Files
rose-ash/tests/playwright/hs-behavioral.spec.js
giles 737964be89 Honest test suite: 424/831 (51%) — all tests run, timeouts fail visibly
Rewrote test architecture: deferred execution. Tests register thunks during
file load (try-call redefined to append to _test-registry), then the
Playwright loop runs each individually with 3s timeout via Promise.race.
Hanging tests (parser infinite loops) fail with TIMEOUT and trigger page
reboot. No tests are hidden or skipped.

Fixed generator: proper quote escaping for HS sources with embedded quotes,
sanitized comments to avoid SX parser special chars.

831 tests registered, 424 pass, 407 fail honestly:
- 22 perfect categories (empty, dialog, morph, default, reset, scroll, etc.)
- Major gaps: if 0/19, wait 0/7, take 0/12, repeat 2/30, set 4/25
- Timeout failures from parser hangs on unsupported syntax

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-10 07:54:01 +00:00

287 lines
11 KiB
JavaScript

// @ts-check
/**
* Hyperscript behavioral tests — SX tests running in Playwright sandbox.
*
* Architecture:
* - SX test file registers test thunks via deftest-deferred (doesn't run them)
* - This spec iterates over registered tests, running each with a 3s timeout
* - Hanging tests (parser infinite loops) fail with TIMEOUT, page is rebooted
* - Results are reported per-category with a hard gate on pass count
*/
const { test, expect } = require('playwright/test');
const fs = require('fs');
const path = require('path');
const PROJECT_ROOT = path.resolve(__dirname, '../..');
const WASM_DIR = path.join(PROJECT_ROOT, 'shared/static/wasm');
const SX_DIR = path.join(WASM_DIR, 'sx');
const SANDBOX_STACKS = {
web: [
'render', 'core-signals', 'signals', 'deps', 'router',
'page-helpers', 'freeze', 'dom', 'browser',
'adapter-html', 'adapter-sx', 'adapter-dom',
'boot-helpers', 'hypersx', 'engine', 'orchestration', 'boot',
],
hs: [
'hs-tokenizer', 'hs-parser', 'hs-compiler', 'hs-runtime', 'hs-integration',
],
};
/**
* Boot WASM kernel with hs stack. Returns loadErrors array.
*/
async function bootSandbox(page) {
await page.goto('about:blank');
const kernelSrc = fs.readFileSync(path.join(WASM_DIR, 'sx_browser.bc.js'), 'utf8');
await page.addScriptTag({ content: kernelSrc });
await page.waitForFunction('!!window.SxKernel', { timeout: 10000 });
// Register FFI + IO driver (compact)
await page.evaluate(() => {
const K = window.SxKernel;
K.registerNative('host-global', a => { const n=a[0]; return (n in globalThis)?globalThis[n]:null; });
K.registerNative('host-get', a => { if(a[0]==null)return null; const v=a[0][a[1]]; return v===undefined?null:v; });
K.registerNative('host-set!', a => { if(a[0]!=null)a[0][a[1]]=a[2]; return a[2]; });
K.registerNative('host-call', a => {
const[o,m,...r]=a;
if(o==null){const f=globalThis[m];return typeof f==='function'?f.apply(null,r):null;}
if(typeof o[m]!=='function')return null;
try{const v=o[m].apply(o,r);return v===undefined?null:v;}catch(e){return null;}
});
K.registerNative('host-new', a => {
const C=typeof a[0]==='string'?globalThis[a[0]]:a[0];
return typeof C==='function'?new C(...a.slice(1)):null;
});
K.registerNative('host-callback', a => {
const fn=a[0];
if(typeof fn==='function'&&fn.__sx_handle===undefined)return fn;
if(fn&&fn.__sx_handle!==undefined){
return function(){
const r=K.callFn(fn,Array.from(arguments));
if(window._driveAsync)window._driveAsync(r);
return r;
};
}
return function(){};
});
K.registerNative('host-typeof', a => {
const o=a[0]; if(o==null)return'nil';
if(o instanceof Element)return'element'; if(o instanceof Text)return'text';
if(o instanceof DocumentFragment)return'fragment'; if(o instanceof Document)return'document';
