feat: move root tags and externalDocs to the info object #794
Conversation
magicmatatjahu
requested review from
fmvilas,
derberg,
dalelane and
asyncapi-bot-eve
as code owners
|
Kudos, SonarCloud Quality Gate passed!
|
There was a problem hiding this comment.
I think this makes sense.
I know I'm normally the one who is pointing out changes that would make AsyncAPI inconsistent with OpenAPI for things that aren't specific to event driven apps, but I think v3 is going to be different enough from OAI that maybe we just do what we think makes the most sense.
|
@dalelane Thanks for your comment! I don't think it causes incompatibility with OpenAPI, because e.g. from the beginning tags in AsyncAPI are used in a different way than in OpenAPI. In OpenAPI you define tags as objects on the root, and in operations you refer to them only by name. In AsyncAPI tags as objects can be used in all places where tags may occur. The only problem may arise if someone wants to refer an Info Object from AsyncAPI in OpenAPI, then yes, this will be a problem, because the validation should inform about the use of additional fields that are not supported ( |
magicmatatjahu
force-pushed
the
next-major/move-tags-external-docs-to-info
branch
from
fb89520
to
92f59cb
Compare
|
@KhudaDad414 Could you check the failing |
|
Is anything stopping us from merging this? 🤔 |
yes, failing |
|
I'll see if we have more luck with re-running the job quickly first |
|
I thought that problem was resolved by this PR f269234 but unfortunately no. |
|
I don't think the problem is actually with the links being broken - the errors are all HTTP-429 - essentially GitHub complaining about too many requests coming too quickly. |
Ideally, the action would accept a header with a valid GH token, so request rate can be higher than actually needed. However there is no such option afaik. We can fix it by follow their recommendations at: https://github.com/gaurav-nelson/github-action-markdown-link-check#too-many-requests |
There was a problem hiding this comment.
LGTM
I recommend ignoring failing links. This CI check is not mandatory as still in testing phase. Pr's can be merged if it gets red. Also Khuda that is keeping an eye on it is not available at the moment. And also these errors are super strange as they never happened to us before, might be some random stuff.
|
/dnm we need changes in JSON Schema and the Parser |
I will create PR for json-schemas. For parser the current "changes" in spec is reflected in the next parserJS version - https://github.com/asyncapi/parser-js/blob/next-major/src/models/v2/info.ts#L64 You can see that it's for 2.X.X, but due to fact that we don't know how 3.0.0 finally will look like the models are only made for version 2.X.X. |
|
Changes in json-schemas asyncapi/spec-json-schemas#244 |
|
Kudos, SonarCloud Quality Gate passed!
|
fmvilas
force-pushed
the
next-major-spec
branch
from
8805c88
to
624ad52
Compare
magicmatatjahu
changed the title
magicmatatjahu
force-pushed
the
next-major/move-tags-external-docs-to-info
branch
from
0a25c50
to
09270e2
Compare
|
Kudos, SonarCloud Quality Gate passed!
|
|
@magicmatatjahu do you want to merge using |
|
@fmvilas by |
|
/rtm |
|
@derberg |
|
@derberg Thanks! |
|
🎉 This PR is included in version 3.0.0-next-major-spec.4 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
Co-authored-by: Jonas Lagoni <jonas-lt@live.dk>
title: "move root tags and externalDocs to the info object"
This PR moves
tagsandexternalDocs(describing applications) from root to theInfo Object. From an application description point of view, this makes much more sense, because all the metadata is in one place. This is more of a stylistic change, however it introduces a breaking change.PR for spec-json-schemas repo asyncapi/spec-json-schemas#244