feat!: introduce split definitions

[jonaslagoni]

Description
This PR introduces split definitions and bundling behavior to provide better maintainability of the JSON Schema files.

Changes:

• Specification extension pattern property regex needs to be valid against u flag. This means that ^x-[\\w\\d\\.\\-\\_]+$ is changed to ^x-[\\w\\d.\\-_]+$.
• Specification components pattern property regex needs to be valid against u flag. This means that ^x-[\\w\\d\\.\\-\\_]+$ is changed to ^x-[\\w\\d.\\-_]+$
• Pattern property regex for parameter and correlationId location needs to be valid against u flag. This means that ^\\$message\\.(header|payload)\\#(\\/(([^\\/~])|(~[01]))*)* is changed to ^\\$message\\.(header|payload)#(\\/(([^\\/~])|(~[01]))*)*
• Fixes $ref is incorrectly defined for a server.
• All schema id's $id now follow http://asyncapi.com/definitions/<version>/<name>.json instead of <name>. The reason for this change is to follow an accurate bundling behavior from JSON Schema.

As this is a breaking change you can read the migration guide to get more information.

Related issue(s)
Fixes #120

[smoya]

@jonaslagoni would you mind changing the PR title to what the main commit says:

feat!: split out definitions

[jonaslagoni]

@dalelane @derberg @fmvilas deciding on whether we should merge this will be better now rather than later (both I and @smoya need it for the next spec versions, or at least would be preferred). Anything you see as being blocked in order to merge this?

[derberg]

@jonaslagoni I'm ok for 3.0 release. I assume it is about splitting definitions. Just please provide more accurate title that will be used in commit and then in release notes. Also make it clear in description, what change is released and what breaking changes are actually expected

[jonaslagoni]

@jonaslagoni I'm ok for 3.0 release. I assume it is about splitting definitions. Just please provide more accurate title that will be used in commit and then in release notes. Also make it clear in description, what change is released and what breaking changes are actually expected

Will do that 👍

I am gonna resolve these conflicts as well, might require one additional PR to fix though.

[jonaslagoni]

PR is up that updates next-major with master #187

[jonaslagoni]

Added a better description of the changes, let me know if anything else have to change 👍

[derberg]

@jonaslagoni looks perfect now, now you just need to solve conflicts

also, can you specify what is next plan, do we just update new 3.0 in parser just like that? no chances are needed? Did you try it already with the release candidate?

[jonaslagoni]

also, can you specify what is next plan, do we just update new 3.0 in parser just like that? no chances are needed? Did you try it already with the release candidate?

Changes are needed but they are simple 🙂 I will make sure to update this PR as soon as it's released: asyncapi/parser-js#423

[jonaslagoni]

@jonaslagoni looks perfect now, now you just need to solve conflicts

@derberg I don't think I can solve those conflicts... I tried with #187 but it does not seem to apply them 🧐 And I do not have write access to the repo, so you will have to resolve them I think...

[jonaslagoni]

Changes are needed but they are simple 🙂 I will make sure to update this PR as soon as it's released: asyncapi/parser-js#423

Or are they 🧐 Thought i did try to use the change, but apparently I did not. Looking into it.

[derberg]

you need to fork the repo, create a branch out from next-major, make sure it is up to date, also make sure your master is up to date. Them merge master to next-major in your fork, solve conflicts and then open a PR from your fork to upstream next-major. Makes sense?

[jonaslagoni]

you need to fork the repo, create a branch out from next-major, make sure it is up to date, also make sure your master is up to date. Them merge master to next-major in your fork, solve conflicts and then open a PR from your fork to upstream next-major. Makes sense?

That's what I did in #187

[derberg]

please double-check, as the PR you link to doesn't show any changes to package.json while here we have conflicts with it

I recommend fresh fork, as something in your current fork might not be up to date with latest master, so you do not have last commit from Sergio from last week

[smoya]

Merge! 😂 You don't even know how much I'm looking forward to these changes (I even started replacing by webpack.config.js your actual released version) :D

As mentioned by @derberg, the release is gonna happen this week, so you won't need to wait so much more for having this PR merged and released. Does it sounds reasonable?

[Dzixxx]

Sounds great, srsly! <3

[jonaslagoni]

Gonna go through the changes in 2.4 and create a pull request so we can finish this 🤞

[jonaslagoni]

PR is up: #212

[jonaslagoni]

@derberg @fmvilas @dalelane
Blocker to finish this PR: #212

[jonaslagoni]

The last conflict does not seem like I will be able to solve.

But I dont think there is anything else that need to happen for this to be ready.

[derberg]

Any help needed? what last conflict you have in mind that can't be solved. We won't be able to merge without solving conflicts 😄

[jonaslagoni]

Any help needed? what last conflict you have in mind that can't be solved. We won't be able to merge without solving conflicts 😄

I do yes, it's the same as last time, where my PRs dont solve the conflicts 🙃 It required codeowner to manually resolve the conflicts.

[sonarcloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
0.0% Duplication

[derberg]

I'm ok to merge and release major

@dalelane @fmvilas ?

[fmvilas]

Yup 🚀

[jonaslagoni]

😮

Is it happening? 😆

[jonaslagoni]

Here goes nothing

/rtm

[jonaslagoni]

Damn, when targeted branches are removed, remaining PRs are closed as well with no way to retarget them...

[asyncapi-bot]

🎉 This PR is included in version 3.0.0 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀

[derberg]

@jonaslagoni oh, bot removes it by default. I restored it