Fun with Python and UTF-8 files! Works on Python 2.7.x.

python.Files.UTF8.py

import codecs
import logging
import os


logger = logging.getLogger(os.path.splitext(os.path.basename(__file__))[0])


# NOTE: Python 3.x doesn't have a `unicode()` built-in function.


def read_utf8_file(path, bom=False):
    """Read a UTF-8 encoded file.

    :param path: Relative, or absolute, path of the file to read.
    :type path: str or unicode
    :param bom: (optional) Read a file and expect a UTF-8 BOM if True otherwise read UTF-8 without BOM. (default: False)
    :type bom: bool
    :return: Data or None
    :rtype: str, unicode, or None
    """
    if bom:
        enc = 'utf-8-sig'
    else:
        enc = 'utf-8'
    try:
        with codecs.open(path, 'r', enc) as f:
            return unicode(f.read().strip())
    except Exception as err:
        logger.exception(err)
        return None


def write_utf8_file(path, data, overwrite_existing=False, bom=False):
    """Write a UTF-8 encoded file.

    :param path: Relative, or absolute, path of the file to write.
    :type path: str or unicode
    :param data: Data to write.
    :type data: *
    :param overwrite_existing: (optional) If the file, overwrite it if set to True otherwise do not write.
                               (default: False)
    :param bom: (optional) Write a UTF-8 BOM if True otherwise write UTF-8 without BOM. (default: False)
    :type bom: bool
    :return: True if successful else False.
    :rtype: bool
    """
    if bom:
        enc = 'utf-8-sig'
    else:
        enc = 'utf-8'
    if os.path.exists(path):
        logger.warning('Output file already exists.')
        if not overwrite_existing:
            logger.warning('Overwrite existing is disabled.')
            return False

    try:
        with codecs.open(path, 'w', enc) as f:
            f.write(data)
    except Exception as err:
        logger.exception(err)
        return False
    return True