prog8/il65/compile.py

550 lines
29 KiB
Python

"""
Programming Language for 6502/6510 microprocessors, codename 'Sick'
This is the compiler of the IL65 code, that prepares the parse tree for code generation.
Written by Irmen de Jong (irmen@razorvine.net) - license: GNU GPL 3.0
"""
import re
import os
import sys
import linecache
from typing import Optional, Tuple, Set, Dict, List, Any, no_type_check
import attr
from .plyparse import *
from .plylex import SourceRef, print_bold
from .datatypes import DataType, VarType
class CompileError(Exception):
pass
class PlyParser:
def __init__(self, imported_module: bool=False) -> None:
self.parse_errors = 0
self.imported_module = imported_module
def parse_file(self, filename: str) -> Module:
print("parsing:", filename)
module = None
try:
module = parse_file(filename, self.lexer_error)
self.check_directives(module)
self.process_imports(module)
self.check_all_symbolnames(module)
self.create_multiassigns(module)
self.check_and_merge_zeropages(module)
self.simplify_some_assignments(module)
if not self.imported_module:
# the following shall only be done on the main module after all imports have been done:
self.apply_directive_options(module)
self.determine_subroutine_usage(module)
self.semantic_check(module)
self.allocate_zeropage_vars(module)
except ParseError as x:
self.handle_parse_error(x)
if self.parse_errors:
print_bold("\nNo output; there were {:d} errors.\n".format(self.parse_errors))
raise SystemExit(1)
return module
def lexer_error(self, sourceref: SourceRef, fmtstring: str, *args: str) -> None:
self.parse_errors += 1
print_bold("ERROR: {}: {}".format(sourceref, fmtstring.format(*args)))
def _check_last_statement_is_return(self, last_stmt: AstNode) -> None:
if isinstance(last_stmt, (Subroutine, Return, Goto)):
return
if isinstance(last_stmt, Directive) and last_stmt.name == "noreturn":
return
if isinstance(last_stmt, InlineAssembly):
for line in reversed(last_stmt.assembly.splitlines()):
line = line.strip()
if line.startswith(";"):
continue
if "jmp " in line or "jmp\t" in line or "rts" in line or "rti" in line:
return
raise ParseError("last statement in a block/subroutine must be a return or goto, "
"(or %noreturn directive to silence this error)", last_stmt.sourceref)
def semantic_check(self, module: Module) -> None:
# perform semantic analysis / checks on the syntactic parse tree we have so far
# (note: symbol names have already been checked to exist when we start this)
previous_stmt = None
encountered_blocks = set() # type: Set[Block]
for node in module.all_nodes():
if isinstance(node, Block):
parentname = (node.parent.name + ".") if node.parent else ""
blockname = parentname + node.name
if blockname in encountered_blocks:
raise ValueError("block names not unique:", blockname)
encountered_blocks.add(blockname)
if isinstance(node, Scope):
if node.nodes and isinstance(node.parent, (Block, Subroutine)):
if isinstance(node.parent, Block) and node.parent.name != "ZP":
self._check_last_statement_is_return(node.nodes[-1])
elif isinstance(node, SubCall):
if isinstance(node.target, SymbolName):
subdef = node.my_scope().lookup(node.target.name)
if isinstance(subdef, Subroutine):
self.check_subroutine_arguments(node, subdef)
elif isinstance(node, Subroutine):
# the previous statement (if any) must be a Goto or Return
if not isinstance(previous_stmt, (Scope, Goto, Return, VarDef, Subroutine)):
raise ParseError("statement preceding subroutine must be a goto or return or another subroutine", node.sourceref)
elif isinstance(node, IncrDecr):
if isinstance(node.target, SymbolName):
symdef = node.my_scope().lookup(node.target.name)
if isinstance(symdef, VarDef) and symdef.vartype == VarType.CONST:
raise ParseError("cannot modify a constant", node.sourceref)
elif isinstance(node, Assignment):
scope = node.my_scope()
for target in node.left.nodes:
if isinstance(target, SymbolName):
symdef = scope.lookup(target.name)
if isinstance(symdef, VarDef) and symdef.vartype == VarType.CONST:
