blacken sources

src/gbasm/arguments/ArgumentParser.py

@@ -0,0 +1,6 @@
+class ArgumentParser(object):
+    def can_parse(token: str) -> bool:
+        raise NotImplementedError()
+
+    def get_name() -> str:
+        raise NotImplementedError()

src/gbasm/arguments/ArgumentParsers.py

@@ -0,0 +1,141 @@
+from .ArgumentParser import ArgumentParser
+from . import Arguments
+
+
+class Address(ArgumentParser):
+
+    NAME = "Address"
+
+    def __init__(self):
+        pass
+
+    def can_parse(self, token: str) -> bool:
+        try:
+            addr = int(token, base=0)
+            return addr < 0x8000 and addr >= 0
+        except ValueError:
+            return False
+
+    def get_name(self) -> str:
+        return Address.NAME
+
+
+class Label(ArgumentParser):
+
+    NAME = "Label"
+
+    def __init__(self):
+        pass
+
+    def can_parse(self, token: str) -> bool:
+        if not token[0] in string.ascii_letters + ["_"]:
+            return False
+        return True
+
+    def parse(self, token) -> Arguments.Register8:
+        return Arguments.Label(token)
+
+    def get_name(self) -> str:
+        return Label.NAME
+
+
+class Register8(ArgumentParser):
+
+    NAME = "Register8"
+
+    def __init__(self, indirect: bool = False, indirect_increment: bool = False):
+        self.indirect = indirect
+
+    def can_parse(self, token: str) -> bool:
+        if token in Arguments.Register8.REGISTERS:
+            return True
+        if self.indirect and token in Arguments.Register8.REGISTERS_INDIRECT:
+            return True
+        if (
+            self.indirect_increment
+            and token in Arguments.Register8.REGISTERS_INDIRECT_INCREMENT
+        ):
+            return True
+        return False
+
+    def parse(self, token) -> Arguments.Register8:
+        return Arguments.Register8(token)
+
+    def get_name(self) -> str:
+        return Register8.NAME
+
+
+class Register16(ArgumentParser):
+
+    NAME = "Immediate8"
+
+    def __init__(self, indirect: bool = False, indirect_increment: bool = False):
+        self.indirect = indirect
+
+    def can_parse(self, token: str) -> bool:
+        return token in Arguments.Register16.REGISTERS
+
+    def to_argument(self, token: str):
+        return None
+
+    def parse(self, token) -> Arguments.Register16:
+        return Arguments.Register16(token)
+
+    def get_name(self) -> str:
+        return Register16.NAME
+
+
+class Immediate8(ArgumentParser):
+
+    NAME = "Immediate8"
+
+    def __init__(self):
+        pass
+
+    def can_parse(self, token: str) -> bool:
+        try:
+            addr = int(token, base=0)
+            return addr <= 0xFF and addr >= 0
+        except ValueError:
+            return False
+
+    def get_name(self) -> str:
+        return Immediate8.NAME
+
+
+class Immediate16(ArgumentParser):
+
+    NAME = "Immediate16"
+
+    def __init__(self):
+        pass
+
+    def can_parse(self, token: str) -> bool:
+        try:
+            addr = int(token, base=0)
+            return addr <= 0xFFFF and addr >= 0
+        except ValueError:
+            return False
+
+    def parse(self, token) -> Arguments.Immediate16:
+        return Arguments.Immediate16(token)
+
+    def get_name(self) -> str:
+        return Immediate16.NAME
+
+
+class Flag(ArgumentParser):
+
+    NAME = "Flag"
+
+    def __init__(self):
+        pass
+
+    def can_parse(self, token: str) -> bool:
+        return token in Arguments.Flag.FLAGS
+
+    def parse(self, token) -> Arguments.Immediate16:
+        return Arguments.Flag(token)
+
+    def get_name(self) -> str:
+        return Flag.NAME

