Add --only-failures flag to bun:test (#23312)

## Summary

Adds a new `--only-failures` flag to `bun test` that only displays test
failures, similar to `--dots` but without printing dots for each test.

## Motivation

When running large test suites or in CI environments, users often only
care about test failures. The existing `--dots` reporter reduces
verbosity by showing dots, but still requires visual scanning to find
failures. The `--only-failures` flag provides a cleaner output by
completely suppressing passing tests.

## Changes

- Added `--only-failures` CLI flag in `Arguments.zig`
- Added `only_failures` boolean to the test reporters struct in
`cli.zig`
- Updated test output logic in `test_command.zig` to skip non-failures
when flag is set
- Updated `jest.zig` and `bun_test.zig` to handle the new flag
- Added comprehensive tests in `only-failures.test.ts`

## Usage

```bash
bun test --only-failures
```

Example output (only shows failures):
```
test/example.test.ts:
(fail) failing test
error: expect(received).toBe(expected)

Expected: 3
Received: 2

5 pass
1 skip
2 fail
Ran 8 tests across 1 file.
```

## Test Plan

- Verified `--only-failures` flag only shows failing tests
- Verified normal test output still works without the flag
- Verified `--dots` reporter still works correctly
- Added regression tests with snapshot comparisons

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Bot <claude-bot@bun.sh>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: pfg <pfg@pfg.pw>

docs/runtime/bunfig.md

@@ -249,6 +249,41 @@ This is useful for:
 
 The `--concurrent` CLI flag will override this setting when specified.
 
+### `test.onlyFailures`
+
+When enabled, only failed tests are displayed in the output. This helps reduce noise in large test suites by hiding passing tests. Default `false`.
+
+```toml
+[test]
+onlyFailures = true
+```
+
+This is equivalent to using the `--only-failures` flag when running `bun test`.
+
+### `test.reporter`
+
+Configure the test reporter settings.
+
+#### `test.reporter.dots`
+
+Enable the dots reporter, which displays a compact output showing a dot for each test. Default `false`.
+
+```toml
+[test.reporter]
+dots = true
+```
+
+#### `test.reporter.junit`
+
+Enable JUnit XML reporting and specify the output file path.
+
+```toml
+[test.reporter]
+junit = "test-results.xml"
+```
+
+This generates a JUnit XML report that can be consumed by CI systems and other tools.
+
 ### `test.randomize`
 
 Run tests in random order. Default `false`.

src/bun.js/test/bun_test.zig

@@ -173,7 +173,7 @@ pub const BunTestRoot = struct {
     pub fn onBeforePrint(this: *BunTestRoot) void {
         if (this.active_file.get()) |active_file| {
             if (active_file.reporter) |reporter| {
-                if (reporter.last_printed_dot and reporter.reporters.dots) {
+                if (reporter.reporters.dots and reporter.last_printed_dot) {
                     bun.Output.prettyError("<r>\n", .{});
                     bun.Output.flush();
                     reporter.last_printed_dot = false;

src/bun.js/test/jest.zig

@@ -15,7 +15,7 @@ const CurrentFile = struct {
         repeat_index: u32,
         reporter: *CommandLineReporter,
     ) void {
-        if (Output.isAIAgent() or reporter.reporters.dots) {
+        if (reporter.reporters.dots or reporter.reporters.only_failures) {
             this.freeAndClear();
             this.title = bun.handleOom(bun.default_allocator.dupe(u8, title));
             this.prefix = bun.handleOom(bun.default_allocator.dupe(u8, prefix));

src/bunfig.zig

@@ -239,6 +239,11 @@ pub const Bunfig = struct {
                         this.ctx.test_options.coverage.enabled = expr.data.e_boolean.value;
                     }
 
