以天津市21年的天气预报为例
import requests
from bs4 import BeautifulSoup
import lxml
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36 SLBrowser/9.0.0.10191 SLBChan/105'
}
for j in range(1, 13):
if j<10:
j = "0"+str(j)
response = requests.get(f"https://lishi.tianqi.com/tianjin/2021{j}.html", headers=headers)
response.encodings = response.apparent_encoding
html = response.text
soup = BeautifulSoup(html, 'lxml')
all_content = soup.find_all(attrs={'class': 'thrui'})
for content in all_content:
data = content.find_all("div")
Length = len(data)
for i in range(Length - 1):
print(data[i].string, end=" ")
if (i + 1) % 5 == 0:
print()
else:
j=str(j)
response = requests.get(f"https://lishi.tianqi.com/tianjin/2021{j}.html",headers=headers)
response.encodings = response.apparent_encoding
html = response.text
soup = BeautifulSoup(html,'lxml')
all_content = soup.find_all(attrs={'class':'thrui'})
for content in all_content:
data = content.find_all("div")
Length = len(data)
for i in range(Length-1):
print(data[i].string,end=" ")
if (i + 1) % 5 == 0:
print()