Refactor data loading

data/provider.py

@@ -0,0 +1,9 @@
+class DataProvider:
+    _instance = None
+
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super(DataProvider, cls).__new__(cls)
+
+        return cls._instance
+

data/raw.py

@@ -0,0 +1,67 @@
+import pandas as pd
+
+from utils.path import data_path
+
+
+valid_data_sources = {
+    'album_artist',
+    'albums',
+    'artists',
+    'audio_features',
+    'liked_tracks',
+    'playlist_track',
+    'playlists',
+    'track_artist',
+    'tracks',
+    'artist_genre',
+    'top_tracks',
+    'top_artists'
+}
+
+df_prefixes = {
+    "albums": "album_",
+    "tracks": "track_",
+    "audio_features": "audio_",
+    "playlists": "playlist_",
+    "artists": "artist_"
+}
+
+
+class RawData:
+    _instance = None
+
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super(RawData, cls).__new__(cls)
+
+        return cls._instance
+
+
+    def __init__(self):
+        self._data = {}
+
+
+    def __getitem__(self, key) -> pd.DataFrame:
+        if key not in valid_data_sources:
+            raise RuntimeError(f'Invalid data source {key}')
+
+        if key not in self._data:
+            df = pd.read_csv(data_path(key))
+            prefix = df_prefixes.get(key, None)
+            if prefix is not None:
+                prefix_df(df, prefix, set(df_prefixes.values()))
+
+            self._data[key] = df
+
+        return self._data[key]
+
+
+def prefix_df(df: pd.DataFrame, prefix: str, prefixes: list[str]):
+    df.columns = [prefix_col(col, prefix, prefixes) for col in df.columns]
+
+
+def prefix_col(col: str, prefix: str, prefixes: list[str]):
+    for other_prefix in prefixes:
+        if col.startswith(other_prefix):
+            return col
+    return prefix + col

summarize/summarize.py

@@ -1,5 +1,6 @@
 import pandas as pd
 
+from data.raw import RawData
 from summarize.pages.artist import make_artist_summary
 from summarize.pages.genre import make_genre_summary
 from summarize.pages.label import make_label_summary
@@ -9,31 +10,25 @@
 from utils.album import short_album_name
 from utils.audio_features import set_tracks_full
 from utils.date import release_year
-from utils.path import clear_markdown, data_path
+from utils.path import clear_markdown
 from utils.record_label import standardize_record_labels
-from utils.util import first, prefix_df
+from utils.util import first
 
 
 def summarize_results():
-    album_artist = pd.read_csv(data_path('album_artist'))
-    albums = pd.read_csv(data_path("albums"))
-    artists = pd.read_csv(data_path("artists"))
-    audio_features = pd.read_csv(data_path("audio_features"))
-    liked_tracks = pd.read_csv(data_path("liked_tracks"))
-    playlist_track = pd.read_csv(data_path("playlist_track"))
-    playlists = pd.read_csv(data_path("playlists"))
-    track_artist = pd.read_csv(data_path("track_artist"))
-    tracks = pd.read_csv(data_path("tracks"))
-    artist_genre = pd.read_csv(data_path("artist_genre"))
-    top_tracks = pd.read_csv(data_path("top_tracks"))
-    top_artists = pd.read_csv(data_path("top_artists"))
-
-    prefixes = ["album_", "track_", "playlist_", "artist_"]
-    prefix_df(albums, "album_", prefixes)
-    prefix_df(tracks, "track_", prefixes)
-    prefix_df(audio_features, "audio_", prefixes)
-    prefix_df(playlists, "playlist_", prefixes)
-    prefix_df(artists, "artist_", prefixes)
+    raw_data = RawData()
+    album_artist = raw_data['album_artist']
+    albums = raw_data["albums"]
+    artists = raw_data["artists"]
+    audio_features = raw_data["audio_features"]
+    liked_tracks = raw_data["liked_tracks"]
+    playlist_track = raw_data["playlist_track"]
+    playlists = raw_data["playlists"]
+    track_artist = raw_data["track_artist"]
+    tracks = raw_data["tracks"]
+    artist_genre = raw_data["artist_genre"]
+    top_tracks = raw_data["top_tracks"]
+    top_artists = raw_data["top_artists"]
 
     albums['album_release_year'] = albums['album_release_date'].apply(release_year)
     albums['album_short_name'] = albums['album_name'].apply(short_album_name)

utils/util.py

@@ -20,16 +20,8 @@ def file_name_friendly(text: str):
     return re.sub(r"[^a-z0-9]", "_", text.lower())
 
 
-def prefix_df(df: pd.DataFrame, prefix: str, prefixes: list[str]):
-    df.columns = [prefix_col(col, prefix, prefixes) for col in df.columns]
 
 
-def prefix_col(col: str, prefix: str, prefixes: list[str]):
-    for other_prefix in prefixes:
-        if col.startswith(other_prefix):
-            return col
-    return prefix + col
-
 
 def first(series: pd.Series):
     return None if len(series) == 0 else series.iloc[0]