How to use the tonic.testing.VICTestError function in tonic

def check_for_nans(df):
    """check if dataframe has nans in it"""
    if df.isnull().any().any():
        raise VICTestError('VIC output has nans in it!')
    else:
        return
# -------------------------------------------------------------------- #

default='$WORKDIR/VIC_tests_{0}'.format(ymd))
    parser.add_argument('--data_dir', type=str,
                        help='directory to find test data',
                        default=os.path.join(test_dir, '../samples/VIC_sample_data'))
    args = parser.parse_args()

    # Define test directories
    data_dir = args.data_dir
    out_dir = os.path.expandvars(args.output_dir)
    os.makedirs(out_dir, exist_ok=True)

    # Validate input directories
    if not (len(args.tests) == 1 and args.tests[0] == 'unit'):
        for d in [data_dir, test_dir]:
            if not os.path.exists(d):
                raise VICTestError('Directory: {0} does not exist'.format(d))

    # Print welcome information
    print(description)
    print('\nStarting tests now...Start Time: {0}\n'.format(starttime))
    print('Running Test Set: {0}'.format(', '.join(args.tests)))

    # Setup VIC executable
    if not (len(args.tests) == 1 and args.tests[0] == 'unit'):
        vic_exe = VIC(args.vic_exe)
        print('VIC version information:\n\n{0}'.format(vic_exe.version.decode()))

    # run test sets
    # unit
    if any(i in ['all', 'unit'] for i in args.tests):
        test_results['unit'] = run_unit_tests(test_dir)

def process_error(error, vic_exe):
    '''Helper function to process possible error raised during testing'''
    tail = None
    if isinstance(error, VICRuntimeError):
        test_comment = 'Test failed during simulation'
        tail = vic_exe.stderr
    elif isinstance(error, VICTestError):
        test_comment = 'Test failed during testing of output files'
    elif isinstance(error, VICValgrindError):
        test_comment = 'Test failed due to memory error detected by valgrind'
        tail = vic_exe.stderr
    elif isinstance(error, VICReturnCodeError):
        test_comment = 'Test failed due to incorrect return code'
        tail = vic_exe.stderr
    elif isinstance(error, AssertionError):
        test_comment = 'AssertionError raised during testing'
    else:
        test_comment = 'Unknown test failure'
        traceback.print_stack()

    print('\t{0}'.format(test_comment))
    print('\t{0}'.format(error))
    if tail is not None:

def check_completed(df, start, end):
    if not df.index[0] == start:
        raise VICTestError(
            'Start dates ({0} and {1}) do not match'.format(df.index[0], start))
    if not df.index[-1] == end:
        raise VICTestError(
            'End dates ({0} and {1}) do not match'.format(df.index[-1], end))
    return

days=1)).strftime('%Y%m%d'),
                0))
        if state_format == 'ASCII':
            states = read_ascii_state(state_fname)
        elif state_format == 'BINARY':
            states = read_binary_state(state_fname)
        # Compare split run states with full run
        # --- If ASCII state file, check if almost the same ---#
        if state_format == 'ASCII':
            np.testing.assert_almost_equal(states, states_full_run, decimal=3,
                                           err_msg='States are not a '
                                                   'close match')
        # --- If BINARY state file, check if exactly the same ---#
        elif state_format == 'BINARY':
            if states != states_full_run:
                raise VICTestError('Restart causes inexact state outputs!')

    elif driver == 'image':
        # Read the state file at the end of the last period of run
        state_fname = os.path.join(
            state_basedir,
            '{}_{}'.format(
                run_last_period_start_date.strftime('%Y%m%d'),
                run_last_period_end_date.strftime('%Y%m%d')),
            'states.{}_{:05d}.nc'.format(
                (run_last_period_end_date +
                 datetime.timedelta(
                     days=1)).strftime('%Y%m%d'),
                0))
        ds_states = xr.open_dataset(state_fname)
        # Compare split run states with full run
        for var in ds_states.data_vars: