improve test

This test was written in such a way that having a wrong data type
for the mp3 samples wasn't detected. Instead of using
np.frombuffer(dtype=int16), just do arithmetic directly on the
samples. During testing time we don't care if it might be a little
slower or use a little more RAM than ulab, and we don't care
whether it's actually an RMS calculation.  Just that it's
consistent and shows the audio data is correct, including its
defined data type.

Author: jepler

tests/circuitpython/issue9705.py

@@ -1,25 +1,19 @@
 import audiomp3, audiocore
-import ulab.numpy as np
 
 TEST_FILE = (
     __file__.rsplit("/", 1)[0]
     + "/../circuitpython-manual/audiocore/jeplayer-splash-44100-stereo.mp3"
 )
 
 
-def normalized_rms_ulab(values):
-    values = np.frombuffer(values, dtype=np.int16)
-    # this function works with ndarrays only
-    minbuf = np.mean(values)
-    values = values - minbuf
-    samples_sum = np.sum(values * values)
-    return (samples_sum / len(values)) ** 0.5
+def loudness(values):
+    return sum(abs(a) for a in values)
 
 
 def print_frame_loudness(decoder, n):
     for i in range(n):
         result, buf = audiocore.get_buffer(decoder)
-        print(f"{i} {result} {normalized_rms_ulab(buf):5.0f}")
+        print(f"{i} {result} {loudness(buf):5.0f}")
     print()