Update app.py

app.py

@@ -4,32 +4,44 @@ import PyPDF2
 import io
 from docx import Document
 import os
-import fitz  # PyMuPDF for better PDF handling
 
 # For PDF generation
 from reportlab.pdfgen import canvas
 from reportlab.lib.pagesizes import letter
+from reportlab.lib import utils
 from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
 from reportlab.lib.styles import getSampleStyleSheet
 
-# Initialize Hugging Face Inference Client with Meta-Llama-3.1-8B-Instruct
-client = InferenceClient(
-    model="meta-llama/Meta-Llama-3.1-8B-Instruct",
-    token=os.getenv("HF_TOKEN")
-)
+# Import for CV Optimizer
+import fitz  # PyMuPDF for PDF handling
+
+# Initialize the inference client from Hugging Face.
+# Updated model to Meta-Llama-3.1-8B-Instruct
+try:
+    client = InferenceClient(
+        model="meta-llama/Meta-Llama-3-8B-Instruct",
+        token=os.getenv("HF_TOKEN")
+    )
+except Exception as e:
+    print(f"Error initializing InferenceClient: {e}")
 
-# Function to extract text from PDF
 
 def extract_text_from_pdf(pdf_file):
+    """Extract text from PDF file."""
     try:
         pdf_reader = PyPDF2.PdfReader(pdf_file)
-        text = "".join(page.extract_text() or "" for page in pdf_reader.pages)
+        text = ""
+        for page in pdf_reader.pages:
+            page_text = page.extract_text()
+            if page_text:
+                text += page_text + "\n"
         return text.strip() or "No text could be extracted from the PDF."
     except Exception as e:
         return f"Error reading PDF: {e}"
 
-# Function to extract text from DOCX
+
 def extract_text_from_docx(docx_file):
+    """Extract text from DOCX file."""
     try:
         doc = Document(docx_file)
         text = "\n".join(para.text for para in doc.paragraphs)
@@ -37,13 +49,14 @@ def extract_text_from_docx(docx_file):
     except Exception as e:
         return f"Error reading DOCX: {e}"
 
-# Function to analyze CV
+
 def parse_cv(file, job_description):
+    """Analyze the CV, show the prompt (debug) and return LLM analysis."""
     if file is None:
         return "Please upload a CV file.", ""
 
     try:
-        file_path = file.name
+        file_path = file.name  # Get the file path
         file_ext = os.path.splitext(file_path)[1].lower()
 
         if file_ext == ".pdf":
@@ -51,13 +64,20 @@ def parse_cv(file, job_description):
         elif file_ext == ".docx":
             extracted_text = extract_text_from_docx(file_path)
         else:
-            return "Unsupported file format. Please upload a PDF or DOCX file.", ""
+            return (
+                "Unsupported file format. Please upload a PDF or DOCX file.",
+                "Unsupported file format.",
+            )
+
     except Exception as e:
-        return f"Error reading file: {e}", ""
+        error_msg = f"Error reading file: {e}"
+        return error_msg, error_msg
 
+    # Check for extraction errors
     if extracted_text.startswith("Error"):
         return extracted_text, "Error during text extraction. Please check the file."
 
+    # Prepare debug prompt
     prompt = (
         f"Analyze the CV against the job description. Provide a summary, assessment, "
         f"and a score 0-10.\n\n"
@@ -65,100 +85,201 @@ def parse_cv(file, job_description):
         f"Candidate CV:\n{extracted_text}\n"
     )
 
+    # Call LLM
     try:
         analysis = client.text_generation(prompt, max_new_tokens=512)
-        return extracted_text, f"--- Analysis Report ---\n{analysis}"
+        # Show both the debug prompt and the LLM analysis in the "Analysis Report"
+        analysis_report = (
+            f"--- DEBUG PROMPT ---\n{prompt}\n"
+            f"--- LLM ANALYSIS ---\n{analysis}"
+        )
+        return extracted_text, analysis_report
     except Exception as e:
         return extracted_text, f"Analysis Error: {e}"
 
