代碼將錯誤的結構 json 插入到文件中import requestsfrom bs4 import BeautifulSoupfrom selenium import webdriverfrom selenium.webdriver.chrome.options import Optionsimport jsonurls = {}urls['Av'] = {'áa', 'Bb'}data = {}for key, value in urls.items(): for x in value: url = 'https://www.google.pt/search?q=' + key + '%20' + x driver = webdriver.Chrome() driver.get(url) html = driver.page_source soup = BeautifulSoup(html, 'html.parser') a = soup.find("body") for child in a.find_all("div", {'class': 'g'}): h2 = child.find("span", {'class': 'Q8LRLc'}) div = child.find("a", {'class': 'Fx4vi'}) data[key] = [] data[key].append({'h2': h2, 'div': div}) print(data) with open("data_file.json", "a") as write_file: json.dump(data, write_file, indent=4) driver.quit()
1 回答
qq_花開花謝_0
TA貢獻1835條經驗 獲得超7個贊
我看到了很多問題，大多數是當它們應該在循環之外時卻在循環內，或者當它們應該在循環中時卻在循環之外。
您在循環內設置變量
h2和，但將它們添加到循環外，因此只會添加最后一個值。divfor child in a.find_all("div", {'class': 'g'}):data此外，您在循環內為每個鍵初始化數據，并且應該在循環外完成，否則每次都會重新初始化。
您還每次都打開文件附加到它，我只做一次。
而且，你在每個循環中初始化你的驅動程序。
requests并且selenium.webdriver.chrome.options.Options都是未使用的進口
所以，我會這樣改變它：
urls = {}
urls['Av'] = {'áa', 'Bb'}
data = {}
driver = webdriver.Chrome()
with open("data_file.json", "a") as write_file:
for key, value in urls.items():
data[key] = []. # initialize only once per key
for x in value:
url = 'https://www.google.pt/search?q=' + key + '%20' + x
driver.get(url)
html = driver.page_source
soup = BeautifulSoup(html, 'html.parser')
a = soup.find("body")
for child in a.find_all("div", {'class': 'g'}):
h2 = child.find("span", {'class': 'Q8LRLc'})
div = child.find("a", {'class': 'Fx4vi'})
data[key].append({'h2': h2, 'div': div}) # update data for every h2/div found
json.dump(data, write_file, indent=4) # This write can be done once, outside all loops!
driver.quit()
對我來說有點難測試，但希望能有所幫助！快樂編碼！
添加回答
舉報
0/150
提交
取消