fix #656: attempt to find runs in subtitle if not in title (#657)

* fix #656: attempt to find runs in subtitle if not in title

* refactor

* remove fix by gotofritz

---------

Co-authored-by: sigma67 <sigma67.github@gmail.com>

tests/conftest.py

@@ -34,12 +34,6 @@ def fixture_sample_video() -> str:
     return "hpSrLjc5SMs"
 
 
-@pytest.fixture(name="sample_playlist")
-def fixture_sample_playlist() -> str:
-    """very large playlist"""
-    return "PL6bPxvf5dW5clc3y9wAoslzqUrmkZ5c-u"
-
-
 @pytest.fixture(name="browser_filepath")
 def fixture_browser_filepath(config) -> str:
     return get_resource(config["auth"]["browser_file"])

tests/mixins/test_playlists.py

@@ -42,32 +42,33 @@ def test_get_playlist_2024(self, yt, test_file, owned):
                 if track["videoType"] == "MUSIC_VIDEO_TYPE_ATV":
                     assert isinstance(track["album"]["name"], str) and track["album"]["name"]
 
-    def test_get_playlist_foreign(self, yt_empty, yt_oauth, sample_playlist):
-        with pytest.raises(Exception):
-            yt_empty.get_playlist("PLABC")
-        playlist = yt_empty.get_playlist(
-            "RDCLAK5uy_nfjzC9YC1NVPPZHvdoAtKVBOILMDOuxOs", limit=300, suggestions_limit=7
-        )
+    @pytest.mark.parametrize(
+        "playlist_id, tracks_len, related_len",
+        [
+            ("RDCLAK5uy_nfjzC9YC1NVPPZHvdoAtKVBOILMDOuxOs", 200, 10),
+            ("PLj4BSJLnVpNyIjbCWXWNAmybc97FXLlTk", 200, 0),  # no related tracks
+            ("PL6bPxvf5dW5clc3y9wAoslzqUrmkZ5c-u", 1000, 10),  # very large
+            ("PLZ6Ih9wLHQ2Hm2d3Cb0iV48Z2hQjGRyNz", 300, 10),  # runs in subtitle, not title
+        ],
+    )
+    def test_get_playlist_foreign(self, yt_oauth, playlist_id, tracks_len, related_len):
+        playlist = yt_oauth.get_playlist(playlist_id, limit=None, related=True)
         assert len(playlist["duration"]) > 5
-        assert playlist["trackCount"] > 200
-        assert len(playlist["tracks"]) > 200
+        assert playlist["trackCount"] > tracks_len
+        assert len(playlist["tracks"]) > tracks_len
+        assert len(playlist["related"]) == related_len
         assert "suggestions" not in playlist
         assert playlist["owned"] is False
 
+    def test_get_playlist_empty(self, yt_empty):
+        with pytest.raises(Exception):
+            yt_empty.get_playlist("PLABC")
+
+    def test_get_playlist_no_track_count(self, yt_oauth):
         playlist = yt_oauth.get_playlist("RDATgXd-")
         assert playlist["trackCount"] is None  # playlist has no trackCount
         assert len(playlist["tracks"]) >= 100
 
-        playlist = yt_oauth.get_playlist("PLj4BSJLnVpNyIjbCWXWNAmybc97FXLlTk", limit=None, related=True)
-        assert playlist["trackCount"] > 200
-        assert len(playlist["tracks"]) > 200
-        assert len(playlist["related"]) == 0
-
-        playlist = yt_oauth.get_playlist(sample_playlist, limit=1100)
-        assert playlist["trackCount"] > 1000
-        assert len(playlist["tracks"]) > 1000
-        assert len(playlist["related"]) == 0
-
     @pytest.mark.parametrize("language", SUPPORTED_LANGUAGES)
     def test_get_playlist_languages(self, language):
         yt = YTMusic(language=language)

ytmusicapi/mixins/playlists.py

@@ -134,28 +134,10 @@ def get_playlist(
             if description_shelf
             else None
         )
-        playlist["thumbnails"] = nav(header, THUMBNAILS)
-        playlist["title"] = nav(header, TITLE_TEXT)
-        playlist.update(parse_song_runs(nav(header, SUBTITLE_RUNS)[2 + playlist["owned"] * 2 :]))
 
