I believe there's some room for experimentation in the best way we can report test results. For example, right now we put in one line all the tests that failed, were ignored, succeeded, etc.
I think it would be interesting to experiment with a better formatted output. For example, an ANSII table that contains this information in a more descriptive way.
Another idea is to use more colors to mark the information that is more relevant and to remove lines that are likely to be polluting the terminal (e.g. things that are obvious from context). I would be grateful if someone is interested in taking a shot at this and propose a nice format.
In the future, I think we should have a web interface that implements BSP (and even creates a superset) so that the test results can be shown in a browser (or vscode directly) in a nice way, trying to cram all the results in a UI instead of printing one line per every test executed. A lot of information gets lost in a terminal with many logs.
@tindzk has kindly submitted a PR to add a nice test report at the end of test execution, thus closing https://github.com/scalacenter/bloop/issues/340 mainly.
I believe there's some room for experimentation in the best way we can report test results. For example, right now we put in one line all the tests that failed, were ignored, succeeded, etc.
I think it would be interesting to experiment with a better formatted output. For example, an ANSII table that contains this information in a more descriptive way.
Another idea is to use more colors to mark the information that is more relevant and to remove lines that are likely to be polluting the terminal (e.g. things that are obvious from context). I would be grateful if someone is interested in taking a shot at this and propose a nice format.
In the future, I think we should have a web interface that implements BSP (and even creates a superset) so that the test results can be shown in a browser (or vscode directly) in a nice way, trying to cram all the results in a UI instead of printing one line per every test executed. A lot of information gets lost in a terminal with many logs.