raise ParseError("cannot modify a constant", target.sourceref)
elif isinstance(node, AugAssignment):
# the assignment target must not be a constant
if isinstance(node.left, SymbolName):
symdef = node.my_scope().lookup(node.left.name)
if isinstance(symdef, VarDef):
if symdef.vartype == VarType.CONST:
raise ParseError("cannot modify a constant", node.sourceref)
elif symdef.datatype not in {DataType.BYTE, DataType.WORD, DataType.FLOAT}:
raise ParseError("cannot modify that datatype ({:s}) in this way"
.format(symdef.datatype.name.lower()), node.sourceref)
# check for divide by (constant) zero
if node.operator in ("/=", "//="):
if isinstance(node.right, LiteralValue) and node.right.value == 0:
raise ParseError("division by zero", node.right.sourceref)
previous_stmt = node
def check_subroutine_arguments(self, call: SubCall, subdef: Subroutine) -> None:
if len(call.arguments.nodes) != len(subdef.param_spec):
raise ParseError("invalid number of arguments ({:d}, required: {:d})"
.format(len(call.arguments.nodes), len(subdef.param_spec)), call.sourceref)
for arg, param in zip(call.arguments.nodes, subdef.param_spec):
if arg.name:
if not param[0]:
raise ParseError("parameter is unnamed but name was used", arg.sourceref)
if arg.name != param[0]:
raise ParseError("parameter name mismatch", arg.sourceref)
def check_and_merge_zeropages(self, module: Module) -> None:
# merge all ZP blocks into one
zeropage = None
for block in module.all_nodes(Block):
if block.name == "ZP": # type: ignore
if zeropage:
# merge other ZP block into first ZP block
for node in block.scope.nodes:
if isinstance(node, Directive):
zeropage.scope.add_node(node, 0)
elif isinstance(node, VarDef):
zeropage.scope.add_node(node)
else:
raise ParseError("only variables and directives allowed in zeropage block", node.sourceref)
else:
zeropage = block
block.parent.remove_node(block)
if zeropage:
# add the zero page again, as the very first block
module.scope.add_node(zeropage, 0)
def allocate_zeropage_vars(self, module: Module) -> None:
# allocate zeropage variables to the available free zp addresses
if not module.scope.nodes:
return
zpnode = module.scope.nodes[0]
if zpnode.name != "ZP":
return
zeropage = Zeropage(module.zp_options)
for vardef in zpnode.all_nodes(VarDef):
if vardef.datatype.isstring():
raise ParseError("cannot put strings in the zeropage", vardef.sourceref)
try:
if vardef.vartype == VarType.VAR:
vardef.zp_address = zeropage.allocate(vardef)
except CompileError as x:
raise ParseError(str(x), vardef.sourceref)
def check_all_symbolnames(self, module: Module) -> None:
for node in module.all_nodes(SymbolName):
check_symbol_definition(node.name, node.my_scope(), node.sourceref) # type: ignore
def simplify_some_assignments(self, module: Module) -> None:
# simplify some assignment statements,
# note taht most of the expression optimization (constant folding etc) is done in the optimizer.
for node in module.all_nodes():
if isinstance(node, IncrDecr) and node.howmuch not in (0, 1):
_, node.howmuch = coerce_constant_value(datatype_of(node.target, node.my_scope()), node.howmuch, node.sourceref)
attr.validate(node)
elif isinstance(node, VarDef):
dtype = DataType.WORD if node.vartype == VarType.MEMORY else node.datatype
try:
_, node.value = coerce_constant_value(dtype, node.value, node.sourceref) # type: ignore
attr.validate(node)
except OverflowError as x:
raise ParseError(str(x), node.sourceref) from None
elif isinstance(node, Assignment):
lvalue_types = set(datatype_of(lv, node.my_scope()) for lv in node.left.nodes)
if len(lvalue_types) == 1:
_, newright = coerce_constant_value(lvalue_types.pop(), node.right, node.sourceref)
if isinstance(newright, Expression):
node.right = newright # type: ignore
else:
raise TypeError("invalid coerced constant type", newright)
else:
for lv_dt in lvalue_types:
coerce_constant_value(lv_dt, node.right, node.sourceref)
@no_type_check
def create_multiassigns(self, module: Module) -> None:
