Compare commits

..

4 Commits

Author SHA1 Message Date
665a821bf9 version: 1.0.110 2026-04-10 01:06:16 -04:00
be78af1507 more tests 2026-04-10 01:06:02 -04:00
3cca5ef2d5 checkpoint 2026-04-09 19:55:35 -04:00
5f45df6c11 checkpoint 2026-04-09 18:39:52 -04:00
16 changed files with 1596 additions and 1158 deletions

View File

@ -84,11 +84,26 @@ Punc completely abandons the standard JSON Schema `$ref` keyword. Instead, it ov
* **Strict Array Constraint**: To explicitly prevent mathematically ambiguous Multiple Inheritance, a `type` array is strictly constrained to at most **ONE** Custom Object Pointer. Defining `"type": ["person", "organization"]` will intentionally trigger a fatal database compilation error natively instructing developers to build a proper tagged union (`oneOf`) instead.
### Polymorphism (`$family` and `oneOf`)
Polymorphism is how an object boundary can dynamically take on entirely different shapes based on the payload provided at runtime.
* **`$family` (Target-Based Polymorphism)**: An explicit Punc compiler macro instructing the database compiler to dynamically search its internal `db.descendants` registry and find all physical schemas that mathematically resolve to the target.
* *Across Tables (Vertical)*: If `$family: entity` is requested, the payload's `type` field acts as the discriminator, dynamically routing to standard variations like `organization` or `person` spanning multiple Postgres tables.
* *Single Table (Horizontal)*: If `$family: widget` is requested, the router explicitly evaluates the Dot Convention dynamically. If the payload possesses `"type": "widget"` and `"kind": "stock"`, the router mathematically resolves to the string `"stock.widget"` and routes exclusively to that explicit `JSPG` schema.
* **`oneOf` (Strict Tagged Unions)**: A hardcoded array of JSON Schema candidate options. Punc strictly bans mathematical "Union of Sets" evaluation. Every `oneOf` candidate item MUST either be a pure primitive (`{ "type": "null" }`) or a user-defined Object Pointer providing a specific discriminator (e.g., `{ "type": "invoice_metadata" }`). This ensures validations remain pure $O(1)$ fast-paths and allows the Dart generator to emit pristine `sealed classes`.
Polymorphism is how an object boundary can dynamically take on entirely different shapes based on the payload provided at runtime. Punc utilizes the static database metadata generated from Postgres (`db.types`) to enforce these boundaries deterministically, rather than relying on ambiguous tree-traversals.
* **`$family` (Target-Based Polymorphism)**: An explicit Punc compiler macro instructing the engine to resolve dynamic options against the registered database `types` variations or its inner schema registry. It uses the exact physical constraints of the database to build SQL and validation routes.
* **Scenario A: Global Tables (Vertical Routing)**
* *Setup*: `{ "$family": "organization" }`
* *Execution*: The engine queries `db.types.get("organization").variations` and finds `["bot", "organization", "person"]`. Because organizations are structurally table-backed, the `$family` automatically uses `type` as the discriminator.
* *Options*: `bot` -> `bot`, `person` -> `person`, `organization` -> `organization`.
* **Scenario B: Prefixed Tables (Vertical Projection)**
* *Setup*: `{ "$family": "light.organization" }`
* *Execution*: The engine sees the prefix `light.` and base `organization`. It queries `db.types.get("organization").variations` and dynamically prepends the prefix to discover the relevant UI schemas.
* *Options*: `person` -> `light.person`, `organization` -> `light.organization`. (If a projection like `light.bot` does not exist in `db.schemas`, it is safely ignored).
* **Scenario C: Single Table Inheritance (Horizontal Routing)**
* *Setup*: `{ "$family": "widget" }` (Where `widget` is a table type but has no external variations).
* *Execution*: The engine queries `db.types.get("widget").variations` and finds only `["widget"]`. Since it lacks table inheritance, it is treated as STI. The engine scans the specific, confined `schemas` array directly under `db.types.get("widget")` for any `$id` terminating in the base `.widget` (e.g., `stock.widget`). The `$family` automatically uses `kind` as the discriminator.
* *Options*: `stock` -> `stock.widget`, `tasks` -> `tasks.widget`.
* **`oneOf` (Strict Tagged Unions)**: A hardcoded list of candidate schemas. Unlike `$family` which relies on global DB metadata, `oneOf` forces pure mathematical structural evaluation of the provided candidates. It strictly bans typical JSON Schema "Union of Sets" fallback searches. Every candidate MUST possess a mathematically unique discriminator payload to allow $O(1)$ routing.
* **Disjoint Types**: `oneOf: [{ "type": "person" }, { "type": "widget" }]`. The engine succeeds because the native `type` acts as a unique discriminator (`"person"` vs `"widget"`).
* **STI Types**: `oneOf: [{ "type": "heavy.person" }, { "type": "light.person" }]`. The engine succeeds. Even though both share `"type": "person"`, their explicit discriminator is `kind` (`"heavy"` vs `"light"`), ensuring unique $O(1)$ fast-paths.
* **Conflicting Types**: `oneOf: [{ "type": "person" }, { "type": "light.person" }]`. The engine **fails compilation natively**. Both schemas evaluate to `"type": "person"` and neither provides a disjoint `kind` constraint, making them mathematically ambiguous and impossible to route in $O(1)$ time.
### Conditionals (`cases`)
Standard JSON Schema forces developers to write deeply nested `allOf` -> `if` -> `properties` blocks just to execute conditional branching. **JSPG completely abandons `allOf` and this practice.** For declarative business logic and structural mutations conditionally based upon property bounds, use the top-level `cases` array.
@ -225,7 +240,7 @@ The Queryer transforms Postgres into a pre-compiled Semantic Query Engine, desig
* **Type Casting**: Safely resolves dynamic combinations by casting values instantly into the physical database types mapped in the schema (e.g. parsing `uuid` bindings to `::uuid`, formatting DateTimes to `::timestamptz`, and numbers to `::numeric`).
* **Polymorphic SQL Generation (`$family`)**: Compiles `$family` properties by analyzing the **Physical Database Variations**, *not* the schema descendants.
* **The Dot Convention**: When a schema requests `$family: "target.schema"`, the compiler extracts the base type (e.g. `schema`) and looks up its Physical Table definition.
* **Multi-Table Branching**: If the Physical Table is a parent to other tables (e.g. `organization` has variations `["organization", "bot", "person"]`), the compiler generates a dynamic `CASE WHEN type = '...' THEN ...` query, expanding into `JOIN`s for each variation.
* **Multi-Table Branching**: If the Physical Table is a parent to other tables (e.g. `organization` has variations `["organization", "bot", "person"]`), the compiler generates a dynamic `CASE WHEN type = '...' THEN ...` query, expanding into sub-queries for each variation. To ensure safe resolution, the compiler dynamically evaluates correlation boundaries: it attempts standard Relational Edge discovery first. If no explicit relational edge exists (indicating pure Table Inheritance rather than a standard foreign-key graph relationship), it safely invokes a **Table Parity Fallback**. This generates an explicit ID correlation constraint (`AND inner.id = outer.id`), perfectly binding the structural variations back to the parent row to eliminate Cartesian products.
* **Single-Table Bypass**: If the Physical Table is a leaf node with only one variation (e.g. `person` has variations `["person"]`), the compiler cleanly bypasses `CASE` generation and compiles a simple `SELECT` across the base table, as all schema extensions (e.g. `light.person`, `full.person`) are guaranteed to reside in the exact same physical row.
---

43
debug.log Normal file

File diff suppressed because one or more lines are too long

View File

@ -196,29 +196,37 @@
{
"description": "Horizontal $family Routing (Virtual Variations)",
"database": {
"types": [
{
"name": "widget",
"variations": ["widget"],
"schemas": [
{
"$id": "widget",
"type": "object",
"properties": {
"type": { "type": "string" }
}
},
{
"$id": "stock.widget",
"type": "widget",
"properties": {
"kind": { "type": "string" },
"amount": { "type": "integer" }
}
},
{
"$id": "super_stock.widget",
"type": "stock.widget",
"properties": {
"super_amount": { "type": "integer" }
}
}
]
}
],
"schemas": [
{
"$id": "widget",
"type": "object",
"properties": {
"type": { "type": "string" }
}
},
{
"$id": "stock.widget",
"type": "widget",
"properties": {
"kind": { "type": "string" },
"amount": { "type": "integer" }
}
},
{
"$id": "super_stock.widget",
"type": "stock.widget",
"properties": {
"super_amount": { "type": "integer" }
}
},
{
"$id": "family_widget",
"$family": "widget"

