Update app.py

app.py

@@ -1,124 +1,37 @@
-import os
 import streamlit as st
+import requests
 from bs4 import BeautifulSoup
-import urllib3
-from langchain_community.vectorstores import FAISS
-from langchain_community.embeddings import HuggingFaceEmbeddings
-from groq import Groq
 
-# Initialize Groq client
-client = Groq(api_key=os.environ.get('GroqApi'))
-
-# Initialize session state
-if "scraped_data" not in st.session_state:
-    st.session_state.scraped_data = []
-if "appliance_data" not in st.session_state:
-    st.session_state.appliance_data = []
-
-def scrape_web_data(url):
-    """Scrape text data from the given URL."""
-    try:
-        http = urllib3.PoolManager()
-        response = http.request('GET', url)
-        if response.status == 200:
-            soup = BeautifulSoup(response.data, 'html.parser')
-            all_text = soup.get_text()
-            return [{'Data': line.strip()} for line in all_text.split('\n') if line.strip()]
-        else:
-            st.write(f"Error: {response.status}")
-    except Exception as e:
-        st.write(f"An error occurred: {e}")
-    return []
-
-def store_tariff_data(data):
-    """Store the scraped data in FAISS vectorstore."""
-    if not data:
-        return
-    combined_text = "\n".join([item["Data"] for item in data])
-    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
-    vectorstore = FAISS.from_texts([combined_text], embeddings)
-    vectorstore.save_local("vectorstore/")
-
-def query_with_groq(prompt):
-    """Query Groq LLM with the provided prompt."""
+def scrape_tariff_data(url):
     try:
-        chat_completion = client.chat.completions.create(
-            messages=[
-                {
-                    "role": "user",
-                    "content": prompt,
-                }
-            ],
-            model="llama3-8b-8192",
-        )
-        return chat_completion.choices[0].message.content
+        response = requests.get(url)
+        response.raise_for_status()  # Raise an error for bad responses
+        soup = BeautifulSoup(response.text, 'html.parser')
+
+        # Example: Find the tariff data in a specific HTML element
+        # You will need to adjust the selector based on the actual website structure
+        tariff_data = soup.find_all('div', class_='tariff')  # Adjust this selector
+        return [tariff.get_text(strip=True) for tariff in tariff_data]
     except Exception as e:
-        return f"Error querying Groq: {e}"
-
-# Streamlit UI
-st.title("Pakistani Electricity Bill Calculator with Groq RAG")
-
-# Step 1: Scraping
-st.subheader("Step 1: Scrape Tariff Data")
-website_url = st.text_input("Enter the tariff URL:")
-if st.button("Scrape Data"):
-    scraped_data = scrape_web_data(website_url)
-    if scraped_data:
-        st.session_state.scraped_data = scraped_data
-        st.success(f"Scraping completed. {len(scraped_data)} items found.")
-        store_tariff_data(scraped_data)
-    else:
-        st.warning("No data found. Please check the URL.")
-
-# Step 2: Appliance Inputs
-st.subheader("Step 2: Enter Appliance Details")
-appliance_name = st.text_input("Appliance Name")
-appliance_load = st.number_input("Load (Watts per Appliance)", min_value=1, value=100)
-appliance_quantity = st.number_input("Quantity", min_value=1, value=1)
-appliance_usage_hours = st.number_input("Usage Hours per Day", min_value=0.0, value=6.0)
-
-if st.button("Add Appliance"):
-    st.session_state.appliance_data.append({
-        "name": appliance_name,
-        "load": appliance_load,
-        "quantity": appliance_quantity,
-        "usage_hours": appliance_usage_hours
-    })
-    st.success(f"Added {appliance_quantity} {appliance_name}(s) to the list!")
-
-if st.session_state.appliance_data:
-    st.subheader("Appliance List")
-    for idx, appliance in enumerate(st.session_state.appliance_data):
-        st.write(f"{idx+1}. {appliance['name']} - {appliance['quantity']} units, "
-                 f"{appliance['load']}W each, {appliance['usage_hours']} hours/day")
-
-# Step 3: Query the RAG system
-st.subheader("Step 3: Query Tariff Information")
-user_query = st.text_input("Enter your query about the tariff:")
-if st.button("Ask Query"):
-    if not st.session_state.scraped_data:
-        st.warning("Please scrape tariff data first.")
-    else:
-        embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
-        vectorstore = FAISS.load_local("vectorstore/", embeddings, allow_dangerous_deserialization=True)
-        retriever = vectorstore.as_retriever()
-
-        context = "\n".join([item["Data"] for item in st.session_state.scraped_data])
-        prompt = f"Context: {context}\n\nUser Query: {user_query}\nAnswer:"
-        response = query_with_groq(prompt)
-        st.write("**Answer:**")
-        st.write(response)
+        return f"An error occurred: {e}"
+
+def main():
+    st.title("Electricity Tariff Scraper")
+    st.write("Enter the URL of the electricity tariff page:")
+
+    url = st.text_input("URL", "")
+    
+    if st.button("Scrape"):
+        if url:
+            with st.spinner("Scraping data..."):
+                data = scrape_tariff_data(url)
+                if isinstance(data, list):
+                    st.success("Data scraped successfully!")
+                    st.write(data)
+                else:
+                    st.error(data)
+        else:
+            st.error("Please enter a valid URL.")
 
-# Step 4: Calculate Bill
-st.subheader("Step 4: Calculate Bill")
-tariff_rate = st.number_input("Enter Rate per Unit (PKR)", min_value=0.0, value=25.0)
-if st.button("Calculate Bill"):
-    if not st.session_state.appliance_data:
-        st.warning("Please add at least one appliance.")
-    else:
-        total_energy_kwh = sum(
-            (appliance["load"] * appliance["quantity"] * appliance["usage_hours"] * 30) / 1000
-            for appliance in st.session_state.appliance_data
-        )
-        monthly_bill = total_energy_kwh * tariff_rate
-        st.write(f"**Total Monthly Electricity Bill: PKR {monthly_bill:.2f}**")
+if __name__ == "__main__":
+    main()