Integrate Merv Report in Your Cucumber Project

Use the merv-client Cucumber plugin to capture every scenario, step, tag, and timing into local HTML and JSON. With Merv-Local, execution data stays on disk — you get a full suite report, a multi-run dashboard (index.html), consolidated history per test case, and KPI charts built from your report folders.

1. Merv-Local with Cucumber

Mode	`merv.properties`	What happens
Merv-Local (offline reports)	Set `merv.local=true`	Execution data stays on disk: timestamped folders under your report root, `merv-report.json`, live and final `merv-report.html`, and an updated local dashboard at `index.html`.

If merv.local is omitted, the client follows its default (see merv-client); set it explicitly to avoid surprises.

2. Add the dependency

Include merv-client-api in the same Maven module where Cucumber runs (or a shared test module):

<dependency>
  <groupId>io.github.techelliiptica</groupId>
  <artifactId>merv-client-api</artifactId>
  <version>4.0.0</version>
</dependency>

Use the current version from Maven Central if newer.

3. Configure `merv.properties`

Copy a sample to your project root (working directory) as merv.properties and edit suite name and paths. For local reports, the essentials are:

merv.local=true
merv.regression_suite=Your Regression Suite Name
merv.report.folder=./merv-reports
merv.execution.parallel=false
merv.screenshot=on

Download sample merv.properties

merv.local=true — enable Merv-Local.
merv.regression_suite=… — title shown on the suite report.
merv.report.folder=… — optional; root directory for all runs (see MervConfig). Each run creates a folder like dd-MM-yyyy HH-mm-ss Merv-Report.
merv.execution.parallel=true|false — matches how you run Cucumber; used for suite behaviour.
merv.screenshot=on — optional; captures per-step screenshots when you bind your driver via MervCucumberHandler.setAutomationToolObject(…).

4. Register the Cucumber plugin

Register the class org.teche.merv.client.plugin.MervCucumberHandler as a Cucumber plugin so it receives test-run events:

JUnit 4 — @CucumberOptions(plugin = { "org.teche.merv.client.plugin.MervCucumberHandler" }) on your runner.
JUnit Platform / Cucumber 7+ — use @ConfigurationParameter(key = PLUGIN_PROPERTY_NAME, value = "org.teche.merv.client.plugin.MervCucumberHandler") or the equivalent in junit-platform.properties / Gradle, per your setup.
CLI — --plugin org.teche.merv.client.plugin.MervCucumberHandler (plus your other options).

The handler implements ConcurrentEventListener, so it is safe to use with parallel scenarios; step and case IDs are tracked per thread.

5. What gets generated on disk (Merv-Local)

Under your configured report root:

Path	Purpose
`index.html`	Local dashboard: all runs, search, execution trend chart, links into each suite, plus Consolidated Report and KPIs (see below).
`<run folder>/html/merv-report-live.html`	Live-updating suite view while tests run (polls JSON).
`<run folder>/html/merv-report.html`	Final suite report after the run (snapshot from live).
`<run folder>/json/merv-report.json`	Source of truth for the dashboard, KPIs, and consolidated views.

Opening index.html in a browser (file or static server) is the recommended entry point for operators and leads.

6. Per-run suite report (HTML)

Each run’s HTML report is designed for drill-down debugging and sign-off:

Sidebar — all scenarios with pass / fail / skip / active chips; search and status filters.
Summary & charts — pass/fail/skip counts, pie chart, collapsible block.
Failure reasons — grouped messages with quick navigation to related cases and screenshot thumbnails.
Run timing — suite start, end, and duration.
Test case detail — status, duration, machine, tags, error text, test steps with status pills (PASS / FAIL / SKIP / IN PROGRESS), step-level duration (gap from previous step end), screenshots, and optional log excerpts.
Smart step status — pending or never-reached steps after a hard failure are shown as SKIP; true in-flight steps use IN PROGRESS styling.
Screenshot movie — optional strip that plays step screenshots in order.
Trend chart (live view) — pass/fail throughput in time buckets while the suite runs.

7. Local dashboard: Test Suites view

The Test Suites tab on index.html lists every run folder with title, pass/fail/skip counts, tag pills, and badges for Completed vs In progress. From each card you can open the suite HTML or copy paths. A live execution chart at the top shows pass (green) and fail (red) activity over a selectable time window (1 hour to custom range), aggregated across listed runs, with periodic refresh while you develop.

8. Consolidated Report

The consolidated views roll up all merv-report.json files under the report root so you see history without opening each run separately.

8.1 TestCase Summary (TestCase View)

A searchable table with one row per logical test case name:

Current status (last run) — outcome of the newest matching execution.
Last 5 runs — compact status history.
Last passed / last failed — timestamps for triage.
Total pass / total fail — cumulative counts across discovered runs.

Expand rows to see per-suite, per-folder breakdowns and deep links where applicable.

8.2 Tag / Group Based Summary

Groups rows by Cucumber tags. A scenario with multiple tags appears under each tag section. Clicking a tag can filter the consolidated search. This matches how teams slice regression (@smoke, @regression, @wip, etc.).

9. KPIs (Chart.js analytics)

The KPIs tab turns the same JSON snapshots into management-friendly charts. You can limit the scope to the last 20 / 30 / 50 / all / custom N builds so trends stay readable in busy pipelines.

Test execution summary — totals for cases passed, failed, skipped; overall pass percentage; doughnut and stacked bar charts for outcome mix and per-build counts.
Execution time & performance — total suite time, average case duration, P95 / P99 suite duration across builds, and charts for suite duration and slowest cases.
Slow test cases — ordered list of the heaviest scenarios (by recorded duration) to target for optimisation.
Flaky test detection — surfaces tests that both passed and failed across builds in scope, with failed-run counts, pass counts, stability, and flip counts (retry semantics are approximated from outcomes stored in JSON).
Trend analysis — build-wise pass % over time, failures per build, and execution time trend (oldest left, newest right).

10. Deleting runs and refreshing the index

The dashboard can integrate with a small local delete API so stale folders can be removed and index.html regenerated; see ReportsDeleteServer and MervReportsIndexHtmlWriter in the merv-client source. After manual folder deletes, regenerate or rerun a suite so the index writer runs, or re-open processes documented in LOCAL_REPORTS_CONTEXT.md.

11. Further reading

merv-client/README.md — client usage, auth, and API overview.
merv-client/LOCAL_REPORTS_CONTEXT.md — report layout, JSON contract, and how other frameworks (e.g. TestNG) should reuse the same dashboard.
Documentation — links to UI, API, and deployment docs.
Documentation — running MERV services locally and choosing framework docs.

← Back to home