Python-Projekte/dada/convert.py

from pathlib import Path
import re
import zipfile


class Converter:
    pass


class DadaConverter:
    current_file: str = ''

    def __init__(self, folder):
        base = Path(folder)
        self.input_dir = base.joinpath('input')
        self.output_dir = base.joinpath('output')
        self.temp_dir = base.joinpath('temp')
        self.logs_dir = base.joinpath('logs')
        self.re_comp = re.compile(r'>([^<]+)<')
        with open(base.joinpath('config/translation.csv'), 'r') as frh:
            self.translation = dict([row.split('\t') for row in frh.read().split('\n') if row.strip() != ''])
        self.translation['\n'] = '\n'
        self.translation[' '] = ' '
        self.missing = []

    def run(self):
        for filename in self.input_dir.glob('*.epub'):
            self.cleanup_temp(self.temp_dir)
            print(filename.name)
            self.epub_workflow(filename)
        self.cleanup_temp(self.temp_dir)
        with open(self.logs_dir.joinpath('missing.txt'), 'w') as fwh:
            fwh.write(','.join(set(self.missing)))

    def cleanup_temp(self, folder):
        for name in folder.glob('*'):
            if name.is_dir():
                self.cleanup_temp(name)
                name.rmdir()
            else:
                name.unlink()

    def epub_workflow(self, filename):
        self.unzip(filename)
        # self.convert_epub_htm_folder()
        self.rezip(self.output_dir.joinpath(filename.name))

    def unzip(self, filename):
        with zipfile.ZipFile(filename, 'r') as zfrh:
            zfrh.extractall(self.temp_dir)

    def convert_epub_htm_folder(self):
        for filename in self.temp_dir.joinpath('OEBPS/Text').glob('*.htm'):
            print(filename)
            self.convert_file(filename)

    def convert_file(self, filename):
        with open(filename, 'r') as frh:
            content = frh.read()

        content = self.re_comp.sub(self.convert_string, content)
        with open(filename, 'w') as frh:
            frh.write(content)

    def convert_string(self, match):
        text = match.group(1)
        try:
            res = [self.translation[c] for c in text]
        except KeyError as e:
            self.missing.append(e.args[0])
            res = list(text)
        return '>' + ''.join(res) + '<'

    def rezip(self, filename):
        file_list = self.file_list(self.temp_dir)
        with zipfile.ZipFile(filename, 'w', compression=0, compresslevel=None) as zfwh:
            for name in file_list:
                zfwh.write(name)

    def file_list(self, folder):
        res = []
        for name in folder.glob('*'):
            if name.is_dir():
                res.extend(self.file_list(name))
            else:
                res.append(name)
        return res

if __name__ == '__main__':
    DadaConverter(str(Path(__file__).parent)).run()