backend/api/selected_tools.py

from flask import Blueprint, jsonify
from datasets import load_dataset
import json

bp = Blueprint("selected_tools", __name__, url_prefix="/api/selected-tools")

HF_REPO = "timchen0618/browsecomp-plus-selected-tools-analysis-v1"

_cache: list | None = None


def _load():
    global _cache
    if _cache is not None:
        return _cache
    ds = load_dataset(HF_REPO, split="train")
    rows = []
    for row in ds:
        tool_counts = {}
        try:
            tool_counts = json.loads(row.get("tool_call_counts") or "{}")
        except Exception:
            pass
        total_tool_calls = sum(tool_counts.values()) if tool_counts else 0
        rows.append({
            "query_id": str(row["query_id"]),
            "rationale": row["rationale"],
            "selected_indices": row["selected_indices"],
            "k_requested": int(row["k_requested"]),
            "k_effective": int(row["k_effective"]),
            "excerpt": row["excerpt"],
            "new_trajectory": row["new_trajectory"],
            "direct_answer": bool(row["direct_answer"]),
            "tool_call_counts": tool_counts,
            "total_tool_calls": total_tool_calls,
            "status": row["status"],
        })
    _cache = rows
    return rows


@bp.get("/")
def get_data():
    try:
        rows = _load()
        return jsonify({"rows": rows})
    except Exception as e:
        return jsonify({"error": str(e)}), 500


@bp.post("/reload")
def reload_data():
    global _cache
    _cache = None
    try:
        rows = _load()
        return jsonify({"status": "ok", "count": len(rows)})
    except Exception as e:
        return jsonify({"error": str(e)}), 500