if(o instanceof Event)return'event'; if(o instanceof Promise)return'promise';
return typeof o;
});
K.registerNative('host-await', a => {
const[p,cb]=a;
if(p&&typeof p.then==='function'){
const f=(cb&&cb.__sx_handle!==undefined)?v=>K.callFn(cb,[v]):()=>{};
p.then(f);
}
});
K.registerNative('load-library!', () => false);
// IO suspension driver
window._driveAsync = function driveAsync(result) {
if(!result||!result.suspended)return;
const req=result.request;const items=req&&(req.items||req);
const op=items&&items[0];const opName=typeof op==='string'?op:(op&&op.name)||String(op);
const arg=items&&items[1];
function doResume(val,delay){
setTimeout(()=>{try{const r=result.resume(val);driveAsync(r);}catch(e){}},delay);
}
if(opName==='io-sleep'||opName==='wait')doResume(null,Math.min(typeof arg==='number'?arg:0,10));
else if(opName==='io-navigate'){}
else if(opName==='io-fetch')doResume({ok:true,text:''},1);
else{}
};
K.eval('(define SX_VERSION "hs-test-1.0")');
K.eval('(define SX_ENGINE "ocaml-vm-sandbox")');
K.eval('(define parse sx-parse)');
K.eval('(define serialize sx-serialize)');
});
// Load web + hs modules
const allModules = [...SANDBOX_STACKS.web, ...SANDBOX_STACKS.hs];
const loadErrors = [];
await page.evaluate(() => { if (window.SxKernel.beginModuleLoad) window.SxKernel.beginModuleLoad(); });
for (const mod of allModules) {
const sxPath = path.join(SX_DIR, mod + '.sx');
const libPath = path.join(PROJECT_ROOT, 'lib/hyperscript', mod.replace(/^hs-/, '') + '.sx');
let src;
try { src = fs.existsSync(sxPath) ? fs.readFileSync(sxPath, 'utf8') : fs.readFileSync(libPath, 'utf8'); }
catch(e) { loadErrors.push(mod + ': file not found'); continue; }
const err = await page.evaluate(s => {
try { window.SxKernel.load(s); return null; }
catch(e) { return e.message; }
}, src);
if (err) loadErrors.push(mod + ': ' + err);
}
await page.evaluate(() => { if (window.SxKernel.endModuleLoad) window.SxKernel.endModuleLoad(); });
// Define deferred test infrastructure
await page.evaluate(() => {
const K = window.SxKernel;
// _test_registry: list of {suite, name, thunk}
K.eval('(define _test-registry (list))');
K.eval('(define _test-suite "")');
// deftest-deferred: register thunk, don't run it
// The SX file uses standard defsuite/deftest but we redefine them to defer
K.eval(`(define push-suite (fn (name) (set! _test-suite name)))`);
K.eval(`(define pop-suite (fn () (set! _test-suite "")))`);
// try-call just runs the thunk — no error handling needed since we defer
K.eval(`(define try-call (fn (thunk)
(set! _test-registry
(append _test-registry
(list {:suite _test-suite :thunk thunk})))
{:ok true}))`);
// report-pass/report-fail record the NAME for the just-registered thunk
// Since try-call always returns {:ok true}, report-pass is always called.
// We patch the last entry with the name.
K.eval(`(define report-pass (fn (name)
(let ((last-idx (- (len _test-registry) 1)))
(when (>= last-idx 0)
(let ((entry (nth _test-registry last-idx)))
(dict-set! entry "name" name))))))`);
K.eval(`(define report-fail (fn (name error)
(let ((last-idx (- (len _test-registry) 1)))
(when (>= last-idx 0)
(let ((entry (nth _test-registry last-idx)))
(dict-set! entry "name" name))))))`);
});
// Load harness (for assert/assert=) + test framework + behavioral tests
for (const f of ['spec/harness.sx', 'spec/tests/test-framework.sx', 'spec/tests/test-hyperscript-behavioral.sx']) {
const src = fs.readFileSync(path.join(PROJECT_ROOT, f), 'utf8');
const err = await page.evaluate(s => {
try { window.SxKernel.load(s); return null; }
catch(e) { return 'LOAD: ' + e.message; }
}, src);
if (err) loadErrors.push(f + ': ' + err);
}
