"""Build search_outcomes.csv (the table in data/) from a searched workspace.

    python outcomes.py ws [OUT.csv]

One row per reaction searched, in the columns of data/search_outcomes.csv. A reaction collected twice
keeps its real result over an empty error row. Needs only Python 3.
"""
import csv
import json
import sys
from pathlib import Path

ws = Path(sys.argv[1])
out = Path(sys.argv[2] if len(sys.argv) > 2 else "search_outcomes.csv")


def jl(p):
    return [json.loads(ln) for ln in p.read_text().splitlines() if ln.strip()]


rx = {r["id"]: r for r in csv.DictReader(open(ws / "reactions.csv"))}
screen = {r["id"]: r for r in jl(ws / "screen/results.jsonl")}
search = {}
for s in jl(ws / "search/results.jsonl"):
    if s["id"] not in search or s.get("status") != "error":
        search[s["id"]] = s
rows = [{"reaction_id": s["id"], "n_atoms": rx[s["id"]]["n_atoms"], "status": s.get("status"),
         "outcome": s.get("outcome") if s.get("status") == "done" else "", "best_kind": s.get("best_kind") or "",
         "ref_barrier_kcal": s.get("ref_barrier_kcal"), "best_barrier_kcal": s.get("best_barrier_kcal"),
         "lower_by_kcal": s.get("lower_by_kcal"), "n_direct": s.get("n_direct"), "n_multistep": s.get("n_multistep"),
         "seconds": s.get("seconds"), "dataset_barrier_kcal": rx[s["id"]]["reference_barrier_kcal"],
         "screen_barrier_kcal": screen[s["id"]].get("barrier_kcal")} for s in search.values()]
with open(out, "w", newline="") as fh:
    w = csv.DictWriter(fh, fieldnames=list(rows[0]))
    w.writeheader()
    w.writerows(rows)
print(f"{out}: {len(rows)} reactions")