Update app.py

app.py

@@ -89,12 +89,23 @@ def download_one(session: requests.Session, url: str, outdir: str, pref: str) ->
                     f.write(chunk)
     return path
 
-def load_excel(xls_path: str, sheet_pref: str | None, pref_name: str) -> pd.DataFrame | None:
+def load_excel(xls_path: str, sheet_pref: str | None, pref_name: str, use_header: bool) -> pd.DataFrame | None:
+    """
+    use_header=True のときのみ上3行を列名として使用
+    use_header=False のときは列名なし（skiprows=3）
+    """
     sheet = pick_sheet_name(xls_path, sheet_pref)
     if not sheet:
         return None
     try:
-        df = pd.read_excel(xls_path, sheet_name=sheet, engine="openpyxl", dtype=str)
+        if use_header:
+            df = pd.read_excel(xls_path, sheet_name=sheet, engine="openpyxl",
+                               header=[0,1,2], dtype=str)
+        else:
+            # 3行スキップしてデータだけ読み込む（列名は後で統一）
+            df = pd.read_excel(xls_path, sheet_name=sheet, engine="openpyxl",
+                               header=None, skiprows=3, dtype=str)
+
         # 前後空白トリム
         for c in df.select_dtypes(include=["object"]).columns:
             df[c] = df[c].str.strip()
@@ -105,6 +116,7 @@ def load_excel(xls_path: str, sheet_pref: str | None, pref_name: str) -> pd.Data
     except Exception:
         return None
 
+
 def zip_paths(paths: list[str], out_zip: str) -> str:
     with zipfile.ZipFile(out_zip, "w", compression=zipfile.ZIP_DEFLATED) as z:
         for p in paths:
@@ -173,10 +185,18 @@ def run_job(sheet_name, sleep_sec, limit, re_download, progress=gr.Progress(trac
     for i, it in enumerate(downloaded, start=1):
         progress(0.72 + 0.18 * i / max(1, len(downloaded)),
                  desc=f"読み込み {i}/{len(downloaded)}: {os.path.basename(it['path'])}")
-        df = load_excel(it["path"], sheet_name if sheet_name else None, it["pref"])
+        df = load_excel(it["path"],
+                        sheet_name if sheet_name else None,
+                        it["pref"],
+                        use_header=(i==1))
         if df is not None and len(df) > 0:
             frames.append(df)
 
+    # 2件目以降は列名が無いため、1件目の列名を上書き
+    if len(frames) > 1:
+        frames[1:] = [f.set_axis(frames[0].columns, axis=1) for f in frames[1:]]
+
+
     if not frames:
         return ("Excelは取得できましたが、読み込めるデータがありませんでした（シート名の指定を見直してください）。",
                 None, None, None, None)