Files
OpenFrame/tests/unit/lib/subtitle-validation.test.ts
T
yusufipek a709ca8544 fix(subtitles): escape a rejected cue tag instead of deleting it
Deleting a tag whole is what lets a filter like this be reassembled around: strip the `<b>` out of `<scr<b>ipt>`
and the two halves close up into a tag nobody wrote. The leftovers are escaped one character at a time instead,
which also covers `-->` in cue text without a second multi-character replacement.

Both are what CodeQL flagged on the branch, js/incomplete-multi-character-sanitization and js/bad-tag-filter.
Neither was reachable as an injection, because the file is served as text/vtt and a cue is parsed by the WebVTT
cue-text parser rather than as HTML, but a sanitiser that cannot be reassembled around is the cheaper thing to own.
2026-08-22 08:01:12 +03:00

212 lines
6.9 KiB
TypeScript
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
import { describe, expect, it } from 'vitest';
import {
decodeSubtitleBuffer,
getSubtitleExtension,
MAX_SUBTITLE_CUES,
normalizeSubtitleFile,
normalizeSubtitleLanguage,
parseSubtitleCues,
sanitizeSubtitleLabel,
SAFE_SUBTITLE_PROXY_PATH,
serializeWebVtt,
subtitleProxyPathToObjectKey,
} from '@/lib/subtitle-validation';
const UUID = '11111111-2222-3333-4444-555555555555';
function encode(text: string): Uint8Array {
return new TextEncoder().encode(text);
}
describe('getSubtitleExtension', () => {
it('accepts the two subtitle formats and nothing else', () => {
expect(getSubtitleExtension('cut.srt')).toBe('srt');
expect(getSubtitleExtension('cut.VTT')).toBe('vtt');
expect(getSubtitleExtension('cut.ass')).toBeNull();
expect(getSubtitleExtension('cut.srt.exe')).toBeNull();
});
});
describe('normalizeSubtitleLanguage', () => {
it('lowercases so a re-upload replaces the track it means to', () => {
expect(normalizeSubtitleLanguage('TR')).toBe('tr');
expect(normalizeSubtitleLanguage(' en-US ')).toBe('en-us');
expect(normalizeSubtitleLanguage('zh-Hant-TW')).toBe('zh-hant-tw');
});
it('rejects anything that is not a language tag', () => {
expect(normalizeSubtitleLanguage('')).toBeNull();
expect(normalizeSubtitleLanguage('t')).toBeNull();
expect(normalizeSubtitleLanguage('tr; drop table')).toBeNull();
expect(normalizeSubtitleLanguage('<script>')).toBeNull();
expect(normalizeSubtitleLanguage(42)).toBeNull();
});
});
describe('sanitizeSubtitleLabel', () => {
it('falls back when the label is empty after cleaning', () => {
expect(sanitizeSubtitleLabel(' ', 'TR')).toBe('TR');
expect(sanitizeSubtitleLabel(undefined, 'TR')).toBe('TR');
});
it('strips control characters and collapses whitespace', () => {
expect(sanitizeSubtitleLabel('Türk\u0000\n çe ', 'TR')).toBe('Türk çe');
});
it('caps the length', () => {
expect(sanitizeSubtitleLabel('a'.repeat(200), 'TR')).toHaveLength(60);
});
});
describe('decodeSubtitleBuffer', () => {
it('reads UTF-8 and drops the byte order mark', () => {
expect(decodeSubtitleBuffer(encode('\uFEFFmerhaba'))).toBe('merhaba');
});
it('falls back to a legacy codepage rather than rejecting the file', () => {
// 0xFD is "ı" in windows-1254 and not valid UTF-8 on its own.