src/gbasm/arguments/Arguments.py

@@ -0,0 +1,72 @@
+class Argument(object):
+    def __init__(self):
+        pass
+
+
+class Address(Argument):
+
+    NAME = "Address"
+
+    def __init__(self, value: int):
+        self.value = value
+
+
+class Label(Argument):
+
+    NAME = "Label"
+
+    def __init__(self, value: str):
+        self.value = value
+
+
+class Register8(Argument):
+
+    NAME = "Register8"
+    REGISTERS = ["A", "B", "C", "D", "E", "H", "L"]
+    REGISTERS_INDIRECT = ["(HL)"]
+    REGISTERS_INDIRECT_INCREMENT = ["(HL)", "(HL+)", "(HL-)"]
+
+    def __init__(self, value: str):
+        if value not in Register8.REGISTERS + Register8.REGISTERS_INDIRECT_INCREMENT:
+            raise ValueError("Unknown Register8: {}".format(value))
+        self.value = value
+
+
+class Register16(Argument):
+
+    NAME = "Immediate8"
+
+    REGISTERS = ["BC", "DE", "DE", "HL", "SP"]
+
+    def __init__(self, value: int):
+        if value not in Register16.REGISTERS:
+            raise ValueError("Unknown Register16: {}".format(value))
+        self.value = value
+
+
+class Immediate8(Argument):
+
+    NAME = "Immediate8"
+
+    def __init__(self, value: int):
+        self.value = value
+
+
+class Immediate16(Argument):
+
+    NAME = "Immediate16"
+
+    def __init__(self, value: int):
+        self.value = value
+
+
+class Flag(Argument):
+
+    NAME = "Immediate8"
+
+    FLAGS = ["Z", "C", "NZ", "NC"]
+
+    def __init__(self, value: str):
+        if value not in Flag.FLAGS:
+            raise ValueError("Unknown Flag: {}".format(value))
+        self.value = value

src/gbasm/arguments/__init__.py

@@ -0,0 +1,10 @@
+from .ArgumentParser import ArgumentParser
+from .ArgumentParsers import (
+    Label,
+    Address,
+    Immediate8,
+    Immediate16,
+    Register8,
+    Register16,
+)
+from .Arguments import Argument

src/gbasm/gbasm.py

@@ -0,0 +1,178 @@
+#!/usr/bin/python3
+
+import argparse
+import logging
+import sys
+
+from .instructions import Instruction
+from .instructions.inc import Inc
+from .instructions.dec import Dec
+from .instructions.nop import Nop
+from .instructions.stop import Stop
+from .instructions.jr import Jr
+from .instructions.ld import Ld
+from .arguments import ArgumentParser, Argument
+
+from typing import Callable, Dict, List, Optional
+
+logger = logging.getLogger(__name__)
+
+COMMENT_CHAR = "#"
+LABEL_SUFFIX = ":"
+
+GB_INSTRUCTIONS = [
+    Nop(),
+    Stop(),
+    Inc(),
+    Dec(),
+    Jr(),
+    Ld(),
+]
+
+
+def build_instruction_map() -> Dict[str, Instruction]:
+    d = {}  # type: Dict[str, Instruction]
+    for i in GB_INSTRUCTIONS:
+        d[i.token] = i
+    return d
+
+
+def try_parse_arguments(
+    args: List[str], arg_types: List[ArgumentParser]
+) -> Optional[List[Argument]]:
+    if len(args) != len(arg_types):
+        return None
+
+    out_args = []
+
+    for (arg, arg_type) in zip(args, arg_types):
+        try:
+            out_args.append(arg_type.parse(arg))
+        except ValueError:
+            return None
+
+    return out_args
+
+
+def parse_line_size(instruction: Instruction, arguments: List[str]) -> bytes:
+
+    for argtype_list in instruction.argument_specs:
+        args = try_parse_arguments(arguments, argtype_list)
+        if args is not None:
+            return instruction.num_bytes(args)
+
+    raise ValueError("Failed to parse line.")
+
+
+def parse_line_bytes(
+    instruction: Instruction,
+    arguments: List[str],
+    instruction_addr: int,
+    label_resolver: Callable[[str], int],
+) -> bytes:
+
+    for argtype_list in instruction.argument_specs:
+        args = try_parse_arguments(arguments, argtype_list)
+        if args is not None:
+            return instruction.to_bytes(args, instruction_addr, label_resolver)
+
+    raise ValueError("Failed to parse line.")
+
+
+def assemble_file(infile) -> bytes:
+    program = infile.readlines()
+    return assemble(program)
+
+
+def assemble(lines: str) -> bytes:
+
+    instruction_map = build_instruction_map()
+    logger.debug("Instruction map: %s", instruction_map)
+
+    labels = {}  # type: Dict[str, int]
+    program = bytes()
+
+    def label_resolver(label: str) -> int:
+        nonlocal labels
+        return labels[label]
+
+    for step in ["SIZE", "CONTENT"]:
+        logger.debug("Starting step: %s", step)
+        byte_offset = 0
+
+        for line_num, line in enumerate(lines):
+            # Remove comments
+            line = line.split(COMMENT_CHAR)[0]
+
+            # Tokenize
+            tokens = line.split()
+
+            if len(tokens) == 0:
+                continue
+
+            instruction_name = tokens[0]
+            args = tokens[1:]
+            try:
+                instruction = instruction_map[instruction_name]
+            except KeyError:
+                if instruction_name[-1] == LABEL_SUFFIX:
+                    if step == "SIZE":
+                        label = instruction_name[:-1]
+                        logger.debug("Found label '%s' at %s", label, byte_offset)
+                        if label in labels.keys():
+                            raise KeyError(
+                                "Label '%s' defined at %s and %s",
+                                label,
+                                labels[label],
+                                line_num,
+                            )
+                        labels[label] = byte_offset
+                    continue
+                raise KeyError(
+                    'Unknown instruction "%s" on line %s', instruction_name, line_num
+                )
+
+            if step == "CONTENT":
+                try:
+                    program += parse_line_bytes(
+                        instruction, args, byte_offset, label_resolver
+                    )
+                except ValueError:
+                    raise ValueError("Failed to parse line %s,\n%s", line_num, line)
+
+            byte_offset += parse_line_size(instruction, args)
+        if step == "SIZE":
+            logger.info("Program size: %s bytes", byte_offset)
+            logger.debug("Found labels: %s", labels)
+
+    return program
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="An assembler for Gameboy assembly")
+
+    parser.add_argument(
+        "--infile", "-i", type=argparse.FileType("r"), default=sys.stdin
+    )
+    parser.add_argument(
+        "--outfile", "-o", type=argparse.FileType("wb", 0), default=sys.stdout
+    )
+    parser.add_argument("--verbose", "-v", action="store_true")
+
+    args = parser.parse_args()
+
+    logging.basicConfig(format="%(levelname)s: %(message)s")
+    logger.setLevel(logging.INFO)
+
+    if args.verbose:
+        logging.basicConfig(
+            format="%(levelname)s: %(filename)s:%(lineno)d: %(message)s"
+        )
+        logger.setLevel(logging.DEBUG)
+
+    program = assemble_file(args.infile)
+    args.outfile.write(program)
+
+
+if __name__ == "__main__":
+    main()