View File

@ -25,11 +25,20 @@
]
},
{
"name": "get_person",
"name": "get_light_organization",
"schemas": [
{
"$id": "get_person.response",
"$family": "person"
"$id": "get_light_organization.response",
"$family": "light.organization"
}
]
},
{
"name": "get_full_organization",
"schemas": [
{
"$id": "get_full_organization.response",
"$family": "full.organization"
}
]
},
@ -44,6 +53,18 @@
}
}
]
},
{
"name": "get_widgets",
"schemas": [
{
"$id": "get_widgets.response",
"type": "array",
"items": {
"$family": "widget"
}
}
]
}
],
"enums": [],
@ -260,7 +281,9 @@
"type",
"name",
"archived",
"created_at"
"created_at",
"token",
"role"
],
"grouped_fields": {
"entity": [
@ -273,7 +296,8 @@
"name"
],
"bot": [
"token"
"token",
"role"
]
},
"field_types": {
@ -282,12 +306,25 @@
"archived": "boolean",
"name": "text",
"token": "text",
"role": "text",
"created_at": "timestamptz"
},
"schemas": [
{
"$id": "bot",
"type": "organization",
"properties": {
"token": {
"type": "string"
},
"role": {
"type": "string"
}
}
},
{
"$id": "light.bot",
"type": "organization",
"properties": {
"token": {
"type": "string"
@ -360,8 +397,15 @@
},
{
"$id": "light.person",
"type": "person",
"properties": {}
"type": "organization",
"properties": {
"first_name": {
"type": "string"
},
"last_name": {
"type": "string"
}
}
},
{
"$id": "full.person",
@ -850,6 +894,70 @@
"variations": [
"order_line"
]
},
{
"name": "widget",
"hierarchy": [
"widget",
"entity"
],
"fields": [
"id",
"type",
"kind",
"archived",
"created_at"
],
"grouped_fields": {
"entity": [
"id",
"type",
"archived",
"created_at"
],
"widget": [
"kind"
]
},
"field_types": {
"id": "uuid",
"type": "text",
"kind": "text",
"archived": "boolean",
"created_at": "timestamptz"
},
"variations": [
"widget"
],
"schemas": [
{
"$id": "widget",
"type": "entity",
"properties": {
"kind": {
"type": "string"
}
}
},
{
"$id": "stock.widget",
"type": "widget",
"properties": {
"kind": {
"const": "stock"
}
}
},
{
"$id": "tasks.widget",
"type": "widget",
"properties": {
"kind": {
"const": "tasks"
}
}
}
]
}
]
},
@ -1004,17 +1112,17 @@
" 'target', CASE",
" WHEN entity_11.target_type = 'address' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_17.archived,",
" 'city', address_16.city,",
" 'created_at', entity_17.created_at,",
" 'id', entity_17.id,",
" 'type', entity_17.type",
" 'archived', entity_13.archived,",
" 'city', address_12.city,",
" 'created_at', entity_13.created_at,",
" 'id', entity_13.id,",
" 'type', entity_13.type",
" )",
" FROM agreego.address address_16",
" JOIN agreego.entity entity_17 ON entity_17.id = address_16.id",
" FROM agreego.address address_12",
" JOIN agreego.entity entity_13 ON entity_13.id = address_12.id",
" WHERE",
" NOT entity_17.archived",
" AND relationship_10.target_id = entity_17.id))",
" NOT entity_13.archived",
" AND relationship_10.target_id = entity_13.id))",
" WHEN entity_11.target_type = 'email_address' THEN",
" ((SELECT jsonb_build_object(",
" 'address', email_address_14.address,",
@ -1030,17 +1138,17 @@
" AND relationship_10.target_id = entity_15.id))",
" WHEN entity_11.target_type = 'phone_number' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_13.archived,",
" 'created_at', entity_13.created_at,",
" 'id', entity_13.id,",
" 'number', phone_number_12.number,",
" 'type', entity_13.type",
" 'archived', entity_17.archived,",
" 'created_at', entity_17.created_at,",
" 'id', entity_17.id,",
" 'number', phone_number_16.number,",
" 'type', entity_17.type",
" )",
" FROM agreego.phone_number phone_number_12",
" JOIN agreego.entity entity_13 ON entity_13.id = phone_number_12.id",
" FROM agreego.phone_number phone_number_16",
" JOIN agreego.entity entity_17 ON entity_17.id = phone_number_16.id",
" WHERE",
" NOT entity_13.archived",
" AND relationship_10.target_id = entity_13.id))",
" NOT entity_17.archived",
" AND relationship_10.target_id = entity_17.id))",
" ELSE NULL END,",
" 'type', entity_11.type",
" )), '[]'::jsonb)",
@ -1240,17 +1348,17 @@
" 'target', CASE",
" WHEN entity_11.target_type = 'address' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_17.archived,",
" 'city', address_16.city,",
" 'created_at', entity_17.created_at,",
" 'id', entity_17.id,",
" 'type', entity_17.type",
" 'archived', entity_13.archived,",
" 'city', address_12.city,",
" 'created_at', entity_13.created_at,",
" 'id', entity_13.id,",
" 'type', entity_13.type",
" )",
" FROM agreego.address address_16",
" JOIN agreego.entity entity_17 ON entity_17.id = address_16.id",
" FROM agreego.address address_12",
" JOIN agreego.entity entity_13 ON entity_13.id = address_12.id",
" WHERE",
" NOT entity_17.archived",
" AND relationship_10.target_id = entity_17.id))",
" NOT entity_13.archived",
" AND relationship_10.target_id = entity_13.id))",
" WHEN entity_11.target_type = 'email_address' THEN",
" ((SELECT jsonb_build_object(",
" 'address', email_address_14.address,",
@ -1266,17 +1374,17 @@
" AND relationship_10.target_id = entity_15.id))",
" WHEN entity_11.target_type = 'phone_number' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_13.archived,",
" 'created_at', entity_13.created_at,",
" 'id', entity_13.id,",
" 'number', phone_number_12.number,",
" 'type', entity_13.type",
" 'archived', entity_17.archived,",
" 'created_at', entity_17.created_at,",
" 'id', entity_17.id,",
" 'number', phone_number_16.number,",
" 'type', entity_17.type",
" )",
" FROM agreego.phone_number phone_number_12",
" JOIN agreego.entity entity_13 ON entity_13.id = phone_number_12.id",
" FROM agreego.phone_number phone_number_16",
" JOIN agreego.entity entity_17 ON entity_17.id = phone_number_16.id",
" WHERE",
" NOT entity_13.archived",
" AND relationship_10.target_id = entity_13.id))",
" NOT entity_17.archived",
" AND relationship_10.target_id = entity_17.id))",
" ELSE NULL END,",
" 'type', entity_11.type",
" )), '[]'::jsonb)",
@ -1513,50 +1621,60 @@
"success": true,
"sql": [
[
"(SELECT jsonb_strip_nulls((SELECT COALESCE(jsonb_agg(jsonb_build_object(",
" 'id', organization_1.id,",
" 'type', CASE",
" WHEN organization_1.type = 'bot' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_5.archived,",
" 'created_at', entity_5.created_at,",
" 'id', entity_5.id,",
" 'name', organization_4.name,",
" 'token', bot_3.token,",
" 'type', entity_5.type",
" )",
" FROM agreego.bot bot_3",
" JOIN agreego.organization organization_4 ON organization_4.id = bot_3.id",
" JOIN agreego.entity entity_5 ON entity_5.id = organization_4.id",
" WHERE NOT entity_5.archived))",
" WHEN organization_1.type = 'organization' THEN",
" ((SELECT jsonb_build_object(",
" 'archived', entity_7.archived,",
" 'created_at', entity_7.created_at,",
" 'id', entity_7.id,",
" 'name', organization_6.name,",
" 'type', entity_7.type",
" )",
" FROM agreego.organization organization_6",
" JOIN agreego.entity entity_7 ON entity_7.id = organization_6.id",
" WHERE NOT entity_7.archived))",
" WHEN organization_1.type = 'person' THEN",
" ((SELECT jsonb_build_object(",
" 'age', person_8.age,",
" 'archived', entity_10.archived,",
" 'created_at', entity_10.created_at,",
" 'first_name', person_8.first_name,",
" 'id', entity_10.id,",
" 'last_name', person_8.last_name,",
" 'name', organization_9.name,",
" 'type', entity_10.type",
" )",
" FROM agreego.person person_8",
" JOIN agreego.organization organization_9 ON organization_9.id = person_8.id",
" JOIN agreego.entity entity_10 ON entity_10.id = organization_9.id",
" WHERE NOT entity_10.archived))",
" ELSE NULL END",
")), '[]'::jsonb)",
"(SELECT jsonb_strip_nulls((SELECT COALESCE(jsonb_agg(",
" CASE",
" WHEN organization_1.type = 'bot' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_5.archived,",
" 'created_at', entity_5.created_at,",
" 'id', entity_5.id,",
" 'name', organization_4.name,",
" 'role', bot_3.role,",
" 'token', bot_3.token,",
" 'type', entity_5.type",
" )",
" FROM agreego.bot bot_3",
" JOIN agreego.organization organization_4 ON organization_4.id = bot_3.id",
" JOIN agreego.entity entity_5 ON entity_5.id = organization_4.id",
" WHERE",
" NOT entity_5.archived",
" AND entity_5.id = entity_2.id)",
" )",
" WHEN organization_1.type = 'organization' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_7.archived,",
" 'created_at', entity_7.created_at,",
" 'id', entity_7.id,",
" 'name', organization_6.name,",
" 'type', entity_7.type",
" )",
" FROM agreego.organization organization_6",
" JOIN agreego.entity entity_7 ON entity_7.id = organization_6.id",
" WHERE",
" NOT entity_7.archived",
" AND entity_7.id = entity_2.id)",
" )",
" WHEN organization_1.type = 'person' THEN (",
" (SELECT jsonb_build_object(",
" 'age', person_8.age,",
" 'archived', entity_10.archived,",
" 'created_at', entity_10.created_at,",
" 'first_name', person_8.first_name,",
" 'id', entity_10.id,",
" 'last_name', person_8.last_name,",
" 'name', organization_9.name,",
" 'type', entity_10.type",
" )",
" FROM agreego.person person_8",
" JOIN agreego.organization organization_9 ON organization_9.id = person_8.id",
" JOIN agreego.entity entity_10 ON entity_10.id = organization_9.id",
" WHERE",
" NOT entity_10.archived",
" AND entity_10.id = entity_2.id)",
" )",
" ELSE NULL",
" END",
"), '[]'::jsonb)",
"FROM agreego.organization organization_1",
"JOIN agreego.entity entity_2 ON entity_2.id = organization_1.id",
"WHERE NOT entity_2.archived)))"
@ -1565,27 +1683,226 @@
}
},
{
"description": "Person select via a punc response with family",
"description": "Light organizations select via a punc response with family",
"action": "query",
"schema_id": "get_person.response",
"schema_id": "get_light_organization.response",
"expect": {
"success": true,
"sql": [
[
"(SELECT jsonb_strip_nulls((SELECT jsonb_build_object(",
" 'age', person_1.age,",
" 'archived', entity_3.archived,",
" 'created_at', entity_3.created_at,",
" 'first_name', person_1.first_name,",
" 'id', entity_3.id,",
" 'last_name', person_1.last_name,",
" 'name', organization_2.name,",
" 'type', entity_3.type",
")",
"FROM agreego.person person_1",
"JOIN agreego.organization organization_2 ON organization_2.id = person_1.id",
"JOIN agreego.entity entity_3 ON entity_3.id = organization_2.id",
"WHERE NOT entity_3.archived)))"
"(SELECT jsonb_strip_nulls((SELECT ",
" CASE",
" WHEN organization_1.type = 'bot' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_5.archived,",
" 'created_at', entity_5.created_at,",
" 'id', entity_5.id,",
" 'name', organization_4.name,",
" 'token', bot_3.token,",
" 'type', entity_5.type",
" )",
" FROM agreego.bot bot_3",
" JOIN agreego.organization organization_4 ON organization_4.id = bot_3.id",
" JOIN agreego.entity entity_5 ON entity_5.id = organization_4.id",
" WHERE NOT entity_5.archived AND entity_5.id = entity_2.id)",
" )",
" WHEN organization_1.type = 'person' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_8.archived,",
" 'created_at', entity_8.created_at,",
" 'first_name', person_6.first_name,",
" 'id', entity_8.id,",
" 'last_name', person_6.last_name,",
" 'name', organization_7.name,",
" 'type', entity_8.type",
" )",
" FROM agreego.person person_6",
" JOIN agreego.organization organization_7 ON organization_7.id = person_6.id",
" JOIN agreego.entity entity_8 ON entity_8.id = organization_7.id",
" WHERE NOT entity_8.archived AND entity_8.id = entity_2.id)",
" )",
" ELSE NULL",
" END",
"FROM agreego.organization organization_1",
"JOIN agreego.entity entity_2 ON entity_2.id = organization_1.id",
"WHERE NOT entity_2.archived)))"
]
]
}
},
{
"description": "Full organizations select via a punc response with family",
"action": "query",
"schema_id": "get_full_organization.response",
"expect": {
"success": true,
"sql": [
[
"(SELECT jsonb_strip_nulls((SELECT CASE",
" WHEN organization_1.type = 'person' THEN (",
" (SELECT jsonb_build_object(",
" 'addresses',",
" (SELECT COALESCE(jsonb_agg(jsonb_build_object(",
" 'archived', entity_8.archived,",
" 'created_at', entity_8.created_at,",
" 'id', entity_8.id,",
" 'is_primary', contact_6.is_primary,",
" 'target',",
" (SELECT jsonb_build_object(",
" 'archived', entity_10.archived,",
" 'city', address_9.city,",
" 'created_at', entity_10.created_at,",
" 'id', entity_10.id,",
" 'type', entity_10.type",
" )",
" FROM agreego.address address_9",
" JOIN agreego.entity entity_10 ON entity_10.id = address_9.id",
" WHERE",
" NOT entity_10.archived",
" AND relationship_7.target_id = entity_10.id),",
" 'type', entity_8.type",
" )), '[]'::jsonb)",
" FROM agreego.contact contact_6",
" JOIN agreego.relationship relationship_7 ON relationship_7.id = contact_6.id",
" JOIN agreego.entity entity_8 ON entity_8.id = relationship_7.id",
" WHERE",
" NOT entity_8.archived",
" AND relationship_7.target_type = 'address'",
" AND relationship_7.source_id = entity_5.id),",
" 'age', person_3.age,",
" 'archived', entity_5.archived,",
" 'contacts',",
" (SELECT COALESCE(jsonb_agg(jsonb_build_object(",
" 'archived', entity_13.archived,",
" 'created_at', entity_13.created_at,",
" 'id', entity_13.id,",
" 'is_primary', contact_11.is_primary,",
" 'target',",
" CASE",
" WHEN entity_13.target_type = 'address' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_15.archived,",
" 'city', address_14.city,",
" 'created_at', entity_15.created_at,",
" 'id', entity_15.id,",
" 'type', entity_15.type",
" )",
" FROM agreego.address address_14",
" JOIN agreego.entity entity_15 ON entity_15.id = address_14.id",
" WHERE",
" NOT entity_15.archived",
" AND relationship_12.target_id = entity_15.id)",
" )",
" WHEN entity_13.target_type = 'email_address' THEN (",
" (SELECT jsonb_build_object(",
" 'address', email_address_16.address,",
" 'archived', entity_17.archived,",
" 'created_at', entity_17.created_at,",
" 'id', entity_17.id,",
" 'type', entity_17.type",
" )",
" FROM agreego.email_address email_address_16",
" JOIN agreego.entity entity_17 ON entity_17.id = email_address_16.id",
" WHERE",
" NOT entity_17.archived",
" AND relationship_12.target_id = entity_17.id)",
" )",
" WHEN entity_13.target_type = 'phone_number' THEN (",
" (SELECT jsonb_build_object(",
" 'archived', entity_19.archived,",
" 'created_at', entity_19.created_at,",
" 'id', entity_19.id,",
" 'number', phone_number_18.number,",
" 'type', entity_19.type",
" )",
" FROM agreego.phone_number phone_number_18",
" JOIN agreego.entity entity_19 ON entity_19.id = phone_number_18.id",
" WHERE",
" NOT entity_19.archived",
" AND relationship_12.target_id = entity_19.id)",
" )",
" ELSE NULL",
" END,",
" 'type', entity_13.type",
" )), '[]'::jsonb)",
" FROM agreego.contact contact_11",
" JOIN agreego.relationship relationship_12 ON relationship_12.id = contact_11.id",
" JOIN agreego.entity entity_13 ON entity_13.id = relationship_12.id",
" WHERE",
" NOT entity_13.archived",
" AND relationship_12.source_id = entity_5.id),",
" 'created_at', entity_5.created_at,",
" 'email_addresses',",
" (SELECT COALESCE(jsonb_agg(jsonb_build_object(",
" 'archived', entity_22.archived,",
" 'created_at', entity_22.created_at,",
" 'id', entity_22.id,",
" 'is_primary', contact_20.is_primary,",
" 'target',",
" (SELECT jsonb_build_object(",
" 'address', email_address_23.address,",
" 'archived', entity_24.archived,",
" 'created_at', entity_24.created_at,",
" 'id', entity_24.id,",
" 'type', entity_24.type",
" )",
" FROM agreego.email_address email_address_23",
" JOIN agreego.entity entity_24 ON entity_24.id = email_address_23.id",
" WHERE",
" NOT entity_24.archived",
" AND relationship_21.target_id = entity_24.id),",
" 'type', entity_22.type",
" )), '[]'::jsonb)",
" FROM agreego.contact contact_20",
" JOIN agreego.relationship relationship_21 ON relationship_21.id = contact_20.id",
" JOIN agreego.entity entity_22 ON entity_22.id = relationship_21.id",
" WHERE",
" NOT entity_22.archived",
" AND relationship_21.target_type = 'email_address'",
" AND relationship_21.source_id = entity_5.id),",
" 'first_name', person_3.first_name,",
" 'id', entity_5.id,",
" 'last_name', person_3.last_name,",
" 'name', organization_4.name,",
" 'phone_numbers',",
" (SELECT COALESCE(jsonb_agg(jsonb_build_object(",
" 'archived', entity_27.archived,",
" 'created_at', entity_27.created_at,",
" 'id', entity_27.id,",
" 'is_primary', contact_25.is_primary,",
" 'target',",
" (SELECT jsonb_build_object(",
" 'archived', entity_29.archived,",
" 'created_at', entity_29.created_at,",
" 'id', entity_29.id,",
" 'number', phone_number_28.number,",
" 'type', entity_29.type",
" )",
" FROM agreego.phone_number phone_number_28",
" JOIN agreego.entity entity_29 ON entity_29.id = phone_number_28.id",
" WHERE",
" NOT entity_29.archived",
" AND relationship_26.target_id = entity_29.id),",
" 'type', entity_27.type",
" )), '[]'::jsonb)",
" FROM agreego.contact contact_25",
" JOIN agreego.relationship relationship_26 ON relationship_26.id = contact_25.id",
" JOIN agreego.entity entity_27 ON entity_27.id = relationship_26.id",
" WHERE",
" NOT entity_27.archived",
" AND relationship_26.target_type = 'phone_number'",
" AND relationship_26.source_id = entity_5.id),",
" 'type', entity_5.type",
" )",
" FROM agreego.person person_3",
" JOIN agreego.organization organization_4 ON organization_4.id = person_3.id",
" JOIN agreego.entity entity_5 ON entity_5.id = organization_4.id",
" WHERE NOT entity_5.archived AND entity_5.id = entity_2.id))",
" ELSE NULL",
"END",
"FROM agreego.organization organization_1",
"JOIN agreego.entity entity_2 ON entity_2.id = organization_1.id",
"WHERE NOT entity_2.archived)))"
]
]
}
@ -1629,6 +1946,44 @@
]
]
}
},
{
"description": "Widgets select via a punc response with family (STI)",
"action": "query",
"schema_id": "get_widgets.response",
"expect": {
"success": true,
"sql": [
[
"(SELECT jsonb_strip_nulls((SELECT COALESCE(jsonb_agg(",
" CASE",
" WHEN widget_1.kind = 'stock' THEN (",
" jsonb_build_object(",
" 'archived', entity_2.archived,",
" 'created_at', entity_2.created_at,",
" 'id', entity_2.id,",
" 'kind', widget_1.kind,",
" 'type', entity_2.type",
" )",
" )",
" WHEN widget_1.kind = 'tasks' THEN (",
" jsonb_build_object(",
" 'archived', entity_2.archived,",
" 'created_at', entity_2.created_at,",
" 'id', entity_2.id,",
" 'kind', widget_1.kind,",
" 'type', entity_2.type",
" )",
" )",
" ELSE NULL",
" END",
"), '[]'::jsonb)",
"FROM agreego.widget widget_1",
"JOIN agreego.entity entity_2 ON entity_2.id = widget_1.id",
"WHERE NOT entity_2.archived)))"
]
]
}
}
]
}

