feat(metadata): rename @dbSetName to @collectionName and add the naming.collections pluralizer

Renames the unreleased @dbSetName to @collectionName (no alias) and registers it in
every port on object.entity, object.projection and object.report, marked
implementedIn csharp: the generated C# context takes it as the DbSet property name.
A project chooses the pluralizer an object without one takes its name from with
naming.collections in .metaobjects/config.json (standard by default, none, or one
the application registers); every port reads the key and refuses an unknown name.

The shared fixture object-port-scoped-attr-dbsetname becomes object-collection-name,
and object-port-scoped-attr-object-adapter keeps the port-scoped marker covered.

Author: dmealing

AGENTS.md

@@ -71,7 +71,7 @@ PyPI has had no product change since `0.25.0` — nothing is broken.
 - **Kotlin** — `codegen-kotlin` (KotlinPoet on JVM): entity + Exposed table + Spring controller + payload + relations + filter allowlist + validator + stored-proc + output-parser generators. `integration-tests-kotlin` runs the persistence-conformance corpus through Exposed against Testcontainers Postgres.
 
 **Cross-port conformance corpora** (every port runs the shared corpus):
-- Metamodel: `fixtures/conformance/` (454 fixtures; 38 shared corpora in total — per-corpus counts + the corpus x port matrix live in `docs/CONFORMANCE.md`). TS / C# / Java / Python all green.
+- Metamodel: `fixtures/conformance/` (455 fixtures; 38 shared corpora in total — per-corpus counts + the corpus x port matrix live in `docs/CONFORMANCE.md`). TS / C# / Java / Python all green.
 - Render: `fixtures/render-conformance/`. TS / C# / Java / Kotlin / Python byte-identical.
 - Persistence: `fixtures/persistence-conformance/`. **Query** scenarios run on every port (TS / C# / Java / Kotlin / Python), each provisioning its test DB by executing the committed, TS-produced `canonical/schema.postgres.sql` (Postgres only — Derby dropped for the cross-port query corpus, ADR-0015; the FR-044 `report-*` scenarios also run on SQLite, MySQL and D1 where a port has the engine, see [docs/CONFORMANCE.md](docs/CONFORMANCE.md)). The **migration** scenarios are exercised by **TS only** (TS owns schema migrations). **The corpus now gates WRITES, not just reads (SP-H):** an `op: roundtrip` scenario type INSERTs through each port's runtime/ORM write codec (NOT raw SQL), reads the row back, and asserts the wire-normalized value. The `AllTypes` entity (`roundtrip-all-types.yaml`) carries one field of **every** persistable `field.*` subtype — string/int/long/double/float/decimal/boolean/date/time/timestamp(+tz)/currency/enum/uuid/object — plus an **array-of-VO** `field.object @isArray @storage:jsonb` column (`labels`, written as 2-element / empty-`[]` / single-element arrays across the three rows) — so every subtype write+read (incl. the array-of-value-object jsonb codec) round-trips through every port against Testcontainers PG. (`field.byte`/`field.short`/`field.class` were cut as non-functional registration-only stubs — the matrix tracks only genuinely-supported subtypes; see `fixtures/registry-conformance/README.md` → "Per-subtype write-round-trip matrix".)
 - API-contract: `fixtures/api-contract-conformance/`. TS / C# / Java / Kotlin / Python all green — each port runs **two lanes**: a hand-rolled reference server AND its **generated** API artifact booted over HTTP (the deployed controller/routes; TS+C# full-stack vs Testcontainers PG, Java/Kotlin/Python generated controller + in-memory repo behind the consumer seam). The generated fan-out found 10 real deployment bugs golden snapshots missed. Three sub-corpora run the **generated lane only, on all five ports** — `write-through/`, `projection/` (F22: a view-only `object.projection` serves GET list + GET by id and answers every write verb with `405 {"error": "method_not_allowed"}`) and `report/` (FR-044: a view-backed `object.report` serves GET list, answers `POST` with that same `405`, and mounts no `/{id}` route). That is deliberate, not a gap: what is under test is whether a port's GENERATOR emits those routes, and a hand-rolled reference server would answer every scenario by construction. The `m2m/` sub-corpus also gates **TPH x M:N together** (base-declared, subtype-declared, abstract-mid-declared, a non-subtype source onto a subtype TARGET, and the cross-subtype source id answering `200 []`) — the two corpora were originally built disjoint (`tph/` had no relationships, `m2m/` no discriminators), which is precisely how that defect class survived.

