Skip to content

Commit e9fc0e6

Browse files
cjihrigtargos
authored andcommitted
test: add initial test module
This commit adds a new 'test' module that exposes an API for creating JavaScript tests. As the tests execute, TAP output is written to standard output. This commit only supports executing individual test files, and does not implement command line functionality for a full test runner. PR-URL: nodejs#42325 Refs: nodejs#40954 Reviewed-By: Matteo Collina <[email protected]> Reviewed-By: Antoine du Hamel <[email protected]>
1 parent 92051cb commit e9fc0e6

17 files changed

+1922
-0
lines changed

doc/api/errors.md

+8
Original file line numberDiff line numberDiff line change
@@ -2548,6 +2548,14 @@ An unspecified or non-specific system error has occurred within the Node.js
25482548
process. The error object will have an `err.info` object property with
25492549
additional details.
25502550

2551+
<a id="ERR_TEST_FAILURE"></a>
2552+
2553+
### `ERR_TEST_FAILURE`
2554+
2555+
This error represents a failed test. Additional information about the failure
2556+
is available via the `cause` property. The `failureType` property specifies
2557+
what the test was doing when the failure occurred.
2558+
25512559
<a id="ERR_TLS_CERT_ALTNAME_FORMAT"></a>
25522560

25532561
### `ERR_TLS_CERT_ALTNAME_FORMAT`

doc/api/index.md

+1
Original file line numberDiff line numberDiff line change
@@ -54,6 +54,7 @@
5454
* [Report](report.md)
5555
* [Stream](stream.md)
5656
* [String decoder](string_decoder.md)
57+
* [Test runner](test.md)
5758
* [Timers](timers.md)
5859
* [TLS/SSL](tls.md)
5960
* [Trace events](tracing.md)

