#!/usr/bin/env python3
"""Generate an RSS 2.0 feed (feed.xml) from a directory of HTML posts.

Expects each post to be a standalone HTML file containing:
  - a <time datetime="..."> element (ISO 8601)   -- required
  - an <h2> title                                -- falls back to filename
  - og:url / og:description meta tags            -- falls back to site URL
                                                   and first <p> of the post
"""

import argparse
import re
from datetime import datetime
from email.utils import format_datetime
from pathlib import Path
from xml.sax.saxutils import escape

TIME_RE    = re.compile(r'<time[^>]+datetime="([^"]+)"', re.I)
OG_DESC_RE = re.compile(r'<meta property="og:description" content="([^"]*)"', re.I)
OG_URL_RE  = re.compile(r'<meta property="og:url" content="([^"]*)"', re.I)
H2_RE      = re.compile(r"<h2[^>]*>(.*?)</h2>", re.I | re.S)
CONTENT_RE = re.compile(r'<div class="post_content">(.*?)</div>', re.I | re.S)
STRIP_TAGS = re.compile(r"<[^>]+>")


def find(pattern, text):
    """Return the first capture group of a regex, or None."""
    m = pattern.search(text)
    return m.group(1).strip() if m else None


def build_entry(path, site_url):
    text = path.read_text(encoding="utf-8", errors="replace")

    title = find(H2_RE, text)
    if title:
        title = STRIP_TAGS.sub("", title).strip()

    url = find(OG_URL_RE, text) or f"{site_url}/posts/{path.name}"

    description = find(OG_DESC_RE, text)
    if not description:
        # fall back to the first <p> inside the post content div
        content = find(CONTENT_RE, text) or ""
        p = re.search(r"<p[^>]*>(.*?)</p>", content, re.I | re.S)
        description = STRIP_TAGS.sub("", p.group(1)).strip()[:300] if p else ""

    date_str = find(TIME_RE, text)
    if not date_str:
        print(f"  ! no <time datetime> in {path.name}, skipping")
        return None
    date = datetime.fromisoformat(date_str)  # preserves original tz offset

    return {
        "title": title or path.stem,
        "url": url.replace("/posts/../", "/posts/"),  # tidy ../ artifacts
        "description": description,
        "date": date,
    }


def build_feed(args):
    post_dir = Path(args.posts_dir)
    paths = sorted(post_dir.glob("*.html"))

    entries = []
    for p in paths:
        entry = build_entry(p, args.site_url)
        if entry:
            entries.append(entry)

    entries.sort(key=lambda e: e["date"], reverse=True)
    entries = entries[:args.max_items]
    latest = entries[0]["date"] if entries else datetime.now()

    items = "\n".join(
        f"""    <item>
      <title>{escape(e["title"])}</title>
      <link>{escape(e["url"])}</link>
      <guid>{escape(e["url"])}</guid>
      <pubDate>{format_datetime(e["date"])}</pubDate>
      <description>{escape(e["description"])}</description>
    </item>"""
        for e in entries
    )

    return f"""<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <title>{escape(args.title)}</title>
    <link>{args.site_url}</link>
    <description>{escape(args.description)}</description>
    <lastBuildDate>{format_datetime(latest)}</lastBuildDate>
{items}
  </channel>
</rss>
"""


def main():
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--site-url", required=True, help="Base URL of the site")
    parser.add_argument("--posts-dir", default="posts",
                        help="Directory containing post HTML files (default: posts/)")
    parser.add_argument("--output", default="feed.xml",
                        help="Output file (default: feed.xml)")
    parser.add_argument("--max-items", type=int, default=20,
                        help="Max entries in the feed (default: 20)")
    parser.add_argument("--title", default=None, help="Feed title (default: site URL)")
    parser.add_argument("--description", default="Latest posts",
                        help="Feed description (default: 'Latest posts')")
    args = parser.parse_args()

    if not args.title:
        args.title = args.site_url

    feed = build_feed(args)
    Path(args.output).write_text(feed, encoding="utf-8")
    n = feed.count("<item>")
    print(f"Wrote {n} entries to {args.output}")


if __name__ == "__main__":
    main()