app.py

import gradio as gr
import argparse
import json
import logging
from typing import List
from scene_gen import *

from pydantic import BaseModel, RootModel, ValidationError
from ollama import chat

# -----------------------------
# Models
# -----------------------------
class QAItem(BaseModel):
    Question: str
    Answer: str
    Voice_Over: str
    include_audio: bool

class QAList(RootModel[List[QAItem]]):
    """RootModel wrapping a list of QAItem"""
    root: List[QAItem]

# -----------------------------
# Configuration
# -----------------------------
MODEL = "qwen3:0.6b"
logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s: %(message)s")

# -----------------------------
# Core logic
# -----------------------------
def generate_qa(topic: str, count: int = 10) -> List[QAItem]:
    """
    Call the Ollama model to generate `count` QA items for a given topic.
    Returns a list of QAItem instances.
    """
    schema = QAList.model_json_schema()
    prompt = (
        f'Given the topic "{topic}", generate {count} entries in JSON format, ' 
        "each with keys Question, Answer, Voice_Over, and include_audio (true/false)."
    )

    response = chat(
        model=MODEL,
        think=False,
        messages=[{"role": "user", "content": prompt}],
        format=schema,
        options={"temperature": 0},
    )

    try:
        qa_list = QAList.model_validate_json(response.message.content)
        return qa_list.root
    except ValidationError as e:
        logging.error("Response validation failed:\n%s", e)
        raise

# -----------------------------
# CLI entrypoint
# -----------------------------
def cli_main():
    parser = argparse.ArgumentParser(description="Generate QA JSON via Ollama")
    parser.add_argument("topic", type=str, help="Topic to generate Q&A for")
    parser.add_argument(
        "-n", "--count", type=int, default=10,
        help="Number of QA items to generate (default: 10)"
    )
    args = parser.parse_args()

    logging.info("Generating %d QA items for topic: %s", args.count, args.topic)
    try:
        items = generate_qa(args.topic, args.count)
    except Exception:
        logging.critical("Aborting due to errors")
        return

    # Convert to plain data
    output = [item.model_dump() for item in items]

    # 1) Pretty-print to stdout
    print(json.dumps(output, indent=2, ensure_ascii=False))

    # 2) Save to file
    filename = f"{args.topic}.json"
    with open(filename, "w", encoding="utf-8") as f:
        json.dump(output, f, indent=2, ensure_ascii=False)
    logging.info("Saved output to %s", filename)

# -----------------------------
# Gradio entrypoint
# -----------------------------
def gradio_generate(topic: str, count: int = 10) -> str:
    """
    Wrapper for Gradio: returns the JSON string.
    """
    items = generate_qa(topic, count)
    output = [item.model_dump() for item in items]
    
    with open("questions.json", "w", encoding="utf-8") as f:
        json.dump(output, f, indent=2, ensure_ascii=False)
        
    return json.dumps(output, indent=2, ensure_ascii=False)

def app():
    demo = gr.Interface(
        fn=gradio_generate,
        inputs=[
            gr.Textbox(label="Topic", placeholder="Enter your topic here"),
            gr.Slider(minimum=1, maximum=50, step=1, label="Number of Q&A items", value=10)
        ],
        outputs=gr.Textbox(label="Generated JSON"),
        title="Transcript Generator for Manim Scene",
        description="Generates JSON transcript for Manim Scene, with voiceover."
    )
    demo.launch(share=True, mcp_server=True)

# -----------------------------
# Bootstrap
# -----------------------------
if __name__ == "__main__":
    # Decide between CLI and UI based on presence of command-line args
    import sys
    if len(sys.argv) > 1:
        cli_main()
    else:
        app()