CHANGELOG.md

@@ -16,12 +16,12 @@ _npm `1.3.0` (full lockstep across all 14 `@metaobjectsdev/*` publish candidates
 
 **`metamodelVersion` moves to `1.3`.** This release adds vocabulary (`field.duration`,
 `validator.exactlyOne`, `@constraintName`, `@condition`, `@maxLength` on `field.enum`,
-`@sequenceStart`, `@sequenceName`, `@flattenedColumnPrefix`, the `pascal` column rule, and
-`@dbSetName`, a C#-only attr) and nothing that loaded before stops loading. The registry gains two
+`@sequenceStart`, `@sequenceName`, `@flattenedColumnPrefix`, the `pascal` column rule and
+`@collectionName`) and nothing that loaded before stops loading. The registry gains two
 fields on a type or attr entry, carried into the registry manifest: `implementedIn` (cross-port
 vocabulary only the named ports implement so far, e.g. the `view.*` controls, `["typescript"]`) and
-`ports` (vocabulary only the named ports ever act on, e.g. `@dbSetName`, `["csharp"]`, and
-`@objectAdapter`, `["java", "kotlin"]`: the JVM runtime both share acts on it). The manifest's row
+`ports` (vocabulary only the named ports ever act on, e.g. `@objectAdapter`, `["java", "kotlin"]`:
+the JVM runtime both share acts on it). The manifest's row
 set grows by the `implementedIn` rows that used to be carved out. A port that reports `1.2` does not
 know that vocabulary; upgrade every port together.
 
@@ -50,6 +50,8 @@ Each item below is described in full under Added, Changed or Fixed; this list is
 - **`@sequenceStart` records a start; in 1.3.0 it does not move an existing counter.** A later
   release is planned to move a counter that sits below the declared start forward to it, never
   backward.
+- **`@dbSetName` is `@collectionName`.** The attr was never published under its first name, so
+  there is no alias: a model that set `@dbSetName` in a pre-release build renames it.
 - **`metamodelVersion` is `1.3`.** Upgrade every port together.
 
 ### Added
@@ -83,22 +85,29 @@ Each item below is described in full under Added, Changed or Fixed; this list is
   canonical grows by those rows. `meta types` marks them `[ts; pending elsewhere]` / `[java;
   pending elsewhere]` (it used to say `[ts-only]`, "the cross-port contract does not carry it"; its
   JSON change is under Changed), and the metamodel docs now cover the view controls.
-  `@objectAdapter` joins `@dbSetName` as port-scoped (`"ports": ["java", "kotlin"]`). C#, Java,
+  `@objectAdapter` is port-scoped (`"ports": ["java", "kotlin"]`, below). C#, Java,
   Kotlin and Python now register the `metaobjects-ui-web` attrs (`view.textarea @rows` and the
   `view.image` attrs), and C# and Python register `@object` / `@objectAdapter`, so a model that
   uses any of them loads strictly in every port. Gated by the conformance fixtures
   `view-kind1-controls`, `object-kind1-jvm-binding-attrs` and
   `error-kind1-view-unregistered-attr`, and by the registry canonical.
-- **Port-scoped vocabulary, and `@dbSetName`.** A registry attr may name the ports
-  that act on it (`"ports": ["csharp"]` on its `spec/metamodel` entry, carried into the registry
+- **Port-scoped vocabulary.** A registry attr may name the ports
+  that act on it (`"ports": ["java", "kotlin"]` on its `spec/metamodel` entry, carried into the registry
   manifest). Every port registers and type-checks it, so a model that carries it loads under every
   port's strict loader; a port not named does nothing with it. An attr the registry does not carry
   still fails, and a wrong value type fails in every port (Java now refuses a list given to a scalar
-  string attr, as the other ports did). `meta types` marks such an attr `[csharp-only]` and its JSON
-  carries `ports`. The first one is `@dbSetName`: the EF Core DbSet property name the generated C#
-  context gives an `object.entity`, `object.projection` or `object.report` (`"@dbSetName":
-  "Diagnosis"`), in place of the derived plural; the routes, the reverse finders and the api
-  reference read the property name through it.
+  string attr, as the other ports did). `meta types` marks such an attr `[java+kotlin]` (one port:
+  `[csharp-only]`) and its JSON carries `ports`. `@objectAdapter` is the one there is.
+- **`@collectionName`, and the `naming.collections` pluralizer.** An `object.entity`,
+  `object.projection` or `object.report` may name its collection in its own casing
+  (`"@collectionName": "CaseHistory"`); it need not be a grammatical plural. The generated C#
+  context uses it as the DbSet property name in place of the derived plural, and the routes, the
+  reverse finders and the api reference read the property through it. Absent, the project's
+  pluralizer derives the name: `"naming": { "collections": "standard" }` (the default) or `"none"`
+  in `.metaobjects/config.json`, or one the application registers (`registerPluralizer` in
+  TypeScript, `NamingStrategies.RegisterPluralizer` in C#, `NamingStrategies.registerPluralizer` in
+  Java, `register_pluralizer` in Python). Every port registers the attr and reads the key, and
+  refuses an unknown pluralizer when it reads the config.
 
