improve printout messages

    with open(cut_file, "r") as f:

        cut = np.fromfile(f, dtype=dtype)

    # remove the duplicates of each value

    # remove the duplicates of each value per frame

    if nsamples_per_frame > 1:

        ref = ref[::nsamples_per_frame]

        cut = cut[::nsamples_per_frame]

    # check the lengths

    N = ref.shape[0] + cut.shape[0]

    reason = f"File {os.path.basename(ref_file)}"

    # check file lengths

    result_len_check = 0

    if len_check and ref.shape != cut.shape:

        print( f"files have different length, ref: {ref.shape[0]}, cut: {cut.shape[0]}" )

        result = 1

        reason = f"files have different length!"

        return result, reason

        print(f"files have different length: Ref {ref.shape[0]}, DuT {cut.shape[0]}, ", end="")

        reason += f" has different length between Ref ({ref.shape[0]}) and DuT ({cut.shape[0]}), "

        # check if threshold has been exceeded

        if abs(ref.shape[0] - cut.shape[0]) / max(ref.shape[0], cut.shape[0]) > min_diff_thr:

            result_len_check = 1

        else:

            result_len_check = 0

    else:

        reason += ": "

    # calculate histograms from data

    unique_values = np.sort(np.unique(np.concatenate([ref, cut])))

    diff = {unique_values[i]: diff_hist[i] for i in range(len(unique_values))}      

    diff = dict(sorted(diff.items(), key=lambda k: abs(k[1]), reverse=True))

    # calcualte the total number of differences

    # calculate the total number of differences

    total_num_diff = sum(np.abs(list(diff.values())))

    total_num_diff_ratio = total_num_diff / N

    total_num_diff_ratio = total_num_diff / max(ref.shape[0], cut.shape[0])

    if total_num_diff_ratio > min_diff_thr:

        print( f"files are different, total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%), exceeding the threshold of {(min_diff_thr*100):.2f}%!" )

        result = 1

        reason = f"files are different, total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%), exceeding the threshold of {(min_diff_thr*100):.2f}%!"

    elif total_num_diff_ratio > 0 and total_num_diff_ratio < min_diff_thr:

        print( f"files are different, total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%)." )

        result = 0

        reason = f"files are different, total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%)."

        msg = f"the total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%), exceeding the threshold of {(min_diff_thr*100):.2f}%!"

        reason += msg

        print(msg, end="")

        result_diff_check = 1

    else:

        print( f"files are bit-exact." )

        msg = f"the total number of differences is {total_num_diff} ({(total_num_diff_ratio*100):.2f}%)."

        reason += msg

        print( msg, end="" )

        result_diff_check = 0

    print("")

    result = 0

        reason = f"files are bit-exact."

    if result_len_check or result_diff_check:

        result = 1

    return result, reason

            print ("=================================\n")

            msg = ""

            max_diff = 0

            max_enc_diff = 0

            enc_test_result = 0

            for f in ENC_AUX_FILES:

                filename = f[0]

                dtype = f[1]

                        min_diff_thr=0.1,

                    )

                    # check if the maximum difference has been exceeded

                    if output_differs:

                        enc_test_result = 1

                        if msg:

                            msg += ", "

                        msg += reason

                        search_result = re.search(MAX_ENC_DIFF_PATTERN, reason)

                    # check if the maximum difference has been exceeded

                    search_result = re.search(MAX_ENC_DIFF_PATTERN, reason)

                    if search_result:

                        diff = search_result.groups(1)[0]

                            diff = np.float(diff)

                            if diff > max_diff:

                                max_diff = diff

                        else:

                            msg = "Error " + MAX_ENC_DIFF_PATTERN + " not found"

                            print(msg)

                            pytest.fail(msg)

                        diff = float(diff)

                        if diff > max_enc_diff:

                            max_enc_diff = diff

            print ("")

            if max_diff > 0:

                record_property("MAXIMUM ENC DIFF", max_diff)

            if msg:

                pytest.fail(reason)

            # remove encoder output files to save disk space

            if ref_enc_dbg_dir:

                shutil.rmtree(ref_enc_dbg_dir.name)

            if dut_enc_dbg_dir:

                shutil.rmtree(dut_enc_dbg_dir.name)

            if enc_test_result:

                record_property("MAXIMUM ENC DIFF", max_enc_diff)

                pytest.fail(msg)

    # check for networkSimulator_g192 command line

    if sim_opts != "":

        sim_split = sim_opts.split()

# regex patterns for parsing the output from cmp_pcm -> mainly for BASOP ci

MLD_PATTERN = r"MLD: ([\d\.]*)"

MAX_DIFF_PATTERN = r"MAXIMUM ABS DIFF: (\d*)"

MAX_ENC_DIFF_PATTERN = r"The total number of differences is \d (\d+\.\d+)%"

MAX_ENC_DIFF_PATTERN = r"total number of differences is \d+ \((\d+\.\d+)%\)"

# list of encoder filename patterns with their data type and number of samples per frame

# note: instead of specifying the number of samples per frame, you can use a formula incl. 'fs', e.g. 'fs/50'