-# Function to optimize resume
+
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    """Generate chatbot response."""
+    messages = [{"role": "system", "content": system_message}]
+    for user_msg, bot_msg in history:
+        if user_msg:
+            messages.append({"role": "user", "content": user_msg})
+        if bot_msg:
+            messages.append({"role": "assistant", "content": bot_msg})
+    messages.append({"role": "user", "content": message})
+
+    response = ""
+    try:
+        for message_chunk in client.text_generation(
+            messages,
+            max_new_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            response += message_chunk
+            yield response
+    except Exception as e:
+        yield f"Error during chat generation: {e}"
+
+
+def create_pdf_report(report_text):
+    """Creates a PDF report using SimpleDocTemplate for better formatting."""
+    if not report_text.strip():
+        report_text = "No analysis report to convert."
+
+    buffer = io.BytesIO()
+    doc = SimpleDocTemplate(buffer, pagesize=letter)
+    styles = getSampleStyleSheet()
+    Story = []
+
+    # Title
+    Story.append(Paragraph("<b>Analysis Report</b>", styles["Title"]))
+    Story.append(Spacer(1, 12))
+
+    # Report Content
+    for line in report_text.split("\n"):
+        Story.append(Paragraph(line, styles["Normal"]))
+        Story.append(Spacer(1, 6))  # Add a small space between lines
+
+    doc.build(Story)
+    buffer.seek(0)
+    return buffer
+
+
+def toggle_download_button(analysis_report):
+    """Toggle the download button."""
+    return gr.update(
+        interactive=bool(analysis_report.strip()),
+        visible=bool(analysis_report.strip()),
+    )
+
+# Function to optimize resume based on job title
 def optimize_resume(resume_text, job_title):
     prompt = f"Optimize the following resume for the job title '{job_title}':\n\n{resume_text}"
     responses = []
     try:
-        for message in client.chat_completion(
-            messages=[{"role": "user", "content": prompt}],
-            max_tokens=1000,
+        for message in client.text_generation(
+            prompt,
+            max_new_tokens=1000,
             stream=True,
         ):
-            responses.append(message.choices[0].delta.content)
+            responses.append(message)
     except Exception as e:
         return f"Error during model inference: {e}"
     
     return ''.join(responses)
 
-# Function to process resume and job title inputs
+# Function to extract text from a PDF file (using PyMuPDF)
+def extract_text_from_pdf_fitz(pdf_file_path):
+    text = ""
+    try:
+        pdf_document = fitz.open(pdf_file_path)
+        for page_num in range(len(pdf_document)):
+            page = pdf_document.load_page(page_num)
+            text += page.get_text()
+    except Exception as e:
+        return f"Error extracting text from PDF: {e}"
+    return text
+
+# Function to process the resume and job title inputs for optimization
 def process_resume(file, job_title):
     try:
         file_name = file.name
         if file_name.endswith(".pdf"):
-            resume_text = extract_text_from_pdf(file.name)
+            # Extract text if the file is a PDF
+            resume_text = extract_text_from_pdf_fitz(file.name)
         elif file_name.endswith(".docx"):
+            # Extract text if the file is a Word document
             resume_text = extract_text_from_docx(file.name)
         else:
-            return "Unsupported file format. Please upload a PDF or DOCX file."
+            # Assume the file is a text file and read it directly
+            with open(file.name, 'r', encoding='utf-8') as f:
+                resume_text = f.read()
         
+        # Optimize the resume
         optimized_resume = optimize_resume(resume_text, job_title)
+        
         return optimized_resume
     except Exception as e:
         return f"Error processing resume: {e}"
 
