【问题标题】:Write Headers Once in Python CSV Writer Loop在 Python CSV Writer 循环中写入标题一次
【发布时间】:2018-12-23 11:25:41
【问题描述】:

下面是一个抓取器,它遍历两个网站,抓取一个团队的花名册信息,将信息放入一个数组中,然后将数组导出到一个 CSV 文件中。一切都很好,但唯一的问题是每次刮板移动到第二个网站时,csv 文件中的 writerow 标题都会重复。是否可以调整代码的 CSV 部分,以便当刮板循环通过多个网站时,标题只出现一次?提前致谢!

import requests
import csv
from bs4 import BeautifulSoup

team_list={'yankees','redsox'}

for team in team_list:
    page = requests.get('http://m.{}.mlb.com/roster/'.format(team))
    soup = BeautifulSoup(page.text, 'html.parser')

    soup.find(class_='nav-tabset-container').decompose()
    soup.find(class_='column secondary span-5 right').decompose()

    roster = soup.find(class_='layout layout-roster')
    names = [n.contents[0] for n in roster.find_all('a')]
    ids = [n['href'].split('/')[2] for n in roster.find_all('a')]
    number = [n.contents[0] for n in roster.find_all('td', index='0')]
    handedness = [n.contents[0] for n in roster.find_all('td', index='3')]
    height = [n.contents[0] for n in roster.find_all('td', index='4')]
    weight = [n.contents[0] for n in roster.find_all('td', index='5')]
    DOB = [n.contents[0] for n in roster.find_all('td', index='6')]
    team = [soup.find('meta',property='og:site_name')['content']] * len(names)

    with open('MLB_Active_Roster.csv', 'a', newline='') as fp:
        f = csv.writer(fp)
        f.writerow(['Name','ID','Number','Hand','Height','Weight','DOB','Team'])
        f.writerows(zip(names, ids, number, handedness, height, weight, DOB, team))

【问题讨论】:

  • 你试过将f.writerow移到for team in team_list上方吗?
  • 只需在for 循环之前写入标题即可。这意味着for 循环应该包装在with 上下文管理器中。

标签: python csv web-scraping beautifulsoup


【解决方案1】:

使用变量检查是否添加了标头可能会有所帮助。如果添加了标题,则不会添加第二次

header_added = False
for team in team_list:
    do_some stuff

    with open('MLB_Active_Roster.csv', 'a', newline='') as fp:
        f = csv.writer(fp)
        if not header_added:
            f.writerow(['Name','ID','Number','Hand','Height','Weight','DOB','Team'])
            header_added = True
        f.writerows(zip(names, ids, number, handedness, height, weight, DOB, team))

【讨论】:

    【解决方案2】:

    另一种方法是在 for 循环之前简单地执行它,这样您就不必检查是否已经编写。

    import requests
    import csv
    from bs4 import BeautifulSoup
    
    team_list={'yankees','redsox'}
    
    with open('MLB_Active_Roster.csv', 'w', newline='') as fp:
        f = csv.writer(fp)
        f.writerow(['Name','ID','Number','Hand','Height','Weight','DOB','Team'])
    
    for team in team_list:
        do_your_bs4_and_parsing_stuff
    
        with open('MLB_Active_Roster.csv', 'a', newline='') as fp:
            f = csv.writer(fp)
            f.writerows(zip(names, ids, number, handedness, height, weight, DOB, team))
    

    您也可以只打开文档一次而不是三次

    import requests
    import csv
    from bs4 import BeautifulSoup
    
    team_list={'yankees','redsox'}
    
    with open('MLB_Active_Roster.csv', 'w', newline='') as fp:
        f = csv.writer(fp)
        f.writerow(['Name','ID','Number','Hand','Height','Weight','DOB','Team'])
    
        for team in team_list:
            do_your_bs4_and_parsing_stuff
    
            f.writerows(zip(names, ids, number, handedness, height, weight, DOB, team))
    

    【讨论】:

      【解决方案3】:

      只需在循环之前写标题,并将循环放在with 上下文管理器中:

      import requests
      import csv
      from bs4 import BeautifulSoup
      
      team_list = {'yankees', 'redsox'}
      
      headers = ['Name', 'ID', 'Number', 'Hand', 'Height', 'Weight', 'DOB', 'Team']
      
      # 1. wrap everything in context manager
      with open('MLB_Active_Roster.csv', 'a', newline='') as fp:
          f = csv.writer(fp)
      
          # 2. write headers before anything else
          f.writerow(headers)
      
          # 3. now process the loop
          for team in team_list:
              # Do everything else...
      

      您也可以在循环外以类似于team_list 的方式定义您的标头,从而使代码更简洁。

      【讨论】:

      • 感谢 RoadRunner 的建议!代码运行了,但不幸的是,它返回了一个空的 CSV 文件。我必须在 for 循环的末尾包含 writerow zip 行吗?
      • @NateWalker 是的,writerrow 应该在循环的末尾。
      • @NateWalker 别担心。
      猜你喜欢
      • 1970-01-01
      • 1970-01-01
      • 2015-04-04
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 2016-11-16
      • 1970-01-01
      相关资源
      最近更新 更多