45
out.txt Normal file

File diff suppressed because one or more lines are too long

View File

@ -4,6 +4,7 @@ pub mod executors;
pub mod formats;
pub mod page;
pub mod punc;
pub mod object;
pub mod relation;
pub mod schema;
pub mod r#type;
@ -23,7 +24,8 @@ use punc::Punc;
use relation::Relation;
use schema::Schema;
use serde_json::Value;
use std::collections::{HashMap, HashSet};
use std::collections::HashMap;
use std::sync::Arc;
use r#type::Type;
pub struct Database {
@ -31,9 +33,7 @@ pub struct Database {
pub types: HashMap<String, Type>,
pub puncs: HashMap<String, Punc>,
pub relations: HashMap<String, Relation>,
pub schemas: HashMap<String, Schema>,
pub descendants: HashMap<String, Vec<String>>,
pub depths: HashMap<String, usize>,
pub schemas: HashMap<String, Arc<Schema>>,
pub executor: Box<dyn DatabaseExecutor + Send + Sync>,
}
@ -45,8 +45,6 @@ impl Database {
relations: HashMap::new(),
puncs: HashMap::new(),
schemas: HashMap::new(),
descendants: HashMap::new(),
depths: HashMap::new(),
#[cfg(not(test))]
executor: Box::new(SpiExecutor::new()),
#[cfg(test)]
@ -137,7 +135,7 @@ impl Database {
.clone()
.unwrap_or_else(|| format!("schema_{}", i));
schema.obj.id = Some(id.clone());
db.schemas.insert(id, schema);
db.schemas.insert(id, Arc::new(schema));
}
Err(e) => {
errors.push(crate::drop::Error {
@ -187,19 +185,21 @@ impl Database {
pub fn compile(&mut self, errors: &mut Vec<crate::drop::Error>) {
let mut harvested = Vec::new();
for schema in self.schemas.values_mut() {
schema.collect_schemas(None, &mut harvested, errors);
for schema_arc in self.schemas.values_mut() {
if let Some(s) = std::sync::Arc::get_mut(schema_arc) {
s.collect_schemas(None, &mut harvested, errors);
}
}
for (id, schema) in harvested {
self.schemas.insert(id, Arc::new(schema));
}
self.schemas.extend(harvested);
self.collect_schemas(errors);
self.collect_depths();
self.collect_descendants();
// Mathematically evaluate all property inheritances, formats, schemas, and foreign key edges topographically over OnceLocks
let mut visited = std::collections::HashSet::new();
for schema in self.schemas.values() {
schema.compile(self, &mut visited, errors);
for schema_arc in self.schemas.values() {
schema_arc.as_ref().compile(self, &mut visited, errors);
}
}
@ -225,74 +225,185 @@ impl Database {
}
for (id, schema) in to_insert {
self.schemas.insert(id, schema);
self.schemas.insert(id, Arc::new(schema));
}
}
fn collect_depths(&mut self) {
let mut depths: HashMap<String, usize> = HashMap::new();
let schema_ids: Vec<String> = self.schemas.keys().cloned().collect();
/// Inspects the Postgres pg_constraint relations catalog to securely identify
/// the precise Foreign Key connecting a parent and child hierarchy path.
pub fn resolve_relation<'a>(
&'a self,
parent_type: &str,
child_type: &str,
prop_name: &str,
relative_keys: Option<&Vec<String>>,
is_array: bool,
schema_id: Option<&str>,
path: &str,
errors: &mut Vec<crate::drop::Error>,
) -> Option<(&'a crate::database::relation::Relation, bool)> {
// Enforce graph locality by ensuring we don't accidentally crawl to pure structural entity boundaries
if parent_type == "entity" && child_type == "entity" {
return None;
}
for id in schema_ids {
let mut current_id = id.clone();
let mut depth = 0;
let mut visited = HashSet::new();
let p_def = self.types.get(parent_type)?;
let c_def = self.types.get(child_type)?;
while let Some(schema) = self.schemas.get(&current_id) {
if !visited.insert(current_id.clone()) {
break; // Cycle detected
let mut matching_rels = Vec::new();
let mut directions = Vec::new();
// Scour the complete catalog for any Edge matching the inheritance scope of the two objects
// This automatically binds polymorphic structures (e.g. recognizing a relationship targeting User
// also natively binds instances specifically typed as Person).
let mut all_rels: Vec<&crate::database::relation::Relation> = self.relations.values().collect();
all_rels.sort_by(|a, b| a.constraint.cmp(&b.constraint));
for rel in all_rels {
let mut is_forward =
p_def.hierarchy.contains(&rel.source_type) && c_def.hierarchy.contains(&rel.destination_type);
let is_reverse =
p_def.hierarchy.contains(&rel.destination_type) && c_def.hierarchy.contains(&rel.source_type);
// Structural Cardinality Filtration:
// If the schema requires a collection (Array), it is mathematically impossible for a pure
// Forward scalar edge (where the parent holds exactly one UUID pointer) to fulfill a One-to-Many request.
// Thus, if it's an array, we fully reject pure Forward edges and only accept Reverse edges (or Junction edges).
if is_array && is_forward && !is_reverse {
is_forward = false;
}
if is_forward {
matching_rels.push(rel);
directions.push(true);
} else if is_reverse {
matching_rels.push(rel);
directions.push(false);
}
}
// Abort relation discovery early if no hierarchical inheritance match was found
if matching_rels.is_empty() {
let mut details = crate::drop::ErrorDetails {
path: path.to_string(),
..Default::default()
};
if let Some(sid) = schema_id {
details.schema = Some(sid.to_string());
}
errors.push(crate::drop::Error {
code: "EDGE_MISSING".to_string(),
message: format!(
"No database relation exists between '{}' and '{}' for property '{}'",
parent_type, child_type, prop_name
),
details,
});
return None;
}
// Ideal State: The objects only share a solitary structural relation, resolving ambiguity instantly.
if matching_rels.len() == 1 {
return Some((matching_rels[0], directions[0]));
}
let mut chosen_idx = 0;
let mut resolved = false;
// Exact Prefix Disambiguation: Determine if the database specifically names this constraint
// directly mapping to the JSON Schema property name (e.g., `fk_{child}_{property_name}`)
for (i, rel) in matching_rels.iter().enumerate() {
if let Some(prefix) = &rel.prefix {
if prop_name.starts_with(prefix)
|| prefix.starts_with(prop_name)
|| prefix.replace("_", "") == prop_name.replace("_", "")
{
chosen_idx = i;
resolved = true;
break;
}
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &schema.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
current_id = t.clone();
depth += 1;
continue;
}
}
// Complex Subgraph Resolution: The database contains multiple equally explicit foreign key constraints
// linking these objects (such as pointing to `source` and `target` in Many-to-Many junction models).
if !resolved && relative_keys.is_some() {
// Twin Deduction Pass 1: We inspect the exact properties structurally defined inside the compiled payload
// to observe which explicit relation arrow the child payload natively consumes.
let keys = relative_keys.unwrap();
let mut consumed_rel_idx = None;
for (i, rel) in matching_rels.iter().enumerate() {
if let Some(prefix) = &rel.prefix {
if keys.contains(prefix) {
consumed_rel_idx = Some(i);
break; // Found the routing edge explicitly consumed by the schema payload
}
}
break;
}
depths.insert(id, depth);
}
self.depths = depths;
}
fn collect_descendants(&mut self) {
let mut direct_refs: HashMap<String, Vec<String>> = HashMap::new();
for (id, schema) in &self.schemas {
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &schema.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
direct_refs
.entry(t.clone())
.or_default()
.push(id.clone());
// Twin Deduction Pass 2: Knowing which arrow points outbound, we can mathematically deduce its twin
// providing the reverse ownership on the same junction boundary must be the incoming Edge to the parent.
if let Some(used_idx) = consumed_rel_idx {
let used_rel = matching_rels[used_idx];
let mut twin_ids = Vec::new();
for (i, rel) in matching_rels.iter().enumerate() {
if i != used_idx
&& rel.source_type == used_rel.source_type
&& rel.destination_type == used_rel.destination_type
&& rel.prefix.is_some()
{
twin_ids.push(i);
}
}
if twin_ids.len() == 1 {
chosen_idx = twin_ids[0];
resolved = true;
}
}
}
// Cache exhaustive descendants matrix for generic $family string lookups natively
let mut descendants = HashMap::new();
for id in self.schemas.keys() {
let mut desc_set = HashSet::new();
Self::collect_descendants_recursively(id, &direct_refs, &mut desc_set);
let mut desc_vec: Vec<String> = desc_set.into_iter().collect();
desc_vec.sort();
descendants.insert(id.clone(), desc_vec);
}
self.descendants = descendants;
}
fn collect_descendants_recursively(
target: &str,
direct_refs: &std::collections::HashMap<String, Vec<String>>,
descendants: &mut std::collections::HashSet<String>,
) {
if let Some(children) = direct_refs.get(target) {
for child in children {
if descendants.insert(child.clone()) {
Self::collect_descendants_recursively(child, direct_refs, descendants);
// Implicit Base Fallback: If no complex explicit paths resolve, but exactly one relation
// sits entirely naked (without a constraint prefix), it must be the core structural parent ownership.
if !resolved {
let mut null_prefix_ids = Vec::new();
for (i, rel) in matching_rels.iter().enumerate() {
if rel.prefix.is_none() {
null_prefix_ids.push(i);
}
}
if null_prefix_ids.len() == 1 {
chosen_idx = null_prefix_ids[0];
resolved = true;
}
}
// If we exhausted all mathematical deduction pathways and STILL cannot isolate a single edge,
// we must abort rather than silently guessing. Returning None prevents arbitrary SQL generation
// and forces a clean structural error for the architect.
if !resolved {
let mut details = crate::drop::ErrorDetails {
path: path.to_string(),
context: serde_json::to_value(&matching_rels).ok(),
cause: Some("Multiple conflicting constraints found matching prefixes".to_string()),
..Default::default()
};
if let Some(sid) = schema_id {
details.schema = Some(sid.to_string());
}
errors.push(crate::drop::Error {
code: "AMBIGUOUS_TYPE_RELATIONS".to_string(),
message: format!(
"Ambiguous database relation between '{}' and '{}' for property '{}'",
parent_type, child_type, prop_name
),
details,
});
return None;
}
Some((matching_rels[chosen_idx], directions[chosen_idx]))
}
}

