Update app.py

app.py

@@ -1,8 +1,12 @@
 import json
 import random
 import os
+import re
 import gradio as gr
+import subprocess
+
 from agents import AnalyzerAgent, CoachAgent, PredictiveAgent
+from ocr_agent import OcrAgent
 
 QUESTIONS_FILE = "questions.json"
 
@@ -10,23 +14,27 @@ if not os.path.exists(QUESTIONS_FILE):
     with open(QUESTIONS_FILE, "w", encoding="utf-8") as f:
         json.dump([], f, indent=2)
 
-with open(QUESTIONS_FILE, "r", encoding="utf-8") as f:
-    QUESTION_BANK = json.load(f)
+def load_question_bank():
+    if os.path.exists(QUESTIONS_FILE):
+        with open(QUESTIONS_FILE, "r", encoding="utf-8") as f:
+            return json.load(f)
+    return []
+
+QUESTION_BANK = load_question_bank()
 
 analyzer = AnalyzerAgent()
 coach_agent = CoachAgent()
 predictor = PredictiveAgent()
+ocr_agent = OcrAgent()
 
+# ----------------- Exam Functions -----------------
 def start_exam(level, subject, num_questions=10, include_predicted=True):
-    # Filter only real past paper questions
     pool = [q for q in QUESTION_BANK if q.get("subject") == f"{level}_{subject}"]
 
-    # Optionally add predictions in memory (not saved)
     predicted_questions = []
     if include_predicted:
         predicted_questions = predictor.generate_predictions(level, subject, n=8)
 
-    # Combine both pools
     combined_pool = pool + predicted_questions
     if not combined_pool:
         return [], gr.update(visible=False), gr.update(visible=False), gr.update(visible=True), []
@@ -65,7 +73,6 @@ def submit_exam(answers, exam_data, level, subject):
 
     analysis = analyzer.analyze(per_question)
     coach = coach_agent.coach(analysis, level, subject)
-
     predictions_summary = predictor.summary(level, subject)
 
     return (
@@ -77,52 +84,96 @@ def submit_exam(answers, exam_data, level, subject):
         gr.update(visible=True)
     )
 
