Efficient database updating

app/model.py

@@ -188,6 +188,12 @@ class Tracks(Base):
             ERROR(f"Can't find track id {id}")
             return None
 
+    @staticmethod
+    def get_all_paths():
+        "Return a list of paths of all tracks"
+
+        return [a[0] for a in session.query(Tracks.path).all()]
+
     @staticmethod
     def get_path(id):
         try:

app/songdb.py

@@ -30,6 +30,26 @@ def main():
     INFO("Finished")
 
 
+def add_path_to_db(path):
+    "Add passed path to database along with metadata"
+
+    track = Tracks.get_or_create(path)
+    tag = TinyTag.get(path)
+    audio = get_audio_segment(path)
+
+    track.title = tag.title
+    track.artist = tag.artist
+    track.duration = int(round(tag.duration,
+                               Config.MILLISECOND_SIGFIGS) * 1000)
+    track.start_gap = leading_silence(audio)
+    track.fade_at = round(fade_point(audio) / 1000,
+                          Config.MILLISECOND_SIGFIGS) * 1000
+    track.silence_at = round(trailing_silence(audio) / 1000,
+                             Config.MILLISECOND_SIGFIGS) * 1000
+    track.mtime = os.path.getmtime(path)
+    session.commit()
+
+
 def get_audio_segment(path):
     try:
         if path.endswith('.mp3'):
@@ -94,40 +114,27 @@ def trailing_silence(audio_segment, silence_threshold=-50.0,
 def update_db():
     """
     Repopulate database
-
-    TODO: remove missing files
     """
 
-    count = 0
+    # Search for tracks in only one of directory and database
+
+    db_paths = set(Tracks.get_all_paths())
+
+    os_paths_list = []
     for root, dirs, files in os.walk(Config.ROOT):
         for f in files:
-            count += 1
             path = os.path.join(root, f)
             ext = os.path.splitext(f)[1]
             if ext in [".flac", ".mp3"]:
-                track = Tracks.get_or_create(path)
-                tag = TinyTag.get(path)
-                audio = get_audio_segment(path)
+                os_paths_list.append(path)
+    os_paths = set(os_paths_list)
 
-                track.title = tag.title
-                track.artist = tag.artist
-                track.duration = int(round(
-                    tag.duration, Config.MILLISECOND_SIGFIGS
-                ) * 1000)
-                track.start_gap = leading_silence(audio)
-                track.fade_at = round(
-                    fade_point(audio) / 1000, Config.MILLISECOND_SIGFIGS
-                ) * 1000
-                track.silence_at = round(
-                    trailing_silence(audio) / 1000, Config.MILLISECOND_SIGFIGS
-                ) * 1000
-                track.mtime = os.path.getmtime(path)
-                session.commit()
+    for path in list(db_paths - os_paths):
+        INFO(f"To remove from database: {path}")
 
-            elif ext not in [".jpg"]:
-                INFO(f"Unrecognised file type: {path}")
-
-    INFO(f"{count} files processed")
+    for path in list(os_paths - db_paths):
+        INFO(f"Adding to datbase: {path}")
+        add_path_to_db(path)
 
 
 if __name__ == '__main__':