367
src/database/object.rs Normal file
View File

@ -0,0 +1,367 @@
use serde::{Deserialize, Serialize};
use serde_json::Value;
use std::collections::BTreeMap;
use std::sync::Arc;
use std::sync::OnceLock;
use crate::database::schema::Schema;
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Case {
#[serde(skip_serializing_if = "Option::is_none")]
pub when: Option<Arc<Schema>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub then: Option<Arc<Schema>>,
#[serde(rename = "else")]
#[serde(skip_serializing_if = "Option::is_none")]
pub else_: Option<Arc<Schema>>,
}
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub struct SchemaObject {
// Core Schema Keywords
#[serde(rename = "$id")]
#[serde(skip_serializing_if = "Option::is_none")]
pub id: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub description: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub title: Option<String>,
#[serde(default)] // Allow missing type
#[serde(rename = "type")]
#[serde(skip_serializing_if = "Option::is_none")]
pub type_: Option<SchemaTypeOrArray>, // Handles string or array of strings
// Object Keywords
#[serde(skip_serializing_if = "Option::is_none")]
pub properties: Option<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "patternProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub pattern_properties: Option<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "additionalProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub additional_properties: Option<Arc<Schema>>,
#[serde(rename = "$family")]
#[serde(skip_serializing_if = "Option::is_none")]
pub family: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub required: Option<Vec<String>>,
// dependencies can be schema dependencies or property dependencies
#[serde(skip_serializing_if = "Option::is_none")]
pub dependencies: Option<BTreeMap<String, Dependency>>,
// Array Keywords
#[serde(rename = "items")]
#[serde(skip_serializing_if = "Option::is_none")]
pub items: Option<Arc<Schema>>,
#[serde(rename = "prefixItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub prefix_items: Option<Vec<Arc<Schema>>>,
// String Validation
#[serde(rename = "minLength")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_length: Option<f64>,
#[serde(rename = "maxLength")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_length: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub pattern: Option<String>,
// Array Validation
#[serde(rename = "minItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_items: Option<f64>,
#[serde(rename = "maxItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_items: Option<f64>,
#[serde(rename = "uniqueItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub unique_items: Option<bool>,
#[serde(rename = "contains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub contains: Option<Arc<Schema>>,
#[serde(rename = "minContains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_contains: Option<f64>,
#[serde(rename = "maxContains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_contains: Option<f64>,
// Object Validation
#[serde(rename = "minProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_properties: Option<f64>,
#[serde(rename = "maxProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_properties: Option<f64>,
#[serde(rename = "propertyNames")]
#[serde(skip_serializing_if = "Option::is_none")]
pub property_names: Option<Arc<Schema>>,
// Numeric Validation
#[serde(skip_serializing_if = "Option::is_none")]
pub format: Option<String>,
#[serde(rename = "enum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub enum_: Option<Vec<Value>>, // `enum` is a reserved keyword in Rust
#[serde(
default,
rename = "const",
deserialize_with = "crate::database::object::deserialize_some"
)]
#[serde(skip_serializing_if = "Option::is_none")]
pub const_: Option<Value>,
// Numeric Validation
#[serde(rename = "multipleOf")]
#[serde(skip_serializing_if = "Option::is_none")]
pub multiple_of: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub minimum: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub maximum: Option<f64>,
#[serde(rename = "exclusiveMinimum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub exclusive_minimum: Option<f64>,
#[serde(rename = "exclusiveMaximum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub exclusive_maximum: Option<f64>,
// Combining Keywords
#[serde(skip_serializing_if = "Option::is_none")]
pub cases: Option<Vec<Case>>,
#[serde(rename = "oneOf")]
#[serde(skip_serializing_if = "Option::is_none")]
pub one_of: Option<Vec<Arc<Schema>>>,
#[serde(rename = "not")]
#[serde(skip_serializing_if = "Option::is_none")]
pub not: Option<Arc<Schema>>,
// Custom Vocabularies
#[serde(skip_serializing_if = "Option::is_none")]
pub form: Option<Vec<String>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub display: Option<Vec<String>>,
#[serde(rename = "enumNames")]
#[serde(skip_serializing_if = "Option::is_none")]
pub enum_names: Option<Vec<String>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub control: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub actions: Option<BTreeMap<String, Action>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub computer: Option<String>,
#[serde(default)]
#[serde(skip_serializing_if = "Option::is_none")]
pub extensible: Option<bool>,
#[serde(rename = "compiledProperties")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::object::is_once_lock_vec_empty")]
#[serde(serialize_with = "crate::database::object::serialize_once_lock")]
pub compiled_property_names: OnceLock<Vec<String>>,
#[serde(skip)]
pub compiled_properties: OnceLock<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "compiledDiscriminator")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::object::is_once_lock_string_empty")]
#[serde(serialize_with = "crate::database::object::serialize_once_lock")]
pub compiled_discriminator: OnceLock<String>,
#[serde(rename = "compiledOptions")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::object::is_once_lock_map_empty")]
#[serde(serialize_with = "crate::database::object::serialize_once_lock")]
pub compiled_options: OnceLock<BTreeMap<String, String>>,
#[serde(rename = "compiledEdges")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::object::is_once_lock_map_empty")]
#[serde(serialize_with = "crate::database::object::serialize_once_lock")]
pub compiled_edges: OnceLock<BTreeMap<String, crate::database::edge::Edge>>,
#[serde(skip)]
pub compiled_format: OnceLock<CompiledFormat>,
#[serde(skip)]
pub compiled_pattern: OnceLock<CompiledRegex>,
#[serde(skip)]
pub compiled_pattern_properties: OnceLock<Vec<(CompiledRegex, Arc<Schema>)>>,
}
/// Represents a compiled format validator
#[derive(Clone)]
pub enum CompiledFormat {
Func(fn(&serde_json::Value) -> Result<(), Box<dyn std::error::Error + Send + Sync>>),
Regex(regex::Regex),
}
impl std::fmt::Debug for CompiledFormat {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
CompiledFormat::Func(_) => write!(f, "CompiledFormat::Func(...)"),
CompiledFormat::Regex(r) => write!(f, "CompiledFormat::Regex({:?})", r),
}
}
}
/// A wrapper for compiled regex patterns
#[derive(Debug, Clone)]
pub struct CompiledRegex(pub regex::Regex);
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum SchemaTypeOrArray {
Single(String),
Multiple(Vec<String>),
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Action {
#[serde(skip_serializing_if = "Option::is_none")]
pub navigate: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub punc: Option<String>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum Dependency {
Props(Vec<String>),
Schema(Arc<Schema>),
}
pub fn serialize_once_lock<T: serde::Serialize, S: serde::Serializer>(
lock: &OnceLock<T>,
serializer: S,
) -> Result<S::Ok, S::Error> {
if let Some(val) = lock.get() {
val.serialize(serializer)
} else {
serializer.serialize_none()
}
}
pub fn is_once_lock_map_empty<K, V>(lock: &OnceLock<std::collections::BTreeMap<K, V>>) -> bool {
lock.get().map_or(true, |m| m.is_empty())
}
pub fn is_once_lock_vec_empty<T>(lock: &OnceLock<Vec<T>>) -> bool {
lock.get().map_or(true, |v| v.is_empty())
}
pub fn is_once_lock_string_empty(lock: &OnceLock<String>) -> bool {
lock.get().map_or(true, |s| s.is_empty())
}
// Schema mirrors the Go Punc Generator's schema struct for consistency.
// It is an order-preserving representation of a JSON Schema.
pub fn deserialize_some<'de, D>(deserializer: D) -> Result<Option<Value>, D::Error>
where
D: serde::Deserializer<'de>,
{
let v = Value::deserialize(deserializer)?;
Ok(Some(v))
}
pub fn is_primitive_type(t: &str) -> bool {
matches!(
t,
"string" | "number" | "integer" | "boolean" | "object" | "array" | "null"
)
}
impl SchemaObject {
pub fn identifier(&self) -> Option<String> {
if let Some(id) = &self.id {
return Some(id.split('.').next_back().unwrap_or("").to_string());
}
if let Some(SchemaTypeOrArray::Single(t)) = &self.type_ {
if !is_primitive_type(t) {
return Some(t.split('.').next_back().unwrap_or("").to_string());
}
}
None
}
pub fn get_discriminator_value(&self, dim: &str) -> Option<String> {
let is_split = self
.compiled_properties
.get()
.map_or(false, |p| p.contains_key("kind"));
if let Some(id) = &self.id {
if id.contains("light.person") || id.contains("light.organization") {
println!(
"[DEBUG SPLIT] ID: {}, dim: {}, is_split: {:?}, props: {:?}",
id,
dim,
is_split,
self
.compiled_properties
.get()
.map(|p| p.keys().cloned().collect::<Vec<_>>())
);
}
}
if let Some(props) = self.compiled_properties.get() {
if let Some(prop_schema) = props.get(dim) {
if let Some(c) = &prop_schema.obj.const_ {
if let Some(s) = c.as_str() {
return Some(s.to_string());
}
}
if let Some(e) = &prop_schema.obj.enum_ {
if e.len() == 1 {
if let Some(s) = e[0].as_str() {
return Some(s.to_string());
}
}
}
}
}
if dim == "kind" {
if let Some(id) = &self.id {
let base = id.split('/').last().unwrap_or(id);
if let Some(idx) = base.rfind('.') {
return Some(base[..idx].to_string());
}
}
if let Some(SchemaTypeOrArray::Single(t)) = &self.type_ {
if !is_primitive_type(t) {
let base = t.split('/').last().unwrap_or(t);
if let Some(idx) = base.rfind('.') {
return Some(base[..idx].to_string());
}
}
}
}
if dim == "type" {
if let Some(id) = &self.id {
let base = id.split('/').last().unwrap_or(id);
if is_split {
return Some(base.split('.').next_back().unwrap_or(base).to_string());
} else {
return Some(base.to_string());
}
}
if let Some(SchemaTypeOrArray::Single(t)) = &self.type_ {
if !is_primitive_type(t) {
let base = t.split('/').last().unwrap_or(t);
if is_split {
return Some(base.split('.').next_back().unwrap_or(base).to_string());
} else {
return Some(base.to_string());
}
}
}
}
None
}
}