# create multi-assign statements from nested assignments (A=B=C=5),
# and optimize TargetRegisters down to single Register if it's just one register.
def reduce_right(assign: Assignment) -> Assignment:
if isinstance(assign.right, Assignment):
right = reduce_right(assign.right)
assign.left.nodes.extend(right.left.nodes)
assign.right = right.right
return assign
for node in module.all_nodes(Assignment):
if isinstance(node.right, Assignment):
multi = reduce_right(node)
assert multi is node and len(multi.left.nodes) > 1 and not isinstance(multi.right, Assignment)
@no_type_check
def apply_directive_options(self, module: Module) -> None:
def set_save_registers(scope: Scope, save_dir: Directive) -> None:
if not scope:
return
if len(save_dir.args) > 1:
raise ParseError("expected zero or one directive argument", save_dir.sourceref)
if save_dir.args:
if save_dir.args[0] in ("yes", "true", True):
scope.save_registers = True
elif save_dir.args[0] in ("no", "false", False):
scope.save_registers = False
else:
raise ParseError("invalid directive args", save_dir.sourceref)
else:
scope.save_registers = True
for directive in module.all_nodes(Directive):
node = directive.my_scope().parent
if isinstance(node, Module):
# process the module's directives
if directive.name == "output":
if len(directive.args) != 1 or not isinstance(directive.args[0], str):
raise ParseError("expected one str directive argument", directive.sourceref)
if directive.args[0] == "raw":
node.format = ProgramFormat.RAW
node.address = 0xc000
elif directive.args[0] == "prg":
node.format = ProgramFormat.PRG
node.address = 0xc000
elif directive.args[0] == "basic":
node.format = ProgramFormat.BASIC
node.address = 0x0801
else:
raise ParseError("invalid directive args", directive.sourceref)
elif directive.name == "address":
if len(directive.args) != 1 or type(directive.args[0]) is not int:
raise ParseError("expected one integer directive argument", directive.sourceref)
if node.format == ProgramFormat.BASIC:
raise ParseError("basic cannot have a custom load address", directive.sourceref)
node.address = directive.args[0]
attr.validate(node)
elif directive.name in "import":
pass # is processed earlier
elif directive.name == "zp":
if len(directive.args) not in (1, 2) or set(directive.args) - {"clobber", "restore"}:
raise ParseError("invalid directive args", directive.sourceref)
if "clobber" in directive.args and "restore" in directive.args:
module.zp_options = ZpOptions.CLOBBER_RESTORE
elif "clobber" in directive.args:
module.zp_options = ZpOptions.CLOBBER
elif "restore" in directive.args:
raise ParseError("invalid directive args", directive.sourceref)
elif directive.name == "saveregisters":
set_save_registers(directive.my_scope(), directive)
else:
raise NotImplementedError(directive.name)
elif isinstance(node, Block):
# process the block's directives
if directive.name == "saveregisters":
set_save_registers(directive.my_scope(), directive)
elif directive.name in ("breakpoint", "asmbinary", "asminclude", "noreturn"):
continue
else:
raise NotImplementedError(directive.name)
elif isinstance(node, Subroutine):
# process the sub's directives
if directive.name == "saveregisters":
set_save_registers(directive.my_scope(), directive)
elif directive.name in ("breakpoint", "asmbinary", "asminclude", "noreturn"):
continue
else:
raise NotImplementedError(directive.name)
@no_type_check
def determine_subroutine_usage(self, module: Module) -> None:
module.subroutine_usage.clear()
for node in module.all_nodes():
if isinstance(node, InlineAssembly):
self._get_subroutine_usages_from_asm(module.subroutine_usage, node, node.my_scope())
elif isinstance(node, SubCall):
self._get_subroutine_usages_from_subcall(module.subroutine_usage, node, node.my_scope())
elif isinstance(node, Goto):
self._get_subroutine_usages_from_goto(module.subroutine_usage, node, node.my_scope())
elif isinstance(node, Return):
