Update app.py

app.py

@@ -1,128 +1,147 @@
-import gradio as gr
-from resume_agent import full_app, scanner_app
-
-def step1_analyze(job_input, resume_file_or_text):
-    """
-    Step 1: Runs ONLY the Scanner to find gaps.
-    Output goes to the 'Human Feedback' box for you to edit.
-    """
-    # Handle File vs Text Input
-    resume_path_or_text = ""
-    if hasattr(resume_file_or_text, "name"):
-        resume_path_or_text = resume_file_or_text.name
-    else:
-        resume_path_or_text = resume_file_or_text
-
-    # Prepare Inputs
-    inputs = {
-        "job_description": job_input,
-        "original_resume": resume_path_or_text,
-        "resume_text": "", 
-        "job_text": "", 
-        "missing_skills": "" 
-    }
-    
-    try:
-        result = scanner_app.invoke(inputs)
-        # Return the missing skills list so you can see/edit it
-        return result['missing_skills']
-        
-    except Exception as e:
-        return f"Error during scan: {str(e)}"
-
-
-def step2_optimize(job_input, resume_file_or_text, user_notes):
-    """
-    Step 2: Runs the Full Optimizer.
-    It uses the 'user_notes' (which you edited in the text box) as context.
-    """
-    # 1. Handle File vs Text Input
-    resume_path_or_text = ""
-    if hasattr(resume_file_or_text, "name"):
-        resume_path_or_text = resume_file_or_text.name
-    else:
-        resume_path_or_text = resume_file_or_text
-
-    # 2. Prepare Inputs
-    # Notice we pass 'user_notes' into 'human_notes'
-    inputs = {
-        "job_description": job_input,
-        "original_resume": resume_path_or_text,
-        "human_notes": user_notes,  # <--- CRITICAL: Uses your edited feedback
-        
-        # Initialize other fields
-        "resume_text": "", "job_text": "", "optimized_resume": "", 
-        "feedback": "", "score": 0, "iteration": 0
-    }
-    
-    try:
-        # Run the "Full Graph" (Improver + Reviewer Loop)
-        final_state = full_app.invoke(inputs)
-        
-        return (
-            final_state.get('optimized_resume', "Error generating resume."), 
-            f"Score: {final_state.get('score', 0)}/100", 
-            final_state.get('feedback', 'No feedback provided.')
-        )
-        
-    except Exception as e:
-        return f"Error: {str(e)}", "Error", "Error"
-
-
-# --- BUILD THE USER INTERFACE ---
-
-with gr.Blocks(title="AI Resume Agent") as demo:
-    gr.Markdown("# AI Resume Optimizer")
-    gr.Markdown("An Agentic workflow: **Scan** for gaps first, **Edit** the plan, then **Generate** the result.")
-    
-    with gr.Row():
-        with gr.Column(scale=1):
-            job_in = gr.Textbox(
-                label="1. Job Description (URL or Text)", 
-                placeholder="Paste Job Text OR a Link (https://...)",
-                lines=3
-            )
-            
-            resume_in = gr.File(
-                label="2. Upload Resume (PDF)", 
-                file_types=[".pdf"],
-                type="filepath"
-            )
-            
-            # Action Button 1
-            btn_analyze = gr.Button("Step 1: Find Missing Skills", variant="secondary")
-            notes_in = gr.Textbox(
-                label="3. Missing Skills & Context (Editable)", 
-                placeholder="Click 'Step 1' to see missing skills here. \nThen DELETE skills you don't have, or ADD context like 'I used this in my internship...'",
-                lines=10, 
-                interactive=True
-            )
-            
-            # Action Button 2
-            btn_optimize = gr.Button("Step 2: Generate Optimized Resume", variant="primary")
-        
-        # (Results)
-        with gr.Column(scale=1):
-            score_out = gr.Label(label="Match Score")
-            feedback_out = gr.Textbox(label="Critique Summary", lines=4)
-            resume_out = gr.Textbox(
-                label="Final Optimized Resume", 
-                lines=25,
-            )
-
-    # 1. Analyze Button
-    btn_analyze.click(
-        fn=step1_analyze, 
-        inputs=[job_in, resume_in], 
-        outputs=[notes_in]
-    )
-    
-    # 2. Optimize Button -> Reads 'notes_in' -> Updates final outputs
-    btn_optimize.click(
-        fn=step2_optimize, 
-        inputs=[job_in, resume_in, notes_in], 
-        outputs=[resume_out, score_out, feedback_out]
-    )
-    
-if __name__ == "__main__":
+import gradio as gr
+import os
+from resume_agent import full_app, scanner_app
+
+def step1_analyze(job_input, resume_file_or_text):
+    """
+    Step 1: Runs ONLY the Scanner to find gaps.
+    Output goes to the 'Human Feedback' box for you to edit.