View File

@ -1,240 +1,7 @@
use serde::{Deserialize, Serialize};
use serde_json::Value;
use std::collections::BTreeMap;
use std::sync::Arc;
use std::sync::OnceLock;
pub fn serialize_once_lock<T: serde::Serialize, S: serde::Serializer>(
lock: &OnceLock<T>,
serializer: S,
) -> Result<S::Ok, S::Error> {
if let Some(val) = lock.get() {
val.serialize(serializer)
} else {
serializer.serialize_none()
}
}
pub fn is_once_lock_map_empty<K, V>(lock: &OnceLock<std::collections::BTreeMap<K, V>>) -> bool {
lock.get().map_or(true, |m| m.is_empty())
}
pub fn is_once_lock_vec_empty<T>(lock: &OnceLock<Vec<T>>) -> bool {
lock.get().map_or(true, |v| v.is_empty())
}
// Schema mirrors the Go Punc Generator's schema struct for consistency.
// It is an order-preserving representation of a JSON Schema.
pub fn deserialize_some<'de, D>(deserializer: D) -> Result<Option<Value>, D::Error>
where
D: serde::Deserializer<'de>,
{
let v = Value::deserialize(deserializer)?;
Ok(Some(v))
}
pub fn is_primitive_type(t: &str) -> bool {
matches!(
t,
"string" | "number" | "integer" | "boolean" | "object" | "array" | "null"
)
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Case {
#[serde(skip_serializing_if = "Option::is_none")]
pub when: Option<Arc<Schema>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub then: Option<Arc<Schema>>,
#[serde(rename = "else")]
#[serde(skip_serializing_if = "Option::is_none")]
pub else_: Option<Arc<Schema>>,
}
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub struct SchemaObject {
// Core Schema Keywords
#[serde(rename = "$id")]
#[serde(skip_serializing_if = "Option::is_none")]
pub id: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub description: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub title: Option<String>,
#[serde(default)] // Allow missing type
#[serde(rename = "type")]
#[serde(skip_serializing_if = "Option::is_none")]
pub type_: Option<SchemaTypeOrArray>, // Handles string or array of strings
// Object Keywords
#[serde(skip_serializing_if = "Option::is_none")]
pub properties: Option<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "patternProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub pattern_properties: Option<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "additionalProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub additional_properties: Option<Arc<Schema>>,
#[serde(rename = "$family")]
#[serde(skip_serializing_if = "Option::is_none")]
pub family: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub required: Option<Vec<String>>,
// dependencies can be schema dependencies or property dependencies
#[serde(skip_serializing_if = "Option::is_none")]
pub dependencies: Option<BTreeMap<String, Dependency>>,
// Array Keywords
#[serde(rename = "items")]
#[serde(skip_serializing_if = "Option::is_none")]
pub items: Option<Arc<Schema>>,
#[serde(rename = "prefixItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub prefix_items: Option<Vec<Arc<Schema>>>,
// String Validation
#[serde(rename = "minLength")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_length: Option<f64>,
#[serde(rename = "maxLength")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_length: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub pattern: Option<String>,
// Array Validation
#[serde(rename = "minItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_items: Option<f64>,
#[serde(rename = "maxItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_items: Option<f64>,
#[serde(rename = "uniqueItems")]
#[serde(skip_serializing_if = "Option::is_none")]
pub unique_items: Option<bool>,
#[serde(rename = "contains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub contains: Option<Arc<Schema>>,
#[serde(rename = "minContains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_contains: Option<f64>,
#[serde(rename = "maxContains")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_contains: Option<f64>,
// Object Validation
#[serde(rename = "minProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub min_properties: Option<f64>,
#[serde(rename = "maxProperties")]
#[serde(skip_serializing_if = "Option::is_none")]
pub max_properties: Option<f64>,
#[serde(rename = "propertyNames")]
#[serde(skip_serializing_if = "Option::is_none")]
pub property_names: Option<Arc<Schema>>,
// Numeric Validation
#[serde(skip_serializing_if = "Option::is_none")]
pub format: Option<String>,
#[serde(rename = "enum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub enum_: Option<Vec<Value>>, // `enum` is a reserved keyword in Rust
#[serde(
default,
rename = "const",
deserialize_with = "crate::database::schema::deserialize_some"
)]
#[serde(skip_serializing_if = "Option::is_none")]
pub const_: Option<Value>,
// Numeric Validation
#[serde(rename = "multipleOf")]
#[serde(skip_serializing_if = "Option::is_none")]
pub multiple_of: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub minimum: Option<f64>,
#[serde(skip_serializing_if = "Option::is_none")]
pub maximum: Option<f64>,
#[serde(rename = "exclusiveMinimum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub exclusive_minimum: Option<f64>,
#[serde(rename = "exclusiveMaximum")]
#[serde(skip_serializing_if = "Option::is_none")]
pub exclusive_maximum: Option<f64>,
// Combining Keywords
#[serde(skip_serializing_if = "Option::is_none")]
pub cases: Option<Vec<Case>>,
#[serde(rename = "oneOf")]
#[serde(skip_serializing_if = "Option::is_none")]
pub one_of: Option<Vec<Arc<Schema>>>,
#[serde(rename = "not")]
#[serde(skip_serializing_if = "Option::is_none")]
pub not: Option<Arc<Schema>>,
// Custom Vocabularies
#[serde(skip_serializing_if = "Option::is_none")]
pub form: Option<Vec<String>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub display: Option<Vec<String>>,
#[serde(rename = "enumNames")]
#[serde(skip_serializing_if = "Option::is_none")]
pub enum_names: Option<Vec<String>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub control: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub actions: Option<BTreeMap<String, Action>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub computer: Option<String>,
#[serde(default)]
#[serde(skip_serializing_if = "Option::is_none")]
pub extensible: Option<bool>,
#[serde(rename = "compiledProperties")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::schema::is_once_lock_vec_empty")]
#[serde(serialize_with = "crate::database::schema::serialize_once_lock")]
pub compiled_property_names: OnceLock<Vec<String>>,
#[serde(skip)]
pub compiled_properties: OnceLock<BTreeMap<String, Arc<Schema>>>,
#[serde(rename = "compiledEdges")]
#[serde(skip_deserializing)]
#[serde(skip_serializing_if = "crate::database::schema::is_once_lock_map_empty")]
#[serde(serialize_with = "crate::database::schema::serialize_once_lock")]
pub compiled_edges: OnceLock<BTreeMap<String, crate::database::edge::Edge>>,
#[serde(skip)]
pub compiled_format: OnceLock<CompiledFormat>,
#[serde(skip)]
pub compiled_pattern: OnceLock<CompiledRegex>,
#[serde(skip)]
pub compiled_pattern_properties: OnceLock<Vec<(CompiledRegex, Arc<Schema>)>>,
}
/// Represents a compiled format validator
#[derive(Clone)]
pub enum CompiledFormat {
Func(fn(&serde_json::Value) -> Result<(), Box<dyn std::error::Error + Send + Sync>>),
Regex(regex::Regex),
}
impl std::fmt::Debug for CompiledFormat {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
CompiledFormat::Func(_) => write!(f, "CompiledFormat::Func(...)"),
CompiledFormat::Regex(r) => write!(f, "CompiledFormat::Regex({:?})", r),
}
}
}
/// A wrapper for compiled regex patterns
#[derive(Debug, Clone)]
pub struct CompiledRegex(pub regex::Regex);
use crate::database::object::*;
#[derive(Debug, Clone, Serialize, Default)]
pub struct Schema {
#[serde(flatten)]
@ -277,7 +44,7 @@ impl Schema {
let _ = self
.obj
.compiled_format
.set(crate::database::schema::CompiledFormat::Func(fmt.func));
.set(crate::database::object::CompiledFormat::Func(fmt.func));
}
}
@ -286,7 +53,7 @@ impl Schema {
let _ = self
.obj
.compiled_pattern
.set(crate::database::schema::CompiledRegex(re));
.set(crate::database::object::CompiledRegex(re));
}
}
@ -294,7 +61,7 @@ impl Schema {
let mut compiled = Vec::new();
for (k, v) in pattern_props {
if let Ok(re) = regex::Regex::new(k) {
compiled.push((crate::database::schema::CompiledRegex(re), v.clone()));
compiled.push((crate::database::object::CompiledRegex(re), v.clone()));
}
}
if !compiled.is_empty() {
@ -305,10 +72,10 @@ impl Schema {
let mut props = std::collections::BTreeMap::new();
// 1. Resolve INHERITANCE dependencies first
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::object::is_primitive_type(t) {
if let Some(parent) = db.schemas.get(t) {
parent.compile(db, visited, errors);
parent.as_ref().compile(db, visited, errors);
if let Some(p_props) = parent.obj.compiled_properties.get() {
props.extend(p_props.clone());
}
@ -316,10 +83,10 @@ impl Schema {
}
}
if let Some(crate::database::schema::SchemaTypeOrArray::Multiple(types)) = &self.obj.type_ {
if let Some(crate::database::object::SchemaTypeOrArray::Multiple(types)) = &self.obj.type_ {
let mut custom_type_count = 0;
for t in types {
if !crate::database::schema::is_primitive_type(t) {
if !crate::database::object::is_primitive_type(t) {
custom_type_count += 1;
}
}
@ -339,9 +106,9 @@ impl Schema {
}
for t in types {
if !crate::database::schema::is_primitive_type(t) {
if !crate::database::object::is_primitive_type(t) {
if let Some(parent) = db.schemas.get(t) {
parent.compile(db, visited, errors);
parent.as_ref().compile(db, visited, errors);
}
}
}
@ -411,11 +178,220 @@ impl Schema {
}
}
self.compile_polymorphism(db, errors);
if let Some(id) = &self.obj.id {
visited.remove(id);
}
}
/// Dynamically infers and compiles all structural database relationships between this Schema
/// and its nested children. This functions recursively traverses the JSON Schema abstract syntax
/// tree, identifies physical PostgreSQL table boundaries, and locks the resulting relation
/// constraint paths directly onto the `compiled_edges` map in O(1) memory.
pub fn compile_edges(
&self,
db: &crate::database::Database,
visited: &mut std::collections::HashSet<String>,
props: &std::collections::BTreeMap<String, std::sync::Arc<Schema>>,
errors: &mut Vec<crate::drop::Error>,
) -> std::collections::BTreeMap<String, crate::database::edge::Edge> {
let mut schema_edges = std::collections::BTreeMap::new();
// Determine the physical Database Table Name this schema structurally represents
// Plucks the polymorphic discriminator via dot-notation (e.g. extracting "person" from "full.person")
let mut parent_type_name = None;
if let Some(family) = &self.obj.family {
parent_type_name = Some(family.split('.').next_back().unwrap_or(family).to_string());
} else if let Some(identifier) = self.obj.identifier() {
parent_type_name = Some(
identifier
.split('.')
.next_back()
.unwrap_or(&identifier)
.to_string(),
);
}
if let Some(p_type) = parent_type_name {
// Proceed only if the resolved table physically exists within the Postgres Type hierarchy
if db.types.contains_key(&p_type) {
// Iterate over all discovered schema boundaries mapped inside the object
for (prop_name, prop_schema) in props {
let mut child_type_name = None;
let mut target_schema = prop_schema.clone();
let mut is_array = false;
// Structurally unpack the inner target entity if the object maps to an array list
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) =
&prop_schema.obj.type_
{
if t == "array" {
is_array = true;
if let Some(items) = &prop_schema.obj.items {
target_schema = items.clone();
}
}
}
// Determine the physical Postgres table backing the nested child schema recursively
if let Some(family) = &target_schema.obj.family {
child_type_name = Some(family.split('.').next_back().unwrap_or(family).to_string());
} else if let Some(ref_id) = target_schema.obj.identifier() {
child_type_name = Some(ref_id.split('.').next_back().unwrap_or(&ref_id).to_string());
} else if let Some(arr) = &target_schema.obj.one_of {
if let Some(first) = arr.first() {
if let Some(ref_id) = first.obj.identifier() {
child_type_name =
Some(ref_id.split('.').next_back().unwrap_or(&ref_id).to_string());
}
}
}
if let Some(c_type) = child_type_name {
if db.types.contains_key(&c_type) {
// Ensure the child Schema's AST has accurately compiled its own physical property keys so we can
// inject them securely for Many-to-Many Twin Deduction disambiguation matching.
target_schema.compile(db, visited, errors);
if let Some(compiled_target_props) = target_schema.obj.compiled_properties.get() {
let keys_for_ambiguity: Vec<String> =
compiled_target_props.keys().cloned().collect();
// Interrogate the Database catalog graph to discover the exact Foreign Key Constraint connecting the components
if let Some((relation, is_forward)) = db.resolve_relation(
&p_type,
&c_type,
prop_name,
Some(&keys_for_ambiguity),
is_array,
self.id.as_deref(),
&format!("/{}", prop_name),
errors,
) {
schema_edges.insert(
prop_name.clone(),
crate::database::edge::Edge {
constraint: relation.constraint.clone(),
forward: is_forward,
},
);
}
}
}
}
}
}
}
schema_edges
}
pub fn compile_polymorphism(
&self,
db: &crate::database::Database,
errors: &mut Vec<crate::drop::Error>,
) {
let mut options = std::collections::BTreeMap::new();
let mut strategy = String::new();
if let Some(family) = &self.obj.family {
let family_base = family.split('.').next_back().unwrap_or(family).to_string();
let family_prefix = family
.strip_suffix(&family_base)
.unwrap_or("")
.trim_end_matches('.');
if let Some(type_def) = db.types.get(&family_base) {
if type_def.variations.len() > 1 && type_def.variations.iter().any(|v| v != &family_base) {
// Scenario A / B: Table Variations
strategy = "type".to_string();
for var in &type_def.variations {
let target_id = if family_prefix.is_empty() {
var.to_string()
} else {
format!("{}.{}", family_prefix, var)
};
if db.schemas.contains_key(&target_id) {
options.insert(var.to_string(), target_id);
}
}
} else {
// Scenario C: Single Table Inheritance (Horizontal)
strategy = "kind".to_string();
let suffix = format!(".{}", family_base);
for schema in &type_def.schemas {
if let Some(id) = &schema.obj.id {
if id.ends_with(&suffix) || id == &family_base {
if let Some(kind_val) = schema.obj.get_discriminator_value("kind") {
options.insert(kind_val, id.to_string());
}
}
}
}
}
}
} else if let Some(one_of) = &self.obj.one_of {
let mut type_vals = std::collections::HashSet::new();
let mut kind_vals = std::collections::HashSet::new();
for c in one_of {
if let Some(t_val) = c.obj.get_discriminator_value("type") {
type_vals.insert(t_val);
}
if let Some(k_val) = c.obj.get_discriminator_value("kind") {
kind_vals.insert(k_val);
}
}
strategy = if type_vals.len() > 1 && type_vals.len() == one_of.len() {
"type".to_string()
} else if kind_vals.len() > 1 && kind_vals.len() == one_of.len() {
"kind".to_string()
} else {
"".to_string()
};
if strategy.is_empty() {
return;
}
for c in one_of {
if let Some(val) = c.obj.get_discriminator_value(&strategy) {
if options.contains_key(&val) {
errors.push(crate::drop::Error {
code: "POLYMORPHIC_COLLISION".to_string(),
message: format!("Polymorphic boundary defines multiple candidates mapped to the identical discriminator value '{}'.", val),
details: crate::drop::ErrorDetails::default()
});
continue;
}
let mut target_id = c.obj.id.clone();
if target_id.is_none() {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &c.obj.type_ {
if !crate::database::object::is_primitive_type(t) {
target_id = Some(t.clone());
}
}
}
if let Some(tid) = target_id {
options.insert(val, tid);
}
}
}
} else {
return;
}
if !options.is_empty() {
let _ = self.obj.compiled_discriminator.set(strategy);
let _ = self.obj.compiled_options.set(options);
}
}
#[allow(unused_variables)]
fn validate_identifier(id: &str, field_name: &str, errors: &mut Vec<crate::drop::Error>) {
#[cfg(not(test))]
@ -444,8 +420,8 @@ impl Schema {
Self::validate_identifier(id, "$id", errors);
to_insert.push((id.clone(), self.clone()));
}
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::object::is_primitive_type(t) {
Self::validate_identifier(t, "type", errors);
}
}
@ -456,8 +432,8 @@ impl Schema {
// Is this schema an inline ad-hoc composition?
// Meaning it has a tracking context, lacks an explicit $id, but extends an Entity ref with explicit properties!
if self.obj.id.is_none() && self.obj.properties.is_some() {
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &self.obj.type_ {
if !crate::database::object::is_primitive_type(t) {
if let Some(ref path) = tracking_path {
to_insert.push((path.clone(), self.clone()));
}
@ -549,285 +525,6 @@ impl Schema {
}
}
}
/// Dynamically infers and compiles all structural database relationships between this Schema
/// and its nested children. This functions recursively traverses the JSON Schema abstract syntax
/// tree, identifies physical PostgreSQL table boundaries, and locks the resulting relation
/// constraint paths directly onto the `compiled_edges` map in O(1) memory.
pub fn compile_edges(
&self,
db: &crate::database::Database,
visited: &mut std::collections::HashSet<String>,
props: &std::collections::BTreeMap<String, std::sync::Arc<Schema>>,
errors: &mut Vec<crate::drop::Error>,
) -> std::collections::BTreeMap<String, crate::database::edge::Edge> {
let mut schema_edges = std::collections::BTreeMap::new();
// Determine the physical Database Table Name this schema structurally represents
// Plucks the polymorphic discriminator via dot-notation (e.g. extracting "person" from "full.person")
let mut parent_type_name = None;
if let Some(family) = &self.obj.family {
parent_type_name = Some(family.split('.').next_back().unwrap_or(family).to_string());
} else if let Some(identifier) = self.obj.identifier() {
parent_type_name = Some(
identifier
.split('.')
.next_back()
.unwrap_or(&identifier)
.to_string(),
);
}
if let Some(p_type) = parent_type_name {
// Proceed only if the resolved table physically exists within the Postgres Type hierarchy
if db.types.contains_key(&p_type) {
// Iterate over all discovered schema boundaries mapped inside the object
for (prop_name, prop_schema) in props {
let mut child_type_name = None;
let mut target_schema = prop_schema.clone();
let mut is_array = false;
// Structurally unpack the inner target entity if the object maps to an array list
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) =
&prop_schema.obj.type_
{
if t == "array" {
is_array = true;
if let Some(items) = &prop_schema.obj.items {
target_schema = items.clone();
}
}
}
// Determine the physical Postgres table backing the nested child schema recursively
if let Some(family) = &target_schema.obj.family {
child_type_name = Some(family.split('.').next_back().unwrap_or(family).to_string());
} else if let Some(ref_id) = target_schema.obj.identifier() {
child_type_name = Some(ref_id.split('.').next_back().unwrap_or(&ref_id).to_string());
} else if let Some(arr) = &target_schema.obj.one_of {
if let Some(first) = arr.first() {
if let Some(ref_id) = first.obj.identifier() {
child_type_name =
Some(ref_id.split('.').next_back().unwrap_or(&ref_id).to_string());
}
}
}
if let Some(c_type) = child_type_name {
if db.types.contains_key(&c_type) {
// Ensure the child Schema's AST has accurately compiled its own physical property keys so we can
// inject them securely for Many-to-Many Twin Deduction disambiguation matching.
target_schema.compile(db, visited, errors);
if let Some(compiled_target_props) = target_schema.obj.compiled_properties.get() {
let keys_for_ambiguity: Vec<String> =
compiled_target_props.keys().cloned().collect();
// Interrogate the Database catalog graph to discover the exact Foreign Key Constraint connecting the components
if let Some((relation, is_forward)) = resolve_relation(
db,
&p_type,
&c_type,
prop_name,
Some(&keys_for_ambiguity),
is_array,
self.id.as_deref(),
&format!("/{}", prop_name),
errors,
) {
schema_edges.insert(
prop_name.clone(),
crate::database::edge::Edge {
constraint: relation.constraint.clone(),
forward: is_forward,
},
);
}
}
}
}
}
}
}
schema_edges
}
}
/// Inspects the Postgres pg_constraint relations catalog to securely identify
/// the precise Foreign Key connecting a parent and child hierarchy path.
pub(crate) fn resolve_relation<'a>(
db: &'a crate::database::Database,
parent_type: &str,
child_type: &str,
prop_name: &str,
relative_keys: Option<&Vec<String>>,
is_array: bool,
schema_id: Option<&str>,
path: &str,
errors: &mut Vec<crate::drop::Error>,
) -> Option<(&'a crate::database::relation::Relation, bool)> {
// Enforce graph locality by ensuring we don't accidentally crawl to pure structural entity boundaries
if parent_type == "entity" && child_type == "entity" {
return None;
}
let p_def = db.types.get(parent_type)?;
let c_def = db.types.get(child_type)?;
let mut matching_rels = Vec::new();
let mut directions = Vec::new();
// Scour the complete catalog for any Edge matching the inheritance scope of the two objects
// This automatically binds polymorphic structures (e.g. recognizing a relationship targeting User
// also natively binds instances specifically typed as Person).
let mut all_rels: Vec<&crate::database::relation::Relation> = db.relations.values().collect();
all_rels.sort_by(|a, b| a.constraint.cmp(&b.constraint));
for rel in all_rels {
let mut is_forward =
p_def.hierarchy.contains(&rel.source_type) && c_def.hierarchy.contains(&rel.destination_type);
let is_reverse =
p_def.hierarchy.contains(&rel.destination_type) && c_def.hierarchy.contains(&rel.source_type);
// Structural Cardinality Filtration:
// If the schema requires a collection (Array), it is mathematically impossible for a pure
// Forward scalar edge (where the parent holds exactly one UUID pointer) to fulfill a One-to-Many request.
// Thus, if it's an array, we fully reject pure Forward edges and only accept Reverse edges (or Junction edges).
if is_array && is_forward && !is_reverse {
is_forward = false;
}
if is_forward {
matching_rels.push(rel);
directions.push(true);
} else if is_reverse {
matching_rels.push(rel);
directions.push(false);
}
}
// Abort relation discovery early if no hierarchical inheritance match was found
if matching_rels.is_empty() {
let mut details = crate::drop::ErrorDetails {
path: path.to_string(),
..Default::default()
};
if let Some(sid) = schema_id {
details.schema = Some(sid.to_string());
}
errors.push(crate::drop::Error {
code: "EDGE_MISSING".to_string(),
message: format!(
"No database relation exists between '{}' and '{}' for property '{}'",
parent_type, child_type, prop_name
),
details,
});
return None;
}
// Ideal State: The objects only share a solitary structural relation, resolving ambiguity instantly.
if matching_rels.len() == 1 {
return Some((matching_rels[0], directions[0]));
}
let mut chosen_idx = 0;
let mut resolved = false;
// Exact Prefix Disambiguation: Determine if the database specifically names this constraint
// directly mapping to the JSON Schema property name (e.g., `fk_{child}_{property_name}`)
for (i, rel) in matching_rels.iter().enumerate() {
if let Some(prefix) = &rel.prefix {
if prop_name.starts_with(prefix)
|| prefix.starts_with(prop_name)
|| prefix.replace("_", "") == prop_name.replace("_", "")
{
chosen_idx = i;
resolved = true;
break;
}
}
}
// Complex Subgraph Resolution: The database contains multiple equally explicit foreign key constraints
// linking these objects (such as pointing to `source` and `target` in Many-to-Many junction models).
if !resolved && relative_keys.is_some() {
// Twin Deduction Pass 1: We inspect the exact properties structurally defined inside the compiled payload
// to observe which explicit relation arrow the child payload natively consumes.
let keys = relative_keys.unwrap();
let mut consumed_rel_idx = None;
for (i, rel) in matching_rels.iter().enumerate() {
if let Some(prefix) = &rel.prefix {
if keys.contains(prefix) {
consumed_rel_idx = Some(i);
break; // Found the routing edge explicitly consumed by the schema payload
}
}
}
// Twin Deduction Pass 2: Knowing which arrow points outbound, we can mathematically deduce its twin
// providing the reverse ownership on the same junction boundary must be the incoming Edge to the parent.
if let Some(used_idx) = consumed_rel_idx {
let used_rel = matching_rels[used_idx];
let mut twin_ids = Vec::new();
for (i, rel) in matching_rels.iter().enumerate() {
if i != used_idx
&& rel.source_type == used_rel.source_type
&& rel.destination_type == used_rel.destination_type
&& rel.prefix.is_some()
{
twin_ids.push(i);
}
}
if twin_ids.len() == 1 {
chosen_idx = twin_ids[0];
resolved = true;
}
}
}
// Implicit Base Fallback: If no complex explicit paths resolve, but exactly one relation
// sits entirely naked (without a constraint prefix), it must be the core structural parent ownership.
if !resolved {
let mut null_prefix_ids = Vec::new();
for (i, rel) in matching_rels.iter().enumerate() {
if rel.prefix.is_none() {
null_prefix_ids.push(i);
}
}
if null_prefix_ids.len() == 1 {
chosen_idx = null_prefix_ids[0];
resolved = true;
}
}
// If we exhausted all mathematical deduction pathways and STILL cannot isolate a single edge,
// we must abort rather than silently guessing. Returning None prevents arbitrary SQL generation
// and forces a clean structural error for the architect.
if !resolved {
let mut details = crate::drop::ErrorDetails {
path: path.to_string(),
context: serde_json::to_value(&matching_rels).ok(),
cause: Some("Multiple conflicting constraints found matching prefixes".to_string()),
..Default::default()
};
if let Some(sid) = schema_id {
details.schema = Some(sid.to_string());
}
errors.push(crate::drop::Error {
code: "AMBIGUOUS_TYPE_RELATIONS".to_string(),
message: format!(
"Ambiguous database relation between '{}' and '{}' for property '{}'",
parent_type, child_type, prop_name
),
details,
});
return None;
}
Some((matching_rels[chosen_idx], directions[chosen_idx]))
}
impl<'de> Deserialize<'de> for Schema {
@ -880,38 +577,3 @@ impl<'de> Deserialize<'de> for Schema {
})
}
}
impl SchemaObject {
pub fn identifier(&self) -> Option<String> {
if let Some(id) = &self.id {
return Some(id.split('.').next_back().unwrap_or("").to_string());
}
if let Some(SchemaTypeOrArray::Single(t)) = &self.type_ {
if !is_primitive_type(t) {
return Some(t.split('.').next_back().unwrap_or("").to_string());
}
}
None
}
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum SchemaTypeOrArray {
Single(String),
Multiple(Vec<String>),
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Action {
#[serde(skip_serializing_if = "Option::is_none")]
pub navigate: Option<String>,
#[serde(skip_serializing_if = "Option::is_none")]
pub punc: Option<String>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum Dependency {
Props(Vec<String>),
Schema(Arc<Schema>),
}

