Resource sending

Author: antisvin

Makefile

@@ -125,6 +125,10 @@ store: patch ## upload and save patch to attached OWL
 	@echo Sending patch $(PATCHNAME) to $(OWLDEVICE) to store in slot $(SLOT)
 	@$(FIRMWARESENDER) -q -in $(BUILD)/$(TARGET).bin -out "$(OWLDEVICE)" -store $(SLOT)
 
+resource: $(RESOURCE) ## upload and save resource to attached OWL
+	@echo Sending resource $(RESOURCE) to $(OWLDEVICE) to store in slot $(SLOT)
+	@$(FIRMWARESENDER) -q -in $(RESOURCE) -out "$(OWLDEVICE)" -store $(SLOT)
+
 docs: ## generate HTML documentation
 	@doxygen Doxyfile
 

Tools/makeresource.py

@@ -0,0 +1,110 @@
+#! /usr/bin/env python3
+
+"""
+    This script adds resource header and converts byte ordering for binary data.
+
+    Input can be obtained from binary file, by parsing WAV files or C sources containing LUTs.
+
+    C format parsing tested with Magus LED RGB tables, power and log LUTs.
+"""
+
+import argparse
+import struct
+import sys
+import wave
+
+parser = argparse.ArgumentParser(
+    description='Convert binary data to OpenWare resource')
+parser.add_argument(
+    '-i', '--input', required=True, help='Input file')
+parser.add_argument(
+    '-o', '--output', required=True, help='Output file')
+parser.add_argument(
+    '-p', '--parse', help='Parse LUT from C header file', action='store_true')
+parser.add_argument(
+    '-w', '--wav', help='Convert WAV file to wavetable', action='store_true')
+parser.add_argument(
+    '-f', '--format', help='WAV file sample wavetable (i.e. <H for 16bit unsigned little-endian)', default='<H')
+parser.add_argument(
+    '-d', '--destination_format',
+    help='Normally this will be retrieved from parsed header. If that fails, use the same format as python\'s "struct" modlue (i.e. <I, <f)')
+parser.add_argument('name', help='Resource name')
+
+args = parser.parse_args()
+
+in_file = open(args.input, 'rb')
+out_file = open(args.output, 'wb')
+
+data = in_file.read()
+
+formats = {
+    'float': '<f',
+    'uint32_t': '<I',
+    'int32_t': '<i',
+    'uint16_t': '<H',
+    'int16_t': '<h',
+    'uint8_t': '<B',
+    'int8_t': '<b',
+    'char': '<c',
+    'size_t': '<N',
+}
+
+if args.parse:
+    values = data.split(b'{', 1)[1].split(b'}', 1)[0].split(b',')
+
+    destination_format = args.destination_format
+    if not args.destination_format:
+        last_line = data.split(b'{', 1)[0].splitlines()[-1]
+        for t in formats:            
+            if f' {t} '.encode('ascii') in last_line:
+                print(f'Found format: {t} -> {formats[t]}')
+                destination_format = formats[t]
+                break
+
+    if destination_format is None:
+        raise ValueError(
+            'Please specify destination format value with -d parameter. '
+            'See https://docs.python.org/3/library/struct.html#format-characters')
+
+
+    if 'f' in destination_format or 'F' in destination_format:
+        values = [float(value.strip()) for value in values if value.strip()]
+        print('Parsing data as floats')
+    else:
+        # Using eval() instead of int() allows us to parse prefixes like 0x or 0b
+        values = [eval(value.strip()) for value in values if value.strip()]
+        print('Parsing data as integers')
+
+    data = b''.join(struct.pack(destination_format, value) for value in values)
+
+elif args.wav:
+    w = wave.open(args.input)
+    n_frames = w.getnframes()
+    data = w.readframes(w.getnframes())
+    input_format = args.format[:-1] + str(n_frames) + args.format[-1]    
+    print(f'Parsing file as {input_format}')
+    values = struct.unpack(input_format, data)
+    destination_format = args.destination_format or args.format
+    data = b''.join(struct.pack(destination_format, value) for value in values)
+
+# Magick
+out_file.write(0xdadadeed.to_bytes(4, 'little'))
+
+# Data size
+size = len(data)
+if size % 4:
+    # Align to 4 bytes
+    size += 4 - size % 4
+print(f'Data size is {size} bytes')
+out_file.write(size.to_bytes(4, 'little'))    
+
+# Resource name
+name = list(struct.pack('24s', args.name.encode('ascii')))
+name[-1] = 0 # Force terminate name with zero
+name = bytes(name)
+for i in range(0, 24, 4):
+    out_file.write(int.from_bytes(name[i : i + 4], sys.byteorder).to_bytes(4, 'little'))
+
+# Data
+for i in range(0, size, 4):
+    out_file.write(int.from_bytes(data[i : i + 4], sys.byteorder).to_bytes(4, 'little'))