self._get_subroutine_usages_from_return(module.subroutine_usage, node, node.my_scope())
elif isinstance(node, Assignment):
self._get_subroutine_usages_from_assignment(module.subroutine_usage, node, node.my_scope())
# print("----------SUBROUTINES IN USE-------------")
# import pprint
# pprint.pprint(module.subroutine_usage)
# print("----------/SUBROUTINES IN USE-------------")
def _get_subroutine_usages_from_subcall(self, usages: Dict[Tuple[str, str], Set[str]],
subcall: SubCall, parent_scope: Scope) -> None:
if isinstance(subcall.target, SymbolName):
usages[(parent_scope.name, subcall.target.name)].add(str(subcall.sourceref))
for arg in subcall.arguments.nodes:
self._get_subroutine_usages_from_expression(usages, arg.value, parent_scope)
def _get_subroutine_usages_from_expression(self, usages: Dict[Tuple[str, str], Set[str]],
expr: Any, parent_scope: Scope) -> None:
if expr is None or isinstance(expr, (int, str, float, bool, Register)):
return
elif isinstance(expr, SubCall):
self._get_subroutine_usages_from_subcall(usages, expr, parent_scope)
elif isinstance(expr, ExpressionWithOperator):
self._get_subroutine_usages_from_expression(usages, expr.left, parent_scope)
self._get_subroutine_usages_from_expression(usages, expr.right, parent_scope)
elif isinstance(expr, LiteralValue):
return
elif isinstance(expr, Dereference):
return self._get_subroutine_usages_from_expression(usages, expr.operand, parent_scope)
elif isinstance(expr, AddressOf):
return self._get_subroutine_usages_from_expression(usages, expr.name, parent_scope)
elif isinstance(expr, SymbolName):
try:
symbol = parent_scope.lookup(expr.name)
if isinstance(symbol, Subroutine):
usages[(parent_scope.name, expr.name)].add(str(expr.sourceref))
except UndefinedSymbolError:
pass
else:
raise TypeError("unknown expr type to scan for sub usages", expr, expr.sourceref)
@no_type_check
def _get_subroutine_usages_from_goto(self, usages: Dict[Tuple[str, str], Set[str]],
goto: Goto, parent_scope: Scope) -> None:
if isinstance(goto.target, SymbolName):
usages[(parent_scope.name, goto.target.name)].add(str(goto.sourceref))
self._get_subroutine_usages_from_expression(usages, goto.condition, parent_scope)
def _get_subroutine_usages_from_return(self, usages: Dict[Tuple[str, str], Set[str]],
returnnode: Return, parent_scope: Scope) -> None:
# node.value_A (expression), value_X (expression), value_Y (expression)
self._get_subroutine_usages_from_expression(usages, returnnode.value_A, parent_scope)
self._get_subroutine_usages_from_expression(usages, returnnode.value_X, parent_scope)
self._get_subroutine_usages_from_expression(usages, returnnode.value_Y, parent_scope)
def _get_subroutine_usages_from_assignment(self, usages: Dict[Tuple[str, str], Set[str]],
assignment: Assignment, parent_scope: Scope) -> None:
# node.right (expression, or another Assignment)
if isinstance(assignment.right, Assignment):
self._get_subroutine_usages_from_assignment(usages, assignment.right, parent_scope)
else:
self._get_subroutine_usages_from_expression(usages, assignment.right, parent_scope)
def _get_subroutine_usages_from_asm(self, usages: Dict[Tuple[str, str], Set[str]],
asmnode: InlineAssembly, parent_scope: Scope) -> None:
# asm can refer to other symbols as well, track subroutine usage
for line in asmnode.assembly.splitlines():
splits = line.split(maxsplit=1)
if len(splits) == 2:
for match in re.finditer(r"(?P<symbol>[a-zA-Z_$][a-zA-Z0-9_\.]+)", splits[1]):
name = match.group("symbol")
if name[0] == '$':
continue
try:
symbol = parent_scope.lookup(name)
except UndefinedSymbolError:
pass
else:
if isinstance(symbol, Subroutine):
if symbol.scope:
namespace = symbol.scope.parent_scope.name
else:
namespace, name = name.rsplit(".", maxsplit=2)
usages[(namespace, symbol.name)].add(str(asmnode.sourceref))
def check_directives(self, module: Module) -> None:
imports = set() # type: Set[str]
for node in module.all_nodes():
