app.py

app.py

@@ -1,30 +1,30 @@
 import streamlit as st
-from openai import OpenAI
 import time
-import pandas as pd
+import fitz  # PyMuPDF for PDF processing
+from openai import OpenAI
 from docx import Document
 from docx.shared import Pt
 import os
 
 # Set the page configuration
-st.set_page_config(page_title="LOR Chat ReportAi")
+st.set_page_config(page_title="Report Generation AI")
 
 # Title and caption
-st.title("LOR Chat Report Ai Assistant")
-st.caption("Chat with an AI Assistant on your LOR Chat Report")
+st.title("AI Report Generator")
+st.caption("Generate Professional Reports from PDF Documents")
 
 # Sidebar for API Key input
 with st.sidebar:
-    OPENAI_API_KEY = st.text_input("Enter your C2 Group of Technologies Access Key", type="password")
+    OPENAI_API_KEY = st.text_input("Enter your Access Key", type="password")
 
 # Check for valid API key
 if OPENAI_API_KEY:
     client = OpenAI(api_key=OPENAI_API_KEY)
 else:
-    st.error("Please enter your C2 Group of Technologies Access Key to continue.")
+    st.error("Please enter your Access Key to continue.")
     st.stop()
 
-ASSISTANT_ID = "asst_lLRD8YOe64ZZ3eYBGP1Ew5y9"
+ASSISTANT_ID = "asst_d68xqxiRtvYkmewBPugwZaGx"
 
 # Initialize session state for chat history
 if "messages" not in st.session_state:
@@ -41,16 +41,23 @@ for message in st.session_state.messages:
     st.chat_message(role).write(content)
 
 # Process user input
-uploaded_file = st.file_uploader("Upload Excel file", type=["xlsx"])
+uploaded_file = st.file_uploader("Upload PDF file", type=["pdf"])
 
 if uploaded_file:
-    # Read the Excel file
-    df = pd.read_excel(uploaded_file)
-    st.write("Data from uploaded Excel file:")
-    st.write(df)
-
-    # Process user input
-    prompt = f"Please analyze the following data: {df.head()}"
+    # Read the PDF file
+    pdf_document = fitz.open(uploaded_file)
+    full_text = ""
+    
+    # Extract text from each page
+    for page_num in range(pdf_document.page_count):
+        page = pdf_document.load_page(page_num)
+        full_text += page.get_text()
+
+    st.write("Text extracted from PDF:")
+    st.text_area("Extracted Text", full_text, height=300)
+
+    # Process user input for assistant
+    prompt = f"Please analyze the following text from the PDF: {full_text[:1000]}"  # Limit to the first 1000 chars for clarity
     st.session_state.messages.append({"role": "user", "content": prompt})
     st.chat_message("user").write(prompt)
 
@@ -98,56 +105,15 @@ if uploaded_file:
         
         # Add Executive Summary
         doc.add_heading('Executive Summary', level=1)
-        doc.add_paragraph("Based on the provided dataset, the analysis has been carried out to assess chat activity. "
-                          "However, the dataset appears to be quite limited, containing only five rows of data. "
-                          "Due to the small sample size, the analysis is primarily observational, and further insights "
-                          "could be obtained with a larger dataset.")
-        
-        # Step 1: Data Processing
-        doc.add_heading('Step 1: Data Processing', level=1)
-        doc.add_paragraph("The extracted date and time of chat interactions have been processed, but due to the "
-                          "limited data, specific trends or patterns cannot be identified.")
-        
-        # Step 2: Data Summary
-        doc.add_heading('Step 2: Data Summary', level=1)
-        doc.add_paragraph("Here are the key observations from the data provided:")
-        doc.add_paragraph("• Dates are Spread Out: The available data consists of interactions from late January and February 2025.")
-        doc.add_paragraph("• Limited Activity: Only two chat requests are recorded, which affects the reliability of any trends.")
+        doc.add_paragraph("Based on the provided PDF content, the analysis has been carried out.")
         
-        # Add Tables (Chat Requests by Day of the Week)
-        doc.add_heading('Chat Requests by Day of the Week', level=2)
-        table = doc.add_table(rows=1, cols=3)
-        hdr_cells = table.rows[0].cells
-        hdr_cells[0].text = 'Day'
-        hdr_cells[1].text = 'Total Requests'
-        hdr_cells[2].text = 'Notes'
-
-        # Populate the table
-        days_data = [
-            ("Monday", 0, "No data"),
-            ("Tuesday", 0, "No data"),
-            ("Wednesday", 0, "No data"),
-            ("Thursday", 0, "No data"),
-            ("Friday", 0, "No data"),
-            ("Saturday", 1, "24 Feb 2025"),
-            ("Sunday", 1, "20 Feb 2025")
-        ]
-        for day, req, note in days_data:
-            row = table.add_row().cells
-            row[0].text = day
-            row[1].text = str(req)
-            row[2].text = note
-
-        # Add observations section
-        doc.add_paragraph("Observations:")
-        doc.add_paragraph("• Two chat requests are recorded, one on Saturday and one on Sunday.")
-        doc.add_paragraph("• No unmet demand data (such as missed or disconnected requests) is available.")
+        # Step 1: Text Analysis
+        doc.add_heading('Step 1: Text Analysis', level=1)
+        doc.add_paragraph(f"The extracted text from the PDF is as follows:\n{full_text[:1500]}")  # Limit preview
         
-        # Add Step 3: Recommendations
-        doc.add_heading('Step 3: Recommendations', level=1)
-        doc.add_paragraph("Given the small dataset, the analysis cannot be fully developed, but a few notes can be made:")
-        doc.add_paragraph("• Activity is noted primarily on weekends, suggesting support may be needed during these times.")
-        doc.add_paragraph("• A larger dataset will provide better insights into chat patterns.")
+        # Step 2: Assistant's Response
+        doc.add_heading('Step 2: Assistant Analysis', level=1)
+        doc.add_paragraph(f"The AI Assistant's insights are:\n{assistant_message}")
         
         # Saving the DOCX file
         word_filename = "AI_Report_Formatted.docx"