#!/usr/bin/env python3
"""PR Newswire press releases for a list of companies or a topic, saved to a CSV (one row per release).

Runs the Apify Actor nexgensignal/press-release-monitor-api once.

    export APIFY_TOKEN=your_apify_token          (Windows: set APIFY_TOKEN=...)
    python prnewswire_to_csv.py --companies "Kimberly-Clark" "Gates Corporation" --per-company 3
    python prnewswire_to_csv.py --topics mergers_acquisitions executive_appointments --max 10
    python prnewswire_to_csv.py --topics venture_capital --keyword AI --since 2026-09-24

A watchlist keeps only releases the company issued itself (not ones that merely mention it). Topics are
PR Newswire's own lists (mergers_acquisitions, earnings, executive_appointments, bankruptcy, contracts,
venture_capital and 60 more; see the Actor's input page). PR Newswire only, English releases.
Cost: $0.05 per release on Apify's Free plan ($0.0335 on Gold); --cap stops the run at that many dollars.
Python 3.8+, nothing to install.
"""
import argparse, csv, json, os, sys, urllib.request

ACTOR = "nexgensignal~press-release-monitor-api"
COLS = ["date", "published_at", "company", "tickers", "title", "location", "topics", "watched_company", "summary", "url",
        "record_id"]


def main():
    p = argparse.ArgumentParser()
    p.add_argument("--companies", nargs="*", default=[], help="company names (watchlist, up to 50)")
    p.add_argument("--topics", nargs="*", default=[], help="PR Newswire topic ids")
    p.add_argument("--keyword", default=None, help="keyword (narrows topics/watchlist to headline or opening lines)")
    p.add_argument("--since", default=None, help="only releases on or after this date (YYYY-MM-DD)")
    p.add_argument("--per-company", type=int, default=3)
    p.add_argument("--max", type=int, default=10, help="most releases returned (each is charged)")
    p.add_argument("--cap", type=float, default=1.00, help="maximum charge for the run, USD")
    p.add_argument("--out", default="press_releases.csv")
    a = p.parse_args()
    token = os.environ.get("APIFY_TOKEN") or sys.exit("Set APIFY_TOKEN to your own Apify API token first.")
    if not (a.companies or a.topics or a.keyword):
        sys.exit("Give --companies, --topics or --keyword.")
    run_input = {"maxReleases": a.max}
    if a.companies:
        run_input.update(companies=a.companies, maxReleasesPerCompany=a.per_company)
    if a.topics:
        run_input["topics"] = a.topics
    if a.keyword:
        run_input["keyword"] = a.keyword
    if a.since:
        run_input["dateFrom"] = a.since
    url = (f"https://api.apify.com/v2/acts/{ACTOR}/run-sync-get-dataset-items"
           f"?maxTotalChargeUsd={a.cap}&clean=true")
    req = urllib.request.Request(url, data=json.dumps(run_input).encode(), method="POST",
                                 headers={"Content-Type": "application/json", "Authorization": "Bearer " + token})
    try:
        rows = json.load(urllib.request.urlopen(req, timeout=320))
    except Exception as e:  # noqa: BLE001
        sys.exit(f"No answer from Apify ({e}). A run may still have started: check Apify Console > Runs "
                 f"before running this again.")
    with open(a.out, "w", newline="", encoding="utf-8") as f:
        w = csv.DictWriter(f, fieldnames=COLS, extrasaction="ignore")
        w.writeheader()
        for r in rows:
            r = dict(r)
            for k in ("tickers", "topics"):
                r[k] = "; ".join(r.get(k) or [])
            w.writerow(r)
    print(f"{len(rows)} release(s) saved to {a.out}")
    for r in rows[:10]:
        print(f"  {r.get('date')}  {r.get('company')}: {r.get('title')}")


if __name__ == "__main__":
    main()