src/gbasm/instructions/Instruction.py

@@ -0,0 +1,19 @@
+from typing import Callable, List
+from ..arguments import Argument, ArgumentParser
+
+
+class Instruction(object):
+    def __init__(self, token: str, argument_specs: List[List[ArgumentParser]]):
+        self.token = token
+        self.argument_specs = argument_specs
+
+    def num_bytes(self, arguments) -> int:
+        raise NotImplementedError()
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_addr: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+        raise NotImplementedError()

src/gbasm/instructions/__init__.py

@@ -0,0 +1 @@
+from .Instruction import Instruction

src/gbasm/instructions/dec.py

@@ -0,0 +1,52 @@
+from .Instruction import Instruction
+from ..arguments.ArgumentParsers import Register8, Register16
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Dec(Instruction):
+    def __init__(self):
+        argtypes = [[Register8()], [Register16()]]
+        super().__init__("DEC", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        return 1
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_addr: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        if len(arguments) != 1:
+            raise ValueError("Incorrect number of arguments")
+
+        value = arguments[0].value
+
+        if value == "BC":
+            return bytes([0x0B])
+        if value == "DE":
+            return bytes([0x1B])
+        if value == "HL":
+            return bytes([0x2B])
+        if value == "SP":
+            return bytes([0x3B])
+        if value == "A":
+            return bytes([0x3D])
+        if value == "B":
+            return bytes([0x05])
+        if value == "C":
+            return bytes([0x0D])
+        if value == "D":
+            return bytes([0x15])
+        if value == "E":
+            return bytes([0x1D])
+        if value == "H":
+            return bytes([0x25])
+        if value == "L":
+            return bytes([0x2D])
+        if value == "(HL)":
+            return bytes([0x35])
+
+        raise ValueError("Unknown value: {}".format(value))

src/gbasm/instructions/inc.py

@@ -0,0 +1,52 @@
+from .Instruction import Instruction
+from ..arguments.ArgumentParsers import Register8, Register16
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Inc(Instruction):
+    def __init__(self):
+        argtypes = [[Register8()], [Register16()]]
+        super().__init__("INC", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        return 1
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_addr: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        if len(arguments) != 1:
+            raise ValueError("Incorrect number of arguments")
+
+        value = arguments[0].value
+
+        if value == "BC":
+            return bytes([0x03])
+        if value == "DE":
+            return bytes([0x13])
+        if value == "HL":
+            return bytes([0x23])
+        if value == "SP":
+            return bytes([0x33])
+        if value == "A":
+            return bytes([0x3C])
+        if value == "B":
+            return bytes([0x04])
+        if value == "C":
+            return bytes([0x0C])
+        if value == "D":
+            return bytes([0x14])
+        if value == "E":
+            return bytes([0x1C])
+        if value == "H":
+            return bytes([0x24])
+        if value == "L":
+            return bytes([0x2C])
+        if value == "(HL)":
+            return bytes([0x34])
+
+        raise ValueError("Unknown value: {}".format(value))

