import pandas as pd

# 读取Excel文件
file_path = "/home/baol/tools/数据草稿-案款1.xls"  # 替换为你的Excel文件路径

# 读取“账户收”工作表，只保留指定列
sheet_account_receipt = pd.read_excel(file_path, sheet_name="一案一账户收")
sheet_account_receipt = sheet_account_receipt[
    ["归属本案金额", "到账金额", "到账日期", "缴款人", "案号"]
]

# 读取“收入”工作表，表头从第二行开始，只保留指定列
sheet_income = pd.read_excel(file_path, sheet_name="银行-收入", header=1)
sheet_income = sheet_income[["对方户名", "交易时间", "金额"]]

# 统一日期格式
sheet_account_receipt["到账日期"] = pd.to_datetime(
    sheet_account_receipt["到账日期"]
).dt.strftime("%Y-%m-%d")
sheet_income["交易时间"] = pd.to_datetime(sheet_income["交易时间"]).dt.strftime(
    "%Y-%m-%d"
)

# 合并两个数据框，保留各自的列名
merged_df = pd.merge(
    sheet_account_receipt,
    sheet_income,
    left_on=["到账金额", "到账日期"],  # 账户收的匹配列
    right_on=["金额", "交易时间"],  # 收入的匹配列
    how="outer",  # 外连接以保留未匹配的记录
    suffixes=("_一案一账户收", "_银行收入"),  # 为重复列添加后缀
)

# 筛选出未匹配的记录
unmatched_records = merged_df[merged_df["到账金额"].isna() | merged_df["金额"].isna()]

# 将结果保存为新的Excel文件
output_file_path_merged = "/home/baol/tools/merged_records.xlsx"  # 合并后的记录
output_file_path_unmatched = "/home/baol/tools/unmatched_records.xlsx"  # 未匹配的记录

merged_df.to_excel(output_file_path_merged, index=False, engine="openpyxl")
unmatched_records.to_excel(output_file_path_unmatched, index=False, engine="openpyxl")

print(f"合并后的记录已保存到 {output_file_path_merged}")
print(f"未匹配的记录已保存到 {output_file_path_unmatched}")