-# Gradio UI
+# ----------------- OCR Upload & Auto Merge -----------------
+def auto_detect(file_path):
+    """Try to detect year + subject from filename like spm_2018_Math.pdf"""
+    fname = os.path.basename(file_path)
+    m = re.match(r"spm_(\d{4})_(\w+)\.pdf", fname, re.IGNORECASE)
+    if m:
+        year, subject = m.groups()
+        return year, subject.capitalize()
+    return None, None
+
+def process_pdf(file, subject, year):
+    raw = ocr_agent.extract_from_pdf(file.name)
+    cleaned = ocr_agent.clean_text(raw)
+    saved_file = ocr_agent.text_to_json(cleaned, subject=subject, year=year, output_dir="data")
+
+    try:
+        subprocess.run(["python", "merge_questions.py"], check=True)
+        global QUESTION_BANK
+        QUESTION_BANK = load_question_bank()
+        return f"✅ OCR complete. Saved: {saved_file}. Dataset merged into {QUESTIONS_FILE}."
+    except subprocess.CalledProcessError as e:
+        return f"⚠️ OCR extracted {saved_file}, but merge failed: {str(e)}"
+
+def prefill_subject_year(file):
+    """Return auto-detected subject/year for UI prefill"""
+    if not file:
+        return "BM", "2018"
+    year, subject = auto_detect(file.name)
+    return subject if subject else "BM", year if year else "2018"
+
+# ----------------- Gradio UI -----------------
 with gr.Blocks() as demo:
-    gr.Markdown("## 📘 SPM Exam Simulator (2018–2024) with AI Predictions")
-
-    with gr.Row():
-        level = gr.Dropdown(["Form5"], value="Form5", label="Level (SPM=Form5)")
-        subject = gr.Dropdown(
-            ["BM", "English", "Math", "History", "Science", "MoralStudies",
-             "Accounting", "Economics", "Business"],
-            value="Math",
-            label="Subject"
+    gr.Markdown("## 📘 SPM Exam Simulator (2018–2024) with AI Predictions + OCR Upload")
+
+    with gr.Tab("📝 Exam Simulator"):
+        with gr.Row():
+            level = gr.Dropdown(["Form5"], value="Form5", label="Level (SPM=Form5)")
+            subject = gr.Dropdown(
+                ["BM", "English", "Math", "History", "Science", "MoralStudies",
+                 "Accounting", "Economics", "Business"],
+                value="Math",
+                label="Subject"
+            )
+            num_q = gr.Slider(5, 50, step=5, value=10, label="Number of Questions")
+            include_pred = gr.Checkbox(True, label="Include AI-predicted")
+            start_btn = gr.Button("Start Exam")
+
+        exam_output = gr.State()
+
+        exam_area = gr.Column(visible=False)
+        with exam_area:
+            gr.Markdown("### Questions")
+            exam_display = gr.JSON(label="Exam")
+            answers_box = gr.JSON(label="Your Answers")
+            submit_btn = gr.Button("Submit Exam")
+
+        results_area = gr.Column(visible=False)
+        with results_area:
+            score_text = gr.Textbox(label="Score", interactive=False)
+            with gr.Tab("Weakness Analysis"):
+                analysis_json = gr.JSON()
+            with gr.Tab("Study Coach"):
+                coach_json = gr.JSON()
+            with gr.Tab("Predictions (Admin)"):
+                predictions_json = gr.JSON()
+
+        start_btn.click(
+            start_exam,
+            inputs=[level, subject, num_q, include_pred],
+            outputs=[exam_display, exam_area, results_area, score_text, exam_output]
         )
-        num_q = gr.Slider(5, 50, step=5, value=10, label="Number of Questions")
-        include_pred = gr.Checkbox(True, label="Include AI-predicted")
-        start_btn = gr.Button("Start Exam")
-
-    exam_output = gr.State()
-
-    exam_area = gr.Column(visible=False)
-    with exam_area:
-        gr.Markdown("### Questions")
-        exam_display = gr.JSON(label="Exam")
-        answers_box = gr.JSON(label="Your Answers")
-        submit_btn = gr.Button("Submit Exam")
-
-    results_area = gr.Column(visible=False)
-    with results_area:
-        score_text = gr.Textbox(label="Score", interactive=False)
-        with gr.Tab("Weakness Analysis"):
-            analysis_json = gr.JSON()
-        with gr.Tab("Study Coach"):
-            coach_json = gr.JSON()
-        with gr.Tab("Predictions (Admin)"):
-            predictions_json = gr.JSON()
-
-    start_btn.click(
-        start_exam,
-        inputs=[level, subject, num_q, include_pred],
-        outputs=[exam_display, exam_area, results_area, score_text, exam_output]
-    )
 
-    submit_btn.click(
-        submit_exam,
-        inputs=[answers_box, exam_output, level, subject],
-        outputs=[score_text, analysis_json, coach_json, predictions_json, exam_area, results_area]
-    )
+        submit_btn.click(
+            submit_exam,
+            inputs=[answers_box, exam_output, level, subject],
+            outputs=[score_text, analysis_json, coach_json, predictions_json, exam_area, results_area]
+        )
+
+    with gr.Tab("📂 Upload Exam PDF"):
+        pdf_file = gr.File(label="Upload SPM PDF", type="file")
+
+        subject_input = gr.Dropdown(["BM", "English", "Math", "History", "Science", "MoralStudies"],
+                                    value="BM", label="Subject")
+        year_input = gr.Dropdown([str(y) for y in range(2018, 2025)], value="2018", label="Year")
+        process_btn = gr.Button("Process PDF → JSON + Merge")
+        ocr_status = gr.Textbox(label="OCR Status", interactive=False)
+
+        # Prefill subject/year when PDF is uploaded
+        pdf_file.change(prefill_subject_year, inputs=[pdf_file], outputs=[subject_input, year_input])
+        process_btn.click(process_pdf, inputs=[pdf_file, subject_input, year_input], outputs=ocr_status)
 
 if __name__ == "__main__":
     demo.launch()
@@ -132,3 +183,4 @@ if __name__ == "__main__":
 
 
 
+