fix脚本添加转写为xlsx

@@ -229,18 +229,49 @@ def main():
     print(f"移除行数: {len(removed)}（同歌手且时长相近）")
     print(f"保留行数: {len(kept)}")
 
-    # 写修复后的 CSV（去掉冗余列）
+    # 写修复后的文件（去掉冗余列）
     DROP_COLS = {"topk_hit", "expected_duplicate", "predicted_duplicate", "sample_class"}
     out_fieldnames = [f for f in fieldnames if f not in DROP_COLS]
     out_path = Path(args.output)
-    with out_path.open("w", newline="", encoding="utf-8") as f:
+
+    # 输出 xlsx（原生支持 Unicode，Excel 打开不乱码）
+    xlsx_path = out_path.with_suffix(".xlsx")
+    import openpyxl
+    from openpyxl.styles import Font, PatternFill, Alignment
+    wb = openpyxl.Workbook()
+    ws = wb.active
+    ws.title = "eval"
+
+    # 表头加粗 + 浅灰背景
+    header_font = Font(bold=True)
+    header_fill = PatternFill("solid", fgColor="D9D9D9")
+    ws.append(out_fieldnames)
+    for cell in ws[1]:
+        cell.font = header_font
+        cell.fill = header_fill
+        cell.alignment = Alignment(horizontal="center")
+
+    # 数据行
+    for r in kept:
+        ws.append([r.get(f, "") for f in out_fieldnames])
+
+    # 自适应列宽（最大 40 字符）
+    for col in ws.columns:
+        max_len = max((len(str(cell.value or "")) for cell in col), default=0)
+        ws.column_dimensions[col[0].column_letter].width = min(max_len + 2, 40)
+
+    wb.save(xlsx_path)
+    print(f"已写入: {xlsx_path}")
+
+    # 同时保留 CSV（utf-8-sig，Excel 双击也能正确识别中文）
+    with out_path.open("w", newline="", encoding="utf-8-sig") as f:
         writer = csv.DictWriter(f, fieldnames=out_fieldnames, extrasaction="ignore")
         writer.writeheader()
         writer.writerows(kept)
     print(f"已写入: {out_path}")
 
     # 重新计算 summary
-    summary = compute_summary(kept, threshold=args.threshold, out_path=str(out_path))
+    summary = compute_summary(kept, threshold=args.threshold, out_path=str(xlsx_path))
     summary_path = out_path.with_suffix(".summary.json")
     with summary_path.open("w", encoding="utf-8") as f:
         json.dump(summary, f, ensure_ascii=False, indent=2)