gateway-gamers.net/scripts/sync-elo.py

"""
sync-elo.py — download the stat-check.com ELO Excel sheet and emit public/elo-data.json.

Column auto-detection: scans the header row for keywords.
If headers change, check the CI log — it always prints what it found.
"""

import json
import sys
from datetime import datetime, timezone
from io import BytesIO

import requests
import openpyxl

DOWNLOAD_URL = (
    "https://excel.officeapps.live.com/x/_layouts/XlFileHandler.aspx?WacUserType=WOPI&usid=4bcd36aa-91bc-4c95-a692-c2ef5de1d13a&NoAuth=1&waccluster=PCA1"
)
OUTPUT_PATH = "public/elo-data.json"


def find_col(headers, *keywords):
    """Return index of first header containing any keyword (case-insensitive)."""
    for i, h in enumerate(headers):
        hl = str(h).lower()
        if any(kw in hl for kw in keywords):
            return i
    return None


def main():
    print(f"Downloading ELO sheet…", flush=True)
    resp = requests.get(DOWNLOAD_URL, timeout=120, headers={"User-Agent": "Mozilla/5.0"})
    resp.raise_for_status()

    content_type = resp.headers.get("Content-Type", "")
    print(f"Downloaded {len(resp.content):,} bytes  Content-Type: {content_type}", flush=True)

    # Diagnose non-Excel responses before attempting to parse
    if "html" in content_type.lower() or not resp.content.startswith(b"PK"):
        print("ERROR: Response is not an Excel/ZIP file. First 500 chars:", flush=True)
        print(resp.content[:500].decode("utf-8", errors="replace"), flush=True)
        sys.exit(1)

    wb = openpyxl.load_workbook(BytesIO(resp.content), read_only=True, data_only=True)
    ws = wb.active

    rows = ws.iter_rows(values_only=True)
    raw_headers = next(rows)
    headers = [str(c).strip() if c is not None else "" for c in raw_headers]
    print(f"Headers ({len(headers)}): {headers}", flush=True)

    rank_col   = find_col(headers, "rank")
    name_col   = find_col(headers, "name", "player")
    rating_col = find_col(headers, "rating", "elo", "score", "points")

    if name_col is None:
        print("ERROR: Could not find a name/player column. Check the headers above.", file=sys.stderr)
        sys.exit(1)

    print(
        f"Mapped  rank→col {rank_col} ('{headers[rank_col] if rank_col is not None else '—'}')  "
        f"name→col {name_col} ('{headers[name_col]}')  "
        f"rating→col {rating_col} ('{headers[rating_col] if rating_col is not None else '—'}')",
        flush=True,
    )

    by_name = {}
    skipped = 0
    for seq, row in enumerate(rows, start=2):
        raw_name = row[name_col] if len(row) > name_col else None
        if not raw_name:
            skipped += 1
            continue

        name    = str(raw_name).strip()
        key     = name.lower()

        raw_rank   = row[rank_col]   if rank_col is not None and len(row) > rank_col   else None
        raw_rating = row[rating_col] if rating_col is not None and len(row) > rating_col else None

        try:
            rank = int(raw_rank) if raw_rank is not None else seq
        except (ValueError, TypeError):
            rank = seq

        try:
            rating = round(float(raw_rating), 1) if raw_rating is not None else None
        except (ValueError, TypeError):
            rating = None

        by_name[key] = {"rank": rank, "rating": rating, "name": name}

    wb.close()

    print(f"Parsed {len(by_name):,} players ({skipped} blank rows skipped)", flush=True)

    output = {
        "updated": datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ"),
        "count": len(by_name),
        "byName": by_name,
    }

    with open(OUTPUT_PATH, "w", encoding="utf-8") as f:
        json.dump(output, f, separators=(",", ":"), ensure_ascii=False)

    print(f"Wrote {OUTPUT_PATH}", flush=True)


if __name__ == "__main__":
    main()