2021-05-30-爬取51job-数据分析

爬取51job-数据分析

import selenium
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
import time
from bs4 import BeautifulSoup
import csv

def page_generator():
    web = webdriver.Chrome()
    web.maximize_window()
    try:
        web.get('https://www.51job.com/')

        input = web.find_element_by_id('kwdselectid')
        input.send_keys('数据分析')
        input.send_keys(Keys.ENTER)

        while True:
            location = web.find_element_by_css_selector('div.clist > a:nth-child(1)')
            if location.text == '全国':
                break
            location.click()
            time.sleep(1)

        old_url = web.current_url
        max_try = 5
        try_count = 0
        while True:
            print(web.current_url)
            # max_height = 600
            # height = 100
            # while True:
            #     web.execute_script(f'window.scrollTo(0, {height})')
            #     height += 100
            #     if height > max_height:
            #         break
            # yield web.execute_script('return window.__SEARCH_RESULT__["engine_search_result"]')
            text = web.page_source
            try:
                # 网络以及翻页速度等原因造成部分页面加载不全的情况，如果下一页标签没出现，则报异常
                next_btn = web.find_element_by_css_selector('li.next')
                next_btn.click()
            except selenium.common.exceptions.ElementNotInteractableException as e:
                # 异常时，最多执行5次重新获取，5次都获取不到可能网页元素有变，这时，不在反复执行
                print('page_url:', web.current_url, '\terror:', e)
                try_count += 1
                if try_count < max_try:
                    # 取不到页面时，停顿一定时间后在尝试，否则时间太短，相同的问题会一直触发
                    time.sleep(1)
                    continue
                else:
                    break
            # 使用生成器获取页面
            yield text
            time.sleep(0.1)
            # 最后一页跳下一页，因为下一页不存在，所以地址和最后一页相同，因此当前页为最后一页的时候跳出
            if old_url == web.current_url:
                break
            old_url = web.current_url
            try_count = 0
    finally:
        web.close()

def analysis_data(content: str):
    res = []
    soup = BeautifulSoup(content, 'lxml')
    boxs = soup.select('div.j_joblist > div')
    for box in boxs:
        job_href = box.select_one('a').attrs['href']
        job_name = box.select_one('a > p.t > span.jname.at').get_text()
        updatedate = box.select_one('a > p.t > span.time').get_text()
        tags = box.select_one('a > p.t > em')
        tags = tags.attrs['alt'] if tags else ''
        # 包含工作地点和条件
        workplace_attribute = box.select_one('a > p.info > span.d.at').get_text()
        salary = box.select_one('a > p.info > span.sal').get_text()
        company_name = box.select_one('div.er > a').get_text()
        company_href = box.select_one('div.er > a').attrs['href']
        # 公司类型以及公司人数规模
        companytype = box.select_one('div.er > p.dc.at').get_text().strip()
        companyind = box.select_one('div.er > p.int.at').get_text()
        jobwelf = box.select_one('a > p.tags')
        jobwelf = jobwelf.attrs['title'] if jobwelf else ''
        res.append([job_name, job_href, updatedate, tags, workplace_attribute, salary, company_name, company_href, companytype, companyind, jobwelf])
        # print(job_name, job_href, updatedate, tags, workplace_attribute, salary, company_name, company_href, companytype, companyind, jobwelf)
    return res


if __name__ == '__main__':
    page = page_generator()
    with open(r'./csvfile/51job.csv', 'w', newline='', encoding='utf-8') as f:
        writer = csv.writer(f)
        writer.writerow(['job_name', 'job_href', 'updatedate', 'tags', 'workplace_attribute', 'salary', 'company_name', 'company_href', 'companytype', 'companyind', 'jobwelf'])

    with open(r'./csvfile/s51job.csv', 'a', newline='', encoding='utf-8') as f:
        writer = csv.writer(f)
        try:
            while True:
                date = next(page)
                res = analysis_data(date)
                writer.writerows(res)
        except StopIteration:
            print('页面加载完毕...')
        finally:
            print('程序运行结束...')
    # print('='*40)
    # print(next(page))
    # with open(r'./temp.html', 'w', encoding='utf-8') as f:
    #     f.write(next(page))
    #
    # with open(r'./temp.html', 'r', encoding='utf-8') as f:
    #     html = f.read()