llm-wiki/harness/adapters/generate_rules.py

#!/usr/bin/env python3
"""Generate platform rule slices from repository-neutral root rule SSOT files."""

from __future__ import annotations

import argparse
import hashlib
import json
from pathlib import Path
import sys
from typing import Any


DEFAULT_ROOT = Path(__file__).resolve().parents[2]
CONFIG = Path("harness/source/rule-adapters.json")
sys.path.insert(0, str(DEFAULT_ROOT / "harness/runtime"))
from fs_transaction import replace_many  # noqa: E402


class RuleGenerationError(ValueError):
    pass


def _safe_path(value: Any) -> Path:
    if not isinstance(value, str) or not value:
        raise RuleGenerationError("path must be a non-empty string")
    path = Path(value)
    if path.is_absolute() or ".." in path.parts:
        raise RuleGenerationError(f"path escapes repository: {value}")
    return path


def _slice(text: str, start: str, end: str | None, source: Path) -> str:
    lines = text.splitlines(keepends=True)
    try:
        start_index = next(index for index, line in enumerate(lines) if line.rstrip("\n") == start)
    except StopIteration as exc:
        raise RuleGenerationError(f"{source}: missing start heading {start!r}") from exc
    end_index = len(lines)
    if end is not None:
        try:
            end_index = next(
                index for index, line in enumerate(lines[start_index + 1 :], start_index + 1)
                if line.rstrip("\n") == end
            )
        except StopIteration as exc:
            raise RuleGenerationError(f"{source}: missing end heading {end!r}") from exc
    return "".join(lines[start_index:end_index]).rstrip() + "\n"


def render(root: Path) -> dict[Path, str]:
    config_path = root / CONFIG
    config = json.loads(config_path.read_text(encoding="utf-8"))
    if config.get("schema_version") != "rule-adapters/v1":
        raise RuleGenerationError("expected rule-adapters/v1")
    groups = config.get("groups")
    if not isinstance(groups, list) or not groups:
        raise RuleGenerationError("groups must be a non-empty list")
    rendered: dict[Path, str] = {}
    for group in groups:
        if not isinstance(group, dict):
            raise RuleGenerationError("group must be an object")
        source_rel = _safe_path(group.get("source"))
        target_dir = _safe_path(group.get("target_dir"))
        source = root / source_rel
        source_text = source.read_text(encoding="utf-8")
        digest = hashlib.sha256(source_text.encode("utf-8")).hexdigest()
        slices = group.get("slices")
        if not isinstance(slices, list) or not slices:
            raise RuleGenerationError(f"{source_rel}: slices must be non-empty")
        links: list[str] = []
        for item in slices:
            target_name = _safe_path(item.get("target"))
            if len(target_name.parts) != 1:
                raise RuleGenerationError("rule slice target must be a filename")
            start = item.get("start")
            end = item.get("end")
            if not isinstance(start, str) or (end is not None and not isinstance(end, str)):
                raise RuleGenerationError("slice start/end must be headings")
            target = target_dir / target_name
            if target in rendered:
                raise RuleGenerationError(f"duplicate rule target: {target}")
            marker = f"<!-- GENERATED from {source_rel.as_posix()} sha256:{digest}; DO NOT EDIT -->"
            rendered[target] = (
                f"{marker}\n\n"
                f"Root SSOT: [`{source_rel.as_posix()}`](../../../../../{source_rel.as_posix()})\n\n"
                f"{_slice(source_text, start, end, source_rel)}"
            )
            links.append(f"- [`{target_name.as_posix()}`]({target_name.as_posix()}): `{start}`")
        index_name = _safe_path(group.get("index"))
        index = target_dir / index_name
        marker = f"<!-- GENERATED from {source_rel.as_posix()} sha256:{digest}; DO NOT EDIT -->"
        rendered[index] = (
            f"{marker}\n\n"
            f"# 생성된 rule index\n\n"
            f"Root SSOT: [`{source_rel.as_posix()}`](../../../../../{source_rel.as_posix()})\n\n"
            "아래 파일은 root SSOT의 heading 구간에서 생성된다. 직접 편집하지 않는다.\n\n"
            + "\n".join(links)
            + "\n"
        )
    return rendered


def generate(root: Path, check: bool) -> tuple[list[str], list[str]]:
    expected = render(root)
    stale = [path.as_posix() for path, text in expected.items() if not (root / path).is_file() or (root / path).read_text(encoding="utf-8") != text]
    written: list[str] = []
    if not check and stale:
        changes = {root / path: expected[path].encode("utf-8") for path in expected if path.as_posix() in stale}
        replace_many(changes)
        written = sorted(stale)
    return sorted(stale), written


def main(argv: list[str] | None = None) -> int:
    parser = argparse.ArgumentParser(description=__doc__)
    mode = parser.add_mutually_exclusive_group(required=True)
    mode.add_argument("--check", action="store_true")
    mode.add_argument("--write", action="store_true")
    parser.add_argument("--root", type=Path, default=DEFAULT_ROOT)
    args = parser.parse_args(argv)
    try:
        root = args.root.resolve(strict=True)
        stale, written = generate(root, args.check)
        result = {
            "schema_version": "rule-adapter-result/v1",
            "status": "DRIFT" if args.check and stale else "UPDATED" if written else "CURRENT",
            "stale": stale,
            "written": written,
            "target_count": len(render(root)),
        }
        json.dump(result, sys.stdout, ensure_ascii=False, indent=2, sort_keys=True)
        sys.stdout.write("\n")
        return 1 if args.check and stale else 0
    except (OSError, UnicodeError, json.JSONDecodeError, RuleGenerationError) as exc:
        json.dump({"schema_version": "rule-adapter-result/v1", "status": "FAIL", "error": str(exc)}, sys.stdout, ensure_ascii=False, indent=2)
        sys.stdout.write("\n")
        return 2


if __name__ == "__main__":
    raise SystemExit(main())