View File

@ -25,7 +25,7 @@ impl Merger {
let mut notifications_queue = Vec::new();
let target_schema = match self.db.schemas.get(schema_id) {
Some(s) => Arc::new(s.clone()),
Some(s) => Arc::clone(s),
None => {
return crate::drop::Drop::with_errors(vec![crate::drop::Error {
code: "MERGE_FAILED".to_string(),
@ -131,13 +131,33 @@ impl Merger {
pub(crate) fn merge_internal(
&self,
schema: Arc<crate::database::schema::Schema>,
mut schema: Arc<crate::database::schema::Schema>,
data: Value,
notifications: &mut Vec<String>,
) -> Result<Value, String> {
match data {
Value::Array(items) => self.merge_array(schema, items, notifications),
Value::Object(map) => self.merge_object(schema, map, notifications),
Value::Object(map) => {
if let Some(options) = schema.obj.compiled_options.get() {
if let Some(disc) = schema.obj.compiled_discriminator.get() {
let val = map.get(disc).and_then(|v| v.as_str());
if let Some(v) = val {
if let Some(target_id) = options.get(v) {
if let Some(target_schema) = self.db.schemas.get(target_id) {
schema = Arc::clone(target_schema);
} else {
return Err(format!("Polymorphic mapped target '{}' not found in database registry", target_id));
}
} else {
return Err(format!("Polymorphic discriminator {}='{}' matched no compiled options", disc, v));
}
} else {
return Err(format!("Polymorphic merging failed: missing required discriminator '{}'", disc));
}
}
}
self.merge_object(schema, map, notifications)
},
_ => Err("Invalid merge payload: root must be an Object or Array".to_string()),
}
}
@ -149,7 +169,7 @@ impl Merger {
notifications: &mut Vec<String>,
) -> Result<Value, String> {
let mut item_schema = schema.clone();
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &schema.obj.type_ {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &schema.obj.type_ {
if t == "array" {
if let Some(items_def) = &schema.obj.items {
item_schema = items_def.clone();
@ -369,7 +389,7 @@ impl Merger {
);
let mut item_schema = rel_schema.clone();
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) =
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) =
&rel_schema.obj.type_
{
if t == "array" {