 - **An int-backed enum can be an array.** `@intValueMap` together with `isArray: true` (the
   map on the field, or inherited from a shared enum) now loads in every port instead of
@@ -635,8 +644,8 @@ Each item below is described in full under Added, Changed or Fixed; this list is
 - **C#: one generated context maps each relation to one object.** Two concrete
   objects of one context that resolve to the same table, view, procedure or function in the same
   `@schema` are refused with `ERR_DUPLICATE_SQL_NAME`, naming both; before, they silently shared it
-  once `@dbSetName` split their DbSet names. The same name in two schemas is two relations, and an
-  object that extends another and inherits its source is one EF hierarchy. A `@dbSetName` that is not
+  once `@collectionName` split their DbSet names. The same name in two schemas is two relations, and an
+  object that extends another and inherits its source is one EF hierarchy. A `@collectionName` that is not
   a C# identifier, or names `Database`, `Model`, `ChangeTracker` or `ContextId`, is refused by name.
 - **TypeScript: the collision refusal checks only the names the run emits.** The pluralized module
   names are checked when entity modules, queries or a UI tier are wired, and the REST segments when
@@ -682,7 +691,7 @@ Each item below is described in full under Added, Changed or Fixed; this list is
   Changed.
 - **A generated name collides only in the scope that carries it**. The run-wide
   refusal of two objects whose short names pluralize alike is gone. C#: DbSet names are unique
-  within one generated context (the error says "rename one, or set @dbSetName"), route segments
+  within one generated context (the error says "rename one, or set @collectionName"), route segments
   within one routes entry, reverse finders within their class. TypeScript: the pluralized names
   within the entity modules' scope (one directory under the flat layout, one per package under
   the package layout), REST segments across the app. Python and JVM keep one run-wide scope,
@@ -858,7 +867,7 @@ Each item below is described in full under Added, Changed or Fixed; this list is
   cross-port wire contract and keeps the shared rule (`/api/diagnosises`) until every port
   changes together. An adopter that aliased the old DbSet name can drop the alias. **Upgrade
   note:** code that names `Diagnosises` or `FindDiagnosisesBy…` stops compiling; use the new name,
-  or set `"@dbSetName": "Diagnosises"` on the entity to keep the old one.
+  or set `"@collectionName": "Diagnosises"` on the entity to keep the old one.
 
 - **TypeScript: a derived flattened `field.object` no longer produces flat member keys (#524
   follow-up).** A flattened field that carries an `origin.*` owns no column (the Drizzle table and

docs/CONFORMANCE.md

@@ -26,7 +26,7 @@ regenerate with `ls -d fixtures/<corpus>/*/ | wc -l` for directory-shaped corpor
 
 | Corpus | Fixtures | TS | Java | Kotlin | C# | Python |
 |---|---|---|---|---|---|---|
-| [`fixtures/conformance/`](../fixtures/conformance/) (metamodel) | 454 | ✓ | ✓ | inherits via `metadata-ktx` | ✓ | ✓ |
+| [`fixtures/conformance/`](../fixtures/conformance/) (metamodel) | 455 | ✓ | ✓ | inherits via `metadata-ktx` | ✓ | ✓ |
 | [`fixtures/yaml-conformance/`](../fixtures/yaml-conformance/) | 16 | 16 / 16 | 15 / 16 (1 ledgered: `yaml-quoted-leading-zero` — Java pipeline strips quotes off `"007"`) | inherits via Java | 15 / 16 (1 ledgered: `error-yaml-coerced-hex-in-string` — YamlDotNet doesn't coerce `0xFF`) | 16 / 16 |
 | [`fixtures/verify-conformance/`](../fixtures/verify-conformance/) | 45 | ✓ | ✓ | inherits via Java | ✓ | ✓ |
 | [`fixtures/verify-strict-conformance/`](../fixtures/verify-strict-conformance/) | 1 | ✓ | — | — | — | ✓ |
@@ -155,7 +155,7 @@ canonical (`fixtures/registry-conformance/expected-registry.json`).
   `registry-conformance.test.ts` in TypeScript; `RegistryManifestConformanceTest` in Java and
   Kotlin; `test_registry_conformance.py` in Python).
 - *Port-scoped, for comparison* (`ports`, language-specific by nature, never "pending"):
-  `@dbSetName` (`["csharp"]`) and `@objectAdapter` (`["java", "kotlin"]`).
+  `@objectAdapter` (`["java", "kotlin"]`).
 
