hr-ocr-regbook/test_addr.py

# 重构代码
import json
import cpca

addr = "江苏省京市永扬大道"

f = open('areas.json', 'r')
content = f.read()
areas = json.loads(content)


def get_cpca(addr):
    df = cpca.transform([addr])
    print(df)
    province = df.iloc[0][0]
    city = df.iloc[0][1] or ""
    region = df.iloc[0][2] or ""
    street = df.iloc[0][3]
    print(province, city, region, street)
    return province, city, region, street


province, city, region, street = get_cpca(addr)


def get_region(city_or_province, temp_addr):
    """
    flag=1：city_or_province is city
    flag=0: city_or_province is province
    """
    maxnum = 0
    for _, json_region in enumerate(city_or_province['children']):
        num = 0
        if len(temp_addr) == len(json_region['name']):
            for i in range(len(temp_addr)):
                if temp_addr[i] == json_region['name'][i]:
                    num += 1
        elif len(temp_addr) < len(json_region['name']):
            for i in range(len(temp_addr)):
                for j in range(len(json_region['name'])):
                    if temp_addr[i] == json_region['name'][j]:
                        num += 1
                        break
        if maxnum <= num:
            maxnum = num
            region = json_region['name']

    return region


def cal_regions(json_province, city, temp_addr, flag):
    """
    flag=0:cpca没有找到"区"，也没有找到"市"，只需要找到市即可
    flag=1:cpca没有找到"区"，需要找到"区"
    """
    if flag == 0:
        return get_region(json_province, temp_addr)
    elif flag == 1:
        for _, json_city in enumerate(json_province['children']):
            if (city and city in json_city['name']) or (not city):
                return get_region(json_city, temp_addr)


def cal_addrs(province, city, region, temp_addr, areas, flag):
    for _, json_province in enumerate(areas):
        if province in json_province['name']:
            region = cal_regions(json_province, city, temp_addr, flag)
            return region


# 第一步，判断 df中的省市区是否存在，存在说明后面的street只要把原来的省市区丢掉即可
if province and city and region:
    print("省市区都存在，只需要切割street中‘区’后面的内容")
    if "区" in street:
        street = street.split("市")[-1].split("区")[-1]
    if "县" in street:
        street = street.split("市")[-1].split("县")[-1]

# 第二步，判断 df中“区”是否存在，如果存在，那么 “省市区” 肯定都存在
elif not region:
    # 判断 df中“市”是否存在，如果存在，那么 ”省市“ 肯定都存在
    if city:
        if '区' in street or '县' in street:
            print("cpca没有检测到‘区’，但是检测到省，市，并且street中含有‘区’，需要校准区")
            flag = 1
            if "区" in street:
                temp_region = street.split("区")[0] + "区"
                street = street.split("区")[-1]
            if "县" in street:
                temp_region = street.split("县")[0] + "县"
                street = street.split("县")[-1]
            region = cal_addrs(province, city, region, temp_region, areas, flag)
        else:
            # （暂时）说明street没有必要切割了
            print("cpca没有检测到‘区’，但是检测到省，市，并且street中没有有‘区’，就不需要校准区")

    # 这里比较复杂
    elif not city:
        # 还是先校准“区”，如果区存在，就直接把区拿到，然后再进行一次cpca，
        # 如果“区”不存在，就直接校准“市”，street直接切割
        if '区' in street or '县' in street:
            print("cpca只检测到了省，但是street中有区，直接分割出区，再做cpca即可")
            flag = 1
            if "区" in street:
                temp_region = street.split('市')[-1].split('区')[0] + "区"
                street = street.split('区')[-1]
            if "县" in street:
                temp_region = street.split('市')[-1].split('县')[0] + "县"
                street = street.split('县')[-1]
            region = cal_addrs(province, city, region, temp_region, areas, flag)
        elif '市' not in street:
            # 市，区都没有，那就直接返回
            print("cpca只检测到了省，并且street中没有市，也没有区，直接跳过")
        elif '市' in street:
            print("cpca只检测到了省，并且street中有市，没有区，就只要校准city")
            flag = 0
            temp_city = street.split('市')[0] + '市'
            street = street.split('市')[-1]
            city = cal_addrs(province, city, region, temp_city, areas, flag)

addr = province + city + region + street
province, city, region, street = get_cpca(addr)