Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import requests
|
2 |
+
from bs4 import BeautifulSoup
|
3 |
+
import pandas as pd
|
4 |
+
import gradio as gr
|
5 |
+
import logging
|
6 |
+
|
7 |
+
# Set up logging
|
8 |
+
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
9 |
+
logger = logging.getLogger()
|
10 |
+
|
11 |
+
def scrape_kosdaq():
|
12 |
+
url = "https://finance.naver.com/sise/sise_quant.naver?sosok=1"
|
13 |
+
headers = {
|
14 |
+
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
|
15 |
+
}
|
16 |
+
|
17 |
+
logger.info("Sending request to URL: %s", url)
|
18 |
+
response = requests.get(url, headers=headers)
|
19 |
+
|
20 |
+
if response.status_code != 200:
|
21 |
+
logger.error("Failed to fetch data. Status code: %d", response.status_code)
|
22 |
+
return f"Error: Unable to fetch data (status code: {response.status_code})"
|
23 |
+
|
24 |
+
soup = BeautifulSoup(response.content, 'html.parser')
|
25 |
+
|
26 |
+
table = soup.find("table", {"class": "type_2"})
|
27 |
+
if not table:
|
28 |
+
logger.error("Failed to find the data table in the HTML.")
|
29 |
+
return "Error: Unable to find the data table."
|
30 |
+
|
31 |
+
rows = table.find_all("tr")
|
32 |
+
data = []
|
33 |
+
|
34 |
+
for row in rows:
|
35 |
+
cols = row.find_all("td")
|
36 |
+
cols = [col.text.strip() for col in cols]
|
37 |
+
if len(cols) > 1:
|
38 |
+
data.append(cols)
|
39 |
+
|
40 |
+
if not data:
|
41 |
+
logger.error("No data found in the table.")
|
42 |
+
return "Error: No data found in the table."
|
43 |
+
|
44 |
+
# Define column names based on the structure
|
45 |
+
columns = [
|
46 |
+
"N", "종목명", "현재가", "전일비", "등락률", "거래량", "거래대금", "매수호가",
|
47 |
+
"매도호가", "시가총액", "PER", "ROE"
|
48 |
+
]
|
49 |
+
|
50 |
+
df = pd.DataFrame(data, columns=columns[:len(data[0])])
|
51 |
+
|
52 |
+
logger.info("Successfully scraped data with %d rows.", len(df))
|
53 |
+
return df
|
54 |
+
|
55 |
+
def interface_function():
|
56 |
+
try:
|
57 |
+
df = scrape_kosdaq()
|
58 |
+
if isinstance(df, str): # If error message
|
59 |
+
return df
|
60 |
+
|
61 |
+
logger.info("Displaying scraped data to user.")
|
62 |
+
return df.to_csv(index=False)
|
63 |
+
except Exception as e:
|
64 |
+
logger.exception("An error occurred during scraping.")
|
65 |
+
return f"An error occurred: {str(e)}"
|
66 |
+
|
67 |
+
def display_interface():
|
68 |
+
gr.Interface(
|
69 |
+
fn=interface_function,
|
70 |
+
inputs=[],
|
71 |
+
outputs="text",
|
72 |
+
title="네이버 증권 코스닥 스크래퍼",
|
73 |
+
description="네이버 증권에서 코스닥 데이터를 스크래핑하여 CSV 형식으로 반환합니다."
|
74 |
+
).launch()
|
75 |
+
|
76 |
+
if __name__ == "__main__":
|
77 |
+
display_interface()
|