Reporters
Here is what each reporter prints, and how you can create your own.
About
A reporter presents the result of a test run. Most of the time the reader is a person, but it can be an IDE or a CI
server too. Trial comes with console reporters, which print while the tests run, and file reporters, which write to the
artifactsLocation folder (.trial by default).
Choose the reporters in the reporters list of your trial.json:
{
"reporters": ["spec", "result", "stats", "html"]
}You can also replace the list for one run with -r, for example dub test -- -r spec,result. See
Command line.
All the samples below come from the same run: a shop.cart module with one unittest block and a Cart spec suite.
The “adds the item prices” test has two steps, and “applies a 10% discount” fails:
private alias suite = Spec!({
describe("Cart", {
it("is empty when created", { ... });
describe("total", {
it("adds the item prices", {
Cart cart;
{
auto step = Step("add two items");
cart.add(10);
cart.add(15);
}
{
auto step = Step("check the total");
cart.total.should.equal(25);
}
});
it("applies a 10% discount", {
Cart cart;
cart.add(50);
cart.totalWithDiscount(10).should.equal(40);
});
});
});
});
/// add keeps the items in order
unittest { ... }Spec
This is the default reporter. It prints the tests nested the same way as the modules and suites. Failed tests get a number that matches the details printed by the result reporter.
Use spec.
shop
cart
✓ add keeps the items in order
Cart
✓ is empty when created
total
✓ adds the item prices
0) applies a 10% discountSpec steps
A flavour of the spec reporter that also prints the steps of each test.
Use spec-steps.
shop
cart
┌ add keeps the items in order
└ ✓ Success
Cart
┌ is empty when created
└ ✓ Success
total
┌ adds the item prices
│ add two items
│ check the total
└ ✓ Success
┌ applies a 10% discount
└ 0) FailureDot Matrix
One character for each test: a . when it passes and a red ! when it fails. Good if you prefer minimal output.
Use dot-matrix.
...!Landing
A gimmick: a plane lands on a runway while the tests run. The plane turns red when a test fails.
Use landing. When the run ends, the runway looks like this:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅⋅✈
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━List
A flat list with the full name of every test, printed as each one passes or fails.
Use list.
✓ shop.cart add keeps the items in order
✓ shop.cart.Cart is empty when created
✓ shop.cart.Cart.total adds the item prices
0) shop.cart.Cart.total applies a 10% discountProgress
A progress bar with the number of finished tests. It turns red when a test fails.
Use progress. When the run ends, it shows:
4/4 ▓▓▓▓Result
Prints the details of every failed test, then a summary of the run. It is part of the default list, and you need it to see why a test failed.
Use result.
0) shop.cart.Cart.total applies a 10% discount:
fluentasserts.core.base.TestException@source/shop/cart.d(53): FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53
----------------
source/shop/cart.d:53 void shop.cart.__lambda_L28_C31().__lambda_L29_C22().__lambda_L34_C25().__lambda_L50_C38() [0x551db6]
...
Executed 4 (1 failed) tests in 3 suites in 7 ms, 978 μs, and 9 hnsecs.TAP
Prints the results for a Test Anything Protocol consumer.
Use tap.
TAP version 13
1..4
ok - shop.cart.add keeps the items in order
ok - shop.cart.Cart.is empty when created
ok - shop.cart.Cart.total.adds the item prices
not ok - shop.cart.Cart.total.applies a 10% discount
# FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53
#
---
message: 'FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53'
severity: failure
location:
fileName: 'source/shop/cart.d'
line: 53HTML
Writes result.html, a page with the duration, the number of passed and failed tests, and every suite with its tests
and steps. The “Failed” button hides the tests that passed, and “details” opens the failure. Publish it with your CI
artifacts to have a report for each build.
Use html.
Allure
Writes one xml file per suite to the allure folder, in the format that the Allure
command line turns into an html report. Steps are kept as Allure steps.