const decoded = decodeSubtitleBuffer(new Uint8Array([0x61, 0xfd, 0x62]));
expect(decoded).not.toBeNull();
expect(decoded).toHaveLength(3);
});
});
describe('parseSubtitleCues', () => {
it('parses SRT, comma decimals and sequence numbers included', () => {
const cues = parseSubtitleCues(
[
'1',
'00:00:01,000 --> 00:00:02,500',
'Merhaba',
'',
'2',
'00:00:03,000 --> 00:00:04,000',
'Dünya',
'',
].join('\n')
);
expect(cues).toEqual([
{ start: 1, end: 2.5, text: 'Merhaba' },
{ start: 3, end: 4, text: 'Dünya' },
]);
});
it('parses WebVTT with cue ids, settings and short timestamps', () => {
const cues = parseSubtitleCues(
['WEBVTT', '', 'intro', '00:01.000 --> 00:02.000 align:start position:10%', 'Hello', ''].join(
'\n'
)
);
expect(cues).toEqual([{ start: 1, end: 2, text: 'Hello' }]);
});
it('skips NOTE, STYLE and REGION blocks', () => {
const cues = parseSubtitleCues(
[
'WEBVTT',
'',
'NOTE this is a comment',
'still the comment',
'',
'STYLE',
'::cue { color: red }',
'',
'00:00:01.000 --> 00:00:02.000',
'Kept',
'',
].join('\n')
);
expect(cues).toEqual([{ start: 1, end: 2, text: 'Kept' }]);
});
it('drops cues that end before they start and cues with no text', () => {
const cues = parseSubtitleCues(
[
'00:00:05,000 --> 00:00:02,000',
'Backwards',
'',
'00:00:06,000 --> 00:00:07,000',
'',
'00:00:08,000 --> 00:00:09,000',
'Good',
'',
].join('\n')
);
expect(cues).toEqual([{ start: 8, end: 9, text: 'Good' }]);
});
it('keeps known cue markup and removes everything else', () => {
const cues = parseSubtitleCues(
['00:00:01,000 --> 00:00:02,000', '<i>tilt</i><script>alert(1)</script>{\\an8}', ''].join(
'\n'
)
);
expect(cues[0].text).toBe('<i>tilt</i>alert(1)');
});
it('escapes the leftovers of a rejected tag so it cannot be reassembled', () => {
// Deleting `<b>` out of the middle would close the two halves into a `<script>` that
// was never written. Escaping what is left over is what stops that.
const cues = parseSubtitleCues(
['00:00:01,000 --> 00:00:02,000', '<scr<b>ipt>alert(1)', ''].join('\n')
);
expect(cues[0].text).toBe('&lt;scr<b>ipt&gt;alert(1)');
expect(cues[0].text).not.toContain('<script');
});
it('neutralises an arrow in cue text so the file cannot be re-split', () => {
const cues = parseSubtitleCues(['00:00:01,000 --> 00:00:02,000', 'a --> b', ''].join('\n'));
expect(cues[0].text).toBe('a --&gt; b');
expect(parseSubtitleCues(serializeWebVtt(cues))).toHaveLength(1);
});
it('stops at the cue ceiling', () => {
const lines: string[] = [];
for (let index = 0; index < MAX_SUBTITLE_CUES + 10; index += 1) {
lines.push(`00:00:0${index % 9}.000 --> 00:00:0${(index % 9) + 1}.000`, `line ${index}`, '');
}
expect(parseSubtitleCues(lines.join('\n'))).toHaveLength(MAX_SUBTITLE_CUES);
});
});
describe('normalizeSubtitleFile', () => {
it('converts SRT to a canonical WebVTT document', () => {
const result = normalizeSubtitleFile(
encode('1\r\n00:00:01,500 --> 00:00:02,000\r\nMerhaba\r\n\r\n')
);
expect(result).toEqual({
ok: true,
cueCount: 1,
vtt: 'WEBVTT\n\n00:00:01.500 --> 00:00:02.000\nMerhaba\n',
});
});
it('refuses an empty file', () => {
const result = normalizeSubtitleFile(new Uint8Array());
expect(result.ok).toBe(false);
});
it('refuses a file with no cues rather than storing an empty track', () => {
const result = normalizeSubtitleFile(encode('this is just prose\nand more prose\n'));
expect(result).toEqual({
ok: false,
error: 'No subtitle cues found. Upload a valid .srt or .vtt file.',
});
});
});
describe('subtitle proxy paths', () => {
it('only recognises a uuid .vtt path', () => {
expect(SAFE_SUBTITLE_PROXY_PATH.test(`/api/upload/subtitle/${UUID}.vtt`)).toBe(true);
expect(SAFE_SUBTITLE_PROXY_PATH.test(`/api/upload/subtitle/${UUID}.srt`)).toBe(false);
expect(SAFE_SUBTITLE_PROXY_PATH.test('/api/upload/subtitle/../../etc/passwd')).toBe(false);
});
it('maps a proxy path to its object key and refuses anything else', () => {
expect(subtitleProxyPathToObjectKey(`/api/upload/subtitle/${UUID}.vtt`)).toBe(
`subtitles/${UUID}.vtt`
);
expect(subtitleProxyPathToObjectKey(`/api/upload/image/${UUID}.png`)).toBeNull();
});
});