【Py】自下而上递归搜寻路径

假设我们有如下行政区划数据

[
  {
    "code": "110000000000",
    "name": "北京市",
    "level": "1",
    "pcode": "0"
  },
  {
    "code": "110100000000",
    "name": "市辖区",
    "level": "2",
    "pcode": "110000000000"
  },
  {
    "code": "110101000000",
    "name": "东城区",
    "level": "3",
    "pcode": "110100000000"
  },
  {
    "code": "110101001000",
    "name": "东华门街道",
    "level": "4",
    "pcode": "110101000000" 
  },
  {
    "code": "110101001001",
    "name": "多福巷社区居委会",
    "level": "5",
    "pcode": "110101001000" 
  },
  {
    "code": "110101001002",
    "name": "银闸社区居委会",
    "level": "5",
    "pcode": "110101001000" 
  },
  {
    "code": "110101001005",
    "name": "东厂社区居委会",
    "level": "5",
    "pcode": "110101001000" 
  }
]

code: 12位，省2位，市2位，县2位，镇3位，村3位
level: 省1，市2，县3，镇4，村5
pcode: 直接父级别的code

我们希望将其变为：

[
  {
    "level": "1",
    "code": "110000000000",
    "name": "北京市",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "",
    "2_code": "",
    "3_name": "",
    "3_code": "",
    "4_name": "",
    "4_code": "",
    "5_name": "",
    "5_code": ""
  },
  {
    "level": "2",
    "code": "110100000000",
    "name": "市辖区",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "",
    "3_code": "",
    "4_name": "",
    "4_code": "",
    "5_name": "",
    "5_code": ""
  },
  {
    "level": "3",
    "code": "110101000000",
    "name": "东城区",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "东城区",
    "3_code": "110101000000",
    "4_name": "",
    "4_code": "",
    "5_name": "",
    "5_code": ""
  },
  {
    "level": "4",
    "code": "110101001000",
    "name": "东华门街道",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "东城区",
    "3_code": "110101000000",
    "4_name": "东华门街道",
    "4_code": "110101001000",
    "5_name": "",
    "5_code": ""
  },
  {
    "level": "5",
    "code": "110101001001",
    "name": "多福巷社区居委会",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "东城区",
    "3_code": "110101000000",
    "4_name": "东华门街道",
    "4_code": "110101001000",
    "5_name": "多福巷社区居委会",
    "5_code": "110101001001"
  },
  {
    "level": "5",
    "code": "110101001002",
    "name": "银闸社区居委会",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "东城区",
    "3_code": "110101000000",
    "4_name": "东华门街道",
    "4_code": "110101001000",
    "5_name": "银闸社区居委会",
    "5_code": "110101001002"
  },
  {
    "level": "5",
    "code": "110101001005",
    "name": "东厂社区居委会",
    "1_name": "北京市",
    "1_code": "110000000000",
    "2_name": "市辖区",
    "2_code": "110100000000",
    "3_name": "东城区",
    "3_code": "110101000000",
    "4_name": "东华门街道",
    "4_code": "110101001000",
    "5_name": "东厂社区居委会",
    "5_code": "110101001005"
  }
]

由于提供有父级编码，所以可以采用自下而上的递归，子节点向上搜寻到根节点

import pandas as pd
df_origin = pd.read_csv('./area_code_2021.csv', names=['code', 'name', 'level', 'pcode'], dtype=str)

def get_df(item, df_item):
	"""
	递归补全数据
	"""
    if item.pcode != '0':
        try:
            item = list(df_origin.loc[df_origin['code']==item.pcode].itertuples(index=False))[0]
            df_item.loc[0, f'{item.level}_code'] = item.code
            df_item.loc[0, f'{item.level}_name'] = item.name
            get_df(item, df_item)
        except:
            import pdb
            pdb.set_trace()
    return df_item

df_province = df_origin.loc[df_origin['level']=='1']
df = df_origin.copy()
for province in df_province.itertuples(index=False):
    df_all = pd.DataFrame()
    df_item = df.copy()
    for item in df_item.loc[df_item['code'].apply(lambda x: x.startswith(province.code[0:2]))].itertuples():
        df_item = pd.DataFrame()
        # 初始化该条数据
        df_item.loc[0, 'level'] = item.level
        df_item.loc[0, 'code'] = item.code
        df_item.loc[0, 'name'] = item.name
        df_item.loc[0, f'{item.level}_name'] = item.name
        df_item.loc[0, f'{item.level}_code'] = item.code
        # 补全该条数据
        df_item = get_df(item, df_item)
        # 移除数据
        df.drop(item.Index, inplace=True)
        # 放入集合中
        df_all = df_all.append(df_item)
    df_all.to_excel(f'{province.name}.xlsx', index=False)

类似的企业微信API中的获取部门列表接口返回的数据结构与上面的类似，稍作修改就能使用