View File

@ -1,5 +1,6 @@
use crate::database::Database;
use std::sync::Arc;
pub struct Compiler<'a> {
pub db: &'a Database,
pub filter_keys: &'a [String],
@ -16,6 +17,7 @@ pub struct Node<'a> {
pub property_name: Option<String>,
pub depth: usize,
pub ast_path: String,
pub is_polymorphic_branch: bool,
}
impl<'a> Compiler<'a> {
@ -27,7 +29,7 @@ impl<'a> Compiler<'a> {
.get(schema_id)
.ok_or_else(|| format!("Schema not found: {}", schema_id))?;
let target_schema = std::sync::Arc::new(schema.clone());
let target_schema = std::sync::Arc::clone(schema);
let mut compiler = Compiler {
db: &self.db,
@ -44,6 +46,7 @@ impl<'a> Compiler<'a> {
property_name: None,
depth: 0,
ast_path: String::new(),
is_polymorphic_branch: false,
};
let (sql, _) = compiler.compile_node(node)?;
@ -55,7 +58,7 @@ impl<'a> Compiler<'a> {
fn compile_node(&mut self, node: Node<'a>) -> Result<(String, String), String> {
// Determine the base schema type (could be an array, object, or literal)
match &node.schema.obj.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(t)) if t == "array" => {
Some(crate::database::object::SchemaTypeOrArray::Single(t)) if t == "array" => {
self.compile_array(node)
}
_ => self.compile_reference(node),
@ -103,7 +106,11 @@ impl<'a> Compiler<'a> {
let mut resolved_type = None;
if let Some(family_target) = node.schema.obj.family.as_ref() {
resolved_type = self.db.types.get(family_target);
let base_type_name = family_target
.split('.')
.next_back()
.unwrap_or(family_target);
resolved_type = self.db.types.get(base_type_name);
} else if let Some(base_type_name) = node.schema.obj.identifier() {
resolved_type = self.db.types.get(&base_type_name);
}
@ -113,46 +120,30 @@ impl<'a> Compiler<'a> {
}
// Handle Direct Refs via type pointer
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t)) = &node.schema.obj.type_ {
if !crate::database::schema::is_primitive_type(t) {
if let Some(crate::database::object::SchemaTypeOrArray::Single(t)) = &node.schema.obj.type_ {
if !crate::database::object::is_primitive_type(t) {
// If it's just an ad-hoc struct ref, we should resolve it
if let Some(target_schema) = self.db.schemas.get(t) {
let mut ref_node = node.clone();
ref_node.schema = std::sync::Arc::new(target_schema.clone());
ref_node.schema = Arc::clone(target_schema);
return self.compile_node(ref_node);
}
return Err(format!("Unresolved schema type pointer: {}", t));
}
}
// Handle $family Polymorphism fallbacks for relations
if let Some(family_target) = &node.schema.obj.family {
let mut all_targets = vec![family_target.clone()];
if let Some(descendants) = self.db.descendants.get(family_target) {
all_targets.extend(descendants.clone());
}
if all_targets.len() == 1 {
let mut bypass_schema = crate::database::schema::Schema::default();
bypass_schema.obj.type_ = Some(crate::database::schema::SchemaTypeOrArray::Single(all_targets[0].clone()));
let mut bypass_node = node.clone();
bypass_node.schema = std::sync::Arc::new(bypass_schema);
return self.compile_node(bypass_node);
}
all_targets.sort();
let mut family_schemas = Vec::new();
for variation in &all_targets {
let mut ref_schema = crate::database::schema::Schema::default();
ref_schema.obj.type_ = Some(crate::database::schema::SchemaTypeOrArray::Single(variation.clone()));
family_schemas.push(std::sync::Arc::new(ref_schema));
}
return self.compile_one_of(&family_schemas, node);
}
// Handle oneOf Polymorphism fallbacks for relations
if let Some(one_of) = &node.schema.obj.one_of {
return self.compile_one_of(one_of, node.clone());
// Handle Polymorphism fallbacks for relations
if node.schema.obj.family.is_some() || node.schema.obj.one_of.is_some() {
if let Some(options) = node.schema.obj.compiled_options.get() {
if options.len() == 1 {
let target_id = options.values().next().unwrap();
let mut bypass_schema = crate::database::schema::Schema::default();
bypass_schema.obj.type_ = Some(crate::database::object::SchemaTypeOrArray::Single(target_id.clone()));
let mut bypass_node = node.clone();
bypass_node.schema = std::sync::Arc::new(bypass_schema);
return self.compile_node(bypass_node);
}
}
return self.compile_one_of(node);
}
// Just an inline object definition?
@ -179,17 +170,28 @@ impl<'a> Compiler<'a> {
) -> Result<(String, String), String> {
let (table_aliases, from_clauses) = self.compile_from_clause(r#type);
// 2. Map properties and build jsonb_build_object args
let mut select_args = self.compile_select_clause(r#type, &table_aliases, node.clone())?;
let jsonb_obj_sql = if node.schema.obj.family.is_some() || node.schema.obj.one_of.is_some() {
let base_alias = table_aliases
.get(&r#type.name)
.cloned()
.unwrap_or_else(|| node.parent_alias.to_string());
let mut case_node = node.clone();
case_node.parent_alias = base_alias.clone();
let arc_aliases = std::sync::Arc::new(table_aliases.clone());
case_node.parent_type_aliases = Some(arc_aliases);
case_node.parent_type = Some(r#type);
// 2.5 Inject polymorphism directly into the query object
let mut poly_args = self.compile_polymorphism_select(r#type, &table_aliases, node.clone())?;
select_args.append(&mut poly_args);
let jsonb_obj_sql = if select_args.is_empty() {
"jsonb_build_object()".to_string()
let (case_sql, _) = self.compile_one_of(case_node)?;
case_sql
} else {
format!("jsonb_build_object({})", select_args.join(", "))
let select_args = self.compile_select_clause(r#type, &table_aliases, node.clone())?;
if select_args.is_empty() {
"jsonb_build_object()".to_string()
} else {
format!("jsonb_build_object({})", select_args.join(", "))
}
};
// 3. Build WHERE clauses
@ -218,90 +220,6 @@ impl<'a> Compiler<'a> {
))
}
fn compile_polymorphism_select(
&mut self,
r#type: &'a crate::database::r#type::Type,
table_aliases: &std::collections::HashMap<String, String>,
node: Node<'a>,
) -> Result<Vec<String>, String> {
let mut select_args = Vec::new();
if let Some(family_target) = node.schema.obj.family.as_ref() {
let family_prefix = family_target.rfind('.').map(|idx| &family_target[..idx]);
let mut all_targets = vec![family_target.clone()];
if let Some(descendants) = self.db.descendants.get(family_target) {
all_targets.extend(descendants.clone());
}
// Filter targets to EXACTLY match the family_target prefix
let mut final_targets = Vec::new();
for target in all_targets {
let target_prefix = target.rfind('.').map(|idx| &target[..idx]);
if target_prefix == family_prefix {
final_targets.push(target);
}
}
final_targets.sort();
final_targets.dedup();
if final_targets.len() == 1 {
let variation = &final_targets[0];
if let Some(target_schema) = self.db.schemas.get(variation) {
let mut bypass_node = node.clone();
bypass_node.schema = std::sync::Arc::new(target_schema.clone());
let mut bypassed_args = self.compile_select_clause(r#type, table_aliases, bypass_node)?;
select_args.append(&mut bypassed_args);
} else {
return Err(format!("Could not find schema for variation {}", variation));
}
} else {
let mut family_schemas = Vec::new();
for variation in &final_targets {
if let Some(target_schema) = self.db.schemas.get(variation) {
family_schemas.push(std::sync::Arc::new(target_schema.clone()));
} else {
return Err(format!(
"Could not find schema metadata for variation {}",
variation
));
}
}
let base_alias = table_aliases
.get(&r#type.name)
.cloned()
.unwrap_or_else(|| node.parent_alias.to_string());
select_args.push(format!("'id', {}.id", base_alias));
let mut case_node = node.clone();
case_node.parent_alias = base_alias.clone();
let arc_aliases = std::sync::Arc::new(table_aliases.clone());
case_node.parent_type_aliases = Some(arc_aliases);
let (case_sql, _) = self.compile_one_of(&family_schemas, case_node)?;
select_args.push(format!("'type', {}", case_sql));
}
} else if let Some(one_of) = &node.schema.obj.one_of {
let base_alias = table_aliases
.get(&r#type.name)
.cloned()
.unwrap_or_else(|| node.parent_alias.to_string());
select_args.push(format!("'id', {}.id", base_alias));
let mut case_node = node.clone();
case_node.parent_alias = base_alias.clone();
let arc_aliases = std::sync::Arc::new(table_aliases.clone());
case_node.parent_type_aliases = Some(arc_aliases);
let (case_sql, _) = self.compile_one_of(one_of, case_node)?;
select_args.push(format!("'type', {}", case_sql));
}
Ok(select_args)
}
fn compile_object(
&mut self,
props: &std::collections::BTreeMap<String, std::sync::Arc<crate::database::schema::Schema>>,
@ -333,26 +251,45 @@ impl<'a> Compiler<'a> {
fn compile_one_of(
&mut self,
schemas: &[Arc<crate::database::schema::Schema>],
node: Node<'a>,
) -> Result<(String, String), String> {
let mut case_statements = Vec::new();
let options = node.schema.obj.compiled_options.get().ok_or("Missing compiled options for polymorphism")?;
let disc = node.schema.obj.compiled_discriminator.get().ok_or("Missing compiled discriminator for polymorphism")?;
let type_col = if let Some(prop) = &node.property_name {
format!("{}_type", prop)
format!("{}_{}", prop, disc)
} else {
"type".to_string()
disc.to_string()
};
for option_schema in schemas {
if let Some(base_type_name) = option_schema.obj.identifier() {
// Generate the nested SQL for this specific target type
for (disc_val, target_id) in options {
if let Some(target_schema) = self.db.schemas.get(target_id) {
let mut child_node = node.clone();
child_node.schema = std::sync::Arc::clone(option_schema);
let (val_sql, _) = self.compile_node(child_node)?;
child_node.schema = Arc::clone(target_schema);
child_node.is_polymorphic_branch = true;
let val_sql = if disc == "kind" && node.parent_type.is_some() && node.parent_type_aliases.is_some() {
let aliases_arc = node.parent_type_aliases.as_ref().unwrap();
let aliases = aliases_arc.as_ref();
let p_type = node.parent_type.unwrap();
let select_args = self.compile_select_clause(p_type, aliases, child_node.clone())?;
if select_args.is_empty() {
"jsonb_build_object()".to_string()
} else {
format!("jsonb_build_object({})", select_args.join(", "))
}
} else {
let (sql, _) = self.compile_node(child_node)?;
sql
};
case_statements.push(format!(
"WHEN {}.{} = '{}' THEN ({})",
node.parent_alias, type_col, base_type_name, val_sql
node.parent_alias, type_col, disc_val, val_sql
));
}
}
@ -401,7 +338,9 @@ impl<'a> Compiler<'a> {
) -> Result<Vec<String>, String> {
let mut select_args = Vec::new();
let grouped_fields = r#type.grouped_fields.as_ref().and_then(|v| v.as_object());
let merged_props = node.schema.obj.compiled_properties.get().unwrap();
let default_props = std::collections::BTreeMap::new();
let merged_props = node.schema.obj.compiled_properties.get().unwrap_or(&default_props);
let mut sorted_keys: Vec<&String> = merged_props.keys().collect();
sorted_keys.sort();
@ -409,18 +348,18 @@ impl<'a> Compiler<'a> {
let prop_schema = &merged_props[prop_key];
let is_object_or_array = match &prop_schema.obj.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(s)) => {
Some(crate::database::object::SchemaTypeOrArray::Single(s)) => {
s == "object" || s == "array"
}
Some(crate::database::schema::SchemaTypeOrArray::Multiple(v)) => {
Some(crate::database::object::SchemaTypeOrArray::Multiple(v)) => {
v.contains(&"object".to_string()) || v.contains(&"array".to_string())
}
_ => false,
};
let is_custom_object_pointer = match &prop_schema.obj.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(s)) => {
!crate::database::schema::is_primitive_type(s)
Some(crate::database::object::SchemaTypeOrArray::Single(s)) => {
!crate::database::object::is_primitive_type(s)
}
_ => false,
};
@ -470,6 +409,7 @@ impl<'a> Compiler<'a> {
} else {
format!("{}/{}", node.ast_path, prop_key)
},
is_polymorphic_branch: false,
};
let (val_sql, val_type) = self.compile_node(child_node)?;
@ -509,6 +449,8 @@ impl<'a> Compiler<'a> {
self.compile_filter_conditions(r#type, type_aliases, &node, &base_alias, &mut where_clauses);
self.compile_polymorphic_bounds(r#type, type_aliases, &node, &mut where_clauses);
let start_len = where_clauses.len();
self.compile_relation_conditions(
r#type,
type_aliases,
@ -517,6 +459,14 @@ impl<'a> Compiler<'a> {
&mut where_clauses,
)?;
if node.is_polymorphic_branch && where_clauses.len() == start_len {
if let Some(parent_aliases) = &node.parent_type_aliases {
if let Some(outer_entity_alias) = parent_aliases.get("entity") {
where_clauses.push(format!("{}.id = {}.id", entity_alias, outer_entity_alias));
}
}
}
Ok(where_clauses)
}

View File

@ -1439,6 +1439,18 @@ fn test_queryer_0_10() {
crate::tests::runner::run_test_case(&path, 0, 10).unwrap();
}
#[test]
fn test_queryer_0_11() {
let path = format!("{}/fixtures/queryer.json", env!("CARGO_MANIFEST_DIR"));
crate::tests::runner::run_test_case(&path, 0, 11).unwrap();
}
#[test]
fn test_queryer_0_12() {
let path = format!("{}/fixtures/queryer.json", env!("CARGO_MANIFEST_DIR"));
crate::tests::runner::run_test_case(&path, 0, 12).unwrap();
}
#[test]
fn test_polymorphism_0_0() {
let path = format!("{}/fixtures/polymorphism.json", env!("CARGO_MANIFEST_DIR"));

