feat(spec,schemas): declare a publisher's event types in the manifest - #106
Open
ucekmez wants to merge 1 commit into
Open
feat(spec,schemas): declare a publisher's event types in the manifest#106ucekmez wants to merge 1 commit into
ucekmez wants to merge 1 commit into
Conversation
9 tasks
The manifest has 24 properties and none of them says what the publisher emits. An agent could discover THAT a publisher speaks EEP and WHERE its endpoints are, but not WHAT it will send — so it had to read human documentation or subscribe to a wildcard and find out empirically. For a protocol whose premise is autonomous agents following entities, that is the wrong shape. Nothing existing filled the gap. §9's "standard event catalog" is a table in the specification: not fetchable, not versioned per publisher, not machine-readable. `capabilities_query_url` covers gated *capabilities*, not event types. The envelope's `eep_known_event_types` puts a discovery concern on the per-event hot path, which is the wrong layer. Changes: - `event_types` on the manifest: `type` (required, §8 reverse-DNS pattern), plus optional `description`, `dataschema`, `since` and `deprecated`. - `event_catalog_url` for publishers whose catalog is too large or too volatile to inline, paginated like §12.4. - New §12.3.2 with a rule worth stating explicitly: declaring a type is not a promise to emit it, omitting one is not a promise never to, and subscribers MUST tolerate an undeclared type rather than rejecting the delivery. A catalog is discovery metadata, not a wire constraint. - `dataschema` here SHOULD match the `dataschema` attribute set on the events themselves (§7.1), so the payload contract is both discoverable and carried with each event. - `@eep-dev/middleware` publishes its configured `eventTypes` in the manifest, with a test asserting the emitted types match the §8 pattern — an invalid type would make the whole manifest fail schema validation, which is worse than having no catalog. - Conformance vectors: the positive discovery manifest gains a catalog, and a negative vector pins that a non-conformant event type is rejected. Optional throughout, so existing manifests stay valid. Refs: EEP audit 2026-08 finding B4 Signed-off-by: Ugur Cekmez <ucekmez@gmail.com>
ucekmez
force-pushed
the
feat/ietf-ratelimit-headers
branch
from
August 26, 2026 19:39
6d6fb69 to
51e2c2b
Compare
ucekmez
force-pushed
the
feat/manifest-event-catalog
branch
from
August 26, 2026 19:39
ca5719c to
a741bb4
Compare
9 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
PR 14 of a stacked series. Base is #105. Not for merge without review.
eep-manifest.jsonhas 24 properties and none of them says what the publisher emits.An agent could discover that a publisher speaks EEP and where its endpoints are, but not what it will send — so it had to read human documentation, or subscribe to a wildcard and find out empirically. For a protocol whose premise is autonomous agents following entities over time, that is the wrong shape.
Nothing existing filled the gap:
capabilities_query_url(§12.4) covers gated capabilities, not event types.eep_known_event_typesputs a discovery concern on the per-event hot path — wrong layer.What changed
{ "event_types": [ { "type": "com.example.entity.updated", "description": "A field on the entity profile changed.", "dataschema": "https://example.com/schemas/entity.updated/v2.json", "since": "0.1" }, { "type": "com.example.trust.changed", "deprecated": true } ] }event_types—typerequired (§8 reverse-DNS pattern), plus optionaldescription,dataschema,since,deprecated.event_catalog_urlfor publishers whose catalog is too large or volatile to inline, paginated like §12.4.dataschemahere SHOULD match thedataschemaattribute set on the events themselves (feat(spec,schemas): adopt the CloudEvents attributes EEP had skipped #100, §7.1), so the payload contract is both discoverable and carried with each event.@eep-dev/middlewarepublishes its configuredeventTypesin the manifest.Scope
Checklist
Verification
@eep-dev/middlewaretests/compliance-cli --fixturestests/cross-impl/test_conformance_fixtures.pycodegen-schema-types --checkNotes for reviewers
One test exists because the feature can backfire. The middleware maps its configured
eventTypesstraight into the manifest — so a publisher configured with an event type that violates §8's reverse-DNS pattern would emit a manifest that fails whole-document schema validation (added in #95). That is a worse outcome than having no catalog at all. There's a test pinning that emitted types match the pattern, and a negative conformance vector (Com.Example.Entity.Updated) pinning that the schema rejects a bad one.eep_known_event_typesis left in place. It is arguably now redundant with the manifest catalog, but removing an envelope attribute is a breaking change and belongs in its own discussion.No Python middleware parity for the manifest change — flagging rather than skipping silently.