feat(clew): extract rpa archives into game tree

.gitignore

@@ -20,6 +20,9 @@ Thumbs.db
 # Bazel convenience symlinks
 /bazel-*
 
+# local generated output
+/output/
+
 # language package-manager caches and outputs
 node_modules/
 dist/

src/clew/BUILD.bazel

@@ -0,0 +1,8 @@
+load("@rules_python//python:defs.bzl", "py_binary")
+
+py_binary(
+    name = "clew",
+    srcs = ["clew.py"],
+    main = "clew.py",
+    deps = ["//third_party/python:rpycdec"],
+)

src/clew/clew.py

@@ -0,0 +1,101 @@
+from __future__ import annotations
+
+import argparse
+import logging
+import os
+import shutil
+import sys
+from collections import deque
+from pathlib import Path
+
+from rpycdec.rpa import extract_rpa
+
+LOG = logging.getLogger(__name__)
+
+def parse_args(argv: list[str]) -> argparse.Namespace:
+    parser = argparse.ArgumentParser(
+        prog="clew",
+        description="Extract Ren'Py game files.",
+    )
+    parser.add_argument(
+        "--game-root",
+        required=True,
+        type=Path,
+        help="Game installation root to scan.",
+    )
+    parser.add_argument(
+        "--output-root",
+        type=Path,
+        default=Path("output"),
+        help="Output root. Defaults to repository output/.",
+    )
+    parser.add_argument(
+        "--overwrite",
+        action="store_true",
+        help="Replace Clew-managed output directories.",
+    )
+    return parser.parse_args(argv)
+
+
+def resolve_output_root(output_root: Path) -> Path:
+    if output_root.is_absolute():
+        return output_root.resolve()
+
+    workspace = os.environ.get("BUILD_WORKSPACE_DIRECTORY")
+    if workspace:
+        return (Path(workspace) / output_root).resolve()
+
+    return output_root.resolve()
+
+
+def get_output_directory(output_root: Path, folder: str) -> Path:
+    directory = output_root / folder
+    directory.mkdir(parents=True, exist_ok=True)
+    return directory
+
+
+def extract_archive(archive: Path, destination: Path) -> None:
+    destination.mkdir(parents=True, exist_ok=True)
+    with archive.open("rb") as handle:
+        extract_rpa(handle, output_dir=str(destination))
+
+
+def extract_game_files(game_root: Path, output_path: Path) -> None:
+    queue = deque(game_root.iterdir())
+
+    while queue:
+        item = queue.popleft()
+        destination = output_path / item.relative_to(game_root)
+
+        if item.is_dir():
+            destination.mkdir(parents=True, exist_ok=True)
+            queue.extend(item.iterdir())
+        elif item.is_file() and item.suffix.lower() == ".rpa":
+            extract_archive(item, destination.with_suffix(""))
+        elif item.is_file():
+            destination.parent.mkdir(parents=True, exist_ok=True)
+            shutil.copy2(item, destination)
+
+
+def run(args: argparse.Namespace) -> None:
+    game_root = args.game_root.resolve()
+    output_root = resolve_output_root(args.output_root)
+
+    if not game_root.is_dir():
+        LOG.critical(f'game root is not a directory: {game_root}')
+        raise SystemExit(2)
+
+    extracted = get_output_directory(output_root, "extracted")
+    extract_game_files(game_root, extracted)
+
+
+if __name__ == "__main__":
+    args = parse_args(sys.argv[1:])
+
+    logging.basicConfig(
+        level=logging.INFO,
+        format="[%(asctime)s] %(levelname)s %(filename)s:%(lineno)d | %(message)s",
+        datefmt="%Y-%m-%d][%H:%M:%S",
+    )
+
+    run(args)