Python实现人民币大写金额转数字和大写规范校验

# -*- coding: utf-8 -*-
# 财务人民币大写金额转换+规范校验完整工具类
import re

# 正则：匹配文本中“大写”到括号的金额片段，用于长文本抽取
re_big_write_content = re.compile('大写[-\d\d]*?）')

def q2b(uchar):
    """单个字符 全角转半角"""
    inside_code = ord(uchar)
    if inside_code == 0x3000:
        inside_code = 0x0020
    else:
        inside_code -= 0xfee0
    if inside_code < 0x0020 or inside_code > 0x7e:
        return uchar
    return chr(inside_code)

def stringq2b(ustring):
    """整段字符串全角转半角，统一文本格式"""
    return "".join([q2b(uchar) for uchar in ustring])

def cnmoney_to_num(cn_str: str) -> float:
    """
    人民币大写转数字小写（单位：元）核心算法
    :param cn_str: 纯财务大写字符串，如 "壹万贰仟叁佰肆拾伍元陆角柒分"
    :return: float 金额（元，保留两位小数）
    """
    digit_map = {
        '零': 0, '壹': 1, '贰': 2, '叁': 3, '肆': 4,
        '伍': 5, '陆': 6, '柒': 7, '捌': 8, '玖': 9
    }
    unit_map = {
        '拾': 10, '佰': 100, '仟': 1000,
        '万': 10000, '亿': 100000000,
        '角': 0.1, '分': 0.01
    }

    # 清理无关字符
    s = cn_str.replace("整", "").replace("圆", "元").strip()
    yuan_part = 0
    jiao_fen = 0.0
    yuan_str = ""

    # 分割整数元 和 角分小数部分
    if "元" in s:
        yuan_str, jf_str = s.split("元", 1)
    else:
        jf_str = s

    # 解析角、分
    jiao = 0
    fen = 0
    if "角" in jf_str:
        jiao_part, fen_part = jf_str.split("角", 1)
        if jiao_part and jiao_part in digit_map:
            jiao = digit_map[jiao_part]
        if "分" in fen_part and fen_part.replace("分", "") in digit_map:
            fen = digit_map[fen_part.replace("分", "")]
    elif "分" in jf_str:
        fen_str = jf_str.replace("分", "")
        if fen_str in digit_map:
            fen = digit_map[fen_str]
    jiao_fen = jiao * 0.1 + fen * 0.01

    # 核心：解析亿、万、仟、佰、拾整数部分
    total = 0
    tmp = 0
    num_buf = none

    for char in yuan_str:
        if char in digit_map:
            num_buf = digit_map[char]
        elif char in unit_map:
            u = unit_map[char]
            if num_buf is not none:
                val = num_buf * u
                # 亿、万属于大单位，需要累加前置临时值
                if u >= 10000:
                    total += (tmp + val)
                    tmp = 0
                else:
                    tmp += val
                num_buf = none
    # 处理末尾无单位数字（例：壹拾万零捌佰，末尾捌佰无后续单位）
    if num_buf is not none:
        tmp += num_buf
    yuan_part = total + tmp

    return round(yuan_part + jiao_fen, 2)

def get_payment_fix(raw_text):
    """
    业务层统一入口：原始长文本抽取大写并转换为金额（支持负数）
    :param raw_text: 原始单据文本，如 "单据大写：负壹万贰仟叁佰肆拾伍元陆角柒分）"
    :return: float 最终金额，负数自动标记
    """
    # 1. 预处理：全角转半角统一格式
    text = stringq2b(raw_text)
    flag = 0
    # 识别负数标识
    if '调减' in text:
        flag = 1
    elif '负' in text:
        flag = 2
    # 正则截取大写片段
    big_match = re_big_write_content.search(text)
    if big_match:
        cn_str = big_match.group(0).replace("大写","").replace("）","").strip()
    else:
        cn_str = text
    # 调用核心转换
    num = cnmoney_to_num(cn_str.replace("圆","元"))
    # 处理负金额
    if flag == 1 or flag == 2:
        num = -num
    return num

