#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import os
import sys
import time
import traceback
import gc

import weasyprint
import pypdf


BATCH_SIZE = 10


def convert(input_file_names: list[str], output_file_name: str, encoding: str = None) -> list[int]:
    # it takes roughly 100ms to generate one document
    page_nums = []
    tmp_file_names = []

    for i in range(0, len(input_file_names), BATCH_SIZE):
        batch = input_file_names[i:i + BATCH_SIZE]
        documents = []
        for file_name in batch:
            html = weasyprint.HTML(file_name, encoding=encoding)
            doc = html.render()
            documents.append(doc)
            del html
        all_pages = [p for doc in documents for p in doc.pages]
        tmp_file_name = f'{output_file_name}.part.{i:0000}'
        documents[0].copy(all_pages).write_pdf(tmp_file_name)
        tmp_file_names.append(tmp_file_name)
        page_nums += [len(doc.pages) for doc in documents]
        del documents
        del all_pages
        gc.collect()

    merger = pypdf.PdfWriter()
    for pdf in tmp_file_names:
        merger.append(pdf)
        os.remove(pdf)
    merger.write(output_file_name)
    merger.close()

    return page_nums


def _wrapper_convert(args: list[str], encoding: str = None) -> None:
    try:
        t0 = time.process_time()
        pages = convert(args[:-1], args[-1], encoding=encoding)
        t1 = time.process_time()
        print(f'{len(args) - 1} documents, '
              f'{sum(pages)} pages ({", ".join(str(p) for p in pages)}), '
              f'{t1 - t0:.1f} sec',
              flush=True)
    except Exception as e:
        msg = str(e).replace('\n', ' ')
        print(f'error: {msg}', flush=True)
        traceback.print_exception(e)


def usage() -> None:
    print(f'usage: {sys.argv[0]} [-h] [-d DIR] [-e ENCODING] [ - | INPUT [INPUT...] OUTPUT ]\n\n'
          'options:\n'
          ' -h, --help       show this help message and exit\n'
          ' -d, --directory  set the working directory\n'
          ' -e, --encoding   encoding of the input files\n'
          '\n'
          ' -                use stdin for retrieving input and output file names (semi-colon-seperated)\n'
          ' INPUT            name of an html input file\n'
          ' OUTPUT           name of an pdf output file', file=sys.stderr)
    sys.exit(1)


def _get_arg(args: list[str], n1: str, n2: str = None) -> str:
    v = None
    for n in [n1] + (n2 and [n2] or []):
        if n in args:
            i = args.index(n)
            if i + 1 >= len(args):
                usage()
            v = args[i + 1]
            args.pop(i)
            args.pop(i)
    return v


def main() -> None:
    args = sys.argv[1:]
    if len(args) == 0 or '-h' in args or '--help' in args:
        usage()

    working_dir = _get_arg(args, '-d', '--directory')
    if working_dir:
        os.chdir(working_dir)
    encoding = _get_arg(args, '-e', '--encoding')

    if args == ['-']:
        for line in sys.stdin:
            _wrapper_convert(line.strip().split(';'), encoding=encoding)
    elif len(args) < 2:
        usage()
    else:
        _wrapper_convert(args, encoding=encoding)


if __name__ == '__main__':
    main()
    sys.exit(0)