#!/usr/bin/env python3
"""Summarize LLM_CONNECT_* records from a DeerFlow backend log."""

from __future__ import annotations

import argparse
import re
import sys
from pathlib import Path


RUN_RE = re.compile(r"run_id=([^ ]+)")
VALUE_RE = re.compile(r"(first_chunk_ms|total_ms)=([^ ]+)")


def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("log", nargs="?", type=Path, help="Log file; omit to read stdin")
    args = parser.parse_args()
    text = args.log.read_text(encoding="utf-8", errors="replace") if args.log else sys.stdin.read()

    calls: dict[str, dict[str, str]] = {}
    for line in text.splitlines():
        if "LLM_" not in line:
            continue
        run_match = RUN_RE.search(line)
        if not run_match:
            continue
        run_id = run_match.group(1)
        row = calls.setdefault(run_id, {"run_id": run_id, "status": "started"})
        if "LLM_CONNECT_OK" in line:
            row["status"] = "connected"
        if "LLM_REQUEST_DONE" in line:
            row["status"] = "success"
        if "LLM_REQUEST_FAILED" in line:
            row["status"] = "failed"
        for key, value in VALUE_RE.findall(line):
            row[key] = value

    print("run_id status first_chunk_ms total_ms")
    for row in calls.values():
        print(
            row["run_id"],
            row.get("status", "-"),
            row.get("first_chunk_ms", "-"),
            row.get("total_ms", "-"),
        )

    succeeded = sum(row.get("status") == "success" for row in calls.values())
    failed = sum(row.get("status") == "failed" for row in calls.values())
    unfinished = len(calls) - succeeded - failed
    print(f"SUMMARY calls={len(calls)} success={succeeded} failed={failed} unfinished={unfinished}")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())