# ------------------- 财务大写合法性校验模块 -------------------
# 1. 定义合法大写汉字
valid_chars = {"零", "壹", "贰", "叁", "肆", "伍", "陆", "柒", "捌", "玖",
               "拾", "佰", "仟", "万", "亿", "元", "角", "分", "整"}

# 2. 正则匹配标准财务大写完整模板
pattern = re.compile(
    r"^人民币"
    r"(零|壹[拾佰仟万亿]*|贰[拾佰仟万亿]*|叁[拾佰仟万亿]*|肆[拾佰仟万亿]*|伍[拾佰仟万亿]*|陆[拾佰仟万亿]*|柒[拾佰仟万亿]*|捌[拾佰仟万亿]*|玖[拾佰仟万亿]*)*"
    r"(元)?"
    r"(零|壹|贰|叁|肆|伍|陆|柒|捌|玖)角?"
    r"(零|壹|贰|叁|肆|伍|陆|柒|捌|玖)分?"
    r"(整)?$"
)

def check_fin_capital(cap_str: str) -> tuple[bool, str]:
    """
    校验财务大写金额是否规范合规
    :param cap_str: 完整大写字符串，例：人民币壹万贰仟叁佰肆拾伍元陆角柒分
    :return: (是否合法, 错误说明)
    """
    s = cap_str.strip()
    if not s:
        return false, "输入不能为空"

    # 规则1：必须以人民币开头
    if not s.startswith("人民币"):
        return false, "规范大写必须以「人民币」开头"

    # 规则2：过滤非法字符
    char_list = list(s)
    for char in char_list:
        if char not in valid_chars and char not in {"人", "民", "币"}:
            return false, f"包含非法字符：{char}，仅允许标准财务大写汉字"

    # 规则3：整体格式正则校验
    if not pattern.match(s):
        return false, "金额格式不符合财务规范，单位语序错误（拾无前壹、单位错乱等）"

    # 规则4：有分不能带整
    if "分" in s and "整" in s:
        return false, "含分位的金额末尾不能添加「整」字"

    # 规则5：拾位前必须带壹，禁止拾元、拾万
    if re.search(r"([零佰仟万亿])拾(元|万|亿)", s):
        pass
    elif re.search(r"^人民币拾", s) or re.search(r"元拾", s) or re.search(r"万拾", s) or re.search(r"亿拾", s):
        return false, "拾位前缺少「壹」，规范应为壹拾元/壹拾万"

    # 规则6：禁止连续多个零
    if "零零" in s:
        return false, "禁止书写连续多个「零」"

    return true, "财务大写格式规范"

# ------------------- 测试入口 -------------------
if __name__ == "__main__":
    print("===== 大写转数字测试 =====")
    print(get_payment_fix("负壹万贰仟叁佰肆拾伍元陆角柒分"))  # -12345.67
    print(get_payment_fix("壹拾万零捌佰元整"))              # 100800.0
    print(get_payment_fix("叁拾元贰角"))                   # 30.2
    print(get_payment_fix("伍分"))                         # 0.05
    print(get_payment_fix("零元捌分"))                     # 0.08
    print(get_payment_fix("贰亿零伍万元整"))               # 200050000.0

    print("\n===== 财务大写规范校验测试 =====")
    test_cases = [
        "人民币壹佰元整",          # 合法
        "人民币壹拾贰元叁角肆分",  # 合法
        "人民币拾元整",            # 非法：缺壹
        "人民币壹佰零拾元整",      # 非法：零拾
        "人民币壹佰元叁角贰分整",  # 非法：有分带整
        "人民币二佰元整",          # 非法：简体二
        "人民币100元整",           # 非法：阿拉伯数字
        "人民币壹万零零伍元整"     # 非法：连续零
    ]
    for case in test_cases:
        res, msg = check_fin_capital(case)
        print(f"{case} → {res} | {msg}")

import pandas as pd
df["数字金额"] = df["原始文本"].apply(get_payment_fix)