pdfmerger.py

"""Merge PDF files from a directory into ``merged.pdf``.

Usage: python3.14 pdfmerger.py DIRECTORY [--sort-type date|prefix] [--no-bookmark]
Requires the ``pypdf`` package.
"""

from __future__ import annotations

import argparse
import sys
from pathlib import Path

from pypdf import PdfWriter


def prefix_key(path: Path) -> tuple[float, str]:
    prefix = path.name.split("_", 1)[0]
    return (float(prefix) if prefix.isdigit() else float("inf"), path.name.casefold())


def date_key(path: Path) -> float:
    return path.stat().st_mtime


def name_key(path: Path) -> str:
    return path.name.casefold()


def merge_pdfs(
    directory: Path,
    *,
    sort_type: str | None = None,
    bookmarks: bool = True,
    output: Path = Path("merged.pdf"),
) -> int:
    """Merge PDFs and return the number of input files."""
    files = [path for path in directory.iterdir() if path.is_file() and path.suffix.lower() == ".pdf"]
    if sort_type == "date":
        files.sort(key=date_key)
    elif sort_type == "prefix":
        files.sort(key=prefix_key)
    else:
        files.sort(key=name_key)
    if not files:
        raise ValueError("no PDF files found")
    writer = PdfWriter()
    try:
        for path in files:
            bookmark = path.stem if bookmarks else None
            print(f"Merging: {path.name}")
            writer.append(path, outline_item=bookmark)
        with output.open("wb") as handle:
            writer.write(handle)
    finally:
        writer.close()
    return len(files)


def main(argv: list[str] | None = None) -> int:
    parser = argparse.ArgumentParser(description="Merge PDF files from a directory.")
    parser.add_argument("path", type=Path, help="directory containing PDF files")
    parser.add_argument("--sort-type", choices=("date", "prefix"))
    parser.add_argument("--no-bookmark", action="store_true")
    args = parser.parse_args(argv)
    if not args.path.is_dir():
        parser.error(f"directory not found: {args.path}")
    try:
        count = merge_pdfs(args.path, sort_type=args.sort_type, bookmarks=not args.no_bookmark)
    except (OSError, ValueError) as exc:
        print(f"Error: {exc}", file=sys.stderr)
        return 1
    print(f"Done. Merged {count} files into merged.pdf")
    return 0


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