safe_fetch.js validates URLs before fetch: rejects non-http(s), literal or DNS-resolved loopback / RFC1918 / link-local / CGNAT / metadata addresses; follows redirects manually with the same checks on each hop. Test fixtures gate the check with VOID_INGEST_ALLOW_PRIVATE for offline fixtures that hit 127.0.0.1. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
45 lines
1.3 KiB
JavaScript
45 lines
1.3 KiB
JavaScript
import crypto from 'node:crypto';
|
|
import { extract } from '../../ingest/readability.js';
|
|
import { safeFetch } from '../../ingest/safe_fetch.js';
|
|
import * as refs from '../../db/repos/refs.js';
|
|
import { pool } from '../../db/pool.js';
|
|
|
|
export const NAME = 'ingest.url';
|
|
|
|
function key(space_id, url) {
|
|
return crypto.createHash('sha256').update(space_id + '\x00' + url).digest('hex');
|
|
}
|
|
|
|
export async function handler(job) {
|
|
const { space_id, url } = job.data;
|
|
const idem = key(space_id, url);
|
|
|
|
const { rows: [existing] } = await pool.query(
|
|
`SELECT id FROM refs WHERE source_kind='url' AND external_id=$1 LIMIT 1`,
|
|
[idem]
|
|
);
|
|
if (existing) return { ref_id: existing.id, idempotent: true };
|
|
|
|
const res = await safeFetch(url, {
|
|
headers: { 'User-Agent': 'void-ingest/2.0' },
|
|
signal: AbortSignal.timeout(15_000)
|
|
});
|
|
if (!res.ok) throw new Error(`fetch ${url} → ${res.status}`);
|
|
const html = await res.text();
|
|
|
|
const parsed = extract(html, url);
|
|
const row = await refs.create({
|
|
space_id,
|
|
kind: 'url',
|
|
source_url: url,
|
|
title: parsed.title || url,
|
|
summary: parsed.excerpt,
|
|
body_text: (parsed.textContent || '').slice(0, 200_000),
|
|
source_kind: 'url',
|
|
external_id: idem,
|
|
metadata: { site_name: parsed.siteName, byline: parsed.byline }
|
|
}, { kind: 'system', id: null });
|
|
|
|
return { ref_id: row.id };
|
|
}
|