Create app.py

app.py

@@ -0,0 +1,116 @@
+import gradio as gr
+import pandas as pd
+from pdf2image import convert_from_path
+import pytesseract
+import google.generativeai as genai
+from io import BytesIO
+
+# Function: Extract text from PDF
+def extract_text_from_pdf(pdf_file):
+    pages = convert_from_path(pdf_file)
+    all_text = ""
+    for page in pages:
+        text = pytesseract.image_to_string(page)
+        all_text += text + "\n"
+    return all_text.strip()
+
+# Function: Chunk text
+def chunk_text(text, chunk_size=1500):
+    words = text.split()
+    for i in range(0, len(words), chunk_size):
+        yield ' '.join(words[i:i+chunk_size])
+
+# Models to try (fallbacks)
+models_to_try = [
+    "gemini-2.5-flash-lite",
+    "gemini-2.5-flash",
+    "gemini-2.5-pro",
+    "gemini-2.0-flash-lite",
+    "gemini-2.0-flash",
+    "gemini-1.5-flash",
+    "gemini-1.5-pro"
+]
+
+# Function: Generate MCQs
+def generate_mcqs(text, api_key):
+    genai.configure(api_key=api_key)
+    chunks = list(chunk_text(text, 1500))
+    mcq_data = []
+
+    for i, chunk in enumerate(chunks, start=1):
+        prompt = f"""
+        Generate 10 MCQs from the following text.
+        Each question must have:
+        - Question
+        - 4 Options (A-D)
+        - Correct Answer
+        Return in CSV format: Question,OptionA,OptionB,OptionC,OptionD,CorrectAnswer.
+        Text:\n{chunk}
+        """
+
+        response = None
+        for model_name in models_to_try:
+            try:
+                model = genai.GenerativeModel(model_name)
+                response = model.generate_content(prompt)
+                if response.text:
+                    break
+            except Exception:
+                continue
+
+        if response and response.text:
+            output = response.text.strip()
+            for line in output.splitlines():
+                parts = line.split(",")
+                if len(parts) >= 6 and parts[0]:
+                    mcq_data.append(parts)
+
+    filtered_mcq_data = [row for row in mcq_data if len(row) == 6]
+    if not filtered_mcq_data:
+        return None, None
+
+    df = pd.DataFrame(filtered_mcq_data, columns=["Question", "OptionA", "OptionB", "OptionC", "OptionD", "CorrectAnswer"])
+    return df, df.head(10).to_markdown()  # Show preview
+
+# Gradio pipeline
+def process_pdf(pdf_file, api_key):
+    if not api_key:
+        return "❌ Please enter your Gemini API key.", None
+
+    try:
+        text = extract_text_from_pdf(pdf_file.name)
+        df, preview = generate_mcqs(text, api_key)
+
+        if df is None:
+            return "❌ No valid MCQs generated.", None
+
+        # Save to Excel in memory
+        output_file = BytesIO()
+        df.to_excel(output_file, index=False)
+        output_file.seek(0)
+
+        return preview, output_file
+    except Exception as e:
+        return f"Error: {str(e)}", None
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## 📘 PDF to MCQ Generator (Gemini AI)")
+    gr.Markdown("Upload a PDF, enter your Gemini API key, extract text with OCR, and generate MCQs saved as Excel.")
+
+    api_key = gr.Textbox(label="Enter your Gemini API Key", type="password")
+    pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
+    generate_btn = gr.Button("Generate MCQs")
+    
+    preview_output = gr.Textbox(label="Preview (First 10 MCQs)", lines=15)
+    excel_output = gr.File(label="Download Excel (.xlsx)")
+
+    generate_btn.click(
+        fn=process_pdf,
+        inputs=[pdf_input, api_key],
+        outputs=[preview_output, excel_output]
+    )
+
+# Run app
+if __name__ == "__main__":
+    demo.launch()