newsranking / app.py
CSB261's picture
Update app.py
d27fedd verified
raw
history blame
3.18 kB
import requests
from bs4 import BeautifulSoup
import pandas as pd
import streamlit as st
import random
import time # time ๋ชจ๋“ˆ์„ ์ž„ํฌํŠธ
# ๋„ค์ด๋ฒ„ ๋ชจ๋ฐ”์ผ ๋‰ด์Šค ๋žญํ‚น URL
url = "https://m.news.naver.com/rankingList"
# ํ—ค๋” ์„ค์ • (User-Agent ๋ฐ Referer ์ถ”๊ฐ€)
headers = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.107 Safari/537.36",
"Referer": "https://m.news.naver.com/"
}
# ๋žœ๋ค ๋”œ๋ ˆ์ด ํ•จ์ˆ˜
def random_delay(min_delay=1, max_delay=3):
delay = random.uniform(min_delay, max_delay)
time.sleep(delay) # time ๋ชจ๋“ˆ์˜ sleep ํ•จ์ˆ˜ ์‚ฌ์šฉ
# ์›น ํŽ˜์ด์ง€ ์š”์ฒญ ๋ฐ ํŒŒ์‹ฑ
response = requests.get(url, headers=headers)
soup = BeautifulSoup(response.content, 'html.parser')
# ๋‰ด์Šค ๋ฆฌ์ŠคํŠธ ์ถ”์ถœ
news_list = []
# ์ƒˆ๋กœ์šด HTML ๊ตฌ์กฐ์— ๋งž๊ฒŒ ๋ฐ์ดํ„ฐ ์ถ”์ถœ
for news_box in soup.select('div.rankingnews_box'):
# ์–ธ๋ก ์‚ฌ ์ด๋ฆ„ ์ถ”์ถœ
press_name = news_box.find('strong', class_='rankingnews_name').text.strip()
# ๊ฐ ๋‰ด์Šค ํ•ญ๋ชฉ์„ ๋ฆฌ์ŠคํŠธ๋กœ ์ถ”์ถœ
for news_item in news_box.select('ul.rankingnews_list li'):
random_delay() # ๋”œ๋ ˆ์ด ์ถ”๊ฐ€
# ์ˆœ์œ„ ๋ฒˆํ˜ธ ์ถ”์ถœ
rank_tag = news_item.find('em', class_='list_ranking_num')
rank = rank_tag.text if rank_tag else 'No Rank'
# ์ œ๋ชฉ ์ถ”์ถœ
title_tag = news_item.find('strong', class_='list_title')
title = title_tag.text.strip() if title_tag else 'No Title'
# ๋งํฌ ์ถ”์ถœ
link = news_item.find('a')['href'] if news_item.find('a') else '#'
# ์‹œ๊ฐ„ ์ถ”์ถœ
time_tag = news_item.find('span', class_='list_time')
time_info = time_tag.text.strip() if time_tag else 'No Time'
# ์ด๋ฏธ์ง€ URL ์ถ”์ถœ
img_tag = news_item.find('img')
image_url = img_tag['src'] if img_tag and 'src' in img_tag.attrs else 'No Image Available'
# ๋ฐ์ดํ„ฐ ๋ฆฌ์ŠคํŠธ์— ์ถ”๊ฐ€
news_list.append({
'Press': press_name,
'Rank': rank,
'Title': title,
'Link': link,
'Time': time_info,
'Image URL': image_url
})
# ๋ฐ์ดํ„ฐํ”„๋ ˆ์ž„์œผ๋กœ ๋ณ€ํ™˜
df = pd.DataFrame(news_list)
# Streamlit์—์„œ ๊ฒฐ๊ณผ ํ‘œ์‹œ
st.title("Naver Mobile Ranking News Scraper")
# ๋ฐ”๋‘‘ํŒ ๋ชจ์–‘์œผ๋กœ ๋‰ด์Šค๋ฅผ 3๊ฐœ์”ฉ ๋ฐฐ์น˜ (3์—ด ๊ทธ๋ฆฌ๋“œ)
columns_per_row = 3
# ๋‰ด์Šค ํ•ญ๋ชฉ์„ ๋ฐ”๋‘‘ํŒ ํ˜•ํƒœ๋กœ ์ถœ๋ ฅ
for i in range(0, len(df), columns_per_row):
cols = st.columns(columns_per_row) # 3์—ด๋กœ ๊ทธ๋ฆฌ๋“œ ์ƒ์„ฑ
# ๊ฐ ์—ด์— ๋‰ด์Šค ๋ฐฐ์น˜
for idx, col in enumerate(cols):
if i + idx < len(df):
row = df.iloc[i + idx]
with col:
# ์ด๋ฏธ์ง€๊ฐ€ ์žˆ์„ ๊ฒฝ์šฐ ์ถœ๋ ฅ
if row['Image URL'] != 'No Image Available':
st.image(row['Image URL'], use_column_width=True)
# ๋‰ด์Šค ์ œ๋ชฉ ํ•˜์ดํผ๋งํฌ
st.markdown(f"**[{row['Title']}]({row['Link']})**")
# ๋‰ด์Šค ์–ธ๋ก ์‚ฌ์™€ ์‹œ๊ฐ„ ์ •๋ณด
st.write(f"Press: {row['Press']} | Time: {row['Time']}")