+    """
+    # Handle File vs Text Input
+    resume_path_or_text = ""
+    if hasattr(resume_file_or_text, "name"):
+        resume_path_or_text = resume_file_or_text.name
+    else:
+        resume_path_or_text = resume_file_or_text
+
+    # Prepare Inputs
+    inputs = {
+        "job_description": job_input,
+        "original_resume": resume_path_or_text,
+        "resume_text": "", 
+        "job_text": "", 
+        "missing_skills": "" 
+    }
+    
+    try:
+        # Run the partial "Scanner Graph"
+        result = scanner_app.invoke(inputs)
+        # Return the missing skills list so you can see/edit it
+        return result['missing_skills']
+        
+    except Exception as e:
+        return f"Error during scan: {str(e)}"
+
+
+def step2_optimize(job_input, resume_file_or_text, user_notes):
+    """
+    Step 2: Runs the Full Optimizer + Document Generator.
+    It returns text previews AND file paths for the PDFs.
+    """
+    # 1. Handle File vs Text Input
+    resume_path_or_text = ""
+    if hasattr(resume_file_or_text, "name"):
+        resume_path_or_text = resume_file_or_text.name
+    else:
+        resume_path_or_text = resume_file_or_text
+
+    # 2. Prepare Inputs
+    inputs = {
+        "job_description": job_input,
+        "original_resume": resume_path_or_text,
+        "human_notes": user_notes,  # Uses your edited feedback
+        
+        # Initialize empty fields for the state
+        "resume_text": "", "job_text": "", "optimized_resume": "", 
+        "feedback": "", "score": 0, "iteration": 0,
+        "cover_letter": "", "resume_pdf_path": "", "cover_letter_pdf_path": ""
+    }
+    
+    try:
+        # Run the "Full Graph" (Improver -> Reviewer -> PDF Gen)
+        final_state = full_app.invoke(inputs)
+        
+        # Extract Results
+        opt_resume_text = final_state.get('optimized_resume', "Error generating resume.")
+        score_text = f"Score: {final_state.get('score', 0)}/100"
+        feedback_text = final_state.get('feedback', 'No feedback provided.')
+        
+        # Get PDF Paths
+        resume_pdf = final_state.get('resume_pdf_path')
+        cover_pdf = final_state.get('cover_letter_pdf_path')
+        
+        return opt_resume_text, score_text, feedback_text, resume_pdf, cover_pdf
+        
+    except Exception as e:
+        return f"Error: {str(e)}", "Error", "Error", None, None
+
+
+# --- BUILD THE USER INTERFACE ---
+
+with gr.Blocks(title="AI Resume Agent") as demo:
+    gr.Markdown("# AI Resume Optimizer")
+    gr.Markdown("An Agentic workflow: **Scan** for gaps, **Edit** the plan, then **Generate** professional PDFs.")
+    
+    with gr.Row():
+        # --- LEFT COLUMN (Inputs) ---
+        with gr.Column(scale=1):
+            job_in = gr.Textbox(
+                label="1. Job Description (URL or Text)", 
+                placeholder="Paste Job Text OR a Link (https://...)",
+                lines=3
+            )
+            
+            resume_in = gr.File(
+                label="2. Upload Resume (PDF)", 
+                file_types=[".pdf"],
+                type="filepath"
+            )
+            
+            # Action Button 1
+            btn_analyze = gr.Button("Step 1: Find Missing Skills", variant="secondary")
+            
+            notes_in = gr.Textbox(
+                label="3. Missing Skills & Context (Editable)", 
+                placeholder="Click 'Step 1' to see missing skills here. \nThen DELETE skills you don't have, or ADD context like 'I used this in my internship...'",
+                lines=10, 
+                interactive=True
+            )
+            
+            # Action Button 2
+            btn_optimize = gr.Button("Step 2: Generate Optimized Resume", variant="primary")
+        
+        # --- RIGHT COLUMN (Outputs) ---
+        with gr.Column(scale=1):
+            score_out = gr.Label(label="Match Score")
+            feedback_out = gr.Textbox(label="Critique Summary", lines=4)
+            
+            # Text Preview - REMOVED 'show_copy_button' to fix crash
+            resume_out = gr.Textbox(
+                label="Final Resume (Text Preview)", 
+                lines=15
+            )
+            
+            # PDF Download Area
+            gr.Markdown("### 📥 Download Documents")
+            with gr.Row():
+                pdf_resume_out = gr.File(label="Optimized Resume PDF")
+                pdf_cover_out = gr.File(label="Cover Letter PDF")
+
+    # --- WIRING THE BUTTONS ---
+    
+    # 1. Analyze Button -> Updates the 'notes_in' box
+    btn_analyze.click(
+        fn=step1_analyze, 
+        inputs=[job_in, resume_in], 
+        outputs=[notes_in]
+    )
+    
+    # 2. Optimize Button -> Returns Text + PDF Files
+    btn_optimize.click(
+        fn=step2_optimize, 
+        inputs=[job_in, resume_in, notes_in], 
+        outputs=[resume_out, score_out, feedback_out, pdf_resume_out, pdf_cover_out]
+    )
+    
+if __name__ == "__main__":
     demo.launch(theme=gr.themes.Soft())