hinamizawa-downporter/src/orig.py

#!/usr/bin/env python3

import os

from unidecode import unidecode

import config
import parser
import flow

japanese_ranges = [
    (0x4E00, 0x9FFF),  # Kanji
    (0x3040, 0x309F),  # Hiragana
    (0x30A0, 0x30FF),  # Katakana
    (0xFF00, 0xFFEF),  # Full-width Roman characters and symbols
    (0x3000, 0x303F),  # CJK symbols and punctuation (including 「」)
]


def process_sections():
    output_filepath = os.path.join(config.get('output_path'), 'out.txt')
    outfile = open(output_filepath, 'w', encoding='shift_jisx0213')
    origfile = open(config.get('original_path'), 'r', encoding='shift_jisx0213')

    sections = dict()

    for line in origfile:
        if line.startswith('*'):
            section_name = line[1:].split(' ', 1)[0].replace('\n','')

            outfile.write(line)

            if section_name in flow.onik:
                print("entering", section_name)
                write_translated(
                    outfile,
                    origfile,
                    flow.onik[section_name],
                )
        else:
            outfile.write(line)

    outfile.close()
    origfile.close()


# Given a set of translation files, the original file and the output file
# replace the japanese lines with the translated ones in a given section.
def write_translated(outfile, origfile, translation_file_paths):
    for transfilepath in translation_file_paths:
        print(f'- reading "{transfilepath}"')
        structure = parser.parse_to_structure(transfilepath)

        for i, line in enumerate(origfile):
            found = False
            for start, end in japanese_ranges:
                if start <= ord(line[0]) <= end:
                    found = True

            if found:
                # The amount of lines may not coincide because the original
                # might have one line for what we have multiple lines. Count
                # the number of appearances of the end-of-command symbols
                # (@ and ¥) to determine how many of the translated lines is
                # equivalent to the given original line.
                amount = line.count("@") + line.count("¥") + line.count('/')

                outfile.write('`')

                _printed_line = ""
                while True:
                    if amount <= 0:
                        break

                    if structure[0].type == 'Line_ContinueAfterTyping':
                        amount += 1

                    _printed_line += structure[0].text_jp
                    outfile.write(
                        unidecode(structure.pop(0).text_en).replace("\\", "¥")
                    )

                    amount -= 1

                outfile.write('\n')

                print("\n-", transfilepath)
                print(">", _printed_line)
                print("<", line, end='')


                # Used up all of the structures, this chapter has ended.
                # Got to the next one
                if len(structure) <= 0:
                    break

            else:
                outfile.write(line)