bf4f895f2c
Some checks failed
Forgejo Android APK / Root app tests (push) Successful in 47s
Forgejo Docker Build / Root app tests (push) Successful in 49s
Forgejo Android APK / Build signed APK (push) Successful in 1m55s
Forgejo Docker Build / Build Docker image (push) Successful in 9s
Forgejo Docker Build / Deploy to the host (push) Failing after 0s
The ceiling was 100 MB per file with ten files allowed at once, and every file is held whole in memory to be parsed — so the old limit let a single request ask for a gigabyte of heap. A source article that size is not a thing anyone uploads here. Now 10 MB, defined once and used by both the multer limit and the post-upload check. The filter accepted `allowed mime OR allowed extension`, so naming a file .pdf was enough on its own, whatever it declared — and the extension is chosen by whoever uploads. Both are required now. Neither of those sees any bytes: multer filters on the headers, before the file has arrived. verifySources() runs once the buffer exists and refuses a file whose contents are not what its type claims, using the same helper as documents, S3 uploads and assistant attachments. It runs before extraction, because an extractor handed a malformed file is where the damage would happen. The CMS screen said 100 MB and listed four of the ten accepted formats; it now says what the server actually does. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Dv6sqaY6Vq3ChZHMem3cnU
42 lines
2.1 KiB
JavaScript
42 lines
2.1 KiB
JavaScript
// Source articles for AI generation are held whole in memory to be parsed, so
|
|
// the ceiling matters, and the declared type is chosen by whoever uploads.
|
|
const test = require('node:test');
|
|
const assert = require('node:assert/strict');
|
|
const fs = require('node:fs');
|
|
const path = require('node:path');
|
|
|
|
const route = fs.readFileSync(path.join(__dirname, '..', 'src/routes/learningAI.js'), 'utf8');
|
|
|
|
test('the per-file ceiling is 10 MB, in one place', () => {
|
|
assert.match(route, /var MAX_SOURCE_BYTES = 10 \* 1024 \* 1024;/);
|
|
assert.match(route, /fileSize: MAX_SOURCE_BYTES/);
|
|
assert.doesNotMatch(route, /100 \* 1024 \* 1024/, 'the old 100 MB ceiling is gone');
|
|
});
|
|
|
|
test('the type and the extension must both be allowed, not either', () => {
|
|
// It was `mime OR extension`, so naming a file .pdf was enough on its own.
|
|
assert.match(route, /ALLOWED_SOURCE_TYPES\.includes\(file\.mimetype\) && ALLOWED_SOURCE_EXTENSIONS\.test\(file\.originalname\)/);
|
|
assert.doesNotMatch(route, /allowed\.includes\(file\.mimetype\) \|\| file\.originalname\.match/);
|
|
});
|
|
|
|
test('the bytes are sniffed once the buffer exists, before anything parses them', () => {
|
|
// multer filters on headers alone, before any byte has arrived.
|
|
assert.match(route, /function verifySources\(files\)/);
|
|
assert.match(route, /fileType\.matches\(file\.mimetype, file\.buffer\)/);
|
|
assert.match(route, /is not the file type it claims to be/);
|
|
// And it runs ahead of extraction, not after.
|
|
const call = route.indexOf('verifySources(req.files)');
|
|
const extract = route.indexOf('await extractText(');
|
|
assert.ok(call > -1 && call < extract, 'verification must precede extraction');
|
|
});
|
|
|
|
test('the size is re-checked on the buffer, not trusted from the header', () => {
|
|
assert.match(route, /file\.size > MAX_SOURCE_BYTES/);
|
|
assert.match(route, /is larger than 10 MB/);
|
|
});
|
|
|
|
test('what the screen promises matches what the server accepts', () => {
|
|
const cms = fs.readFileSync(path.join(__dirname, '..', 'public/components/cms.html'), 'utf8');
|
|
assert.match(cms, /max 10 MB each, up to 10 files/);
|
|
assert.doesNotMatch(cms, /100 MB/);
|
|
});
|