Changelog

We are happy to announce that we’ve enhanced our SSO integration to support automated users provisioning using SCIM protocol.

Check out the documentation to discover how to enable SCIM.

Improved History Browser

The new history browser allows more convenient navigation and exploration of historical items

Inline browsing for Test Explorer

Test Explorer view allows exploring test performance trends, past executions and most common errors. In past, exploring past executions required navigating to a different view, causing confusion and context switch.

With the recent change the executions are shown inline for more convenient and speedy exploration.

Better navigation for multi-group runs

We’ve improved the layout and filtering options for runs with multiple groups. The new variant allows more intuitive filtering of groups when browsing Spec Files view and Live View.

Inline links to Explorer, Run and Live View

When browsing spec or test results we’ve added a shortcut links for more convenient navigation

• A link to Spec or Test Explorer for the associated entities

• A link to the associated Run View

• A link to the associated Live View (Playwright and Jest only)

Misc

• An improved list items layout to better support long names and inline actions

• Added list line item numbers

• Split complex views into sub-views accessible via tabs

• Better axis labels that auto-adjust to timeline duration

As a follow up on our previous release that included support for Test Annotations - a test owner will be notified via a Slack notification when their tests fail.

To activate the notifications, make sure to enable Slack integration and set owner field value to match Slack username.

See the documentation for more details.

📝 Test Annotations

Playwright Test Annotations allows augmenting test results with additional metadata, for example: a Jira ticket number or Test Rail tracking id.

Starting from version @currents/playwright@1.5.0 , Currents will store test annotations for the recorded tests, the annotations will show up in the dashboard.

Use case - test owners

Some test annotations have a special purpose, for example:

testInfo.annotations.push({ 
  "type": "owner",
  "description": "johnSmith"
})

Using type:owner has a special meaning in the dashboard - it will designate… a test owner. Test owner can be a person or a team.

We are working on incorporating the concept into our integrations and across the dashboard - e.g. sending notifications to a person or a team when their test fails, allowing filtering tests by their owner.

Let us know if you have more use cases that involve annotations.

👟 Inline Steps View

One of the unique features of Currents reporter for Playwright is how we upload the results. Instead of waiting for the whole test or a spec file to complete, we stream step-level information. That allows to capture very detailed information - even if playwright or CI machine crashes, we still send and present the last reported step.

The recent release exposes step-level information within test details view for more convenient and effective troubleshooting.

For example, one of our clients uses test steps as a “documentation” - they give steps descriptive names, so team members can quickly understand what’s happening in a test just by glancing at the reported steps.

🌼 UI changes

We revamped the UI for more convenient browsing:

• split test attempt details into tabs (error details, screenshots, videos, traces, steps)

• improved the onboarding and setup instructions to help new users start using the platform

New

Playwright Test Annotations allows augmenting test results with additional metadata, for example: a Jira ticket number or Test Rail tracking id. Starting from version 1.5.0 Currents will store test annotations for the recorded tests, the annotations will show up in the dashboard.

Use case: test owners
Some test annotations have a special purpose, for example:

 testInfo.annotations.push({ 
    "type": "owner",
    "description": "johnSmith"
 })

Using type:owner has a special meaning in the dashboard - it will designate… a test owner. Test owner can be a person or a team.

We are working on incorporating the concept into our integrations and across the dashboard - e.g. sending notifications to a person or a team when their test fails, allowing filtering tests by their owner.

Let us know if you have more use cases that involve annotations.

Fixes

• Retry reporting results when certain network requests fail

• Better support for Fail Fast strategy, which allows to automatically interrupt CI runs when a certain number of tests failed. Can be configured in Project Settings or via CLI using --pwc-cancel-after-failures <number | false>

This is a minor release that fixes a few bugs and mostly contains backend changes to support upcoming features

• Fix: use consistent representation of git commit sha (8 first characters)

• Fix: correctly select machine when machine has tests from multiple groups

• Fix: sanitize and store test annotations, in preparation for using annotations for test owners and notification rules

• Fix: fetch up to 198 nested steps for playwright live view

Customize the metric for runs duration chart: average, P99, P95, P50, Max and Min.

• All the metrics are also available when you export CSV or JSON

• All the metrics appear in the breakdown table

• The selected metric will be used for the calculating the trend line

This release is packed with improvements and bug fixes. See below👇

💅🏻 Run Summary UI

Machines view visualizes the allocation of projects (groups) and spec files on CI machines, using a single SVG pane with a unified timeline.

Selecting a particular machine shows the spec files that ran on the selected machine and execution details:

• Spin-up time (1) - time it took the machine to start reporting results

• Active time (2) - time it took the machine to run its specs

Pin and unpin spec files for more convenient triaging.

🐞 Top run errors

Run Errors tab show failed and flaky tests grouped by error message. Failure and test details show up inline, so you can troubleshoot issues without extra navigation steps.

✨ Similar Errors Detection

When browsing failed or flaky tests, we show a warning if we find that the test failed because of a similar reason on other branches and runs.

🎁 Misc

• You can export data from the dashboard JSON format, in addition to already supported CSV

• Customize Run title for run created via GitHub Pull Request in project settings - use commit message or PR title

• Supporting up to 98 nested steps in Live Results

• Opening a deleted run shows user-friendly error message

• Currents will updating GitHub comment for a successful run with new data even if Failed Only setting is enabled

Leave us a feedback and share your opinion in the comment section below.