doc/api/test.md

+316
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,316 @@
1+
# Test runner
2+
3+
<!--introduced_in=REPLACEME-->
4+
5+
> Stability: 1 - Experimental
6+
7+
<!-- source_link=lib/test.js -->
8+
9+
The `node:test` module facilitates the creation of JavaScript tests that
10+
report results in [TAP][] format. To access it:
11+
12+
```mjs
13+
import test from 'node:test';
14+
```
15+
16+
```cjs
17+
const test = require('node:test');
18+
```
19+
20+
This module is only available under the `node:` scheme. The following will not
21+
work:
22+
23+
```mjs
24+
import test from 'test';
25+
```
26+
27+
```cjs
28+
const test = require('test');
29+
```
30+
31+
Tests created via the `test` module consist of a single function that is
32+
processed in one of three ways:
33+
34+
1. A synchronous function that is considered failing if it throws an exception,
35+
and is considered passing otherwise.
36+
2. A function that returns a `Promise` that is considered failing if the
37+
`Promise` rejects, and is considered passing if the `Promise` resolves.
38+
3. A function that receives a callback function. If the callback receives any
39+
truthy value as its first argument, the test is considered failing. If a
40+
falsy value is passed as the first argument to the callback, the test is
41+
considered passing. If the test function receives a callback function and
42+
also returns a `Promise`, the test will fail.
43+
44+
The following example illustrates how tests are written using the
45+
`test` module.
46+
47+
```js
48+
test('synchronous passing test', (t) => {
49+
// This test passes because it does not throw an exception.
50+
assert.strictEqual(1, 1);
51+
});
52+
53+
test('synchronous failing test', (t) => {
54+
// This test fails because it throws an exception.
55+
assert.strictEqual(1, 2);
56+
});
57+
58+
test('asynchronous passing test', async (t) => {
59+
// This test passes because the Promise returned by the async
60+
// function is not rejected.
61+
assert.strictEqual(1, 1);
62+
});
63+
64+
test('asynchronous failing test', async (t) => {
65+
// This test fails because the Promise returned by the async
66+
// function is rejected.
67+
assert.strictEqual(1, 2);
68+
});
69+
70+
test('failing test using Promises', (t) => {
71+
// Promises can be used directly as well.
72+
return new Promise((resolve, reject) => {
73+
setImmediate(() => {
74+
reject(new Error('this will cause the test to fail'));
75+
});
76+
});
77+
});
78+
79+
test('callback passing test', (t, done) => {
80+
// done() is the callback function. When the setImmediate() runs, it invokes
81+
// done() with no arguments.
82+
setImmediate(done);
83+
});
84+
85+
test('callback failing test', (t, done) => {
86+
// When the setImmediate() runs, done() is invoked with an Error object and
87+
// the test fails.
88+
setImmediate(() => {
89+
done(new Error('callback failure'));
90+
});
91+
});
92+
```
93+
94+
As a test file executes, TAP is written to the standard output of the Node.js
95+
process. This output can be interpreted by any test harness that understands
96+
the TAP format. If any tests fail, the process exit code is set to `1`.
97+
98+
## Subtests
99+
100+
The test context's `test()` method allows subtests to be created. This method
101+
behaves identically to the top level `test()` function. The following example
102+
demonstrates the creation of a top level test with two subtests.
103+
104+
```js
105+
test('top level test', async (t) => {
106+
await t.test('subtest 1', (t) => {
107+
assert.strictEqual(1, 1);
108+
});
109+
110+
await t.test('subtest 2', (t) => {
111+
assert.strictEqual(2, 2);
112+
});
113+
});
114+
```
115+
116+
In this example, `await` is used to ensure that both subtests have completed.
117+
This is necessary because parent tests do not wait for their subtests to
118+
complete. Any subtests that are still outstanding when their parent finishes
119+
are cancelled and treated as failures. Any subtest failures cause the parent
120+
test to fail.
121+
122+
## Skipping tests
123+
124+
Individual tests can be skipped by passing the `skip` option to the test, or by
125+
calling the test context's `skip()` method. Both of these options support
126+
including a message that is displayed in the TAP output as shown in the
127+
following example.
128+
129+
```js
130+
// The skip option is used, but no message is provided.
131+
test('skip option', { skip: true }, (t) => {
132+
// This code is never executed.
133+
});
134+
135+
// The skip option is used, and a message is provided.
136+
test('skip option with message', { skip: 'this is skipped' }, (t) => {
137+
// This code is never executed.
138+
});
139+
140+
test('skip() method', (t) => {
141+
// Make sure to return here as well if the test contains additional logic.
142+
t.skip();
143+
});
144+
145+
test('skip() method with message', (t) => {
146+
// Make sure to return here as well if the test contains additional logic.
147+
t.skip('this is skipped');
148+
});
149+
```
150+
151+
## Extraneous asynchronous activity
152+
153+
Once a test function finishes executing, the TAP results are output as quickly
154+
as possible while maintaining the order of the tests. However, it is possible
155+
for the test function to generate asynchronous activity that outlives the test
156+
itself. The test runner handles this type of activity, but does not delay the
157+
reporting of test results in order to accommodate it.
158+
159+
In the following example, a test completes with two `setImmediate()`
160+
operations still outstanding. The first `setImmediate()` attempts to create a
161+
new subtest. Because the parent test has already finished and output its
162+
results, the new subtest is immediately marked as failed, and reported in the
163+
top level of the file's TAP output.
164+
165+
The second `setImmediate()` creates an `uncaughtException` event.
166+
`uncaughtException` and `unhandledRejection` events originating from a completed
167+
test are handled by the `test` module and reported as diagnostic warnings in
168+
the top level of the file's TAP output.
169+
170+
```js
171+
test('a test that creates asynchronous activity', (t) => {
172+
setImmediate(() => {
173+
t.test('subtest that is created too late', (t) => {
174+
throw new Error('error1');
175+
});
176+
});
177+
178+
setImmediate(() => {
179+
throw new Error('error2');
180+
});
181+
182+
// The test finishes after this line.
183+
});
184+
```
185+
186+
## `test([name][, options][, fn])`
187+
188+
<!-- YAML
189+
added: REPLACEME
190+
-->
191+
192+
