Rename and re-encode Nulab Backlog CSV files in a directory.

ren_n_utf8_bklog_csv.py

#!/usr/bin/env python3
import re
from pathlib import Path

import pandas as pd


def glob_backlog_csvs(csv_dir):
    csv_files = list(Path(csv_dir).glob('Backlog-Issues*.csv'))
    if csv_files:
        return csv_files
    else:
        exit('\033[93mNo CSV files found.\033[0m')


def try_open_with_encoding(file_path, encoding):
    """Try to open the file with a specific encoding.
    """
    try:
        with file_path.open(encoding=encoding) as f:
            f.read(64)  # Sample read to validate encoding
            return True
    except UnicodeDecodeError:
        return False


def detect_file_encoding(file_path):
    """Detect the encoding of a given file.
    """
    for encoding in ['UTF8', 'CP932']:
        if try_open_with_encoding(file_path, encoding):
            return encoding
    return 'unknown'


def rename_csv(csv_file_path):
    """Rename and re-encode a backlog CSV file.
    Returns:
        fname_ (str): New file name.
        df_ (DataFrame): DataFrame in utf-8.
    Example:
        Backlog-Issues-19991231-2359.csv -> backlog_LONG_NOW_19991231-2359.csv
    """
    dttm = re.search(r'-(\d{8}-\d{4})', csv_file_path.stem).group(1)
    encoding = detect_file_encoding(csv_file_path)

    if encoding == 'UTF8':
        df_ = pd.read_csv(csv_file_path, encoding='UTF8')
        project_key = df_.at[0, 'Key'].rsplit('-', maxsplit=1)[0]
    elif encoding == 'CP932':
        df_ = pd.read_csv(csv_file_path, encoding='CP932')
        project_key = df_.at[0, 'キー'].rsplit('-', maxsplit=1)[0]
    else:
        exit(f'\033[93mUnknown encoding for: {csv_file_path}\033[0m')

    fname_ = f'backlog_{project_key}_{dttm}.csv'
    return fname_, df_


if __name__ == '__main__':
    source_dir = Path(input('Backlog CSVs DIR? >> '))
    if not source_dir.is_dir():
        exit('\033[93mNot a directory.\033[0m')

    # Prep a directory for original files.
    old_dir = source_dir / 'old'
    if not old_dir.is_dir():
        old_dir.mkdir()

    # Rename and re-encode.
    for orig_csv in glob_backlog_csvs(source_dir):
        fname, df = rename_csv(orig_csv)
        fpath = source_dir / fname
        df.to_csv(fpath, index=False)
        print(f'\033[93mSaved: {fpath}\033[0m')
        orig_csv.rename(old_dir / orig_csv.name)