RFC #1 - test entry point #20
Conversation
The RFC proposes we specify the behavior of a test entry point. [Fix RetailMeNotSandbox#6]
| script MUST exit with a non-zero exit code. Otherwise, it MUST exit with an exit | ||
| code of zero. | ||
|
|
||
| The test entry point MAY receive configuration via its `process.argv` and |
There was a problem hiding this comment.
I think this needs to be fleshed out a bit further. If I am invoking tests for multiple ingredients from the command line, will all arguments passed to process.argv be passed to all the ingredients? What if two ingredients allow for the same flags but interpret them to mean different things?
Random thought that is probably a bad idea: Perhaps using a tool like commander to roll up, parse, and generate a help message for the options allowed by each ingredient being run for this invocation of the test runner would be in order? That would give us a chance to prevent two ingredients from expecting the same flag with a different meaning.
| The test entry point MAY receive configuration via its `process.argv` and | ||
| `process.env` properties. | ||
|
|
||
| The test entry point MUST NOT assume a Roux server is running. If a Roux server |
There was a problem hiding this comment.
Its not clear to me what the test would do differently in a case where the server is running vs. where it is not? I'm still on board with the server being running being optional, however if a test.js bypasses an entire subset of its test suite because the server is not running, I would not be particularly comfortable with the test.js file exiting 0 in this case, as it is not fully confident that the ingredient passes all of its test.
5 participants
The RFC proposes we specify the behavior of a test entry point.
[Fix #6]