#!/usr/bin/env python3
"""Extract the committed file tree of the troupe repo into a verified manifest.

Source of truth: `git ls-files` (tracked files) in the repo root.
Outputs (in the same directory as this script, wizard-work/ledger/):
  - 2026-10-29-git-tree-manifest.csv  : relpath, ext, size_bytes, sha256
  - 2026-10-29-git-tree-manifest.json : structured manifest
  - 2026-10-29-git-tree-manifest.md   : human-readable table

The independent re-derivation lives in verify_git_tree_manifest.py;
the committed CHECK runs that verifier and exits 0 iff it matches.
"""
import hashlib
import json
import os
import pathlib

REPO_ROOT = pathlib.Path(__file__).resolve().parent.parent.parent  # troupe-standup
HERE = pathlib.Path(__file__).resolve().parent
OUT_DIR = HERE


def find_repo_root(start: pathlib.Path) -> pathlib.Path:
    """Walk up from `start` until a directory containing `.git` is found."""
    current = start.resolve()
    for _ in range(12):
        if (current / ".git").exists():
            return current
        if current.parent == current:
            break
        current = current.parent
    raise SystemExit(f"no .git found above {start}")


def sha256_of(path: pathlib.Path) -> str:
    h = hashlib.sha256()
    with path.open("rb") as fh:
        for chunk in iter(lambda: fh.read(1024 * 1024), b""):
            h.update(chunk)
    return h.hexdigest()


def main() -> None:
    REPO_ROOT = find_repo_root(HERE)
    import subprocess
    out = subprocess.run(
        ["git", "ls-files"], cwd=REPO_ROOT, check=True,
        capture_output=True, text=True, env=os.environ,
    ).stdout.splitlines()

    rows = []
    for rel in sorted(out):
        p = (REPO_ROOT / rel).resolve()
        ext = p.suffix.lower() if p.suffix else "-"
        rows.append({
            "relpath": rel,
            "ext": ext,
            "size_bytes": p.stat().st_size,
            "sha256": sha256_of(p),
        })

    ext_counts = {}
    total_bytes = 0
    for r in rows:
        ext_counts[r["ext"]] = ext_counts.get(r["ext"], 0) + 1
        total_bytes += r["size_bytes"]

    manifest = {
        "repo": str(REPO_ROOT),
        "generated_at": __import__("datetime").datetime.now(__import__("datetime").UTC).isoformat(),
        "tracked_file_count": len(rows),
        "total_bytes": total_bytes,
        "extension_counts": ext_counts,
        "top_extensions": sorted(ext_counts.items(), key=lambda kv: (-kv[1], kv[0]))[:15],
        "files": rows,
    }

    (OUT_DIR / "2026-10-29-git-tree-manifest.json").write_text(
        json.dumps(manifest, indent=2) + "\n", encoding="utf-8"
    )

    with (OUT_DIR / "2026-10-29-git-tree-manifest.csv").open("w", encoding="utf-8",
                                                             newline="") as fh:
        fh.write("relpath\text\tytes_sha256\n")
        for r in rows:
            fh.write(f'{r["relpath"]}\t{r["ext"]}\t{r["size_bytes"]}\t{r["sha256"]}\n')

    lines = [
        "# Git tree manifest — 2026-10-29",
        "",
        f"- repo: `{REPO_ROOT}`",
        f"- tracked files: {manifest['tracked_file_count']}",
        f"- total bytes: {manifest['total_bytes']:,}",
        "",
        "## Top extensions",
        "",
        "| extension | files |",
        "|---|---|",
    ]
    for ext, cnt in manifest["top_extensions"]:
        lines.append(f"| {ext or '(no extension)'} | {cnt} |")
    lines += ["", "## Full file list", "", "| # | relpath | ext | bytes |", "|---|---|---|---|"]
    for i, r in enumerate(manifest["files"], 1):
        lines.append(f"| {i} | {r['relpath']} | {r['ext']} | {r['size_bytes']} |")
    (OUT_DIR / "2026-10-29-git-tree-manifest.md").write_text("\n".join(lines) + "\n", encoding="utf-8")

    print(f"wrote {len(rows)} files -> {OUT_DIR}")


if __name__ == "__main__":
    main()