"""
Generate Lib/_opcode_metadata.py for RustPython bytecode.

This file generates opcode metadata that is compatible with CPython 3.13.
"""

import itertools
import pathlib
import re
import typing

ROOT = pathlib.Path(__file__).parents[1]
BYTECODE_FILE = (
    ROOT / "crates" / "compiler-core" / "src" / "bytecode" / "instruction.rs"
)
OPCODE_METADATA_FILE = ROOT / "Lib" / "_opcode_metadata.py"


class Opcode(typing.NamedTuple):
    rust_name: str
    id: int

    @property
    def cpython_name(self) -> str:
        name = re.sub(r"(?<=[a-z0-9])([A-Z])", r"_\1", self.rust_name)
        return re.sub(r"(\D)(\d+)$", r"\1_\2", name).upper()

    @classmethod
    def from_str(cls, body: str):
        raw_variants = re.split(r"(\d+),", body.strip())
        raw_variants.remove("")
        for raw_name, raw_id in itertools.batched(raw_variants, 2):
            name = re.findall(r"\b[A-Z][A-Za-z]*\d*\b(?=\s*[\({=])", raw_name)[0]
            yield cls(rust_name=name.strip(), id=int(raw_id))

    def __lt__(self, other: typing.Self) -> bool:
        return self.id < other.id


def extract_enum_body(contents: str, enum_name: str) -> str:
    res = re.search(f"pub enum {enum_name} " + r"\{(.+?)\n\}", contents, re.DOTALL)
    if not res:
        raise ValueError(f"Could not find {enum_name} enum")

    return "\n".join(
        line.split("//")[0].strip()  # Remove any comment. i.e. "foo // some comment"
        for line in res.group(1).splitlines()
        if not line.strip().startswith("//")  # Ignore comment lines
    )


contents = BYTECODE_FILE.read_text()
enum_body = "\n".join(
    extract_enum_body(contents, enum_name)
    for enum_name in ("Instruction", "PseudoInstruction")
)
opcodes = list(Opcode.from_str(enum_body))

# Generate the output file
output = """# This file is generated by scripts/generate_opcode_metadata.py
# for RustPython bytecode format (CPython 3.13 compatible opcode numbers).
# Do not edit!

_specializations = {}

_specialized_opmap = {}

opmap = {
"""

for opcode in sorted(opcodes):
    output += f"    '{opcode.cpython_name}': {opcode.id},\n"

output += """}

# CPython 3.13 compatible: opcodes < 44 have no argument
HAVE_ARGUMENT = 44
MIN_INSTRUMENTED_OPCODE = 236
"""

OPCODE_METADATA_FILE.write_text(output)