#!/usr/bin/env python3
"""Independent verification of checkin-inventory.csv.

Two independent implementations of the same extraction spec:
  build : os.listdir() + open/read + re.match() + str.count()
  verify: os.scandir() + Path.read_text() + re.fullmatch() + str.count()

Exits 0 only if the CSV's rows equal a fresh recomputation from the source files.
"""
import os
import re
import sys
import csv

SRC = "/root/troupe-standup/checkins"
CSV_PATH = "/root/troupe-standup/wizard-work/ledger/checkin-inventory.csv"
PAT = re.compile(r"^(\d{4}-\d{2}-\d{2})-(.+)$")
FIELDS = ["date", "wizard", "filename", "n_lines", "n_did", "n_blocked", "n_next", "n_ask"]


def build_expected():
    rows = []
    with os.scandir(SRC) as it:
        for entry in it:
            if not entry.is_file():
                continue
            name = entry.name
            if not name.endswith(".md"):
                continue
            m = PAT.fullmatch(name)
            if not m:
                continue
            date, wizard = m.group(1), m.group(2)
            with open(entry.path, encoding="utf-8", errors="replace") as fh:
                data = fh.read()
            rows.append((date, wizard, name, data.count("\n"),
                         data.count("Did"), data.count("Blocked"),
                         data.count("Next"), data.count("Ask")))
    rows.sort(key=lambda r: (r[0], r[1]))
    return rows





def _norm(row):
    d = dict(row)
    for k in ("n_lines", "n_did", "n_blocked", "n_next", "n_ask"):
        d[k] = int(d[k])
    return tuple(d[h] for h in FIELDS)


def read_csv():
    with open(CSV_PATH, newline="", encoding="utf-8") as fh:
        reader = csv.DictReader(fh)
        assert reader.fieldnames == FIELDS, f"header mismatch: {reader.fieldnames}"
        return [_norm(r) for r in reader]


def main():
    ok = True
    try:
        expected = build_expected()
        actual = read_csv()
    except Exception as exc:  # noqa: BLE001
        print("FAIL: could not load source or artifact:", exc)
        return 1

    if len(expected) != len(actual):
        ok = False
        print(f"FAIL: row count mismatch: expected {len(expected)}, actual {len(actual)}")

    diff = 0
    for i in range(max(len(expected), len(actual))):
        exp = expected[i] if i < len(expected) else None
        act = actual[i] if i < len(actual) else None
        if exp != act:
            ok = False
            diff += 1
            print(f"FAIL row {i + 1}: expected={exp} actual={act}")

    if ok:
        print(f"PASS: {len(actual)} check-in rows match the source files "
              f"({len(set(r[0] for r in actual))} distinct dates, "
              f"{sorted(set(r[1] for r in actual))})")
    else:
        print(f"FAIL: {diff} differing rows")
    return 0 if ok else 1


if __name__ == "__main__":
    sys.exit(main())