Use allure. Part of the file for the shop.cart.Cart.total suite:
<ns2:test-suite start="1791110623829" stop="1791110623829" version="1.5.2" xmlns:ns2="urn:model.allure.qatools.yandex.ru">
<name>shop.cart.Cart.total</name>
<title>shop.cart.Cart.total</title>
<test-cases>
<test-case start="1791110623829" stop="1791110623829" status="passed">
<name>adds the item prices</name>
<steps>
<step start="1791110623829" stop="1791110623829" status="passed">
<name>add two items</name>
</step>
<step start="1791110623829" stop="1791110623829" status="passed">
<name>check the total</name>
</step>
</steps>
</test-case>
<test-case start="1791110623829" stop="1791110623829" status="failed">
<name>applies a 10% discount</name>
<failure>
<message>FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53</message>
...To turn the files into an html report:
allure generate -o allure-html .trial/allureXUnit
Writes one JUnit xml file per suite to the xunit folder. Most CI servers read this format; in GitLab add
.trial/xunit/*.xml to artifacts:reports:junit.
Use xunit. The file for the shop.cart.Cart.total suite:
<?xml version="1.0" encoding="UTF-8"?>
<testsuites>
<testsuite name="shop.cart.Cart.total" errors="0" skipped="0" tests="2" failures="1" time="0" timestamp="2026-10-04T12:43:43.8297456">
<testcase name="adds the item prices">
</testcase>
<testcase name="applies a 10% discount">
<failure message="FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53">fluentasserts.core.base.TestException@source/shop/cart.d(53): ...</failure>
</testcase>
</testsuite>
</testsuites>Stats
Writes stats.csv with the start time, end time and status of every suite, test and step, and the file and line of
each test. The spec progress reporter reads it on the next run to estimate how long the tests take.
Use stats.
shop.cart.add keeps the items in order,2026-10-04T12:43:43.8297084,2026-10-04T12:43:43.8297097,success,source/shop/cart.d,61
shop.cart,2026-10-04T12:43:43.8297061,2026-10-04T12:43:43.8297112,unknown,,0
shop.cart.Cart.is empty when created,2026-10-04T12:43:43.8297117,2026-10-04T12:43:43.8297446,success,source/shop/cart.d,30
shop.cart.Cart,2026-10-04T12:43:43.8297115,2026-10-04T12:43:43.8297455,unknown,,0
shop.cart.Cart.total.adds the item prices.add two items,2026-10-04T12:43:43.8297462,2026-10-04T12:43:43.82975,unknown,,0
shop.cart.Cart.total.adds the item prices.check the total,2026-10-04T12:43:43.8297505,2026-10-04T12:43:43.8297578,unknown,,0
shop.cart.Cart.total.adds the item prices,2026-10-04T12:43:43.8297458,2026-10-04T12:43:43.8297586,success,source/shop/cart.d,35
shop.cart.Cart.total.applies a 10% discount,2026-10-04T12:43:43.8297594,2026-10-04T12:43:43.8298288,failure,source/shop/cart.d,50
shop.cart.Cart.total,2026-10-04T12:43:43.8297456,2026-10-04T12:43:43.8298293,unknown,,0Spec progress
An experimental reporter that extends the spec reporter with the running time of the current suite and test, and the
time left until the run ends. It is meant for slow tests, like UI tests written with selenium or appium, together
with the parallel executor and the stats reporter.
Use spec-progress. Part of the output:
*[0s]shop.cart.Cart.total *[0s]adds the item prices
total
✓ adds the item pricesAgent
A plain text reporter for AI coding agents. It prints nothing for the tests that pass. For each failed test it prints the
test name, where the test and the failure are, the failure message, and the -f filter that runs that test again. The
run ends with one summary line.
Use agent.
FAIL shop.cart.Cart.total applies a 10% discount
test: source/shop/cart.d:50
at: source/shop/cart.d:53
FAIL: 45 should equal 40. | actual=45 expected=40 | source/shop/cart.d:53
rerun: -f "=shop.cart.Cart.total applies a 10% discount"
RESULT failed=1 passed=3 pending=0 skipped=0It is enabled automatically when an agent runs the tests. See Agent mode.
Extending
If you want to write a custom reporter, have a look at the Lifecycle interface that trial provides and implement the methods that you need.
If you want to use your custom reporter, you can add it to the LifeCycleListeners:
static this() {
LifeCycleListeners.instance.add(new MyCustomReporter);
}