#! /usr/bin/env python

# Copyright 2021, 2022 Steinar Knutsen
#
# Licensed under the EUPL, Version 1.2 or - as soon they will be approved by the
# European Commission - subsequent versions of the EUPL (the "Licence"); You may
# not use this work except in compliance with the Licence. You may obtain a copy
# of the Licence at:
#
# https://joinup.ec.europa.eu/collection/eupl/eupl-text-eupl-12
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the Licence is distributed on an "AS IS" basis, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# Licence for the specific language governing permissions and limitations under
# the Licence.

import sys, getopt

HELP_TEXT = """txt2itxt [-E] [-S] [-h] [-l] [-m] [-s]

-E
    only use explicit rules
-S
    treat paragraphs of short lines as preformatted (default off)
-h
    print this text and exit
-m
    treat mailquoting as monospaced (leading "> ") (default on)
-s
    treat paragraphs where some line has "many" consecutive space as monospaced
    (default on)

Read from STDIN and write to STDOUT, heuristically converting from plain text to
limited ITXT.
"""

MONO = "--------------------------------------------------------------------------------"

def series_of_short_lines(raw_lines):
    linelengths = [len(line) for line in raw_lines]
    chars = sum(linelengths)
    max_line = max(linelengths)
    if chars / len(raw_lines) < 55 and max_line < 66:
        return True
    return False

def many_consecutive_spaces(raw_lines):
    for line in raw_lines:
        if "   " in line:
            return True
    return False

def mail_quoting(raw_lines):
    for line in raw_lines:
        if line[0] != '>':
            return False
    return True

def space_leading_any_line(raw_lines):
    for line in raw_lines:
        if line[0].isspace():
            return True
    return False

def use_proportional(raw_lines, rules):
    if space_leading_any_line(raw_lines):
        return False
    if series_of_short_lines in rules:
        if series_of_short_lines(raw_lines):
            return False
    if many_consecutive_spaces in rules:
        if many_consecutive_spaces(raw_lines):
            return False
    if mail_quoting in rules:
        if mail_quoting(raw_lines):
            return False
    return True

def convert_paragraph(lines, rules):
    if use_proportional(lines, rules):
        return lines

    new_lines = [ "-" * 80 + "\n"]
    new_lines.extend(lines)
    new_lines.append(new_lines[0])
    return new_lines

def convert(src, dst, rules):
    paragraph = []
    for line in src:
        if len(line.strip()) == 0:
            if len(paragraph) > 0:
                dst.writelines(convert_paragraph(paragraph, rules))
                paragraph.clear()
            dst.write(line)
        elif line.rstrip() == MONO:
            raise ValueError("ITXT monospace marker in input, not handled.")
        else:
            paragraph.append(line)
    if len(paragraph) > 0:
        dst.writelines(convert_paragraph(paragraph, rules))
        paragraph.clear()

def main():
    rules = set()
    rules.add(many_consecutive_spaces)
    rules.add(mail_quoting)
    options, arguments = getopt.gnu_getopt(sys.argv[1:], "EShms")
    for option, value in options:
        if option == "-E":
            rules.clear()
    for option, value in options:
        if option == "-h":
            print(HELP_TEXT)
            sys.exit(0)
        elif option == "-S":
            rules.add(series_of_short_lines)
        elif option == "-m":
            rules.add(mail_quoting)
        elif option == "-s":
            rules.add(many_consecutive_spaces)
    convert(sys.stdin, sys.stdout, rules)

if __name__ == "__main__":
    main()