return loadErrors;
}
/**
* Get list of registered tests: [{suite, name, index}]
*/
async function getTestList(page) {
return page.evaluate(() => {
const K = window.SxKernel;
const count = K.eval('(len _test-registry)');
const tests = [];
for (let i = 0; i < count; i++) {
const suite = K.eval(`(get (nth _test-registry ${i}) "suite")`) || '';
const name = K.eval(`(get (nth _test-registry ${i}) "name")`) || `test-${i}`;
tests.push({ suite, name, index: i });
}
return tests;
});
}
/**
* Run a single test by index. Returns {pass, error}.
*/
async function runTest(page, index) {
return page.evaluate(idx => {
const K = window.SxKernel;
// Clean DOM between tests
const body = document.querySelector('body');
if (body) body.innerHTML = '';
const entry = K.eval(`(nth _test-registry ${idx})`);
const thunk = K.eval(`(get (nth _test-registry ${idx}) "thunk")`);
if (!thunk) return { pass: false, error: 'no thunk registered' };
let lastError = null;
const origError = console.error;
console.error = function() {
const msg = Array.from(arguments).join(' ');
if (msg.startsWith('[sx]')) lastError = msg;
origError.apply(console, arguments);
};
try {
K.callFn(thunk, []);
console.error = origError;
if (lastError) return { pass: false, error: lastError.slice(0, 200) };
return { pass: true, error: null };
} catch(e) {
console.error = origError;
return { pass: false, error: (e.message || String(e)).slice(0, 200) };
}
}, index);
}
// ===========================================================================
// Test suite
// ===========================================================================
test.describe('Hyperscript behavioral tests', () => {
test.describe.configure({ timeout: 600000 }); // 10 min for 831 tests
test('upstream conformance', async ({ browser }) => {
let page = await browser.newPage();
const loadErrors = await bootSandbox(page);
expect(loadErrors).toEqual([]);
const testList = await getTestList(page);
console.log(`\n Registered: ${testList.length} tests`);
// Run each test with a 3s timeout — hang = fail + page reboot
const results = [];
for (const t of testList) {
let result;
try {
result = await Promise.race([
runTest(page, t.index),
new Promise(resolve =>
setTimeout(() => resolve({ pass: false, error: 'TIMEOUT: test hung (>3s)' }), 3000))
]);
} catch(e) {
result = { pass: false, error: 'CRASH: ' + (e.message || '').slice(0, 100) };
}
// If test timed out, the page is stuck — reboot and re-register
if (result.error && (result.error.startsWith('TIMEOUT') || result.error.startsWith('CRASH'))) {
await page.close().catch(() => {});
page = await browser.newPage();
const rebootErrors = await bootSandbox(page);
if (rebootErrors.length > 0) {
// Can't recover — mark remaining tests as failed
for (let j = testList.indexOf(t) + 1; j < testList.length; j++) {
results.push({ suite: testList[j].suite, name: testList[j].name, pass: false, error: 'SKIPPED: page reboot failed' });
}
break;
}
}
results.push({ suite: t.suite, name: t.name, pass: result.pass, error: result.error });
}
await page.close();
// Tally
let passed = 0, failed = 0;
const cats = {};
for (const r of results) {
if (r.pass) passed++; else failed++;
if (!cats[r.suite]) cats[r.suite] = { p: 0, f: 0 };
if (r.pass) cats[r.suite].p++; else cats[r.suite].f++;
}
console.log(`\n Upstream conformance: ${passed}/${results.length} (${(100*passed/results.length).toFixed(0)}%)`);
for (const [cat, s] of Object.entries(cats).sort((a,b) => b[1].p - a[1].p)) {
const mark = s.f === 0 ? `${s.p}` : `${s.p}/${s.p+s.f}`;
console.log(` ${cat}: ${mark}`);
}
// Hard gate
expect(results.length).toBeGreaterThan(0);
expect(passed).toBeGreaterThanOrEqual(420);
});
});