Add script for loading digital data from older example worklfow

Older workflow used CsvWriter instead of MatrixWriter

Author: cjsha

articles/hardware/breakout/load-data.md

@@ -5,6 +5,14 @@ title: Load Data
 
 The following python script can be used to load and plot the data produced by the Breakout Board [example workflow](xref:breakout_workflow).
 
+> [!NOTE]
+> Previous versions of the example workflow used CsvWriter instead of MatrixWriter to write digital
+> input data. You can download the script for loading digital data saved using CsvWriter 
+> <a href="/workflows/hardware/breakout/load-breakoutboard_csvwriter-digital-inputs.py"
+> download>here</a>. With that said, using the updated example workflow is recommended because the
+> resulting data occupies less storage.
+
+
 [!code-python[](../../../workflows/hardware/breakout/load-breakoutboard.py)]
 
 > [!NOTE]

workflows/hardware/breakout/load-breakoutboard_csvwriter-digital-inputs.py

@@ -0,0 +1,74 @@
+import numpy as np
+import matplotlib.pyplot as plt
+import csv
+
+# Load data from breakout board tutorial workflow
+suffix = '0'; # Change to match file names' suffix
+plt.close('all')
+
+#%% Metadata
+dt = {'names': ('time', 'acq_clk_hz', 'block_read_sz', 'block_write_sz'),
+      'formats': ('datetime64[us]', 'u4', 'u4', 'u4')}
+meta = np.genfromtxt('start-time_' + suffix + '.csv', delimiter=',', dtype=dt)
+print(f"Recording was started at {meta['time']} GMT")
+
+#%% Analog Inputs
+analog_input = {}
+analog_input['time'] = np.fromfile('analog-clock_' + suffix + '.raw', dtype=np.uint64) / meta['acq_clk_hz']
+analog_input['data'] = np.reshape(np.fromfile('analog-data_' + suffix + '.raw', dtype=np.float32), (-1, 12))
+
+plt.figure()
+plt.plot(analog_input['time'], analog_input['data'])
+plt.xlabel("time (sec)")
+plt.ylabel("volts")
+plt.legend(['Ch. 0', 'Ch. 1', 'Ch. 2', 'Ch. 3', 'Ch. 4', 'Ch. 5', 
+            'Ch. 6', 'Ch. 7', 'Ch. 8', 'Ch. 9', 'Ch. 10', 'Ch. 11'])
+
+#%% Digital Inputs
+digital_input_flags = ['Pin0', 'Pin1', 'Pin2', 'Pin3', 'Pin4', 'Pin5', 'Pin6', 'Pin7',
+                       'Moon', 'Triangle', 'X', 'Check', 'Circle', 'Square',
+                       'Reserved0', 'Reserved1', 'PortDOn', 'PortCOn', 'PortBOn', 'PortAOn']
+digital_input_data = []
+with open("digital-data_" + suffix + '.csv', 'r') as digital_input_file:
+      digital_input_reader = csv.reader(digital_input_file)
+      row = [entry.strip() for entry in next(digital_input_reader)]
+      time = int(row[0]) / meta['acq_clk_hz']
+      for digital_input_flag in digital_input_flags:
+            if digital_input_flag in [entry.strip() for entry in row[1:]]:
+                  digital_input_data.append((time, digital_input_flag, True))
+            else:
+                  digital_input_data.append((time, digital_input_flag, False))
+      row_previous = row
+      for row in digital_input_reader:
+            time = int(row[0]) / meta['acq_clk_hz']
+            row = [entry.strip() for entry in row]
+            for entry in (set(row) | set(row_previous)):
+                  if (entry in row_previous[1:]) and (not entry in row[1:]):
+                        digital_input_data.append((time, entry, False))
+                  elif (not entry in row_previous[1:]) and (entry in row[1:]): 
+                        digital_input_data.append((time, entry, True))
+            row_previous = row
+
+dt = np.dtype([('Clock', np.float64), ('Digital Input', '<U10'), ('State', np.bool_)])
+digital_input_data = np.array(digital_input_data, dtype=dt)
+
+plt.figure()
+line_spacing = 1.25
+for i, digital_input_flag in enumerate(digital_input_flags):
+      index_mask = digital_input_data['Digital Input'] == digital_input_flag
+      line = plt.step(digital_input_data['Clock'][index_mask], digital_input_data['State'][index_mask] + line_spacing * i, color="black")
+plt.ylim(-0.5, len(digital_input_flags) * line_spacing + 0.5)
+plt.yticks(np.arange(0.5, len(digital_input_flags) * line_spacing, line_spacing), digital_input_flags)
+plt.tight_layout()
+
+plt.show()
+
+#%% Hardware FIFO buffer use
+dt = {'names': ('clock', 'bytes', 'percent'),
+      'formats': ('u8', 'u4', 'f8')}
+memory_use = np.genfromtxt('memory-use_' + suffix + '.csv', delimiter=',', dtype=dt)
+
+plt.figure()
+plt.plot(memory_use['clock'] / meta['acq_clk_hz'], memory_use['percent'])
+plt.xlabel("time (sec)")
+plt.ylabel("FIFO used (%)")