#!/usr/bin/env python3
"""CLI: crawl Confluence into Markdown files.

Reads CONFLUENCE_PAT and CONFLUENCE_URL from .env.

Examples:
    python main.py --space KEY --out output/KEY
    python main.py --all --out output
    python main.py --all --max-spaces 3 --verbose
"""
import argparse
import json
import logging
import sys
from pathlib import Path

from confluence_crawler.config import get_config
from confluence_crawler.crawler import ConfluenceCrawler


def main(argv: list[str] | None = None) -> int:
    parser = argparse.ArgumentParser(description="Crawl Confluence pages to Markdown.")
    parser.add_argument("--space", help="crawl only this space key")
    parser.add_argument("--all", action="store_true", help="crawl all visible spaces")
    parser.add_argument("--out", default="output", help="output directory (default: output)")
    parser.add_argument("--max-spaces", type=int, help="limit number of spaces with --all")
    parser.add_argument("--max-pages", type=int, help="limit pages per space (dry run)")
    parser.add_argument("--verbose", action="store_true", help="debug logging")
    args = parser.parse_args(argv)

    logging.basicConfig(level=logging.DEBUG if args.verbose else logging.INFO)

    if not args.space and not args.all:
        parser.error("specify --space KEY or --all")

    try:
        cfg = get_config()
    except RuntimeError as exc:
        print(f"error: {exc}", file=sys.stderr)
        return 2

    crawler = ConfluenceCrawler(cfg["url"], cfg["pat"])
    out = Path(args.out)
    if args.space:
        paths = crawler.crawl_space(args.space, out, max_pages=args.max_pages)
        print(f"wrote {len(paths)} pages to {out}")
    else:
        result = crawler.crawl_all(out, max_spaces=args.max_spaces)
        total = sum(len(p) for p in result.values())
        print(f"wrote {total} pages across {len(result)} spaces to {out}")
    return 0


if __name__ == "__main__":
    sys.exit(main())