Create app.py

app.py

@@ -0,0 +1,121 @@
+import boto3
+import json
+import time
+from datetime import datetime, timedelta
+import random
+import requests
+from bs4 import BeautifulSoup
+from flask import Flask, render_template, jsonify
+import threading
+
+# AWS Credentials and Region
+AWS_ACCESS_KEY_ID = "AKIATFBMPLLTCCPHI56U"  # Verify or replace
+AWS_SECRET_ACCESS_KEY = "SzHSXlxassWvt5eRGHRoDTxStXH+Ay670QRrMQ44"  # Verify or replace
+AWS_REGION = "ap-south-1"  # Try 'us-east-1' if issues persist
+
+# Bedrock client setup
+bedrock = boto3.client(
+    service_name='bedrock-runtime',
+    region_name=AWS_REGION,
+    aws_access_key_id=AWS_ACCESS_KEY_ID,
+    aws_secret_access_key=AWS_SECRET_ACCESS_KEY
+)
+
+# Flask app setup
+app = Flask(__name__, template_folder='templates', static_folder='static')
+
+# List of topics
+TOPICS = [
+    "Market trends",
+    "IPO",
+    "Mutual funds",
+    "Cryptocurrency",
+    "Economy",
+    "Banking",
+    "Government policies and budget",
+    "Gold and Silver"
+]
+
+# Model configurations
+MODELS = [
+    {
+        "modelId": "mistral.mixtral-8x7b-instruct-v0:1",
+        "contentType": "application/json",
+        "accept": "application/json",
+        "body_template": {
+            "prompt": "<s>[INST] {prompt} [/INST]",
+            "max_tokens": 100,
+            "temperature": 0.5,
+            "top_p": 0.9,
+            "top_k": 50
+        }
+    },
+    {
+        "modelId": "mistral.mistral-7b-instruct-v0:2",
+        "contentType": "application/json",
+        "accept": "application/json",
+        "body_template": {
+            "prompt": "<s>[INST] {prompt} [/INST]",
+            "max_tokens": 100,
+            "temperature": 0.5,
+            "top_p": 0.9,
+            "top_k": 50
+        }
+    },
+    {
+        "modelId": "meta.llama3-8b-instruct-v1:0",
+        "contentType": "application/json",
+        "accept": "application/json",
+        "body_template": {
+            "prompt": "{prompt}",
+            "max_gen_len": 128,
+            "temperature": 0.5,
+            "top_p": 0.9
+        }
+    }
+]
+
+WEBSITES = [
+    "https://economictimes.indiatimes.com/markets/stocks/news",
+    "https://www.moneycontrol.com/news/business/markets/",
+    "https://www.business-standard.com/markets-news",
+    "https://www.hindustantimes.com/business",
+    "https://indianexpress.com/section/business/",
+    "https://www.ndtvprofit.com/markets",
+    "https://www.zeebiz.com/markets",
+    "https://www.livemint.com/market/market-stats",
+    "https://www.financialexpress.com/market/",
+    "https://www.cnbctv18.com/market/",
+    "https://www.businesstoday.in/markets",
+    "https://economictimes.indiatimes.com/markets",
+    "https://www.indiainfoline.com/news/market",
+    "https://www.reuters.com/markets/asia/",
+    "https://www.bloomberg.com/asia"
+]
+
+# Global variable to store news articles
+latest_articles = []
+
+def fetch_from_websites(topic, current_time, time_24h_ago):
+    articles = []
+    headers = {'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'}
+    used_headlines = set()
+    
+    for url in WEBSITES:
+        try:
+            response = requests.get(url, headers=headers, timeout=10)
+            response.raise_for_status()
+            soup = BeautifulSoup(response.text, 'html.parser')
+            
+            news_items = soup.select('h1, h2, h3, .story, .article, .headline, .title')
+            for item in news_items:
+                if len(articles) >= 10:
+                    break
+                title = item.get_text(strip=True)[:30]
+                if title and title not in used_headlines and topic.lower() in title.lower():
+                    used_headlines.add(title)
+                    time_diff = random.randint(0, 24 * 60 * 60)
+                    pub_date = (current_time - timedelta(seconds=time_diff)).strftime("%Y-%m-%dT%H:%M:%SZ")
+                    
+                    desc_elem = item.find_next('p') or item.find_next('div', class_=['summary', 'content', 'desc'])
+                    raw_desc = desc_elem.get_text(strip=True) if desc_elem else f"Latest {topic} update from {url