Greg Dennis

I've created a branch (see commit ☝️) in my implementation that does the "[alternate iteration](https://github.com/json-schema-org/JSON-Schema-Test-Suite/issues/657#issuecomment-1504198836)" just above. The equality code wasn't hard to write, but I did need to do...

Just toying with this again, and expanding on the "minimal match" (I think I like "subset equality" better), I came up with this: ```json [ { "description": "readOnly generates its...

Maybe instead of moving tests, we can copy them in the new organization scheme to a new folder so that older test suite runners can still operate.

FWIW, mine results in this: ```json { "$id": "https://bundler.hyperjump.io/main(bundled)", "$defs": { "0708a427dd": { "$id": "https://bundler.hyperjump.io/main", "$schema": "https://json-schema.org/draft/2020-12/schema", "type": "object", "properties": { "foo": { "$ref": "/string" } } }, "7082beaf59": {...

> `$id`s identify a schema resource, which doesn't include embedded schemas. - @jdesrosiers But a schema resource does include _keys_. In your example, you've added `$defs`, which is _changing_ the...

> That sounds like the desired behavior to me. If I ask for _X_, I should get _X_. This is expected behavior. Getting _X-with-all-of-its-references_ is unexpected and surprising behavior.

I'm starting to reconsider whether bundling actually needs to be part of the core spec. Having different (especially _incompatibly_ different) approaches to bundling isn't great. --- The primary contradiction here...

> they don't produce identical output This is true, but then that's not a requirement of the spec. The spec does say this, though (running out of markdown formats): _A...

> However, that shouldn't mean that a schema couldn't be bundled in such a way that it could also be used as a drop-in replacement for a schema. If we...

> I would want to do some analysis to find out what percentage of "current" (2019-09 and above) implementations support a registry. @Julian is finishing up that work currently.