Once finished, check our implementation against the final version of our api.raml spec

[forevermatt]

Once we believe the implementation of this is done, let's do a final comparison (ideally using an automated tool, even if only run once) between the implementation and the RAML spec.

[longrunningprocess]

I believe this will be done after #41 is merged.

[forevermatt]

If it's not hard, I'd still be interested in some software comparing the RAML spec file against the working version, if only because I have such a low level of confidence in my own ability to spot possible differences.

[lvail]

Maybe it would be worth a code review meeting Friday together. I'd be interested in participating in a walkthrough of the api.raml compared with the implementation.

…

On Wed, Apr 5, 2017 at 1:43 PM, forevermatt ***@***.***> wrote: If it's not hard, I'd still be interested in some software comparing the RAML spec file against the working version, if only because I have such a low level of confidence in my own ability to spot possible differences. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#32 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACA627coqcSKflMY7FeQ19hQTDk-CODwks5rs9KzgaJpZM4MugU4> .

-- Larry Vail JAARS Volunteer, DevOps Engineer JAARS Inc. Waxhaw, North Carolina

[longrunningprocess]

Neither @forevermatt nor I will be there on Fri, would it be worth it to you to do it over video conference?

That part aside, are we suggesting not closing this until we've reviewed it together? If so, I need to remove a comment from my current PR that will resolve this issue.

[lvail]

Not needed. You and Matt have stayed very close on this via github. I think it is more important to keep moving forward at this point.

…

On Wed, Apr 5, 2017 at 2:32 PM, billy clark ***@***.***> wrote: Neither @forevermatt <https://github.com/forevermatt> nor I will be there on Fri, would it be worth it to you to do it over video conference? That part aside, are we suggesting not closing this until we've reviewed it together? If so, I need to remove a comment from my current PR that will resolve this issue. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#32 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACA62z7oQPaTjpQHL5QE1ZfaduMSunukks5rs95IgaJpZM4MugU4> .

-- Larry Vail JAARS Volunteer, DevOps Engineer JAARS Inc. Waxhaw, North Carolina

[longrunningprocess]

@forevermatt I'll dig around a bit here and see if anything exists that will take in RAML and some local endpoint configs, e.g., hostname, and run some tests. I don't suppose you've seen anything crop up since you had this thought have you?

[forevermatt]

@wcjr That page is all I've really had to go on. Postman might be useful, but you have more experience than I do with that.

[longrunningprocess]

trying Postman and Abao on https://github.com/silinternational/idp-id-broker/tree/feature/battering-raml

[longrunningprocess]

well, after investigating some tools and considering the 1.0 version of RAML, I'm not sold this will be worth it in the end. The RAML file is a nice communication/collaboration tool when building up the API and possibly even documenting to new users (provided it were prettied up a bit) but the idea of running more tests against it and the infrastructure needed to accomplish that just isn't worth it to me. In the end, we have some good progress on the Behat tests for coverage. This just feels like another maintenance burden when/if things change. I'd like to either close this or leave it to you @forevermatt if you're compelled.

What do you think @fillup and @forevermatt ?

[forevermatt]

@wcjr Okay that's fine. Thanks again for looking into this.