fixed bug with parsing labels starting with BYTE

Author: michaelkamprath

CHANGELOG.md

@@ -15,10 +15,12 @@ Changes that are planned but not implemented yet:
   * unknown labels
 * Add named label scopes. This would allow a label to be defined in a specific scope that can be shared across files.
 * Create a "align if needed" preprocessor directive paid that generates an `.align` directive if the bytecode in between the pair isn't naturally on the same page and can fit on the same page if aligned. An error would be benerated if the block of code can't fit on the same page regardless of alignment.
+* Update the `#ifdef` and related preprocessor directives to include detection of labels and constants.
 
 ## [Unreleased]
 * Upgrade python version requirements to 3.11
 * Fixed a bug where embedded stringsd weren't properly parsed if they contained a newline character or there were multiple embedded strings per line
+* Fixed a bug where parsing properly discriminate between labels starting with `BYTE` string and the `BYTEx()` operator.
 
 ## [0.4.2]
 *  Added support for The Minimal 64x4 Home Computer with an example and updated assembler functionality to support it.

src/bespokeasm/__init__.py

@@ -1,4 +1,4 @@
-BESPOKEASM_VERSION_STR = '0.4.3b1'
+BESPOKEASM_VERSION_STR = '0.4.3b2'
 
 # if a cconfig file requires a certain bespoke ASM version, it should be at least this version.
 BESPOKEASM_MIN_REQUIRED_STR = '0.3.0'

src/bespokeasm/assembler/bytecode/generator/macro.py

@@ -39,7 +39,7 @@ def generate_bytecode_parts(
             if assembled_instruction is not None:
                 return assembled_instruction
 
-        sys.exit(f'ERROR: {line_id} - Instruction "{mnemonic}" has no valid operands configured.')
+        sys.exit(f'ERROR: {line_id} - Macro "{mnemonic}" has no valid operands configured.')
 
     @classmethod
     def generate_variant_bytecode_parts(

src/bespokeasm/expression/__init__.py

@@ -185,7 +185,7 @@ def _lexical_analysis(line_id: LineIdentifier, s: str) -> list[ExpressionNode]:
         if part in TOKEN_MAPPINGS:
             token_type = TOKEN_MAPPINGS[part]
             token = ExpressionNode(token_type, value=part)
-        elif part.startswith('BYTE'):
+        elif re.match(r'^BYTE\d\(', part):
             token = ExpressionNode(TokenType.T_BYTE, value=part)
         elif is_string_numeric(part):
             token = ExpressionNode(TokenType.T_NUM, value=parse_numeric_string(part))

test/config_files/test_instruction_operands.yaml

@@ -31,6 +31,15 @@ operand_sets:
         bytecode:
           value: 3
           size: 4
+  immediate_8bit:
+    # one byte is interpreted as an immediate value
+    operand_values:
+      int8:
+        type: numeric
+        argument:
+          size: 8
+          byte_align: true
+          endian: little
   enums:
     operand_values:
       register_enum:
@@ -141,3 +150,13 @@ instructions:
                 byte_align: true
                 max: 128
                 min: -127
+  adi:
+    # Add immediate to A: A += imm
+    bytecode:
+      value: 0xa2
+      size: 8
+    operands:
+      count: 1
+      operand_sets:
+        list:
+          - immediate_8bit

test/test_expression.py

@@ -11,10 +11,12 @@ class TestExpression(unittest.TestCase):
     @classmethod
     def setUpClass(cls):
         cls.label_values = GlobalLabelScope(set())
-        cls.label_values.set_label_value('value_1', 12, 1)
-        cls.label_values.set_label_value('the_8_ball', 8, 2)
-        cls.label_values.set_label_value('MixedCase', 2, 3)
-        cls.label_values.set_label_value('MAX_N', 20, 1)
+        line = LineIdentifier(0, 'setUpClass')
+        cls.label_values.set_label_value('value_1', 12, line)
+        cls.label_values.set_label_value('the_8_ball', 8, line)
+        cls.label_values.set_label_value('MixedCase', 2, line)
+        cls.label_values.set_label_value('MAX_N', 20, line)
+        cls.label_values.set_label_value('BYTECODE_LL0', 0x07, line)
 
     def test_expression_parsing(self):
         line_id = LineIdentifier(1212, 'test_expression_parsing')
@@ -59,6 +61,10 @@ def test_expression_parsing(self):
             parse_expression(line_id, '(MAX_N + 3) % %101').get_value(TestExpression.label_values, line_id),
             3, 'test handling of modulo: (MAX_N + 3) % %101 = 3'
         )
+        self.assertEqual(
+            parse_expression(line_id, 'BYTECODE_LL0').get_value(TestExpression.label_values, line_id),
+            0x07, 'test handling of label with prefix: BYTECODE_LL0'
+        )
 
         with self.assertRaises(SyntaxError, msg='only integer numeric values are supported'):
             parse_expression(1212, '(value_1/5)/2.4').get_value(TestExpression.label_values, 100)

test/test_instruction_parsing.py

@@ -695,6 +695,30 @@ def test_operand_order(self):
         t3.generate_bytes()
         self.assertEqual(list(t3.get_bytes()), [0x81, 0x32, 0x10], 'instruction byte should match')
 
+    def test_operand_expression(self):
+        fp = pkg_resources.files(config_files).joinpath('test_instruction_operands.yaml')
+        isa_model = AssemblerModel(str(fp), 0)
+        memzone_mngr = MemoryZoneManager(
+            isa_model.address_size,
+            isa_model.default_origin,
+            isa_model.predefined_memory_zones,
+        )
+
+        lineid = LineIdentifier(42, 'test_operand_order')
+
+        t1 = InstructionLine.factory(
+            lineid,
+            'ADI var1w+0',
+            'comment',
+            isa_model,
+            memzone_mngr.global_zone,
+            memzone_mngr,
+        )
+
+        t1.label_scope = TestInstructionParsing.label_values
+        self.assertIsInstance(t1, InstructionLine)
+        self.assertEqual(t1.byte_size, 2, 'has 2 bytes')
+
     def test_relative_address_operand(self):
         fp = pkg_resources.files(config_files).joinpath('test_instruction_operands.yaml')
         isa_model = AssemblerModel(str(fp), 0)