john
5f8ef9ddb0
Avoid importing mindspace-public-finish-sync in production runtime so release link repair can copy oa docx files or strip broken anchors without extra deps. Co-authored-by: Cursor <cursoragent@cursor.com>
107 lines
3.7 KiB
JavaScript
Executable File
107 lines
3.7 KiB
JavaScript
Executable File
#!/usr/bin/env node
|
|
/**
|
|
* Repair missing public/*.docx download targets before release link checks.
|
|
* Runtime-safe: no imports from the full MindSpace finish-sync graph.
|
|
*/
|
|
import fs from 'node:fs';
|
|
import path from 'node:path';
|
|
import { fileURLToPath } from 'node:url';
|
|
|
|
const DOCX_HREF_RE = /<a\b[^>]*\bhref=(["'])([^"']+\.docx)\1[^>]*>[\s\S]*?<\/a>/gi;
|
|
const DOCX_HREF_CAPTURE_RE = /<a\b[^>]*\bhref=(["'])([^"']+\.docx)\1/gi;
|
|
|
|
function listDocxHrefBasenames(html) {
|
|
const basenames = new Set();
|
|
let match = DOCX_HREF_CAPTURE_RE.exec(html);
|
|
while (match) {
|
|
const base = path.basename(String(match[2] ?? '')).trim();
|
|
if (base) basenames.add(base);
|
|
match = DOCX_HREF_CAPTURE_RE.exec(html);
|
|
}
|
|
DOCX_HREF_CAPTURE_RE.lastIndex = 0;
|
|
return [...basenames];
|
|
}
|
|
|
|
function listMissingDocxTargets(html, publicDir) {
|
|
return listDocxHrefBasenames(html).filter((basename) => {
|
|
const docxPath = path.join(publicDir, basename);
|
|
return !fs.existsSync(docxPath);
|
|
});
|
|
}
|
|
|
|
function stripBrokenDocxAnchors(html, missingBasenames) {
|
|
const missing = new Set(missingBasenames.map((item) => String(item ?? '').toLowerCase()));
|
|
if (!missing.size) return { html, changed: false };
|
|
let changed = false;
|
|
const next = html.replace(DOCX_HREF_RE, (match, _quote, hrefValue) => {
|
|
const base = path.basename(String(hrefValue ?? '')).toLowerCase();
|
|
if (!missing.has(base)) return match;
|
|
changed = true;
|
|
return '';
|
|
});
|
|
return { html: next, changed };
|
|
}
|
|
|
|
function tryCopyDocxFromOa(userRoot, publicDir, basename) {
|
|
const destination = path.join(publicDir, basename);
|
|
if (fs.existsSync(destination)) return false;
|
|
const source = path.join(userRoot, 'oa', basename);
|
|
if (!fs.existsSync(source) || !fs.statSync(source).isFile()) return false;
|
|
fs.mkdirSync(path.dirname(destination), { recursive: true });
|
|
fs.copyFileSync(source, destination);
|
|
return true;
|
|
}
|
|
|
|
export function repairMindspacePublicDownloads({ publishDir } = {}) {
|
|
const root = path.resolve(String(publishDir ?? ''));
|
|
if (!root || !fs.existsSync(root)) {
|
|
return { synced: [], stripped: [] };
|
|
}
|
|
|
|
const synced = [];
|
|
const stripped = [];
|
|
|
|
for (const userDir of fs.readdirSync(root, { withFileTypes: true })) {
|
|
if (!userDir.isDirectory()) continue;
|
|
const userRoot = path.join(root, userDir.name);
|
|
const publicDir = path.join(userRoot, 'public');
|
|
if (!fs.existsSync(publicDir)) continue;
|
|
for (const entry of fs.readdirSync(publicDir, { withFileTypes: true })) {
|
|
if (!entry.isFile() || !entry.name.endsWith('.html')) continue;
|
|
const htmlPath = path.join(publicDir, entry.name);
|
|
const html = fs.readFileSync(htmlPath, 'utf8');
|
|
for (const basename of listDocxHrefBasenames(html)) {
|
|
if (tryCopyDocxFromOa(userRoot, publicDir, basename)) {
|
|
synced.push(path.join(userDir.name, 'public', basename));
|
|
}
|
|
}
|
|
const missingBasenames = listMissingDocxTargets(html, publicDir);
|
|
const { html: nextHtml, changed } = stripBrokenDocxAnchors(html, missingBasenames);
|
|
if (!changed) continue;
|
|
fs.writeFileSync(htmlPath, nextHtml, 'utf8');
|
|
stripped.push(path.relative(root, htmlPath));
|
|
}
|
|
}
|
|
|
|
return { synced, stripped };
|
|
}
|
|
|
|
function main() {
|
|
const args = process.argv.slice(2);
|
|
let publishDir = path.join(path.dirname(fileURLToPath(import.meta.url)), '..', 'MindSpace');
|
|
for (let i = 0; i < args.length; i += 1) {
|
|
if (args[i] === '--root' && args[i + 1]) {
|
|
publishDir = args[i + 1];
|
|
i += 1;
|
|
}
|
|
}
|
|
const result = repairMindspacePublicDownloads({ publishDir });
|
|
console.log(JSON.stringify(result, null, 2));
|
|
}
|
|
|
|
const isMain = process.argv[1]
|
|
&& fileURLToPath(import.meta.url) === path.resolve(process.argv[1]);
|
|
if (isMain) {
|
|
main();
|
|
}
|