"""
Extract embedded images from each PPT in C:\\Users\\admin\\Downloads\\illustrations
into images/<topic_slug>/NN.<ext>.

Topic slug is derived from the PPT filename. Internal ordering within a topic
is not semantic (the team curated each PPT as a topic set), so images are
numbered 01..NN in the order they appear inside the PPT's ppt/media/ folder.

Original extension (png / jpeg) is preserved. No resizing is performed.

Safety: refuses to overwrite an existing target file. Refuses to write into
a topic folder that already contains files (so this script can't clobber
prior work).
"""
from __future__ import annotations

import re
import sys
import zipfile
from pathlib import Path

SRC      = Path(r"C:\Users\admin\Downloads\illustrations")
DST_ROOT = Path(r"C:\Users\admin\Desktop\v2\images")
IMAGE_PREFIX = "ppt/media/"
IMAGE_EXTS = {".png", ".jpg", ".jpeg"}


def topic_slug(stem: str) -> str:
    s = stem.lower()
    s = re.sub(r"[&]", "and", s)
    s = re.sub(r"[^a-z0-9]+", "_", s).strip("_")
    return s


def extract(pptx_path: Path, dst_dir: Path) -> tuple[int, int]:
    """Returns (files_written, bytes_written)."""
    if dst_dir.exists() and any(dst_dir.iterdir()):
        print(f"  [skip] {dst_dir} already has files; refusing to clobber.")
        return (0, 0)

    dst_dir.mkdir(parents=True, exist_ok=True)
    written = 0
    bytes_written = 0
    with zipfile.ZipFile(pptx_path) as zf:
        media = [
            i for i in zf.infolist()
            if i.filename.startswith(IMAGE_PREFIX)
            and Path(i.filename).suffix.lower() in IMAGE_EXTS
        ]
        for idx, info in enumerate(media, start=1):
            ext = Path(info.filename).suffix.lower()
            if ext == ".jpg":
                ext = ".jpeg"
            target = dst_dir / f"{idx:02d}{ext}"
            if target.exists():
                print(f"  [skip] {target} exists.")
                continue
            with zf.open(info) as src, open(target, "wb") as dst:
                data = src.read()
                dst.write(data)
                bytes_written += len(data)
                written += 1
    return written, bytes_written


def main() -> int:
    if not SRC.exists():
        print(f"[error] source folder not found: {SRC}", file=sys.stderr)
        return 1

    ppts = sorted(SRC.glob("*.pptx"))
    if not ppts:
        print("[warn] no .pptx files found.")
        return 0

    total_files = 0
    total_bytes = 0
    print("EXTRACTION REPORT")
    print("-" * 78)
    print(f"{'topic_slug':<48} {'files':>6} {'size_MB':>10}")
    print("-" * 78)
    for p in ppts:
        slug = topic_slug(p.stem)
        dst  = DST_ROOT / slug
        files, byts = extract(p, dst)
        total_files += files
        total_bytes += byts
        print(f"{slug:<48} {files:>6} {byts/1024/1024:>10.2f}")
    print("-" * 78)
    print(f"{'TOTAL':<48} {total_files:>6} {total_bytes/1024/1024:>10.2f}")
    print()
    print(f"Destination: {DST_ROOT}")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())