* `name` {string} The name of the test, which is displayed when reporting test
193+
results. **Default:** The `name` property of `fn`, or `'<anonymous>'` if `fn`
194+
does not have a name.
195+
* `options` {Object} Configuration options for the test. The following
196+
properties are supported:
197+
* `concurrency` {number} The number of tests that can be run at the same time.
198+
If unspecified, subtests inherit this value from their parent.
199+
**Default:** `1`.
200+
* `skip` {boolean|string} If truthy, the test is skipped. If a string is
201+
provided, that string is displayed in the test results as the reason for
202+
skipping the test. **Default:** `false`.
203+
* `todo` {boolean|string} If truthy, the test marked as `TODO`. If a string
204+
is provided, that string is displayed in the test results as the reason why
205+
the test is `TODO`. **Default:** `false`.
206+
* `fn` {Function|AsyncFunction} The function under test. This first argument
207+
to this function is a [`TestContext`][] object. If the test uses callbacks,
208+
the callback function is passed as the second argument. **Default:** A no-op
209+
function.
210+
* Returns: {Promise} Resolved with `undefined` once the test completes.
211+
212+
The `test()` function is the value imported from the `test` module. Each
213+
invocation of this function results in the creation of a test point in the TAP
214+
output.
215+
216+
The `TestContext` object passed to the `fn` argument can be used to perform
217+
actions related to the current test. Examples include skipping the test, adding
218+
additional TAP diagnostic information, or creating subtests.
219+
220+
`test()` returns a `Promise` that resolves once the test completes. The return
221+
value can usually be discarded for top level tests. However, the return value
222+
from subtests should be used to prevent the parent test from finishing first
223+
and cancelling the subtest as shown in the following example.
224+
225+
```js
226+
test('top level test', async (t) => {
227+
// The setTimeout() in the following subtest would cause it to outlive its
228+
// parent test if 'await' is removed on the next line. Once the parent test
229+
// completes, it will cancel any outstanding subtests.
230+
await t.test('longer running subtest', async (t) => {
231+
return new Promise((resolve, reject) => {
232+
setTimeout(resolve, 1000);
233+
});
234+
});
235+
});
236+
```
237+
238+
## Class: `TestContext`
239+
240+
<!-- YAML
241+
added: REPLACEME
242+
-->
243+
244+
An instance of `TestContext` is passed to each test function in order to
245+
interact with the test runner. However, the `TestContext` constructor is not
246+
exposed as part of the API.
247+
248+
### `context.diagnostic(message)`
249+
250+
<!-- YAML
251+
added: REPLACEME
252+
-->
253+
254+
* `message` {string} Message to be displayed as a TAP diagnostic.
255+
256+
This function is used to write TAP diagnostics to the output. Any diagnostic
257+
information is included at the end of the test's results. This function does
258+
not return a value.
259+
260+
### `context.skip([message])`
261+
262+
<!-- YAML
263+
added: REPLACEME
264+
-->
265+
266+
* `message` {string} Optional skip message to be displayed in TAP output.
267+
268+
This function causes the test's output to indicate the test as skipped. If
269+
`message` is provided, it is included in the TAP output. Calling `skip()` does
270+
not terminate execution of the test function. This function does not return a
271+
value.
272+
273+
### `context.todo([message])`
274+
275+
<!-- YAML
276+
added: REPLACEME
277+
-->
278+
279+
* `message` {string} Optional `TODO` message to be displayed in TAP output.
280+
281+
This function adds a `TODO` directive to the test's output. If `message` is
282+
provided, it is included in the TAP output. Calling `todo()` does not terminate
283+
execution of the test function. This function does not return a value.
284+
285+
### `context.test([name][, options][, fn])`
286+
287+
<!-- YAML
288+
added: REPLACEME
289+
-->
290+
291+
* `name` {string} The name of the subtest, which is displayed when reporting
292+
test results. **Default:** The `name` property of `fn`, or `'<anonymous>'` if
293+
`fn` does not have a name.
294+
* `options` {Object} Configuration options for the subtest. The following
295+
properties are supported:
296+
* `concurrency` {number} The number of tests that can be run at the same time.
297+
If unspecified, subtests inherit this value from their parent.
298+
**Default:** `1`.
299+
* `skip` {boolean|string} If truthy, the test is skipped. If a string is
300+
provided, that string is displayed in the test results as the reason for
301+
skipping the test. **Default:** `false`.
302+
* `todo` {boolean|string} If truthy, the test marked as `TODO`. If a string
303+
is provided, that string is displayed in the test results as the reason why
304+
the test is `TODO`. **Default:** `false`.
305+
* `fn` {Function|AsyncFunction} The function under test. This first argument
306+
to this function is a [`TestContext`][] object. If the test uses callbacks,
307+
the callback function is passed as the second argument. **Default:** A no-op
308+
function.
309+
* Returns: {Promise} Resolved with `undefined` once the test completes.
310+
311+
This function is used to create subtests under the current test. This function
312+
behaves in the same fashion as the top level [`test()`][] function.
313+
314+
[TAP]: https://testanything.org/
315+
[`TestContext`]: #class-testcontext
316+
[`test()`]: #testname-options-fn

lib/internal/errors.js

+12
Original file line numberDiff line numberDiff line change
@@ -1557,6 +1557,18 @@ E('ERR_STREAM_WRAP', 'Stream has StringDecoder set or is in objectMode', Error);
15571557
E('ERR_STREAM_WRITE_AFTER_END', 'write after end', Error);
15581558
E('ERR_SYNTHETIC', 'JavaScript Callstack', Error);
15591559
E('ERR_SYSTEM_ERROR', 'A system error occurred', SystemError);
1560+
E('ERR_TEST_FAILURE', function(error, failureType) {
1561+
hideInternalStackFrames(this);
1562+
assert(typeof failureType === 'string',
1563+
"The 'failureType' argument must be of type string.");
1564+
1565+
const msg = error?.message ?? lazyInternalUtilInspect().inspect(error);
1566+
1567+
this.failureType = error?.failureType ?? failureType;
1568+
this.cause = error;
1569+
1570+
return msg;
1571+
}, Error);
15601572
E('ERR_TLS_CERT_ALTNAME_FORMAT', 'Invalid subject alternative name string',
15611573
SyntaxError);
15621574
E('ERR_TLS_CERT_ALTNAME_INVALID', function(reason, host, cert) {

0 commit comments

Comments
 (0)