消息关闭
    暂无新消息!

爬虫生成dict后,想将其写入csv文件,却出错
使用jupyter notebook,window环境。

具体代码如下

import requests

from multiprocessing.dummy import Pool as ThreadPool

from lxml import etree

import sys

import time

import random

import csv


def spider(url):

    header={

        'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'

    }

    timeout=random.choice(range(31,50))

    html = requests.get(url,header,timeout=timeout)

    time.sleep(random.choice(range(8,16)))

    selector = etree.HTML(html.text)

    content_field = selector.xpath('//*[@class="inner"]/div[3]/div[2]/ul/li')

    item ={}

    for each in content_field:

        g = each.xpath('a/div[1]/div[1]/h3/span/text()')

        go = each.xpath('a/div[1]/div[2]/div/h3/text()')

        h = each.xpath('a/div[1]/div[2]/div/p/text()[1]')

        j= each.xpath('a/div[1]/div[1]/p/text()[2]')

        ge = each.xpath('a/div[1]/div[2]/div/p/text()[3]')

        x = each.xpath('a/div[1]/div[1]/p/text()[3]')

        city = each.xpath('a/div[1]/div[1]/p/text()[1]')

        gg = each.xpath('a/div[2]/span/text()')

        item['city']="".join(city)

        item['hangye']="".join(hangye)

        item['guimo']="".join(guimo)

        item['gongsi']="".join(gongsi)

        item['gongzi']="".join(gongzi)

        item['jingyan']="".join(jingyan)

        item['xueli']="".join(xueli)

        item['gongzuoneirong']="".join(gongzuoneirong)

        fieldnames =['city','hangye','guimo','gongsi','gongzi','jingyan','xueli','gongzuoneirong']

        with open('bj.csv','a',newline='',errors='ignore')as f:

            f_csv=csv.DictWriter(f,fieldnames=fieldnames)

            f_csv.writeheader()

            f_csv.writerow(item)


if __name__ == '__main__':
    pool = ThreadPool(4)
    f=open('bj.csv','w')
    page = []
    for i in range(1,100):
        newpage = 'https://www.zhipin.com/c101010100/h_101010100/?query=%E6%95%B0%E6%8D%AE%E8%BF%90%E8%90%A5&page='+str(i) + '&ka=page-' + str(i)
        page.append(newpage)
        
    results = pool.map(spider,page)
    pool.close()
    pool.join()
    f.close()

运行上面代码,提示错误为

ValueError: too many values to unpack (expected 2)
通过查询原因是要将dict遍历,需要dict.items()的形式。但在上述代码中如何实现,一直没有理顺,求教各位


3个回答

︿ 0

不好意思哈,现在才有时间来回答你的问题,看到你根据我的建议把代码改过来了,下面我把改过的代码贴出来,我运行过,是没问题的

import requests
from multiprocessing.dummy import Pool
from lxml import etree
import time
import random
import csv


def spider(url):
    header = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
    }

    timeout = random.choice(range(31, 50))
    html = requests.get(url, headers=header, timeout=timeout)

    time.sleep(random.choice(range(8, 16)))

    selector = etree.HTML(html.text)

    content_field = selector.xpath('//*[@class="inner"]/div[3]/div[2]/ul/li')

    item = {}

    for each in content_field:
        g = each.xpath('a/div[1]/div[1]/h3/span/text()')

        go = each.xpath('a/div[1]/div[2]/div/h3/text()')

        h = each.xpath('a/div[1]/div[2]/div/p/text()[1]')

        j = each.xpath('a/div[1]/div[1]/p/text()[2]')

        ge = each.xpath('a/div[1]/div[2]/div/p/text()[3]')

        x = each.xpath('a/div[1]/div[1]/p/text()[3]')

        city = each.xpath('a/div[1]/div[1]/p/text()[1]')

        gg = each.xpath('a/div[2]/span/text()')

        item['city'] = "".join(city)

        item['hangye'] = "".join(g)

        item['guimo'] = "".join(go)

        item['gongsi'] = "".join(h)

        item['gongzi'] = "".join(j)

        item['jingyan'] = "".join(ge)

        item['xueli'] = "".join(x)

        item['gongzuoneirong'] = "".join(gg)

        fieldnames = ['city', 'hangye', 'guimo', 'gongsi', 'gongzi', 'jingyan', 'xueli', 'gongzuoneirong']

        with open('bj.csv', 'a', newline='', errors='ignore')as f:
            f_csv = csv.DictWriter(f, fieldnames=fieldnames)

            f_csv.writeheader()

            f_csv.writerow(item)


if __name__ == '__main__':
    f = open('bj.csv', 'w')
    page = []
    for i in range(1, 100):
        newpage = 'https://www.zhipin.com/c101010100/h_101010100/?query=%E6%95%B0%E6%8D%AE%E8%BF%90%E8%90%A5&page=' + str(
            i) + '&ka=page-' + str(i)
        page.append(newpage)
    print(page)
    pool = Pool(4)
    results = pool.map(spider, page)
    pool.close()
    pool.join()
    f.close()

这里主要是header,你原来是set类型,我修改后是dict类型

这里还需要给你一些建议

  1. 你的代码是放到ide还是文本编辑器中运行的?有的东西在ide下明显会报错啊

  2. 建议新手从开始学的时候就遵守PEP8规范,别养成了坏习惯,你看看你的命名

︿ 0
item = {'a':1, 'b':2}
fieldnames = ['a', 'b']

with open('test.csv', 'a') as f:
    f_csv = DictWriter(f, fieldnames=fieldnames)
    f_csv.writeheader()
    f_csv.writerow(item)

我这样写并没报错喔

writerow就是直接接收dict的吧,你这个问题,我感觉是因为item的key与你表头不对应