v0.6.21 — Since last run, and the command to run a test again

Release notes

Node.js · npm reporting-labs

v0.6.21 — Since last run, and the command to run a test again

Tag v0.6.21 on commit 8296880.

New

• Since last run: one line under the tiles with what changed against the previous run: new failures, fixed, still failing, got slower and new tests. Click a number to see those tests. Turn it off with widgets: { changes: false }.
• Run again: every test shows the command that runs just that test, with a Copy button, e.g. npx playwright test tests/checkout.spec.ts:12 --project=chromium. playwright-bdd gets npx bddgen && first. Cypress and WebdriverIO re-run the spec file (--spec).
• Copy re-run command: one command for all failed tests, on the Overview and the Failures tab.

Masking fixes

• Type names and Cucumber parameters are code, not secrets: password: string, password {string}. Before, string was learned and blanked in code snippets.
• Format placeholders are code, not secrets: password={password}, password={} were shown as password=****}. A real value in braces ({S3cret!}) is still masked.

Same report as npm reporting-labs 0.6.21, PyPI reporting-labs 0.1.11 and Maven dev.reportinglabs 0.1.32.

npm i -D reporting-labs@0.6.21

Docs: https://reportinglabs.dev/features/run-again · https://reportinglabs.dev/features/history-trend#since-last-run

v0.6.20 — Playwright BDD (playwright-bdd)

Tag v0.6.20 on commit 58dfa4b.

Playwright with Cucumber, through playwright-bdd: keep playwright.config.ts, fixtures, projects and retries, and get a Cucumber-style report.

New

• Each scenario at its .feature file and line, under its Feature and Rule, so grouping, editor links and history follow the feature file.
• Background steps next to the scenario's own, and the steps a failure left unrun, greyed out.
• Scenario Outline rows named after their values, with the row as an Examples table. Data tables and doc strings as test data. A password column is masked in the table and wherever the value shows up.
• Tags that became meta (@P1, @critical, @owner:asha) are not repeated as tags.
• An error in the generated spec (a missing step) points at the step in the feature file. "Missing step" is explained as Step not defined, "Multiple definitions matched" as Step defined twice.
• examples-playwright-bdd: a runnable example with two feature files.

Works with playwright-bdd 7, 8 and 9 (7 gives the location, Feature / Rule and Outline rows), on Playwright 1.56 to 1.64. Specs that are not playwright-bdd output are unchanged.

npm i -D reporting-labs@0.6.20

Guide: https://reportinglabs.dev/get-started/nodejs#playwright-bdd

v0.6.2 — Race Condition Bug Fixe and added support for custom links that need more than one value

0.6.2

Fixed

• import 'reporting-labs/auto' could fail every test with
  apiRequestContext._wrapApiCall: Target page, context or browser has been closed
  when tracing was enabled. The auto module created a throwaway request context at
  import time; when its disposal raced with Playwright's trace recorder, the recorder
  started a trace on a closed context. The module now hooks context creation and
  patches on first use, no throwaway context. (#29)
• links keys are matched case-insensitively, so links: { testCaseId: ... } works.
• Array meta values (story: ['SHOP-1', 'SHOP-2']) render one chip and one link per
  value, as documented.

Added

• Multi-parameter links: a links entry may be { url, display }. Placeholders in
  url are filled from an object passed to meta(), display (default {id}) is
  what the report shows, the other fields only build the URL. Works for any tool:
  links: { workItem: { url: 'https://dev.azure.com/{org}/{project}/_workitems/edit/{id}', display: 'AB#{id}' } }
  with meta({ workItem: { id: '1234', org: 'acme', project: 'shop-web' } }). (#29)