpe?g',
         },
     }, {
@@ -1336,6 +1350,7 @@ class DouyinIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:https?://.+\.jpe?g',
         },
     }, {
@@ -1353,6 +1368,7 @@ class DouyinIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
         },
         'skip': 'No longer available',
     }, {
@@ -1377,6 +1393,7 @@ class DouyinIE(TikTokBaseIE):
             'like_count': int,
             'repost_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:https?://.+\.jpe?g',
         },
     }]
@@ -1437,6 +1454,7 @@ class TikTokVMIE(InfoExtractor):
             'view_count': int,
             'like_count': int,
             'comment_count': int,
+            'save_count': int,
             'thumbnail': r're:https://.+\.webp.*',
             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAdZ_NcPPgMneaGrW0hN8O_J_bwLshwNNERRF5DxOw2HKIzk0kdlLrR8RkVl1ksrMO',
             'duration': 29,