+                    if (test_.get("onlyFailures")) |expr| {
+                        try this.expect(expr, .e_boolean);
+                        this.ctx.test_options.reporters.only_failures = expr.data.e_boolean.value;
+                    }
+
                     if (test_.get("reporter")) |expr| {
                         try this.expect(expr, .e_object);
                         if (expr.get("junit")) |junit_expr| {

src/cli.zig

@@ -355,6 +355,7 @@ pub const Command = struct {
 
         reporters: struct {
             dots: bool = false,
+            only_failures: bool = false,
             junit: bool = false,
         } = .{},
         reporter_outfile: ?[]const u8 = null,

src/cli/Arguments.zig

@@ -209,6 +209,7 @@ pub const test_only_params = [_]ParamType{
     clap.parseParam("--reporter <STR>                 Test output reporter format. Available: 'junit' (requires --reporter-outfile), 'dots'. Default: console output.") catch unreachable,
     clap.parseParam("--reporter-outfile <STR>         Output file path for the reporter format (required with --reporter).") catch unreachable,
     clap.parseParam("--dots                           Enable dots reporter. Shorthand for --reporter=dots.") catch unreachable,
+    clap.parseParam("--only-failures                  Only display test failures, hiding passing tests.") catch unreachable,
     clap.parseParam("--max-concurrency <NUMBER>        Maximum number of concurrent tests to execute at once. Default is 20.") catch unreachable,
 };
 pub const test_params = test_only_params ++ runtime_params_ ++ transpiler_params_ ++ base_params_;
@@ -463,6 +464,11 @@ pub fn parse(allocator: std.mem.Allocator, ctx: Command.Context, comptime cmd: C
             ctx.test_options.reporters.dots = true;
         }
 
+        // Handle --only-failures flag
+        if (args.flag("--only-failures")) {
+            ctx.test_options.reporters.only_failures = true;
+        }
+
         if (args.option("--coverage-dir")) |dir| {
             ctx.test_options.coverage.reports_directory = dir;
         }

src/cli/test_command.zig

@@ -579,6 +579,7 @@ pub const CommandLineReporter = struct {
 
     reporters: struct {
         dots: bool = false,
+        only_failures: bool = false,
         junit: ?*JunitReporter = null,
     } = .{},
 
@@ -874,8 +875,8 @@ pub const CommandLineReporter = struct {
                             },
                         }
                         buntest.reporter.?.last_printed_dot = true;
-                    } else if (Output.isAIAgent() and (comptime result.basicResult()) != .fail) {
-                        // when using AI agents, only print failures
+                    } else if (((comptime result.basicResult()) != .fail) and (buntest.reporter != null and buntest.reporter.?.reporters.only_failures)) {
+                        // when using --only-failures, only print failures
                     } else {
                         buntest.bun_test_root.onBeforePrint();
 
@@ -900,7 +901,7 @@ pub const CommandLineReporter = struct {
 
         var this: *CommandLineReporter = buntest.reporter orelse return; // command line reporter is missing! uh oh!
 
-        if (!this.reporters.dots) switch (sequence.result.basicResult()) {
+        if (!this.reporters.dots and !this.reporters.only_failures) switch (sequence.result.basicResult()) {
             .skip => bun.handleOom(this.skips_to_repeat_buf.appendSlice(bun.default_allocator, output_buf.items[initial_length..])),
             .todo => bun.handleOom(this.todos_to_repeat_buf.appendSlice(bun.default_allocator, output_buf.items[initial_length..])),
             .fail => bun.handleOom(this.failures_to_repeat_buf.appendSlice(bun.default_allocator, output_buf.items[initial_length..])),
@@ -1362,6 +1363,11 @@ pub const TestCommand = struct {
         if (ctx.test_options.reporters.dots) {
             reporter.reporters.dots = true;
         }
+        if (ctx.test_options.reporters.only_failures) {
+            reporter.reporters.only_failures = true;
+        } else if (Output.isAIAgent()) {
+            reporter.reporters.only_failures = true; // only-failures defaults to true for ai agents
+        }
 
         js_ast.Expr.Data.Store.create();
         js_ast.Stmt.Data.Store.create();

test/js/bun/test/only-failures.fixture.ts

@@ -0,0 +1,27 @@
+import { expect, test } from "bun:test";
+
+test("passing test 1", () => {
+  expect(1 + 1).toBe(2);
+});
+
+test("passing test 2", () => {
+  expect(2 + 2).toBe(4);
+});
+
+test("failing test", () => {
+  expect(1 + 1).toBe(3);
+});
+
+test("passing test 3", () => {
+  expect(3 + 3).toBe(6);
+});
+
+test.skip("skipped test", () => {
+  expect(true).toBe(false);
+});
+
+test.todo("todo test");
+
+test("another failing test", () => {
+  throw new Error("This test fails");
+});

