Deprecated clarfications around arrays

[philsturgeon]

No description provided.

[philsturgeon]

I think I fixed it?

[Relequestual]

I'm unsure what it ought to mean...

My feeling is that people will only look to use this if they know what they want to indicate.

@philsturgeon Say the first item in items is deprecated. Does that mean the second item then becomes the first version? I can't see it means anything else. Right?

[handrews]

@Relequestual some languages allow undefined array positions.

But really it means "it won't look like this anymore soon", it does not say what it will look like. The default assumption is that it goes away, but you could also just use it to indicate an overall change. I guess.

[handrews]

@philsturgeon @Relequestual what still needs to be done here?

[philsturgeon]

I am not sure, Ben thought there was further work to be done but I thought you answered his questions, so... break the deadlock with a thumb? Or, Ben, if you can let me know of a specific actionable thing?

…

-- Phil Sturgeon @philsturgeon

On Jul 11, 2019, at 22:50, Henry Andrews ***@***.***> wrote: @philsturgeon @Relequestual what still needs to be done here? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[Relequestual]

@philsturgeon I guess the objective of this PR is to align with the expectations of people who see this when enclosed in an OpenAPI spec document, right?

As such, can you encourage someone from the OpenAPI spec team to take a little look and agree that this change meets said objective? I took a little look at the spec, but I can't see this expectation. Am I reading it wrong? Does this expectation come from somewhere else?

I've made a further review. One question, one suggested language change.
I think this is certinly clearer from where we started! Direction of travel is good! =]

[handrews]

@philsturgeon @Relequestual any update on this?

[philsturgeon]

I was trying to get an OAI person to review this. I'll bump it on their slack. We'll give it a few more days then just merge.

[Relequestual]

I've added a comment on the review.

My query is regarding, if deprecated changes behaviour when items is an array of schemas, to indicate that one of the subschemas will be deprecated, how is the position known?

See my review comment.