Testing/E2E: Guides and Practical Know-how | Page 15 | mironsoft Blog
Testing/E2E
-
Jul 12, 2025
A cookie consent banner gets set up once in many projects, visually signed off, and then barely treated as its own, test-worthy functional area again, even though this exact area has a direct legal dimension: if a tracking script loads despite an actively declined consent, that's not a cosmetic detail, it's an actual violation of the General Data Protection Regulation with potential fine consequences. An automated E2E test that deliberately simulates different consent states and verifies whether tracking scripts actually load depending on that state makes this legally relevant functionality just as reliably regression-safe as any other critical business process in the store.
-
Jul 9, 2025
Dockerfiles that are never statically checked accumulate silent problems: outdated base images with critical CVEs, invalid compose configurations, and structural issues that only surface once they hit production. Automated linting and testing closes that gap right inside the CI pipeline.
-
Jul 8, 2025
A CSS refactor nudges the cart button three pixels to the left, every unit and functional test stays green, and yet the checkout suddenly looks broken. Visual regression testing closes exactly that gap, comparing screenshots automatically against a stored baseline, separating real layout bugs from harmless rendering noise, and reliably warning developers before customers ever notice the break.
-
Jun 30, 2025
Playwright has established itself alongside Cypress as a second central solution for end to end tests and ships with multi browser support, a built in Trace Viewer and a free worker model right out of the box. This article walks through installation, the configuration file and a first real test against a Magento and Hyva storefront, so teams can make an informed choice between Cypress and Playwright.
-
Jun 24, 2025
Magento 2 ships its own test infrastructure that differs substantially from a standard PHPUnit project. If you want to run integration tests directly from PHPStorm, see coverage reports in the IDE, and debug individual test methods with a single click, you need specific configuration, which this article covers in full.
-
Jun 17, 2025
Automated accessibility tests with axe-core reliably catch roughly a third of all WCAG violations and integrate directly into an existing Playwright or Cypress suite. With a baseline strategy the build only fails on new violations, while known legacy issues stay visible instead of blocking the pipeline permanently. Manual review remains essential for keyboard operation and screen readers.
-
Jun 15, 2025
AI-assisted test generation promises to automatically turn a recorded user flow or a plain, natural-language description into runnable, syntactically correct test code, considerably speeding up the tedious, manual writing of test cases for many standard scenarios. That promise holds up reasonably well for obvious, repetitive UI interactions, but it hides a central problem: a generated test that runs syntactically correctly doesn't automatically check what's actually relevant from a business standpoint.
-
Jun 13, 2025
A test matrix with Docker checks an application against several PHP or Node versions at the same time, instead of relying on a single local developer version. Containers make every combination reproducible, executable in parallel, and clearly separable in the results, without maintaining multiple physical test machines.
-
Jun 6, 2025
Most developers know set -x only as a quick switch for more output and never use more than the default setting. With a customized PS4, targeted redirection of the trace output and selective switching on and off, set -x turns from a blunt tool into a precise Bash tracing technique that can also be safely used in CI pipelines.
-
Jun 4, 2025
Self-healing tests promise that a selector broken by a harmless layout change gets automatically detected and swapped for a matching alternative, without a human ever manually touching the test code. That promise sounds tempting given the well-known maintenance burden of brittle E2E tests, but it hides a fundamental misunderstanding of what a selector actually does, and where automatic heuristics hit their real limits the moment not just the structure but an element's actual meaning changes.