Python如何爬取京東的評價信息

Python如何爬取京東的評價信息

模塊：requests，BeautifulSoup

import re
import time
import csv
import requests
from bs4 import BeautifulSoup

def write_a_row_in_csv(data, csv_doc):
    "save good information into a row in csv document"
    with open(csv_doc, 'a', newline='') as f:
        writer = csv.writer(f)
        writer.writerow(data)

# add headers, download page, check status code, return page
url = 'https://search.jd.com/Search?keyword=%E5%8D%8E%E4%B8%BAp20&enc=utf-8&suggest=1.def.0.V13&wq=%E5%8D%8E%E4%B8%BA&pvid=f47b5d05bba84d9dbfabf983575a6875'
headers = {
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 SE 2.X MetaSr 1.0"
    }
response = requests.get(url, headers=headers)
print(response.status_code)

# save as html document
with open('html.html', 'w', encoding='utf8') as f:
    f.write(response.text)

# save as csv document
with open('phone.csv', 'w', newline='') as f:
    writer = csv.writer(f)
    fields = ('id', '名稱', '價格', '評價人數(shù)', '好評率')
    writer.writerow(fields)

# find elements, such as name, item, price, comment, goodrate, comment count
soup_all = BeautifulSoup(response.content, 'lxml')
sp_all_items = soup_all.find_all('li', attrs={'class': 'gl-item'})
for soup in sp_all_items[:3]:
    print('-' * 50)
    name = soup.find('div', attrs={'class': 'p-name p-name-type-2'}).find('em').text
    print('name: ', name)
    item = soup.find('div', attrs={'class': 'p-name p-name-type-2'}).find('a')
    print('item: ', item['href'], re.search(r'(\d+)', item['href']).group())
    price = soup.find_all('div', attrs={'class': 'p-price'})
    print('price:', price[0].i.string)
    comment = soup.find_all('div', attrs={'class': 'p-commit'})
    print('comment url:', comment[0].find('a').attrs['href'])
    time.sleep(0.2)

    # need add referer into headers
    item_id = re.search(r'(\d+)', item['href']).group()
    url = f'https://sclub.jd.com/comment/productPageComments.action?productId={item_id}&score=0&sortType=5&page=0&pageSize=10&isShadowSku=0&fold=1'
    headers = {
        "referer": f"https://item.jd.com/{item_id}.html",
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 SE 2.X MetaSr 1.0"
    }
    response = requests.get(url, headers=headers)
    with open('html.json', 'w', encoding='utf8') as f:
        f.write(response.text)
    data = response.json()
    comment_count = data['productCommentSummary']['commentCount']
    print('評價人數(shù)：', comment_count)
    good_rate = data['productCommentSummary']['goodRate']
    print('好評率：', good_rate)

    # record data into CSV sheet
    write_a_row_in_csv(('id'+item_id, name, price[0].i.string, comment_count, good_rate), 'phone.csv')