crowdin_parse.py

#!/usr/bin/python3

import csv
import os
import re
import shutil
import subprocess
import configparser

import click

config = configparser.RawConfigParser()
config.read('crowdin.cfg')

try:
    TEMPLATE = config.get('TEMPLATES', 'template')
    ITEM_TEMPLATE = config.get('TEMPLATES', 'item-template')
    KEYS = dict(config.items('KEYS'))
    BASE_PATH = config.get('PATHS', 'base-path')
    PATH = config.get('PATHS', 'path')
    FILENAME = config.get('FILE', 'filename')
    FILE_EXT = config.get('FILE', 'extension')
    ONLY_FILES = config.get('FILE', 'only_files')
except configparser.NoOptionError as e:
    click.secho(e.message, fg='red')
    exit()


@click.group()
def cli():
    """Simple Crowdin parser"""
    pass


def _mkdir(dir):
    if not os.path.exists(dir):
        os.makedirs(dir)


@cli.command()
@click.argument("namespace", nargs=1)
def download(namespace):
    """Download and extract project"""
    if namespace in KEYS:
        directory = os.path.join(BASE_PATH, 'translations_source', namespace)
        url = "https://api.crowdin.com/api/project/%s/download/all.zip?key=%s" % (namespace, KEYS[namespace])
        _mkdir(directory)
        subprocess.Popen(['wget', url, '-O', namespace + '.zip'], cwd=directory).wait()
        subprocess.Popen(['unzip', '-o', '-q', namespace + '.zip'], cwd=directory).wait()
    else:
        click.secho("%s not found in projects.json file." % namespace, fg='red')
        exit()


def parse(path, namespace, filename):
    with open(path) as csv_file:

        if ONLY_FILES and filename not in ONLY_FILES:
            return

        reader = csv.DictReader(csv_file)
        path_list = path.split('/')
        lang = path_list[path_list.index('translations_source') + 2]
        directory = BASE_PATH + PATH.format(lang=lang)

        if not FILENAME:
            filename = re.sub(r'(\'|&| )', '', filename)
            result = os.path.join(directory, filename.replace('csv', FILE_EXT))
        else:
            result = os.path.join(directory, FILENAME.format(
                lang=lang
            )) + '.' + FILE_EXT

        _mkdir(directory)

        with open(result, 'w') as file:
            str_keys = ''
            for row in reader:
                key = row['key-id'].replace('"', "")
                translated = row['source'] if not row['translation'] else row['translation']
                str_keys += ITEM_TEMPLATE.format(
                    key=key,
                    value=translated.replace('"', "'").splitlines()[0],
                    n='\n',
                    t='\t'
                )

            file.write(TEMPLATE.format(
                array=str_keys,
                n='\n'
            ))


@cli.command()
@click.argument('namespace', nargs=1)
def convert(namespace):
    """Convert project to language files"""
    path = os.path.join(BASE_PATH, 'translations_source', namespace)
    if os.path.isdir(path):
        for (dirpath, dirnames, filenames) in os.walk(path):
            for filename in [f for f in filenames if f.endswith(".csv")]:
                parse(os.path.join(dirpath, filename), namespace, filename)
        click.secho("%s successfully converted." % namespace, fg='green')
    else:
        click.secho("%s not found in source folder. You need to download in first." % namespace, fg='red')
        exit()


@cli.command()
def cleanup():
    """Delete 'translations_source' folder"""
    SOURCE_DIR = os.path.join(BASE_PATH, 'translations_source')
    if os.path.exists(SOURCE_DIR):
        shutil.rmtree(SOURCE_DIR)


@cli.command()
@click.pass_context
def run(ctx):
    """Download and convert all projects"""
    for namespace in KEYS:
        ctx.invoke(download, namespace=namespace)
        ctx.invoke(convert, namespace=namespace)
        ctx.invoke(cleanup)
    click.secho("All done!", fg='green', bold=True)


if __name__ == '__main__':
    run()