if isinstance(node, Directive):
assert isinstance(node.parent, Scope)
if node.parent.level == "module":
if node.name not in {"output", "zp", "address", "import", "saveregisters", "noreturn"}:
raise ParseError("invalid directive in module", node.sourceref)
if node.name == "import":
if imports & set(node.args):
raise ParseError("duplicate import", node.sourceref)
imports |= set(node.args)
else:
if node.name not in {"asmbinary", "asminclude", "breakpoint", "saveregisters", "noreturn"}:
raise ParseError("invalid directive in " + node.parent.__class__.__name__.lower(), node.sourceref)
if node.name == "saveregisters":
# it should be the first node in the scope
if node.parent.nodes[0] is not node:
raise ParseError("saveregisters directive must be first in this scope", node.sourceref)
def process_imports(self, module: Module) -> None:
# (recursively) imports the modules
imported = []
for directive in module.all_nodes(Directive):
if directive.name == "import": # type: ignore
if len(directive.args) < 1: # type: ignore
raise ParseError("missing argument(s) for import directive", directive.sourceref)
for arg in directive.args: # type: ignore
filename = self.find_import_file(arg, directive.sourceref.file)
if not filename:
raise ParseError("imported file not found", directive.sourceref)
imported_module, import_parse_errors = self.import_file(filename)
imported_module.scope.parent_scope = module.scope
imported.append(imported_module)
self.parse_errors += import_parse_errors
if not self.imported_module:
# compiler support library is always imported (in main parser)
filename = self.find_import_file("il65lib", module.sourceref.file)
if filename:
imported_module, import_parse_errors = self.import_file(filename)
imported_module.scope.parent_scope = module.scope
imported.append(imported_module)
self.parse_errors += import_parse_errors
else:
raise FileNotFoundError("missing il65lib")
# append the imported module's contents (blocks) at the end of the current module
for block in (node for imported_module in imported
for node in imported_module.scope.nodes
if isinstance(node, Block)):
module.scope.add_node(block)
def import_file(self, filename: str) -> Tuple[Module, int]:
sub_parser = PlyParser(imported_module=True)
return sub_parser.parse_file(filename), sub_parser.parse_errors
def find_import_file(self, modulename: str, sourcefile: str) -> Optional[str]:
candidates = [modulename+".ill", modulename]
filename_at_source_location = os.path.join(os.path.split(sourcefile)[0], modulename)
if filename_at_source_location not in candidates:
candidates.append(filename_at_source_location+".ill")
candidates.append(filename_at_source_location)
filename_at_libs_location = os.path.join(os.path.split(__file__)[0], "lib", modulename)
if filename_at_libs_location not in candidates:
candidates.append(filename_at_libs_location+".ill")
candidates.append(filename_at_libs_location)
for filename in candidates:
if os.path.isfile(filename):
return filename
return None
def handle_parse_error(self, exc: ParseError) -> None:
self.parse_errors += 1
out = sys.stdout
if out.isatty():
print("\x1b[1m", file=out)
if self.imported_module:
print("Error (in imported file):", str(exc), file=out)
else:
print("Error:", str(exc), file=out)
sourcetext = linecache.getline(exc.sourceref.file, exc.sourceref.line).rstrip()
if sourcetext:
print(" " + sourcetext.expandtabs(8), file=out)
if exc.sourceref.column:
print(' ' * (1+exc.sourceref.column) + '^', file=out)
if out.isatty():
print("\x1b[0m", file=out, end="", flush=True)
raise exc # XXX temporary to see where the error occurred
class Zeropage:
SCRATCH_B1 = 0x02
SCRATCH_B2 = 0x03
SCRATCH_W1 = 0xfb # $fb/$fc
SCRATCH_W2 = 0xfd # $fd/$fe
def __init__(self, options: ZpOptions) -> None:
self.free = [] # type: List[int]
self.allocations = {} # type: Dict[int, Tuple[str, DataType]]
if options in (ZpOptions.CLOBBER_RESTORE, ZpOptions.CLOBBER):
