Spaces:
Runtime error
Runtime error
requirements.txt
#1
by AnesKAM - opened
app.py
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# app.py داخل Hugging Face Space لخدمة البحث
|
| 2 |
+
import gradio as gr
|
| 3 |
+
import requests
|
| 4 |
+
from bs4 import BeautifulSoup
|
| 5 |
+
import urllib.parse
|
| 6 |
+
import re
|
| 7 |
+
|
| 8 |
+
# دالة البحث والزحف
|
| 9 |
+
def perform_duckduckgo_scrape(query, max_chars=1500):
|
| 10 |
+
if not query:
|
| 11 |
+
return "الرجاء تقديم استعلام بحث."
|
| 12 |
+
|
| 13 |
+
encoded_query = urllib.parse.quote(query)
|
| 14 |
+
search_url = f"https://duckduckgo.com/html/?q={encoded_query}"
|
| 15 |
+
|
| 16 |
+
headers = {
|
| 17 |
+
'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'
|
| 18 |
+
}
|
| 19 |
+
|
| 20 |
+
print(f"--- تلقي طلب بحث عن: '{query}' ---")
|
| 21 |
+
|
| 22 |
+
scraped_content = ""
|
| 23 |
+
try:
|
| 24 |
+
response = requests.get(search_url, headers=headers, timeout=20)
|
| 25 |
+
response.raise_for_status()
|
| 26 |
+
soup = BeautifulSoup(response.text, 'html.parser')
|
| 27 |
+
|
| 28 |
+
result_snippets = soup.find_all('a', class_='result__snippet')
|
| 29 |
+
if not result_snippets:
|
| 30 |
+
result_snippets = soup.find_all('div', class_='result__body')
|
| 31 |
+
|
| 32 |
+
for snippet in result_snippets:
|
| 33 |
+
text = snippet.get_text().strip()
|
| 34 |
+
if text:
|
| 35 |
+
scraped_content += text + "\n"
|
| 36 |
+
|
| 37 |
+
if not scraped_content and soup.find('meta', {'name': 'description'}):
|
| 38 |
+
scraped_content = soup.find('meta', {'name': 'description'})['content'] + "\n"
|
| 39 |
+
|
| 40 |
+
if len(scraped_content) > max_chars:
|
| 41 |
+
scraped_content = scraped_content[:max_chars] + "..."
|
| 42 |
+
|
| 43 |
+
if scraped_content:
|
| 44 |
+
print("معلومات تم استخراجها بنجاح.")
|
| 45 |
+
return scraped_content
|
| 46 |
+
else:
|
| 47 |
+
print("لم يتم العثور على مقتطفات نصية ذات صلة.")
|
| 48 |
+
return "لم يتم العثور على معلومات ذات صلة عبر DuckDuckGo."
|
| 49 |
+
|
| 50 |
+
except Exception as e:
|
| 51 |
+
print(f"خطأ أثناء البحث/الزحف: {e}")
|
| 52 |
+
return f"حدث خطأ أثناء البحث في الويب: {str(e)}"
|
| 53 |
+
|
| 54 |
+
# إنشاء واجهة Gradio (هذا هو ما سيوفر نقطة نهاية الـ API)
|
| 55 |
+
iface = gr.Interface(
|
| 56 |
+
fn=perform_duckduckgo_scrape,
|
| 57 |
+
inputs=gr.Textbox(lines=2, label="استعلام البحث (باللغة العربية)"),
|
| 58 |
+
outputs=gr.Textbox(lines=10, label="النتائج المستخرجة"),
|
| 59 |
+
title="أداة البحث في DuckDuckGo (مقدمة من Pi-1)"
|
| 60 |
+
)
|
| 61 |
+
|
| 62 |
+
iface.launch(share=False)
|