-        playlist["views"] = None
-        playlist["duration"] = None
-        if "runs" in header["secondSubtitle"]:
-            second_subtitle_runs = header["secondSubtitle"]["runs"]
-            has_views = (len(second_subtitle_runs) > 3) * 2
-            playlist["views"] = None if not has_views else to_int(second_subtitle_runs[0]["text"])
-            has_duration = (len(second_subtitle_runs) > 1) * 2
-            playlist["duration"] = (
-                None if not has_duration else second_subtitle_runs[has_views + has_duration]["text"]
-            )
-            song_count_text = second_subtitle_runs[has_views + 0]["text"]
-            song_count_search = re.findall(r"\d+", song_count_text)
-            # extract the digits from the text, return 0 if no match
-            song_count = to_int("".join(song_count_search)) if song_count_search is not None else None
-        else:
-            song_count = None
+        playlist.update(parse_playlist_header_meta(header))
 
-        playlist["trackCount"] = song_count
+        playlist.update(parse_song_runs(nav(header, SUBTITLE_RUNS)[2 + playlist["owned"] * 2 :]))
 
         request_func = lambda additionalParams: self._send_request(endpoint, body, additionalParams)
 

ytmusicapi/parsers/playlists.py

@@ -19,10 +19,9 @@ def parse_playlist_header(response: dict) -> dict[str, Any]:
                 response, [*TWO_COLUMN_RENDERER, *TAB_CONTENT, *SECTION_LIST_ITEM, *RESPONSIVE_HEADER]
             )
 
-    playlist["title"] = nav(header, TITLE_TEXT)
-    playlist["thumbnails"] = nav(header, THUMBNAIL_CROPPED, True)
+    playlist.update(parse_playlist_header_meta(header))
     if playlist["thumbnails"] is None:
-        playlist["thumbnails"] = nav(header, THUMBNAILS)
+        playlist["thumbnails"] = nav(header, THUMBNAIL_CROPPED, True)
     playlist["description"] = nav(header, DESCRIPTION, True)
     run_count = len(nav(header, SUBTITLE_RUNS))
     if run_count > 1:
@@ -33,24 +32,33 @@ def parse_playlist_header(response: dict) -> dict[str, Any]:
         if run_count == 5:
             playlist["year"] = nav(header, SUBTITLE3)
 
-    playlist["views"] = None
-    playlist["duration"] = None
-    playlist["trackCount"] = None
+    return playlist
+
+
+def parse_playlist_header_meta(header: dict[str, Any]) -> dict[str, Any]:
+    playlist_meta = {
+        "views": None,
+        "duration": None,
+        "trackCount": None,
+        "title": nav(header, TITLE_TEXT, none_if_absent=True),
+        "thumbnails": nav(header, THUMBNAILS),
+    }
     if "runs" in header["secondSubtitle"]:
         second_subtitle_runs = header["secondSubtitle"]["runs"]
         has_views = (len(second_subtitle_runs) > 3) * 2
-        playlist["views"] = None if not has_views else to_int(second_subtitle_runs[0]["text"])
+        playlist_meta["views"] = None if not has_views else to_int(second_subtitle_runs[0]["text"])
         has_duration = (len(second_subtitle_runs) > 1) * 2
-        playlist["duration"] = (
+        playlist_meta["duration"] = (
             None if not has_duration else second_subtitle_runs[has_views + has_duration]["text"]
         )
         song_count_text = second_subtitle_runs[has_views + 0]["text"]
-        song_count_search = re.search(r"\d+", song_count_text)
+        song_count_search = re.findall(r"\d+", song_count_text)
         # extract the digits from the text, return 0 if no match
-        song_count = to_int(song_count_search.group()) if song_count_search is not None else 0
-        playlist["trackCount"] = song_count
+        playlist_meta["trackCount"] = (
+            to_int("".join(song_count_search)) if song_count_search is not None else None
+        )
 
-    return playlist
+    return playlist_meta
 
 
 def parse_playlist_items(results, menu_entries: Optional[list[list]] = None, is_album=False):