-# Function to generate a PDF report
-def create_pdf_report(report_text):
-    buffer = io.BytesIO()
-    doc = SimpleDocTemplate(buffer, pagesize=letter)
-    styles = getSampleStyleSheet()
-    Story = [Paragraph("<b>Analysis Report</b>", styles["Title"]), Spacer(1, 12)]
-    
-    for line in report_text.split("\n"):
-        Story.append(Paragraph(line, styles["Normal"]))
-        Story.append(Spacer(1, 6))
-    
-    doc.build(Story)
-    buffer.seek(0)
-    return buffer
-
-# Function to toggle the download button
-def toggle_download_button(analysis_report):
-    return gr.update(interactive=bool(analysis_report.strip()), visible=bool(analysis_report.strip()))
-
 # Build the Gradio UI
 demo = gr.Blocks()
 with demo:
-    gr.Markdown("## AI-powered CV Analyzer, Optimizer, and Chatbot")
+    gr.Markdown("## AI-powered CV Analyzer and Chatbot")
 
     with gr.Tab("Chatbot"):
         chat_interface = gr.ChatInterface(
-            lambda message, history: client.chat_completion(
-                messages=[{"role": "user", "content": message}],
-                max_tokens=512,
-            ),
-            chatbot=gr.Chatbot(label="Chatbot"),
+            respond,
+            chatbot=gr.Chatbot(value=[], label="Chatbot"),
             textbox=gr.Textbox(placeholder="Enter your message here...", label="Message"),
+            additional_inputs=[
+                gr.Textbox(
+                    value="You are a friendly Chatbot.", label="System message"
+                ),
+                gr.Slider(
+                    minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"
+                ),
+                gr.Slider(
+                    minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"
+                ),
+                gr.Slider(
+                    minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"
+                ),
+            ],
         )
 
     with gr.Tab("CV Analyzer"):
         gr.Markdown("### Upload your CV and provide the job description")
         file_input = gr.File(label="Upload CV", file_types=[".pdf", ".docx"])
         job_desc_input = gr.Textbox(label="Job Description", lines=5)
-        extracted_text = gr.Textbox(label="Extracted CV Content", lines=10, interactive=False)
-        analysis_output = gr.Textbox(label="Analysis Report", lines=10, interactive=False)
-        download_pdf_button = gr.Button("Download Analysis as PDF", visible=False, interactive=False)
+        extracted_text = gr.Textbox(
+            label="Extracted CV Content", lines=10, interactive=False
+        )
+        analysis_output = gr.Textbox(
+            label="Analysis Report", lines=10, interactive=False
+        )
+        download_pdf_button = gr.Button(
+            "Download Analysis as PDF", visible=False, interactive=False
+        )
         pdf_file = gr.File(label="Download PDF", interactive=False)
+
         analyze_button = gr.Button("Analyze CV")
-        
-        analyze_button.click(parse_cv, [file_input, job_desc_input], [extracted_text, analysis_output])
-        analyze_button.then(toggle_download_button, [analysis_output], [download_pdf_button])
-        download_pdf_button.click(create_pdf_report, [analysis_output], [pdf_file])
 
+        analyze_button.click(
+            parse_cv,
+            inputs=[file_input, job_desc_input],
+            outputs=[extracted_text, analysis_output],
+        ).then(
+            toggle_download_button,
+            inputs=[analysis_output],
+            outputs=[download_pdf_button],
+        )
+
+        download_pdf_button.click(
+            create_pdf_report, inputs=[analysis_output], outputs=[pdf_file]
+        )
+    
     with gr.Tab("CV Optimizer"):
-        gr.Markdown("### Upload your Resume and Enter Job Title")
-        resume_file = gr.File(label="Upload Resume (PDF or Word)")
-        job_title_input = gr.Textbox(label="Job Title", lines=1)
+        gr.Markdown("### Upload your CV and enter the job title to optimize your resume")
+        cv_file_input = gr.File(label="Upload CV (PDF or DOCX)", file_types=[".pdf", ".docx"])
+        job_title_input = gr.Textbox(label="Job Title", placeholder="Enter the job title...")
         optimized_resume_output = gr.Textbox(label="Optimized Resume", lines=20)
-        optimize_button = gr.Button("Optimize Resume")
-        
-        optimize_button.click(process_resume, [resume_file, job_title_input], [optimized_resume_output])
+        optimize_button = gr.Button("Optimize CV")
+
+        optimize_button.click(
+            process_resume,
+            inputs=[cv_file_input, job_title_input],
+            outputs=[optimized_resume_output]
+        )
 
 if __name__ == "__main__":
     demo.queue().launch()