src/gbasm/instructions/jr.py

@@ -0,0 +1,60 @@
+from .Instruction import Instruction
+from ..arguments.ArgumentParsers import Flag, Label
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Jr(Instruction):
+
+    INSTRUCTION_SIZE = 2
+
+    def __init__(self):
+        argtypes = [[Label()], [Flag(), Label()]]
+        super().__init__("JR", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        return Jr.INSTRUCTION_SIZE
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_address: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        out_bytes = bytearray()
+
+        if len(arguments) == 1:
+            out_bytes.append(0x18)
+            dest_arg = arguments[0].value
+
+        elif len(arguments) == 2:
+            flag_dict = {"NZ": 0x20, "NC": 0x30, "Z": 0x28, "C": 0x38}
+            flag = arguments[0].value
+            try:
+                out_bytes.append(flag_dict[flag])
+            except KeyError:
+                logger.exception("Instruction JR does not accept flag %s", arg)
+            dest_arg = arguments[1].value
+
+        else:
+            raise ValueError("Incorrect number of arguments")
+
+        if isinstance(dest_arg, str):
+            label = dest_arg
+            dest_addr = label_resolver(label)
+        else:
+            dest_addr = dest_arg
+
+        addr_offset = dest_addr - (instruction_address + Jr.INSTRUCTION_SIZE)
+        if addr_offset > 127 or addr_offset < -128:
+            raise ValueError(
+                "JR instruction cannot jump to {} ({}), from {}, {} bytes away".format(
+                    label, dest_addr, instruction_address, addr_offset
+                )
+            )
+        if addr_offset < 0:
+            addr_offset += 256
+
+        out_bytes.append(addr_offset)
+        return bytes(out_bytes)

src/gbasm/instructions/ld.py

@@ -0,0 +1,40 @@
+from .Instruction import Instruction
+from ..arguments.ArgumentParsers import Register16, Immediate16
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Ld(Instruction):
+    def __init__(self):
+        argtypes = [[Register16(), Immediate16()]]
+        super().__init__("LD", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        # the instruction_address and label_resolver are unused
+        return len(self.to_bytes(arguments, 0, lambda x: None))
+
+    def encode_reg_16(self, register: Register16, immediate: Immediate16) -> bytes:
+        reg_dict = {"BC": 0x01, "DE": 0x11, "HL": 0x21, "SP": 0x31}
+        imm = int(immediate.value, 0)
+        print("immediate", imm)
+        out = bytearray()
+        out.append(reg_dict[register.value])
+        out.extend(imm.to_bytes(2, "little", signed=imm < 0))
+        return bytes(out)
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_address: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        # print(arguments)
+        # print("Register16", type(arguments[0]))
+        # print("Immediate16", isinstance(arguments[1], Immediate16))
+
+        # if isinstance(arguments[0], Register16) and \
+        #    isinstance(arguments[1], Immediate16):
+        return self.encode_reg_16(arguments[0], arguments[1])
+
+        raise TypeError("Unhandled argument types")

src/gbasm/instructions/nop.py

@@ -0,0 +1,24 @@
+from .Instruction import Instruction
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Nop(Instruction):
+    def __init__(self):
+        argtypes = [[]]
+        super().__init__("NOP", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        return 1
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_addr: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        if len(arguments) != 0:
+            raise ValueError("Incorrect number of arguments")
+
+        return bytes([0x00])

src/gbasm/instructions/stop.py

@@ -0,0 +1,24 @@
+from .Instruction import Instruction
+from ..arguments import Argument
+from typing import Callable, List
+
+
+class Stop(Instruction):
+    def __init__(self):
+        argtypes = [[]]
+        super().__init__("STOP", argtypes)
+
+    def num_bytes(self, arguments) -> int:
+        return 1
+
+    def to_bytes(
+        self,
+        arguments: List[Argument],
+        instruction_addr: int,
+        label_resolver: Callable[[str], int],
+    ) -> bytes:
+
+        if len(arguments) != 0:
+            raise ValueError("Incorrect number of arguments")
+
+        return bytes([0x10])