James Addison

> Err... why not. It's just that `testroot` is a common name when taking about tests and their static content. It wasn't for the testroot `test-root`. OK, I didn't realize...

FYI @picnixz: I'm planning to take a more detailed code-review/local-exploration of these changes later today.

In the back of my mind I am thinking: the cost and time savings from this change could be significant - so is it a risk to delay/block it? It...

> > the cost and time savings from this change could be significant - so is it a risk to delay/block it? It could be. > > Not really. It's...

@picnixz I wonder whether we could do this by matching on the source-name branch instead? (that would indirectly also ensure that people (such as me, for example) trying to fix/identify...

> @picnixz I wonder whether we could do this by matching on the source-name branch instead? > > (that would indirectly also ensure that people (such as me, for example)...

> For fairness I'd suggest adding both `if linux` and `if windows` expressions -- assuming this approach is possible. Perhaps those should be (pseudocode again): `if branch_os is empty ||...

I'm not sure that my suggestion is feasible; I think that when a commit is pushed to a pull request branch, it is the [`pull_request` event](https://docs.github.com/en/actions/using-workflows/events-that-trigger-workflows#pull_request) that is initiated. The...

> > I think that the [on.[push|pull_request]: ... condition](https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#onpushpull_requestpull_request_targetpathspaths-ignore) may be limited to positive-filtering (branch name like x) but here we want to disable other tests when a branch name...

Do we need to update `ENV_VERSION` with this change?