app.py

import os, io, time, base64, random, subprocess
from typing import Optional
from urllib.parse import quote

import requests
from PIL import Image
import gradio as gr

# -------- Modal inference endpoint (dev) --------
INFERENCE_URL = "https://moonmath-ai-dev--moonmath-i2v-backend-moonmathinference-run.modal.run"

# -------- small helpers --------
def _save_video_bytes(data: bytes, tag: str) -> str:
    os.makedirs("/tmp", exist_ok=True)
    path = f"/tmp/{tag}_{int(time.time())}.mp4"
    with open(path, "wb") as f:
        f.write(data)
    return path

def _png_bytes(img: Image.Image) -> bytes:
    buf = io.BytesIO()
    img.save(buf, format="PNG")
    return buf.getvalue()

def _download_to_bytes(url: str) -> bytes:
    r = requests.get(url, timeout=180)
    r.raise_for_status()
    return r.content

def stitch_call(start_img: Image.Image, end_img: Image.Image, prompt: str, seed: Optional[int]) -> Optional[str]:
    if start_img is None or end_img is None:
        return None

    if seed in (None, 0, -1):
        seed = random.randint(1, 2**31 - 1)

    url = f"{INFERENCE_URL}?prompt={quote(prompt or '')}&seed={seed}"

    files = {
        "image_bytes": ("start.png", _png_bytes(start_img), "image/png"),
        "image_bytes_end": ("end.png", _png_bytes(end_img), "image/png"),
    }
    headers = {"accept": "application/json"}

    try:
        resp = requests.post(url, files=files, headers=headers, timeout=600)
        ctype = (resp.headers.get("content-type") or "").lower()

        # Raw video bytes
        if "application/json" not in ctype:
            resp.raise_for_status()
            return _save_video_bytes(resp.content, "stitch")

        # JSON with url or base64
        data = resp.json()
        video_url = data.get("video_url") or data.get("url") or data.get("result")
        if isinstance(video_url, str) and video_url.startswith("http"):
            b = _download_to_bytes(video_url)
            return _save_video_bytes(b, "stitch")

        video_b64 = data.get("video_b64")
        if isinstance(video_b64, str):
            pad = (-len(video_b64)) % 4
            if pad:
                video_b64 += "=" * pad
            b = base64.b64decode(video_b64)
            return _save_video_bytes(b, "stitch")

    except Exception as e:
        print("stitch_call error:", e)

    return None

# -------- FFmpeg-based concatenation --------
def concat_videos(vid1: str, vid2: str) -> Optional[str]:
    if not vid1 or not vid2:
        return None
    try:
        os.makedirs("/tmp", exist_ok=True)
        out_path = f"/tmp/final_{int(time.time())}.mp4"

        # Create a temporary file list for ffmpeg
        list_file = f"/tmp/list_{int(time.time())}.txt"
        with open(list_file, "w") as f:
            f.write(f"file '{vid1}'\n")
            f.write(f"file '{vid2}'\n")

        # Run ffmpeg concat
        subprocess.run(
            ["ffmpeg", "-y", "-f", "concat", "-safe", "0", "-i", list_file, "-c", "copy", out_path],
            check=True,
            stdout=subprocess.PIPE,
            stderr=subprocess.PIPE,
        )

        return out_path
    except Exception as e:
        print("concat_videos error:", e)
        return None

# -------- Gradio callbacks --------
def stitch_12(prompt12, seed, img1, img2):
    path = stitch_call(img1, img2, prompt12 or "", int(seed or 0))
    return path

def stitch_23(prompt23, seed, img2, img3):
    path = stitch_call(img2, img3, prompt23 or "", int(seed or 0))
    return path

def stitch_all(vid12, vid23):
    if vid12 is None or vid23 is None:
        gr.Warning("Generate both videos first before stitching all.")
        return None
    return concat_videos(vid12, vid23)

# -------- UI --------
CSS = """
.gradio-container { padding: 24px; }
"""

with gr.Blocks(css=CSS, title="Stitch — 3 uploads, 2 stitches") as demo:
    gr.Markdown("## Stitch — Upload 3 images → Generate 1→2, 2→3, then combine.")

# --- Uploads row (side-by-side) ---
with gr.Row():
    with gr.Column(scale=1, min_width=280):
        img1 = gr.Image(label="Image 1 upload", type="pil")
    with gr.Column(scale=1, min_width=280):
        img2 = gr.Image(label="Image 2 upload", type="pil")
    with gr.Column(scale=1, min_width=280):
        img3 = gr.Image(label="Image 3 upload", type="pil")

    with gr.Row():
        # Prompts + buttons
        with gr.Column():
            seed = gr.Number(value=0, precision=0, label="Seed (0 = random)")
            prompt12 = gr.Textbox(placeholder="Prompt for stitching 1→2", lines=2, label="Prompt (1→2)")
            btn12 = gr.Button("Stitch 1&2")
            prompt23 = gr.Textbox(placeholder="Prompt for stitching 2→3", lines=2, label="Prompt (2→3)")
            btn23 = gr.Button("Stitch 2&3")
            btn_all = gr.Button("Stitch All (combine 1→2 and 2→3)")

        with gr.Column():
            vid12 = gr.Video(label="Video (1→2)")
            vid23 = gr.Video(label="Video (2→3)")
            vid_all = gr.Video(label="Final Combined Video")

    # Wire
    btn12.click(stitch_12, inputs=[prompt12, seed, img1, img2], outputs=[vid12])
    btn23.click(stitch_23, inputs=[prompt23, seed, img2, img3], outputs=[vid23])
    btn_all.click(stitch_all, inputs=[vid12, vid23], outputs=[vid_all])

if __name__ == "__main__":
    demo.queue().launch()