Question

我有一个csv网址列表，我需要抓取并组织成一个csv文件。我希望每个url中的数据都是csv文件中的一行。我有大约19000个网址要刮，但我试图用很少的东西解决这个问题。我能够抓取文件并在终端中查看它们，但是当我将它们导出到csv文件时，只显示最后一个文件。

网址在csv文件中显示为：

http://www.gpo.gov/fdsys/pkg/CREC-2005-01-26/html/CREC-2005-01-26-pt1-PgH199-6.htm

http://www.gpo.gov/fdsys/pkg/CREC-2005-01-26/html/CREC-2005-01-26-pt1-PgH200-3.htm

我有一种感觉，我在循环中做错了什么，但似乎无法弄清楚在哪里。任何帮助将不胜感激！

到目前为止，我正在与之合作：

import urllib
from bs4 import BeautifulSoup
import csv
import re
import pandas as pd
import requests

with open('/Users/test/Dropbox/one_minute_json/Extracting Data/a_2005_test.csv') as f:
reader = csv.reader(f)

for row in reader:
    html = urllib.urlopen(row[0])
    r = requests.get(html)
    soup = BeautifulSoup(r, "lxml")

for item in soup:

volume = int(re.findall(r"Volume (\d{1,3})", soup.title.text)[0])
print(volume)

issue = int(re.findall(r"Issue (\d{1,3})", soup.title.text)[0])
print(issue)



date = re.findall(r"\((.*?)\)", soup.title.text)[0]
print(date)

page = re.findall(r"\[Page (.*?)]", soup.pre.text.split('\n')[3])[0]
print(page)

title = soup.pre.text.split('\n\n  ')[1].strip()
print(title)

name = soup.pre.text.split('\n ')[2]
print(name)

text = soup.pre.text.split(')')[2]
print(text)

df = pd.DataFrame()
df['volume'] = [volume]
df['issue'] = [issue]
df['date'] = [date]
df['page'] = [page]
df['title'] = [title]
df['name'] = [name]
df['text'] = [text]

df.to_csv('test_scrape.csv', index=False)

谢谢！

Answer 1

您的缩进完全关闭，请尝试以下操作：

from bs4 import BeautifulSoup
import csv
import re
import pandas as pd
import requests

with open('/Users/test/Dropbox/one_minute_json/Extracting Data/a_2005_test.csv') as f:
    reader = csv.reader(f)

    index = 0    
    df = pd.DataFrame(columns=["Volume", "issue", "date", "page", "title", "name", "text"])

    for row in reader:
        r = requests.get(row[0])
        soup = BeautifulSoup(r.text, "lxml")

        for item in soup:
            volume = int(re.findall(r"Volume (\d{1,3})", soup.title.text)[0])
            issue = int(re.findall(r"Issue (\d{1,3})", soup.title.text)[0])
            date = re.findall(r"\((.*?)\)", soup.title.text)[0]
            page = re.findall(r"\[Page (.*?)]", soup.pre.text.split('\n')[3])[0]
            title = soup.pre.text.split('\n\n  ')[1].strip()
            name = soup.pre.text.split('\n ')[2]
            text = soup.pre.text.split(')')[2]
            row = [volume, issue, date, page, title, name, text]

            df.loc[index] = row
            index += 1

    df.to_csv('test_scrape.csv', index=False)

将python中的多个已删除文件从漂亮的汤导出到cvs文件

1 个答案: