python+playwright获取国家统计局行政区划数据

hongyinvjianke

已于 2024-03-15 11:50:31 修改

阅读量749

点赞数 17

文章标签： python 开发语言

于 2024-03-15 11:44:24 首次发布

本文链接：https://blog.csdn.net/hongyinvjianke/article/details/136734791

版权

实现目标：

使用python+playwright从国家统计局获取天津市（示例）所有行政区划数据

前置条件：

1.获取数据的网站：

2023年统计用区划代码和城乡划分代码

2.编辑器：

VScode

3.第三方库下载：

pip install playwight

playwright install

pip pandas

实现步骤：

1.region_1.py：待运行的python文件

# 导入
from playwright.sync_api import Playwright, sync_playwright
import pandas as pd
from urllib.parse import urljoin
import os

BASE_URL_1 = 'https://www.stats.gov.cn/sj/tjbz/tjyqhdmhcxhfdm/2023/12/1201.html'
BASE_URL_2 = 'https://www.stats.gov.cn/sj/tjbz/tjyqhdmhcxhfdm/2023/12/01/120101.html'
url_list = [BASE_URL_1]
count = [0,0,0]
result = pd.DataFrame()

# 抓取网页内容
def scrape_page(page, url):
    page.goto(url)
    page.wait_for_load_state('networkidle')

# 获取解析内容
def parse_index(page):
    # 获取网页内容请求
    elements = page.query_selector_all('tr td:nth-child(2) a')
    if page.query_selector('table.towntable'):
       BASE_URL = BASE_URL_2
    else:
       BASE_URL = BASE_URL_1
    # 获取元素信息
    for element in elements:
        part_of_url = element.get_attribute('href')
        detail_url = urljoin(BASE_URL, part_of_url)
        url_list.append(detail_url)
    parse_detail(page)

def parse_detail(page):
    global result
    global count
    data = page.query_selector_all('.villagetable tbody tr, .countytable tbody tr, .towntable tbody tr')
    for i in range(1,len(data)):
      acode = data[i].query_selector('td:nth-child(1)').text_content()
      region = data[i].query_selector('td:nth-child(3)').text_content() if data[i].get_attribute('class') == 'villagetr' else data[i].query_selector('td:nth-child(2)').text_content()
      #社区
      if data[i].get_attribute('class') == 'villagetr':
        facode = acode[:9] 
        count[2] += 1
      #街道
      elif  data[i].get_attribute('class') == 'towntr':
        acode = acode[:9]
        facode = acode[:6]
        count[1] += 1
      #区县
      else:
        acode = acode[:6]
        facode = acode[:4]
        count[0] += 1
      busi_data = [{
          'acode': acode,
          'region': region,
          'facode':facode
      }]
      df = pd.DataFrame(busi_data)
      result = pd.concat([result, df], axis=0, ignore_index=True)
    url_list.pop(0)

def run (playwright: Playwright) -> None:
    print("---------------行政区划获取中(来源：国家统计局)-------------------")
    browser = playwright.chromium.launch(headless=False)
    content = browser.new_context()
    page = content.new_page()
    while len(url_list):
       scrape_page(page, url_list[0])
       parse_index(page)
    # 将脚本所在路径作为excel输出路径
    output_path = os.getcwd() + os.sep + "output_region_1.xlsx"
    # 将结果写入到output_path 所在的excel中
    result.to_excel(output_path, index=False)
    print('共统计出 %d 条数据:' %len(result))
    print('———————————————————————————')
    print('     区（县）   |   %d 个  ' %count[0])
    print('  街道（乡/镇） |   %d 个  ' %count[1])
    print('    社区（村）  |   %d 个  ' %count[2])
    print('———————————————————————————')
    print("---------------------------获取完成------------------------------")
    content.close()       
    browser.close()
   

with sync_playwright() as playwright:
  run(playwright)

2.运行region_1.py