新手求教python3如何把dict循環寫入csv檔（進行爬蟲時遇到的問題）？

Question

爬蟲生成dict後，想將其寫入csv文件，卻出錯
使用jupyter notebook,window環境。

具體程式碼如下

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"]/p[3]/p[2]/ul/li')

    item ={}

    for each in content_field:

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

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

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

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

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

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

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

        gg = each.xpath('a/p[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()

執行上面程式碼，提示錯誤為

ValueError: too many values to unpack (expected 2)
透過查詢原因是要將dict遍歷，需要dict.items()的形式。但在上述程式碼中如何實現，一直沒有理順，求教各位

Answer 1

不好意思哈，現在才有時間來回答你的問題，看到你根據我的建議把代碼改過來了，下面我把改過的代碼貼出來，我運行過，是沒問題的

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"]/p[3]/p[2]/ul/li')

    item = {}

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

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

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

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

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

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

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

        gg = each.xpath('a/p[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()

這裡主要是header,你原來是set类型，我修改后是dict類型

這裡還需要給你一些建議

1. 你的程式碼是放到ide還是文字編輯器中運行的？有的東西在ide下明顯會報錯啊

2. 建議新手從開始學的時候就遵守PEP8規範，別養成了壞習慣，你看看你的命名

Answer 2

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)

我這樣寫並沒報錯

writerow就是直接接收dict的吧，你這個問題，我感覺是因為item的key與你表頭不對應

Answer 3

因為在 fields 中指定的某些列名在 item 中不存在