还在为写调查问卷发愁的你赶快来看看这个自动填写问卷（问卷星版）

Posted 2021-06-07 adventure.Li

一、背景

马克思实践报告需要，由于并不是需要严谨深入的研究调查，因此想到使用自动填写来完成，于是打开pycharm开整吧。

二、实现过程

1. chromedriver+selenium填写（可以进行填写，但是发现问卷星反爬虫了，于是采用第二种办法）

import time
from selenium import webdriver
import asyncio
from pyppeteer import launch
from pyppeteer_stealth import stealth    # 反爬虫第三方库
chrome_driver = r"E:\\PycharmProjects\\PythonBase\\crawler\\driver\\chromedriver.exe"
# 调用浏览器
browser = webdriver.Chrome(executable_path=chrome_driver)
url = "https://www.wjx.cn/vm/PG4tFZd.aspx"
page = browser.get(url)
stealth(page)
# browser.get(url)
# print("val:", browser.find_element_by_class_name("field ui-field-contain"))
while 1 == 1:
    for i in range(10):
        i = i + 1
        idName = "div" + str(i)
        print(idName)
        try:
            e = browser.find_element_by_id(idName)
            # e.find_element_by_class_name("jqradio").click()
            # radio类, 1:男女性别, 2:所在年级, 4:选择？ 6:是否听说校园贷, 7:处理方式？ 8:是否担心
            # jcheck类:  3:事件（多选）
            # 9:打分 rate-off
            if i == 9:
                print("打分rate-off类型")
                elements = e.find_elements_by_class_name("rate-off")
                j = 1
                for detailE in elements:
                    print("测试" + str(j), detailE.text)
                    if j == 5:
                        detailE.click()
                    j = j + 1
            elif i == 3:
                print("check多选类型")
                elements = e.find_elements_by_class_name("jqcheck")
                j = 1
                for detailE in elements:
                    print("测试" + str(j))
                    detailE.click()
                    j = j + 1
            else:
                print("radio类型")
                elements = e.find_elements_by_class_name("jqradio")
                j = 1
                for detailE in elements:
                    print("测试" + str(j))
                    detailE.click()
                    j = j + 1
        except Exception:
            print("error")

        print("enter....")
        print(e.id)
    time.sleep(3000)

2. pyppeteer +stealth
   分析页面：
   进行试探点击看源代码反应，明确点击地方，然后获取下来id或class进行定位。（可能会出现断点阻断需要在source处点击一个如下图按钮然后点击调试）
   

import asyncio
from random import randint
from pyppeteer import launch
from pyppeteer_stealth import stealth  # 反爬虫第三方库
import time


async def main():
    # launch方法会新建一个browser对象,然后赋值给browser
    browser = await launch({
        # 路径就是你的谷歌浏览器的安装路径
        'executablePath': 'C:\\\\Program Files\\\\Google\\\\Chrome\\\\Application\\\\chrome.exe',
        # Pyppeteer 默认使用的是无头浏览器,所以要显示需要给False
        'headless': False,
        # 设置Windows-size和Viewport大小来实现网页完整显示
        'args': ['--no-sandbox', '--window-size=1366,850']
    })
    url = "https://www.wjx.cn/vm/PG4tFZd.aspx"
    # 调用 newPage 方法相当于浏览器中新建了一个选项卡,同时新建了一个Page对象
    page = await browser.newPage()
    await page.setViewport({'width': 1366, 'height': 768})
    # 防止页面识别出脚本(反爬虫关键语句)
    await stealth(page)

    # 调用了Page对象的goto方法就相当于在浏览器中输入问卷的网址,浏览器跳转到了对应的页面进行加载
    await page.goto(url)

    # 单选题：先用page.querySelector(selector)找到指定的元素,再调用元素的click()方法, await?
    # buttons = await page.querySelectorAll("#div1 .jqradio")
    i = 1

    for i in range(9):
        i = i + 1
        # 确定区域
        id_name = "#div" + str(i)
        print(id_name)
        if i == 3:
            # jqcheck
            click_toolkit = id_name + ' .jqcheck'
            checks = await page.querySelectorAll(click_toolkit)
            for check in checks:
                await check.click()
        elif i == 9:
            # rate-off
            click_toolkit = id_name + ' .rate-off'
            checks = await page.querySelectorAll("#div9 .rate-off")
            j = 1
            ack = randint(3,8)
            for check in checks:
                # 4 + randint(1, 3)
                if j == ack:
                    await check.click()
                j = j + 1

        else:
            # radio类型 .jqradio
            # "#div1 .jqradio"
            click_toolkit = id_name + " .jqradio"
            print("test ", click_toolkit)
            checks = await page.querySelectorAll(click_toolkit)

            if i == 1 or i == 2:
                k = randint(1, 2 * i)
                l = 1
                for check in checks:
                    if k == l:
                        await check.click()
                    l = l + 1
            elif i == 6 or i == 8:
                l = 1
                for check in checks:
                    if l == 1:
                        print(l, "打印")
                        await check.click()
                    l = l + 1
                    print(l, "打印")
            elif i == 4:
                k = randint(2, 5)
                l = 1
                for check in checks:
                    if k == l:
                        await check.click()
                    l = l + 1
            else:
                for check in checks:
                    await check.click()

    # 找到提交按钮提交
    # await asyncio.sleep(1000)
    submit = await page.querySelector('#divSubmit #ctlNext')
    await submit.click()
    await asyncio.sleep(2)  # 页面延迟2s看是否提交成功
    await browser.close()


while 1 == 1:
    asyncio.get_event_loop().run_until_complete(main())
    wait = randint(5 , 10)
    time.sleep(wait)

3. 常用模板

# 填空题：page.type(selector,text),在指定selector的元素上填写text
    # await page.type('#q1', '姓名')
    # await page.type('#q2', '学号')
    # await page.type('#divquestion5 > table > tbody > tr:nth-child(1) > td > div > textarea', '体温')

    # 单选题：先用page.querySelector(selector)找到指定的元素,再调用元素的click()方法, await?
    button = await page.querySelector("#div1 .jqradio")
    buttons = await page.querySelectorAll("#div1 .jqradio")
    i = 1
    for b in buttons:
        if i == 2:
            await b.click()
        i = i+1



    # 地址题：先点击手动填写地址,再在地址框内填写相应地址
    # address = await page.querySelector("#divquestion7 > ul > li:nth-child(1) > label")
    # await address.click()
    # await page.type('#q9', '地址')

    # 日期选择题：先点击日期选择框,在出现的iframe寻找元素并调用click()方法
    # date1 = await page.querySelector("#q4")
    # await date1.click()

    # frame = page.frames
    # date2 = await frame[1].querySelector('#selectTodayButton')
    # await date2.click()

    # 找到提交按钮提交
    submit = await page.querySelector('#submit_button')
    # await submit.click()
    await asyncio.sleep(200)  # 页面延迟2s看是否提交成功
    # await browser.close()