#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Benchmark of html-to-pdf converters.
(c) @justhabrauser, GPLv3.
"""
# 1. system
import os
import sys
from time import time
# 2. 3rd
from pdfkit import from_string          # https://github.com/JazzCore/python-pdfkit
from weasyprint import HTML             # https://github.com/Kozea/WeasyPrint
from xhtml2pdf.pisa import CreatePDF    # https://github.com/xhtml2pdf/xhtml2pdf
def __pdfkit(html: str) -> bytes:
    return from_string(html, False, options={'quiet': ''})
def __weasy(html: str) -> bytes:
    return HTML(string=html).write_pdf()
def __pisa(html: str) -> bytes:
    pdf = CreatePDF(html)
    if not pdf.err:
        pdf.dest.seek(0)
        return pdf.dest.read()
def main(indir: str, outdir: str, count: int) -> None:
    # 1. Load all htmls
    modules = (__pdfkit, __weasy, __pisa)
    html_list = list()      # (filename, content)[]
    dir_list = os.listdir(indir)
    dir_list.sort()
    for i, fn in enumerate(dir_list):
        fpath = os.path.join(indir, fn)
        if os.path.isfile(fpath) and fpath.endswith(".html"):
            print("Load '{}'".format(fn), file=sys.stderr)
            with open(fpath, "rt") as i_f:
                html = i_f.read()
                html_list.append(html)
                # 2. write results (and warm up)
                for j, m in enumerate(modules):
                    with open(os.path.join(outdir, "%d_%d.pdf" % (i, j)), "wb") as o_f:
                        o_f.write(m(html))
    # 2. for C times x I pages x J engines:
    print("Count\tPage\tEngine\tTime\n=====\t====\t======\t====")
    for c in range(count):                   # count
        for i, h in enumerate(html_list):    # html page
            for j, m in enumerate(modules):  # engine
                t0 = time()
                m(h)
                t1 = time()
                print("{}\t{}\t{}\t{}".format(c, i, j, t1-t0))
if __name__ == '__main__':
    if len(sys.argv) != 3:
        print("Usage: {} <dir_with_htmls> <output_dir_for_pdfs>".format(sys.argv[0]), file=sys.stderr)
    elif not os.path.isdir(sys.argv[1]):
        print("Input '{}' is not dir or not exists.".format(sys.argv[1]), file=sys.stderr)
    elif not os.path.isdir(sys.argv[2]):
        print("Output '{}' is not dir or not exists.".format(sys.argv[2]), file=sys.stderr)
    else:
        main(sys.argv[1], sys.argv[2], 5)