 **Native index vocabulary and `@description` in the ORM models.** `meta migrate` creates
 every index (`identity.secondary`, `index.lookup`, a field's `@unique`, with `@where`, `@using`,
@@ -351,7 +351,7 @@ unit-test runners (`bun test`, `dotnet test`, `pytest`, `mvn test`) pull Docker.
 
 ## Fixture-to-doc mapping
 
-### `fixtures/conformance/` — metamodel loader + canonical serializer (454)
+### `fixtures/conformance/` — metamodel loader + canonical serializer (455)
 
 | Fixture prefix | Feature doc |
 |---|---|
@@ -376,6 +376,7 @@ unit-test runners (`bun test`, `dotnet test`, `pytest`, `mvn test`) pull Docker.
 | `requirement-*`, `error-unknown-attr-requirement` | [features/requirements.md](features/requirements.md) (vocabulary only — the gate has its own corpus, `requirement-check-conformance/`; see "Split coverage" above) |
 | `reporting-*`, `error-dimension-*`, `error-measure-*`, `error-ratio-*`, `error-segment-*`, `error-report-*`, `error-relative-date-*` | [features/reporting.md](features/reporting.md) (the `dimension` / `measure` / `segment` / `object.report` vocabulary and its load-time rules; design in the [FR-044 spec](superpowers/specs/2026-10-02-fr-044-core-reporting-design.md)) |
 | `provider-extension-*`, `provider-custom-vocabulary-*` | [features/extending-with-providers.md](features/extending-with-providers.md) (a project's own provider: a new subtype plus attributes on an existing type load with it and are refused without it, every port writing the provider in its own language) |
+| `object-collection-name` | [features/migrations/physical-naming.md](features/migrations/physical-naming.md) (`@collectionName` and the `collections` pluralizer) |
 | `smoke-empty-metadata` | [features/entities.md](features/entities.md) |
 
 ### `fixtures/yaml-conformance/` (16)
@@ -712,7 +713,7 @@ JSON and verify payload feeds it unchanged.
 
 ## Orphaned fixtures (tested but not yet documented)
 
-The fixtures in the nine corpora mapped above (metamodel 454 + yaml 16 + verify 45
+The fixtures in the nine corpora mapped above (metamodel 455 + yaml 16 + verify 45
 + render 18 + persistence 62 + api-contract 123 + source-resolution 25 + scope 10 +
 dependency 47) each map to a feature doc, and so do the two requirement corpora, `shared-model-conformance`, the
 `cube-model` corpus and `explorer-conformance`, whose case lists live in their own READMEs. None are orphaned today. The remaining

docs/features/migrations/physical-naming.md

@@ -27,6 +27,7 @@ every runtime `ObjectManager` and `verify` read it, so they always produce the s
 |---|---|---|---|
 | `columns` | `snake_case`, `literal`, `kebab-case`, `pascal`, or a name your application registered | a field with no `@column` | `snake_case` |
 | `tables` | `snake_case`, `literal`, or a name your application registered | a source with no `@table`/`@view`/`@proc`/`@function`: its `name`, else the object's name made plural | `snake_case` |
+| `collections` | `standard`, `none`, or a name your application registered | an object with no `@collectionName`: the collection name the generated C# DbSet takes | `standard` |
 
 An unknown key fails when the file is read, and so does a name no registered strategy answers.
 Index and constraint names are unchanged.
@@ -61,7 +62,9 @@ and is `pascal` otherwise:
 
