pptList/scripts/inspect_pptx_picture_layouts.py

from __future__ import annotations

import json
import sys
from pathlib import Path

from pptx import Presentation
from pptx.enum.shapes import MSO_SHAPE_TYPE


MANIFEST = Path(sys.argv[1] if len(sys.argv) > 1 else "artifacts/music-template-extraction/review/manifest.json")
TARGETS = set(sys.argv[2:] or ["candidate-05", "candidate-20", "candidate-26", "candidate-30", "candidate-31", "candidate-33"])


def main() -> None:
    items = json.loads(MANIFEST.read_text(encoding="utf-8"))["items"]
    for item in items:
        if item["id"] not in TARGETS:
            continue
        deck = Presentation(item["source"])
        print(f"\n{item['id']} {item['name']}")
        for sample in item["samples"]:
            number = int(Path(sample).stem.split("-")[1])
            slide = deck.slides[number - 1]
            print(f" slide={number}")
            for index, shape in enumerate(slide.shapes):
                if shape.shape_type != MSO_SHAPE_TYPE.PICTURE:
                    continue
                left = shape.left / deck.slide_width
                top = shape.top / deck.slide_height
                width = shape.width / deck.slide_width
                height = shape.height / deck.slide_height
                print(f"  picture={index:02d} box=({left:.3f},{top:.3f},{width:.3f},{height:.3f}) area={width * height:.3f}")


if __name__ == "__main__":
    main()