andwati.com/scripts/migrate-posts-to-writings.mjs

// One-off migration: converts content/posts/*.md (legacy Zola posts) into
// content/writings/<slug>/index.md (new schema). Additive only — never
// touches content/posts/. Re-runnable: overwrites its own prior output.
import fs from "node:fs";
import path from "node:path";
import {
  parse as parseToml,
  stringify as stringifyToml,
  TomlDate,
} from "smol-toml";

const root = process.cwd();
const postsDir = path.join(root, "content", "posts");
const writingsDir = path.join(root, "content", "writings");

const FRONTMATTER_RE = /^\+\+\+\r?\n([\s\S]*?)\r?\n\+\+\+\r?\n?([\s\S]*)$/;

function walk(dir) {
  return fs.readdirSync(dir, { withFileTypes: true }).flatMap((entry) => {
    const full = path.join(dir, entry.name);
    if (entry.isDirectory()) return walk(full);
    return entry.isFile() && entry.name.endsWith(".md") ? [full] : [];
  });
}

function slugFor(file) {
  const rel = path.relative(postsDir, file).split(path.sep).join("/");
  return rel.endsWith("/index.md")
    ? rel.slice(0, -"/index.md".length)
    : rel.slice(0, -".md".length);
}

function compact(obj) {
  const out = {};
  for (const [key, value] of Object.entries(obj)) {
    if (value === undefined || value === null || value === "") continue;
    if (Array.isArray(value) && value.length === 0) continue;
    if (
      typeof value === "object" &&
      !Array.isArray(value) &&
      !(value instanceof Date)
    ) {
      const nested = compact(value);
      if (Object.keys(nested).length === 0) continue;
      out[key] = nested;
      continue;
    }
    out[key] = value;
  }
  return out;
}

function toIsoDate(value) {
  if (value instanceof Date)
    return new TomlDate(value.toISOString().slice(0, 10));
  if (typeof value === "string") return new TomlDate(value.slice(0, 10));
  return undefined;
}

/** Converts the one unambiguous Zola shortcode in use (youtube embeds) to a
 * plain HTML iframe, and strips Zola's `{% raw %}`/`{% endraw %}` escape
 * tags (meaningless now — there's no template engine processing the body).
 * Anything else Zola-specific (note/tip/warning/danger callouts) is left
 * as-is: those need real Astro components before they can be translated,
 * so they're flagged for manual follow-up instead of guessed at. */
function migrateBody(body) {
  const flags = [];
  const out = body
    .replace(/\{%\s*raw\s*%\}\r?\n?/g, "")
    .replace(/\{%\s*endraw\s*%\}\r?\n?/g, "")
    .replace(/\{\{<\s*youtube\s+id="([^"]+)"[\s\S]*?\/>\s*\}\}/g, (_m, id) => {
      return `<iframe width="560" height="315" src="https://www.youtube-nocookie.com/embed/${id}" title="YouTube video" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>`;
    });

  if (/\{%\s*<?\/?(note|tip|warning|danger)>?\s*%\}/.test(out)) {
    flags.push("callout-shortcode");
  }
  if (/\{\{|\{%/.test(out)) {
    flags.push("unresolved-template-syntax");
  }
  return { body: out, flags };
}

const files = walk(postsDir).filter((f) => path.basename(f) !== "_index.md");
fs.mkdirSync(writingsDir, { recursive: true });

let migrated = 0;
const flaggedFiles = [];
const migratedSlugs = [];

for (const file of files) {
  const raw = fs.readFileSync(file, "utf8");
  const match = raw.match(FRONTMATTER_RE);
  if (!match) continue;
  const [, fmSource, body] = match;
  const fm = parseToml(fmSource);
  const slug = slugFor(file);

  const { body: newBody, flags } = migrateBody(body.trim());
  if (flags.length) flaggedFiles.push({ slug, flags });

  const frontmatter = stringifyToml(
    compact({
      title: fm.title,
      description: fm.description,
      author: fm.author,
      date: toIsoDate(fm.date),
      updated: toIsoDate(fm.updated),
      draft: fm.draft ?? false,
      canonical_url: fm.canonical_url,
      taxonomies: { tags: fm.taxonomies?.tags ?? [] },
      extra: {
        series: fm.extra?.series,
        series_index: fm.extra?.series_index,
        llms_description: fm.extra?.llms_description,
      },
    }),
  );

  const outDir = path.join(writingsDir, slug);
  fs.mkdirSync(outDir, { recursive: true });
  fs.writeFileSync(
    path.join(outDir, "index.md"),
    `+++\n${frontmatter}\n+++\n\n${newBody}\n`,
  );
  migrated += 1;
  migratedSlugs.push(slug);
}

// Keep old /posts/<slug>/ URLs alive (RSS subscribers, inbound links) by
// redirecting to the new /writings/<slug>/ path. Regenerated idempotently
// inside marker comments so re-running this script doesn't duplicate or
// clobber the rest of sws.toml.
const swsPath = path.join(root, "sws.toml");
const BEGIN =
  "# BEGIN migrated-posts-redirects (generated by scripts/migrate-posts-to-writings.mjs)";
const END = "# END migrated-posts-redirects";
const redirectBlock = [
  BEGIN,
  ...migratedSlugs
    .sort()
    .flatMap((slug) => [
      "[[advanced.redirects]]",
      `source = "/posts/${slug}/"`,
      `destination = "/writings/${slug}/"`,
      "kind = 301",
      "",
    ]),
  END,
]
  .join("\n")
  .trimEnd();

function escapeRegExp(s) {
  return s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
}

const swsSource = fs.readFileSync(swsPath, "utf8");
const markerRe = new RegExp(
  `${escapeRegExp(BEGIN)}[\\s\\S]*?${escapeRegExp(END)}`,
);
const updatedSws = markerRe.test(swsSource)
  ? swsSource.replace(markerRe, redirectBlock)
  : `${swsSource.trimEnd()}\n\n${redirectBlock}\n`;
fs.writeFileSync(swsPath, `${updatedSws.trimEnd()}\n`);

console.log(
  `Migrated ${migrated} post(s) from content/posts/ to content/writings/.`,
);
console.log(`Wrote ${migratedSlugs.length} redirect(s) to sws.toml.`);
if (flaggedFiles.length) {
  console.log(
    "\nNeeds manual follow-up (Zola-specific syntax with no Astro equivalent yet):",
  );
  for (const { slug, flags } of flaggedFiles) {
    console.log(`  - ${slug}: ${flags.join(", ")}`);
  }
}