Problem Statement
Test results running running on GHA are nice, but they could be super fancy.
Proposed Solution
Two things have happened recently that I think have created a great opportunity:
I think this creates a great opportunity to produce a rich, single page HTML report, which is a combination of the test results, but also has a trace viewing capability.
The dream is you open the trace from a link from the failing CI summary, you go straight into the HTML report, you expand a failing test, you see it tooks 30 seconds, and you can see it spend 30 seconds timing out on an API call, and you see in the structured logs the wrong config was used.
Obviously this is a bit ambitious, but I think this could be yet another killer feature that makes TUnit stand out from the rest.
Alternatives Considered
I've played around with my own set of tools here, including creating a trace artifact, and then replaying it into a standalone Aspire instance (with dnx aspirerunner.tool).
It works just fine, and would solve the "it fails in CI, now let's put it down and investigate". However I know I would use this approach way more if there was no extra steps required, and I think a dedicated view is required to get the best experience.
Feature Category
Test Output / Reporting
How important is this feature to you?
Important - significantly impacts my workflow
Additional Context
No response
Contribution
Problem Statement
Test results running running on GHA are nice, but they could be super fancy.
Proposed Solution
Two things have happened recently that I think have created a great opportunity:
I think this creates a great opportunity to produce a rich, single page HTML report, which is a combination of the test results, but also has a trace viewing capability.
The dream is you open the trace from a link from the failing CI summary, you go straight into the HTML report, you expand a failing test, you see it tooks 30 seconds, and you can see it spend 30 seconds timing out on an API call, and you see in the structured logs the wrong config was used.
Obviously this is a bit ambitious, but I think this could be yet another killer feature that makes TUnit stand out from the rest.
Alternatives Considered
I've played around with my own set of tools here, including creating a trace artifact, and then replaying it into a standalone Aspire instance (with
dnx aspirerunner.tool).It works just fine, and would solve the "it fails in CI, now let's put it down and investigate". However I know I would use this approach way more if there was no extra steps required, and I think a dedicated view is required to get the best experience.
Feature Category
Test Output / Reporting
How important is this feature to you?
Important - significantly impacts my workflow
Additional Context
No response
Contribution