Goldebek/scripts/check-download-metadata.mjs

import fs from "node:fs";
import path from "node:path";

const DOWNLOADS_DIR = path.resolve(import.meta.dirname, "../public/downloads");
const PDF_METADATA_PATTERN =
  /\/(?:Title|Author|Subject|Keywords|Creator|CreationDate|ModDate)\s*\(|<xmp:(?:CreatorTool|CreateDate|ModifyDate)>|<dc:(?:creator|title)>|<pdf:Keywords>/i;

function collectPdfs(directory) {
  if (!fs.existsSync(directory)) return [];

  return fs
    .readdirSync(directory, { withFileTypes: true })
    .filter((entry) => entry.isFile() && entry.name.toLowerCase().endsWith(".pdf"))
    .map((entry) => path.join(directory, entry.name));
}

const files = collectPdfs(DOWNLOADS_DIR);
const filesWithMetadata = files.filter((file) =>
  PDF_METADATA_PATTERN.test(fs.readFileSync(file, "latin1")),
);

for (const file of filesWithMetadata) {
  console.error(`[downloads:check-metadata] PDF metadata found: ${file}`);
}

if (filesWithMetadata.length > 0) {
  process.exitCode = 1;
} else {
  console.log(`[downloads:check-metadata] ${files.length} PDFs checked; no document metadata found.`);
}