How to parse xls data including merged cells using python pandas

How to parse xls data to this struct, both row and column have merged cells, simply use df.index.to_series().ffill() cannot handle.

{
  "time": "time",
  "category": "A",
  "variety": "A1",
  "specification": "S1",
  "unit": "U1",
  "average": 1.25,
  "region": "RegionA",
  "market": "MarketA",
  "price": 1.1,
}

I figured out this solution:

def test_xls_parse():
    file_path = 'test.xls'
    df = pd.read_excel(file_path, engine='xlrd')

    time_label = df.iloc[0, 0]


    categories = df.iloc[1, 2:]
    varieties = df.iloc[2, 2:]
    specifications = df.iloc[3, 2:]
    units = df.iloc[4, 2:]
    averages = df.iloc[5, 2:]


    regions = df.iloc[6:, 0].ffill()
    markets = df.iloc[6:, 1]
    prices = df.iloc[6:, 2:]

    result = []
    for i in range(len(categories)):
        for j in range(len(regions)):
            obj = {
                "date": time_label,
                "category": categories.iloc[i],
                "variety": varieties.iloc[i],
                "specification": specifications.iloc[i],
                "unit": units.iloc[i],
                "average": None if averages.iloc[i] == '-' else float(averages.iloc[i]),
                "region": regions.iloc[j],
                "market": markets.iloc[j],
                "price": None if prices.iloc[j, i] == '-' else float(prices.iloc[j, i])
            }
            result.append(obj)

    return pd.DataFrame(result)