# clobber the zp, more free storage, yay!
self.free = list(range(0x04, 0xfb)) + [0xff]
for updated_by_irq in [0xa0, 0xa1, 0xa2, 0x91, 0xc0, 0xc5, 0xcb, 0xf5, 0xf6]:
self.free.remove(updated_by_irq)
else:
# these are valid for the C-64 (when no RS232 I/O is performed):
# ($02, $03, $fb-$fc, $fd-$fe are reserved as scratch addresses for various routines)
self.free = [0x04, 0x05, 0x06, 0x2a, 0x52, 0xf7, 0xf8, 0xf9, 0xfa]
assert self.SCRATCH_B1 not in self.free
assert self.SCRATCH_B2 not in self.free
assert self.SCRATCH_W1 not in self.free
assert self.SCRATCH_W2 not in self.free
def allocate(self, vardef: VarDef) -> int:
assert not vardef.name or vardef.name not in {a[0] for a in self.allocations.values()}, "var name is not unique"
assert vardef.vartype == VarType.VAR, "can only allocate var"
def sequential_free(location: int) -> bool:
return all(location + i in self.free for i in range(size))
def lone_byte(location: int) -> bool:
return (location-1) not in self.free and (location+1) not in self.free and location in self.free
def make_allocation(location: int) -> int:
for loc in range(location, location + size):
self.free.remove(loc)
self.allocations[location] = (vardef.name or "<unnamed>", vardef.datatype)
return location
if vardef.datatype == DataType.BYTE:
size = 1
elif vardef.datatype == DataType.WORD:
size = 2
elif vardef.datatype == DataType.FLOAT:
print_bold("warning: {}: allocating a large datatype in zeropage".format(vardef.sourceref))
size = 5
elif vardef.datatype == DataType.BYTEARRAY:
print_bold("warning: {}: allocating a large datatype in zeropage".format(vardef.sourceref))
size = vardef.size[0]
elif vardef.datatype == DataType.WORDARRAY:
print_bold("warning: {}: allocating a large datatype in zeropage".format(vardef.sourceref))
size = vardef.size[0] * 2
elif vardef.datatype == DataType.MATRIX:
print_bold("warning: {}: allocating a large datatype in zeropage".format(vardef.sourceref))
size = vardef.size[0] * vardef.size[1]
elif vardef.datatype.isstring():
print_bold("warning: {}: allocating a large datatype in zeropage".format(vardef.sourceref))
size = vardef.size[0]
else:
raise CompileError("cannot put datatype {:s} in ZP".format(vardef.datatype.name))
if len(self.free) > 0:
if size == 1:
for candidate in range(min(self.free), max(self.free)+1):
if lone_byte(candidate):
return make_allocation(candidate)
return make_allocation(self.free[0])
for candidate in range(min(self.free), max(self.free)+1):
if sequential_free(candidate):
return make_allocation(candidate)
raise CompileError("ERROR: no more free space in ZP to allocate {:d} sequential bytes".format(size))
def available(self) -> int:
return len(self.free)