Convert Anki decks from JSON to YAML and vice versa.

convert_yaml_json.py

#! /usr/bin/env python3
# -*- coding: utf-8 -*-
import os
import sys
import timeit
import datetime
import argparse
import tempfile
import pathlib

# python3 -m pip install pyyaml
# import yaml
# https://stackoverflow.com/questions/71836675/force-pyyaml-to-write-multiline-string-literals-regardless-of-string-content
# python3 -m pip install ruamel.yaml
import ruamel.yaml as yaml
import sys
import json

from pathlib import Path
from typing import Optional

g_start_time = timeit.default_timer()
SCRIPT_DIRECTORY = pathlib.Path(__file__).parent.resolve()


def main() -> Optional[int]:
    argumentsNamespace = g_argumentParser.parse_args()
    if argumentsNamespace.help:
        g_argumentParser.print_help()
        return

    to_json = argumentsNamespace.to_json
    fast_scandir(SCRIPT_DIRECTORY, max_recursion=1, to_json=to_json)


def fast_scandir(current_directory: Path, max_recursion=1, to_json=False, recursion_depth=0):
    subfolders = []
    for subpath in os.scandir(current_directory):
        fullpath = current_directory / subpath

        if fullpath.is_dir():
            if recursion_depth < max_recursion:
                subfolders.extend(
                    fast_scandir(
                        fullpath,
                        max_recursion=max_recursion,
                        to_json=to_json,
                        recursion_depth=recursion_depth + 1
                        )
                    )
        else:
            if to_json:
                if fullpath.name == "deck.yml":
                    newjsonpath = fullpath.with_suffix(".json")
                    with open(fullpath, 'r') as readyml, \
                            tempfile.NamedTemporaryFile(
                                mode="w",
                                dir=current_directory,
                                suffix=".ignore",
                                newline='\r\n',
                                encoding='utf8',
                            ) as temporaryfile:
                        print(f"Converting to json {fullpath}.", file=sys.stderr)
                        json.dump(yaml.safe_load(readyml), temporaryfile, indent=4, ensure_ascii=False)
                        temporaryfile.flush()  # atomic write for safety
                        os.replace(temporaryfile.name, newjsonpath)
                        print_time()
                        # NamedTemporaryFile with delete=True should not fail if file already deleted
                        # https://bugs.python.org/issue29573
                        with open(temporaryfile.name, 'w') as fake:
                            fake.write("nothing")
            else:
                if fullpath.name == "deck.json":
                    newymlpath = fullpath.with_suffix(".yml")
                    with open(fullpath, 'r') as readjson, \
                            tempfile.NamedTemporaryFile(
                                mode="w",
                                dir=current_directory,
                                suffix=".ignore",
                                newline='\n',
                                encoding='utf8',
                            ) as temporaryfile:
                        print(f"Converting to yml {fullpath}.", file=sys.stderr)
                        yaml.dump(json.load(readjson), temporaryfile,
                            default_flow_style=False,
                            default_style='|',
                            allow_unicode=True)
                        temporaryfile.flush()  # atomic write for safety
                        os.replace(temporaryfile.name, newymlpath)
                        print_time()
                        # NamedTemporaryFile with delete=True should not fail if file already deleted
                        # https://bugs.python.org/issue29573
                        with open(temporaryfile.name, 'w') as fake:
                            fake.write("nothing")

    return subfolders


# Call it without the start parameter to print the time since the last call
def print_time(start=None):
    global g_start_time
    start = g_start_time if start is None else start
    end = timeit.default_timer()
    print("Time elapsed %s... " % datetime.timedelta(seconds=end-start),
        end="\n", file=sys.stderr)
    g_start_time = end


# Simple python program example to run tests together with your code using pytest
# https://gist.github.com/evandrocoan/69aeaee3e16ad77fc5c889cfd55fc698
g_argumentParser = argparse.ArgumentParser(
        description = \
"""
Convert Anki decks from JSON to YAML and vice versa.
""",
        formatter_class=argparse.RawTextHelpFormatter,
        add_help=False,
    )

g_argumentParser.add_argument( "-h", "--help", action="store_true",
help= """
Show this help message and exits.
""" )

g_argumentParser.add_argument( "-j", "--to-json", action="store_true", default=None,
        help=
"""
Convert Anki deck back from YAML to json.
""" )


if __name__ == '__main__':
    start = timeit.default_timer()
    code = main()
    print_time(start)
    sys.exit(code if code else 0)