 Table rules register the same way (`namingStrategies.tables`, `RegisterTables`, `registerTables`,
 `register_table_naming_rule`) and name a source with no physical name; the entity-name fallback
-is still made plural after the rule. A built-in name cannot be replaced. Register the same
+is still made plural after the rule. Pluralizers register the same way (`registerPluralizer`, `RegisterPluralizer`,
+`registerPluralizer`, `register_pluralizer`): an object name in, in its declared casing, its
+collection name out. `standard` is the cross-port plural and `none` keeps the name as it is. A built-in name cannot be replaced. Register the same
 strategy in every port that reads the database, because each port resolves the names itself.
 Every port looks a strategy up by name each time it applies one, so registering again under the
 same name replaces it for everything that runs afterwards. A test that registers its own calls

examples/showcase/site-payload.json

@@ -7,7 +7,7 @@
     "metamodel": "1.3"
   },
   "counts": {
-    "fixtures": 454,
+    "fixtures": 455,
     "corpora": 38,
     "baseTypes": 17
   },

fixtures/conformance/error-port-scoped-attr-wrong-type/input/meta.diagnosis.json

@@ -2,7 +2,7 @@
   "metadata.root": {
     "package": "acme::clinical",
     "children": [
-      { "object.entity": { "name": "Diagnosis", "@dbSetName": ["Diagnosis"], "children": [
+      { "object.entity": { "name": "Diagnosis", "@objectAdapter": ["com.acme.clinical.DiagnosisAdapter"], "children": [
         { "field.long": { "name": "id" } },
         { "identity.primary": { "name": "pk", "@fields": ["id"] } }
       ] } }

…port-scoped-attr-dbsetname/expected.json …nce/object-collection-name/expected.jsonfixtures/conformance/object-port-scoped-attr-dbsetname/expected.json renamed to fixtures/conformance/object-collection-name/expected.json fixtures/conformance/object-port-scoped-attr-dbsetname/expected.json renamed to fixtures/conformance/object-collection-name/expected.json

@@ -5,7 +5,7 @@
       {
         "object.entity": {
           "name": "Diagnosis",
-          "@dbSetName": "Diagnosis",
+          "@collectionName": "Diagnosis",
           "children": [
             {
               "source.rdb": {
@@ -51,7 +51,7 @@
       {
         "object.projection": {
           "name": "DiagnosisSummary",
-          "@dbSetName": "DiagnosisSummary",
+          "@collectionName": "DiagnosisSummary",
           "children": [
             {
               "source.rdb": {
@@ -70,7 +70,7 @@
       {
         "object.report": {
           "name": "DiagnosisCounts",
-          "@dbSetName": "DiagnosisCounts",
+          "@collectionName": "DiagnosisCounts",
           "@dimensions": [
             "code"
           ],

…attr-dbsetname/input/meta.diagnosis.json …ollection-name/input/meta.diagnosis.jsonfixtures/conformance/object-port-scoped-attr-dbsetname/input/meta.diagnosis.json renamed to fixtures/conformance/object-collection-name/input/meta.diagnosis.json fixtures/conformance/object-port-scoped-attr-dbsetname/input/meta.diagnosis.json renamed to fixtures/conformance/object-collection-name/input/meta.diagnosis.json

@@ -2,19 +2,19 @@
   "metadata.root": {
     "package": "acme::clinical",
     "children": [
-      { "object.entity": { "name": "Diagnosis", "@dbSetName": "Diagnosis", "children": [
+      { "object.entity": { "name": "Diagnosis", "@collectionName": "Diagnosis", "children": [
         { "source.rdb": { "@table": "diagnoses" } },
         { "field.long": { "name": "id" } },
         { "field.string": { "name": "code" } },
         { "identity.primary": { "name": "pk", "@fields": ["id"] } },
         { "dimension.attribute": { "name": "code", "@of": "Diagnosis.code" } },
         { "measure.aggregate": { "name": "diagnoses", "@agg": "count", "@of": "Diagnosis.id" } }
       ] } },
-      { "object.projection": { "name": "DiagnosisSummary", "@dbSetName": "DiagnosisSummary", "children": [
+      { "object.projection": { "name": "DiagnosisSummary", "@collectionName": "DiagnosisSummary", "children": [
         { "source.rdb": { "@kind": "view", "@view": "v_diagnosis_summary" } },
         { "field.string": { "name": "code" } }
       ] } },
-      { "object.report": { "name": "DiagnosisCounts", "@dbSetName": "DiagnosisCounts", "@from": "Diagnosis",
+      { "object.report": { "name": "DiagnosisCounts", "@collectionName": "DiagnosisCounts", "@from": "Diagnosis",
         "@dimensions": ["code"], "@measures": ["diagnoses"], "children": [
         { "source.rdb": { "@kind": "view", "@view": "v_diagnosis_counts" } }
       ] } }