Has capability test

[nichoth]

This PR adds some tests for the hasCapability method. These were things that I was curious about when I was looking at how to validate permissions via UCAN. There is still one question I had, about the originator field. I made a test that currently fails -- https://github.com/nichoth/ts-ucan/blob/f2fde61b72adcd0a434e5ce1ba91cfe1d834fba1/tests/attenuation.test.ts#L296 -- the question is how is originator used? It doesn't seem to affect the return value of hasCapability. I'm not sure if this is not implemented yet or if my test is wrong in some way.

Thanks. Hopefully this is helpful and not creating needless work 😬

[matheus23]

Thanks, that's an amazing contribution.

the question is how is originator used? It doesn't seem to affect the return value of hasCapability. I'm not sure if this is not implemented yet or if my test is wrong in some way.

Yes, this is a bug. Whoops! hasCapability should check the originator to make sure it matches what's passed in and also check that the time bounds work out. All of that instead of running delegateCapabilityInfo... My bad.

Your contribution is amazing. You've created a failing test case. We'll use that to implement a fix for this.

[icidasset]

@nichoth Any chance you can rebase this on main? It should only be two merge conflicts, because the path of ./emailCapabilities has changed to ./capabilities/email, and I changed emailCaps = Array.from( to be multiline.

Would love to get this in so we can fix hasCapability 🙏

[nichoth]

Ok I think it has been rebased & merged properly. My brain started overheating thinking about the branches, PRs, etc, but hopefully I did that correctly

[icidasset]

Thanks! 🙌 I'll merge this next Monday and will make sure all the tests pass.

[nichoth]

I just fixed the type checking with the updates, however there is one type I wasn't sure about -- https://github.com/nichoth/ts-ucan/blob/hasCapability-test/tests/attenuation.test.ts#L234

It is telling me

Property 'email' does not exist on type 'Capability'.ts(2339)

[nichoth]

Update -- I have fixed the above error. It was a mixup with the expected object shapes for the capabilities.

[matheus23]

@nichoth this PR is almost good to go. I proposed some changes to this PR in another PR (see the mention above this comment 👆 ).

Once we've merged that PR, we can merge this PR 🎉

[matheus23]

Okay. I'd love to merge this.

With my changes this also fixes #60

I'd love to see another quick look at this @icidasset 🙏

[icidasset]

Looks good! 👏