# /// script
# dependencies = ["icalendar==7.3.0", "recurring-ical-events==3.8.2"]
# ///
"""Expected VEVENT instances for tests/corpus.rs, from recurring-ical-events.

    uv run oracle.py <fixture dir> > oracle.tsv          # compact, checked in
    uv run oracle.py <fixture dir> --full > full.tsv     # every start, for review

Conventions shared with corpus.rs: starts in [1970, 2040) as unix seconds,
floating times and dates in UTC, X-WR-TIMEZONE ignored.
"""
import datetime as dt
import os
import sys
from concurrent.futures import ProcessPoolExecutor

UTC = dt.timezone.utc
LO, HI = dt.datetime(1970, 1, 1, tzinfo=UTC), dt.datetime(2040, 1, 1, tzinfo=UTC)


def fnv(data: bytes) -> str:
    h = 0xCBF29CE484222325
    for b in data:
        h = ((h ^ b) * 0x100000001B3) & 0xFFFFFFFFFFFFFFFF
    return f"{h:016x}"


def expand(path):
    # One process per file: icalendar caches TZIDs globally, so a VTIMEZONE
    # from one fixture would leak into the next.
    import icalendar
    import recurring_ical_events

    raw = open(path, "rb").read()
    try:
        txt = raw.decode("utf-8")
    except UnicodeDecodeError:
        txt = raw.decode("latin-1")
    starts = []
    try:
        for cal in icalendar.Calendar.from_ical(txt, multiple=True):
            cal.pop("X-WR-TIMEZONE", None)
            for ev in recurring_ical_events.of(cal, components=["VEVENT"]).between(LO, HI):
                s = ev.get("DTSTART")
                if s is None:
                    continue
                s = s.dt
                if not isinstance(s, dt.datetime):
                    s = dt.datetime(s.year, s.month, s.day)
                if s.tzinfo is None:
                    s = s.replace(tzinfo=UTC)
                if LO <= s < HI:
                    starts.append(int(s.timestamp()))
        status = "ok"
    except Exception as e:
        status = "error:" + type(e).__name__
    starts.sort()
    return fnv(raw), status, starts


def main():
    d, full = sys.argv[1], "--full" in sys.argv
    names = sorted(n for n in os.listdir(d) if n.endswith(".ics"))
    with ProcessPoolExecutor(max_tasks_per_child=1) as pool:
        for name, (file_hash, status, starts) in zip(
            names, pool.map(expand, [os.path.join(d, n) for n in names])
        ):
            text = ",".join(map(str, starts))
            last = text if full else fnv(text.encode())
            print(f"{name}\t{file_hash}\t{status}\t{len(starts)}\t{last}", flush=True)


if __name__ == "__main__":
    main()