#!/usr/bin/env python3
"""Per-tid truncation for canary JSONL logs.

Canary's full boot log can exceed 800 MB; the diff tool loads the
entire file into RAM. We only need enough events per tid to walk past
the first divergence — anything beyond is dead weight. Cap each tid at
a configurable max (default: 250k for tid=6 main, 20k for others)."""

import json
import sys
from pathlib import Path

MAIN_CAP = 250_000  # tid=6 (canary's main chain — mapped to ours tid=1)
SISTER_CAP = 20_000  # everything else


def main() -> int:
    src = Path(sys.argv[1])
    dst = Path(sys.argv[2])
    counts: dict[int, int] = {}
    kept = 0
    total = 0
    with src.open("r", encoding="utf-8") as fin, dst.open("w", encoding="utf-8") as fout:
        for lineno, line in enumerate(fin, start=1):
            if lineno == 1:
                fout.write(line)
                continue
            total += 1
            try:
                ev = json.loads(line)
            except json.JSONDecodeError:
                continue
            tid = ev.get("tid", 0)
            cap = MAIN_CAP if tid == 6 else SISTER_CAP
            c = counts.get(tid, 0)
            if c >= cap:
                continue
            counts[tid] = c + 1
            fout.write(line)
            kept += 1
    print(f"kept {kept}/{total} events across {len(counts)} tids")
    for tid in sorted(counts):
        print(f"  tid={tid:4d}  {counts[tid]}")
    return 0


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