#!/usr/bin/env python3
# build_dir_census.py -- regenerate the ledger dir census CSV (extract lane).
#
# Schema (must match verify_ledger_dir_census.py):
#   rel_path,size_bytes,sha256
#   rel_path is relative to the troupe-standup repo root, sorted ascending.
#   The artifact itself is excluded from BOTH the re-derivation and the
#   committed file, so the check is self-consistent across regeneration.
#
# Consumes (real tool output): the live filesystem of wizard-work/ledger/.
# Produces: the committed artifact at
#   wizard-work/ledger/2026-10-30-ledger-dir-census.csv
# and the primary deliverable at
#   /root/.hermes/wizard-outputs/ledger/2026-10-30-ledger-dir-census.csv
import os
import csv
import hashlib

HERE = os.path.dirname(os.path.abspath(__file__))            # wizard-work/ledger
REPO_ROOT = os.path.abspath(os.path.join(HERE, "..", ".."))  # troupe-standup root
SRC = HERE
OUT = "/root/troupe-standup/wizard-work/ledger/2026-10-30-ledger-dir-census.csv"


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


def re_derive():
    rows = []
    for dirpath, dirnames, filenames in os.walk(SRC):
        dirnames.sort()
        for fn in sorted(filenames):
            full = os.path.join(dirpath, fn)
            if os.path.abspath(full) == os.path.abspath(OUT):
                continue
            rel = os.path.relpath(full, REPO_ROOT)
            size = os.path.getsize(full)
            rows.append((rel, size, sha256_of(full)))
    rows.sort(key=lambda r: r[0])
    return rows


def main():
    rows = re_derive()
    os.makedirs(os.path.dirname(OUT), exist_ok=True)
    with open(OUT, "w", newline="", encoding="utf-8") as f:
        w = csv.writer(f)
        w.writerow(["rel_path", "size_bytes", "sha256"])
        w.writerows(rows)
    print(f"rows: {len(rows)}")
    print(f"wrote: {OUT}")


if __name__ == "__main__":
    main()