test/js/bun/test/only-failures.test.ts

@@ -0,0 +1,120 @@
+import { expect, test } from "bun:test";
+import { bunEnv, bunExe, normalizeBunSnapshot, tempDir } from "harness";
+
+test.concurrent("only-failures flag should show only failures", async () => {
+  const result = await Bun.spawn({
+    cmd: [bunExe(), "test", import.meta.dir + "/only-failures.fixture.ts", "--only-failures"],
+    stdout: "pipe",
+    stderr: "pipe",
+    env: bunEnv,
+  });
+  const exitCode = await result.exited;
+  const stdout = await result.stdout.text();
+  const stderr = await result.stderr.text();
+  expect({
+    exitCode,
+    stdout: normalizeBunSnapshot(stdout),
+    stderr: normalizeBunSnapshot(stderr),
+  }).toMatchInlineSnapshot(`
+    {
+      "exitCode": 1,
+      "stderr": 
+    "test/js/bun/test/only-failures.fixture.ts:
+     7 | test("passing test 2", () => {
+     8 |   expect(2 + 2).toBe(4);
+     9 | });
+    10 | 
+    11 | test("failing test", () => {
+    12 |   expect(1 + 1).toBe(3);
+                         ^
+    error: expect(received).toBe(expected)
+
+    Expected: 3
+    Received: 2
+        at <anonymous> (file:NN:NN)
+    (fail) failing test
+    21 | });
+    22 | 
+    23 | test.todo("todo test");
+    24 | 
+    25 | test("another failing test", () => {
+    26 |   throw new Error("This test fails");
+                                            ^
+    error: This test fails
+        at <anonymous> (file:NN:NN)
+    (fail) another failing test
+
+     3 pass
+     1 skip
+     1 todo
+     2 fail
+     4 expect() calls
+    Ran 7 tests across 1 file."
+    ,
+      "stdout": "bun test <version> (<revision>)",
+    }
+  `);
+});
+
+test.concurrent("only-failures flag should work with multiple files", async () => {
+  const result = await Bun.spawn({
+    cmd: [
+      bunExe(),
+      "test",
+      import.meta.dir + "/printing/dots/dots1.fixture.ts",
+      import.meta.dir + "/only-failures.fixture.ts",
+      "--only-failures",
+    ],
+    stdout: "pipe",
+    stderr: "pipe",
+    env: bunEnv,
+  });
+  const exitCode = await result.exited;
+  const stdout = await result.stdout.text();
+  const stderr = await result.stderr.text();
+  expect(exitCode).toBe(1);
+  expect(normalizeBunSnapshot(stderr)).toContain("(fail) failing test");
+  expect(normalizeBunSnapshot(stderr)).toContain("(fail) another failing test");
+  expect(normalizeBunSnapshot(stderr)).not.toContain("(pass)");
+});
+
+test.concurrent("only-failures should work via bunfig.toml", async () => {
+  using dir = tempDir("bunfig-only-failures", {
+    "bunfig.toml": `
+[test]
+onlyFailures = true
+`,
+    "my.test.ts": `
+import { test, expect } from "bun:test";
+
+test("passing test", () => {
+  expect(1 + 1).toBe(2);
+});
+
+test("failing test", () => {
+  expect(1 + 1).toBe(3);
+});
+
+test("another passing test", () => {
+  expect(true).toBe(true);
+});
+`,
+  });
+
+  const result = await Bun.spawn({
+    cmd: [bunExe(), "test"],
+    stdout: "pipe",
+    stderr: "pipe",
+    env: bunEnv,
+    cwd: String(dir),
+  });
+
+  const exitCode = await result.exited;
+  const stderr = await result.stderr.text();
+
+  expect(exitCode).toBe(1);
+  // Should only show the failing test
+  expect(normalizeBunSnapshot(stderr, dir)).toContain("(fail) failing test");
+  // Should not show passing tests
+  expect(normalizeBunSnapshot(stderr, dir)).not.toContain("(pass)");
+});