typo

app.py

@@ -337,7 +337,64 @@ def extract_notes_for_comparison(data: Union[str, dict]) -> list[str]:
 
 from rapidfuzz import fuzz
 
-notes': ', '.join(unique_matched_notes),
+def find_best_perfumes_from_json(data: Union[str, dict], top_n: int = 5, threshold: int = 80) -> pd.DataFrame:
+    """
+    Finds top N matching perfumes using fuzzy matching on notes, with purity factor.
+
+    Args:
+        data (Union[str, dict]): The input JSON or dict.
+        top_n (int): Number of results to return.
+        threshold (int): Minimum fuzz ratio for note match.
+
+    Returns:
+        pd.DataFrame: Matching perfumes.
+    """
+    try:
+        user_notes = extract_notes_for_comparison(data)
+    except (KeyError, ValueError) as e:
+        return pd.DataFrame([{
+            'brand': 'N/A',
+            'perfume': 'N/A',
+            'matching_notes': str(e),
+            'match_count': 0,
+            'similarity_score': 0,
+            'purity': 0,
+            'adjusted_score': 0
+        }])
+
+    user_notes_clean = [n.strip().lower() for n in user_notes]
+
+    matches = []
+
+    for _, row in df.iterrows():
+        perfume_notes = [n.strip().lower() for n in row['notes'].split(',')]
+        total_perfume_notes = len(perfume_notes)
+
+        matched_notes = []
+        total_ratio = 0
+
+        for u_note in user_notes_clean:
+            best_ratio = 0
+            best_p_note = None
+            for p_note in perfume_notes:
+                ratio = fuzz.partial_ratio(u_note, p_note)
+                if ratio > best_ratio:
+                    best_ratio = ratio
+                    best_p_note = p_note
+            if best_ratio >= threshold and best_p_note:
+                matched_notes.append(best_p_note)
+                total_ratio += best_ratio
+
+        unique_matched_notes = sorted(set(matched_notes))
+        match_count = len(unique_matched_notes)
+
+        purity = match_count / total_perfume_notes if total_perfume_notes > 0 else 0
+        adjusted_score = purity * total_ratio
+
+        matches.append({
+            'brand': row['brand'],
+            'perfume': row['perfume'],
+            'matching_notes': ', '.join(unique_matched_notes),
             'match_count': match_count,
             'similarity_score': round(total_ratio, 2),
             'purity': round(purity, 2),