soapbox/soapbox/build.py

"""Statisk generering av bloggen til ``public/``.

Kjøres ved hver lagring i editoren (tar millisekunder for en vanlig blogg).
Utdata er rene filer som kan serveres av hva som helst – appen selv, Caddy
eller en CDN.

Alle lenker i genererte sider er relative. Malene får en ``rel()``-funksjon
som regner ut stien fra *denne* siden til et mål, så ``rel("theme/style.css")``
blir ``theme/style.css`` på forsiden og ``../../theme/style.css`` i et
gjesteinnlegg. Dermed kan hele ``public/`` flyttes til hvilken som helst
sub-path uten rebuild. Unntaket er ``feed.xml``, der RSS krever absolutte
URL-er; den bruker ``SOAPBOX_SITE_URL``.
"""

from __future__ import annotations

import shutil
from datetime import UTC, datetime
from email.utils import format_datetime
from pathlib import Path
from xml.sax.saxutils import escape

from jinja2 import Environment, FileSystemLoader, select_autoescape

from .config import Config, load_site_settings
from .content import IMAGE_EXT, Post, list_posts
from .render import excerpt, relative_to, render_markdown


def _env(cfg: Config) -> Environment:
    env = Environment(
        loader=FileSystemLoader(cfg.theme_dir / "templates"),
        autoescape=select_autoescape(["html"]),
    )
    env.filters["date"] = lambda d, fmt="%d.%m.%Y": d.strftime(fmt)
    return env


def _write(path: Path, content: str) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    path.write_text(content, encoding="utf-8")


def build_site(cfg: Config) -> int:
    """Genererer hele siden. Returnerer antall publiserte innlegg."""
    site = load_site_settings(cfg)
    env = _env(cfg)
    posts = list_posts(cfg)
    out = cfg.public_dir

    # Bygg i en midlertidig katalog og bytt atomisk, så besøkende aldri ser
    # en halvferdig side.
    tmp = out.with_name(out.name + ".tmp")
    if tmp.exists():
        shutil.rmtree(tmp)
    tmp.mkdir(parents=True)

    def render(template: str, page_path: str, **ctx) -> str:
        return env.get_template(template).render(
            site=site,
            page_path=page_path,
            rel=lambda target: relative_to(page_path, target),
            now=datetime.now(UTC),
            **ctx,
        )

    guests: dict[str, list[Post]] = {}
    for p in posts:
        if p.guest:
            guests.setdefault(p.guest, []).append(p)

    # Forside: alle innlegg, også gjestenes, nyeste først
    _write(tmp / "index.html", render("index.html", "", posts=posts, heading=None))

    # Innlegg + bilder
    for p in posts:
        html = render_markdown(p.body, p.url_path)
        _write(tmp / p.url_path / "index.html", render("post.html", p.url_path, post=p, html=html))
        for f in p.path.parent.iterdir():
            if f.suffix.lower() in IMAGE_EXT:
                shutil.copy2(f, tmp / p.url_path / f.name)

    # Én oversiktsside per gjest
    for guest, gposts in guests.items():
        _write(
            tmp / guest / "index.html",
            render("index.html", f"{guest}/", posts=gposts, heading=gposts[0].author or guest),
        )

    _write(tmp / "feed.xml", _feed(cfg, site, posts[:20]))

    theme_static = cfg.theme_dir / "static"
    if theme_static.exists():
        shutil.copytree(theme_static, tmp / "theme")

    if out.exists():
        old = out.with_name(out.name + ".old")
        if old.exists():
            shutil.rmtree(old)
        out.rename(old)
        tmp.rename(out)
        shutil.rmtree(old)
    else:
        tmp.rename(out)
    return len(posts)


def _feed(cfg: Config, site: dict, posts: list[Post]) -> str:
    base = cfg.site_url + "/"
    items = []
    for p in posts:
        link = base + p.url_path
        items.append(
            "<item>"
            f"<title>{escape(p.title)}</title>"
            f"<link>{escape(link)}</link>"
            f"<guid>{escape(link)}</guid>"
            f"<pubDate>{format_datetime(p.date)}</pubDate>"
            + (f"<dc:creator>{escape(p.author)}</dc:creator>" if p.author else "")
            + f"<description>{escape(excerpt(p.body))}</description>"
            "</item>"
        )
    return (
        '<?xml version="1.0" encoding="UTF-8"?>\n'
        '<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" '
        'xmlns:atom="http://www.w3.org/2005/Atom"><channel>'
        f"<title>{escape(site['title'])}</title>"
        f"<link>{escape(base)}</link>"
        f'<atom:link href="{escape(base)}feed.xml" rel="self" type="application/rss+xml"/>'
        f"<description>{escape(site.get('description', ''))}</description>"
        f"<language>{escape(site.get('language', 'nb'))}</language>"
        + "".join(items)
        + "</channel></rss>\n"
    )