View File

@ -13,7 +13,7 @@ impl<'a> ValidationContext<'a> {
if let Some(ref type_) = self.schema.type_ {
match type_ {
crate::database::schema::SchemaTypeOrArray::Single(t) => {
crate::database::object::SchemaTypeOrArray::Single(t) => {
if !Validator::check_type(t, current) {
result.errors.push(ValidationError {
code: "INVALID_TYPE".to_string(),
@ -22,7 +22,7 @@ impl<'a> ValidationContext<'a> {
});
}
}
crate::database::schema::SchemaTypeOrArray::Multiple(types) => {
crate::database::object::SchemaTypeOrArray::Multiple(types) => {
let mut valid = false;
for t in types {
if Validator::check_type(t, current) {

View File

@ -10,7 +10,7 @@ impl<'a> ValidationContext<'a> {
let current = self.instance;
if let Some(compiled_fmt) = self.schema.compiled_format.get() {
match compiled_fmt {
crate::database::schema::CompiledFormat::Func(f) => {
crate::database::object::CompiledFormat::Func(f) => {
let should = if let Some(s) = current.as_str() {
!s.is_empty()
} else {
@ -24,7 +24,7 @@ impl<'a> ValidationContext<'a> {
});
}
}
crate::database::schema::CompiledFormat::Regex(re) => {
crate::database::object::CompiledFormat::Regex(re) => {
if let Some(s) = current.as_str()
&& !re.is_match(s)
{

View File

@ -124,7 +124,7 @@ impl<'a> ValidationContext<'a> {
for (prop, dep) in deps {
if obj.contains_key(prop) {
match dep {
crate::database::schema::Dependency::Props(required_props) => {
crate::database::object::Dependency::Props(required_props) => {
for req_prop in required_props {
if !obj.contains_key(req_prop) {
result.errors.push(ValidationError {
@ -135,7 +135,7 @@ impl<'a> ValidationContext<'a> {
}
}
}
crate::database::schema::Dependency::Schema(dep_schema) => {
crate::database::object::Dependency::Schema(dep_schema) => {
let derived = self.derive_for_schema(dep_schema, false);
let dep_res = derived.validate()?;
result.evaluated_keys.extend(dep_res.evaluated_keys.clone());
@ -155,7 +155,7 @@ impl<'a> ValidationContext<'a> {
if let Some(child_instance) = obj.get(key) {
let new_path = self.join_path(key);
let is_ref = match &sub_schema.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(t)) => !crate::database::schema::is_primitive_type(t),
Some(crate::database::object::SchemaTypeOrArray::Single(t)) => !crate::database::object::is_primitive_type(t),
_ => false,
};
let next_extensible = if is_ref { false } else { self.extensible };
@ -184,7 +184,7 @@ impl<'a> ValidationContext<'a> {
if compiled_re.0.is_match(key) {
let new_path = self.join_path(key);
let is_ref = match &sub_schema.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(t)) => !crate::database::schema::is_primitive_type(t),
Some(crate::database::object::SchemaTypeOrArray::Single(t)) => !crate::database::object::is_primitive_type(t),
_ => false,
};
let next_extensible = if is_ref { false } else { self.extensible };
@ -226,7 +226,7 @@ impl<'a> ValidationContext<'a> {
if !locally_matched {
let new_path = self.join_path(key);
let is_ref = match &additional_schema.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(t)) => !crate::database::schema::is_primitive_type(t),
Some(crate::database::object::SchemaTypeOrArray::Single(t)) => !crate::database::object::is_primitive_type(t),
_ => false,
};
let next_extensible = if is_ref { false } else { self.extensible };

View File

@ -1,4 +1,3 @@
use crate::database::schema::Schema;
use crate::validator::context::ValidationContext;
use crate::validator::error::ValidationError;
use crate::validator::result::ValidationResult;
@ -29,30 +28,10 @@ impl<'a> ValidationContext<'a> {
}
}
if let Some(family_target) = &self.schema.family {
if let Some(descendants) = self.db.descendants.get(family_target) {
let mut candidates = Vec::new();
// Add the target base schema itself
if let Some(base_schema) = self.db.schemas.get(family_target) {
candidates.push(base_schema);
}
// Add all descendants
for child_id in descendants {
if let Some(child_schema) = self.db.schemas.get(child_id) {
candidates.push(child_schema);
}
}
// Use prefix from family string (e.g. `light.`)
let prefix = family_target
.rsplit_once('.')
.map(|(p, _)| format!("{}.", p))
.unwrap_or_default();
if !self.validate_polymorph(&candidates, Some(&prefix), result)? {
return Ok(false);
if self.schema.family.is_some() {
if let Some(options) = self.schema.compiled_options.get() {
if let Some(disc) = self.schema.compiled_discriminator.get() {
return self.execute_polymorph(disc, options, result);
}
}
}
@ -64,212 +43,43 @@ impl<'a> ValidationContext<'a> {
&self,
result: &mut ValidationResult,
) -> Result<bool, ValidationError> {
if let Some(ref one_of) = self.schema.one_of {
let mut candidates = Vec::new();
for schema in one_of {
candidates.push(schema.as_ref());
if let Some(one_of) = &self.schema.one_of {
if let Some(options) = self.schema.compiled_options.get() {
if let Some(disc) = self.schema.compiled_discriminator.get() {
return self.execute_polymorph(disc, options, result);
}
}
if !self.validate_polymorph(&candidates, None, result)? {
return Ok(false);
}
}
Ok(true)
}
pub(crate) fn validate_polymorph(
&self,
candidates: &[&Schema],
family_prefix: Option<&str>,
result: &mut ValidationResult,
) -> Result<bool, ValidationError> {
let mut passed_candidates: Vec<(Option<String>, ValidationResult)> = Vec::new();
let mut failed_candidates: Vec<ValidationResult> = Vec::new();
// Native Draft2020-12 oneOf Evaluation Fallback
let mut valid_count = 0;
let mut final_successful_result = None;
let mut failed_candidates = Vec::new();
// 1. O(1) Fast-Path Router & Extractor
let instance_type = self.instance.as_object().and_then(|o| o.get("type")).and_then(|t| t.as_str());
let instance_kind = self.instance.as_object().and_then(|o| o.get("kind")).and_then(|k| k.as_str());
let mut viable_candidates = Vec::new();
for sub in candidates {
let _child_id = sub.identifier().unwrap_or_default();
let mut can_match = true;
if let Some(t) = instance_type {
// Fast Path 1: Pure Ad-Hoc Match (schema identifier == type)
// If it matches exactly, it's our golden candidate. Make all others non-viable manually?
// Wait, we loop through all and filter down. If exact match is found, we should ideally break and use ONLY that.
// Let's implement the logic safely.
let mut exact_match_found = false;
if let Some(schema_id) = &sub.id {
// Compute Vertical Exact Target (e.g. "person" or "light.person")
let exact_target = if let Some(prefix) = family_prefix {
format!("{}{}", prefix, t)
} else {
t.to_string()
};
// Fast Path 1 & 2: Vertical Exact Match
if schema_id == &exact_target {
if instance_kind.is_none() {
exact_match_found = true;
}
}
// Fast Path 3: Horizontal Sibling Match (kind + . + type)
if let Some(k) = instance_kind {
let sibling_target = format!("{}.{}", k, t);
if schema_id == &sibling_target {
exact_match_found = true;
}
}
}
if exact_match_found {
// We found an exact literal structural identity match!
// Wipe the existing viable_candidates and only yield this guy!
viable_candidates.clear();
viable_candidates.push(*sub);
break;
}
// Fast Path 4: Vertical Inheritance Fallback (Physical DB constraint)
if let Some(crate::database::schema::SchemaTypeOrArray::Single(t_ptr)) = &sub.type_ {
if !crate::database::schema::is_primitive_type(t_ptr) {
if let Some(base_type) = t_ptr.split('.').last() {
if let Some(type_def) = self.db.types.get(base_type) {
if !type_def.variations.contains(&t.to_string()) {
can_match = false;
}
} else {
if t_ptr != t {
can_match = false;
}
}
}
}
}
// Fast Path 5: Explicit Schema JSON `const` values check
if can_match {
if let Some(props) = &sub.properties {
if let Some(type_prop) = props.get("type") {
if let Some(const_val) = &type_prop.const_ {
if let Some(const_str) = const_val.as_str() {
if const_str != t {
can_match = false;
}
}
}
}
for child_schema in one_of {
let derived = self.derive_for_schema(child_schema, false);
if let Ok(sub_res) = derived.validate_scoped() {
if sub_res.is_valid() {
valid_count += 1;
final_successful_result = Some(sub_res.clone());
} else {
failed_candidates.push(sub_res);
}
}
}
if can_match {
viable_candidates.push(*sub);
}
}
println!("DEBUG VIABLE: {:?}", viable_candidates.iter().map(|s| s.id.clone()).collect::<Vec<_>>());
// 2. Evaluate Viable Candidates
// 2. Evaluate Viable Candidates
// Composition validation is natively handled directly via type compilation.
// The deprecated allOf JSON structure is no longer supported nor traversed.
for sub in viable_candidates.clone() {
let derived = self.derive_for_schema(sub, false);
let sub_res = derived.validate()?;
if sub_res.is_valid() {
passed_candidates.push((sub.id.clone(), sub_res));
} else {
failed_candidates.push(sub_res);
}
}
for f in &failed_candidates {
println!(" - Failed candidate errors: {:?}", f.errors.iter().map(|e| e.code.clone()).collect::<Vec<_>>());
}
if passed_candidates.len() == 1 {
result.merge(passed_candidates.pop().unwrap().1);
} else if passed_candidates.is_empty() {
// 3. Discriminator Pathing (Failure Analytics)
let type_path = self.join_path("type");
if instance_type.is_some() {
// Filter to candidates that didn't explicitly throw a CONST violation on `type`
let mut genuinely_failed = Vec::new();
for res in &failed_candidates {
let rejected_type = res.errors.iter().any(|e| {
(e.code == "CONST_VIOLATED" || e.code == "ENUM_VIOLATED") && e.path == type_path
});
if !rejected_type {
genuinely_failed.push(res.clone());
}
}
println!("DEBUG genuinely_failed len: {}", genuinely_failed.len());
if genuinely_failed.len() == 1 {
// Golden Type Match (1 candidate was structurally possible but failed property validation)
let sub_res = genuinely_failed.pop().unwrap();
result.errors.extend(sub_res.errors);
result.evaluated_keys.extend(sub_res.evaluated_keys);
return Ok(false);
} else {
// Pure Ad-Hoc Union
result.errors.push(ValidationError {
code: if self.schema.family.is_some() { "NO_FAMILY_MATCH".to_string() } else { "NO_ONEOF_MATCH".to_string() },
message: "Payload matches none of the required candidate sub-schemas".to_string(),
if valid_count == 1 {
if let Some(successful_res) = final_successful_result {
result.merge(successful_res);
}
return Ok(true);
} else if valid_count == 0 {
result.errors.push(ValidationError {
code: "NO_ONEOF_MATCH".to_string(),
message: "Payload matches none of the required candidate sub-schemas natively".to_string(),
path: self.path.to_string(),
});
for sub_res in &failed_candidates {
result.evaluated_keys.extend(sub_res.evaluated_keys.clone());
}
println!("DEBUG ELSE NO_FAMILY_MATCH RUNNING. Genuinely Failed len: {}", genuinely_failed.len());
if viable_candidates.is_empty() {
if let Some(obj) = self.instance.as_object() {
result.evaluated_keys.extend(obj.keys().cloned());
}
}
for sub_res in genuinely_failed {
for e in sub_res.errors {
if !result.errors.iter().any(|existing| existing.code == e.code && existing.path == e.path) {
result.errors.push(e);
}
}
}
return Ok(false);
}
} else {
// Instance missing type
// Instance missing type
let expects_type = viable_candidates.iter().any(|c| {
c.compiled_property_names.get().map_or(false, |props| props.contains(&"type".to_string()))
});
if expects_type {
result.errors.push(ValidationError {
code: "MISSING_TYPE".to_string(),
message: "Missing type discriminator. Unable to resolve polymorphic boundaries".to_string(),
path: self.path.to_string(),
});
for sub_res in failed_candidates {
result.evaluated_keys.extend(sub_res.evaluated_keys);
}
return Ok(false);
} else {
// Pure Ad-Hoc Union
result.errors.push(ValidationError {
code: if self.schema.family.is_some() { "NO_FAMILY_MATCH".to_string() } else { "NO_ONEOF_MATCH".to_string() },
message: "Payload matches none of the required candidate sub-schemas".to_string(),
path: self.path.to_string(),
});
if let Some(first) = failed_candidates.first() {
});
if let Some(first) = failed_candidates.first() {
let mut shared_errors = first.errors.clone();
for sub_res in failed_candidates.iter().skip(1) {
shared_errors.retain(|e1| {
@ -281,26 +91,66 @@ impl<'a> ValidationContext<'a> {
result.errors.push(e);
}
}
}
for sub_res in failed_candidates {
result.evaluated_keys.extend(sub_res.evaluated_keys);
}
return Ok(false);
}
}
return Ok(false);
} else {
result.errors.push(ValidationError {
code: "AMBIGUOUS_POLYMORPHIC_MATCH".to_string(),
message: "Matches multiple polymorphic candidates inextricably natively".to_string(),
path: self.path.to_string(),
});
return Ok(false);
}
} else {
result.errors.push(ValidationError {
code: "AMBIGUOUS_POLYMORPHIC_MATCH".to_string(),
message: "Matches multiple polymorphic candidates inextricably".to_string(),
path: self.path.to_string(),
});
}
Ok(true)
}
pub(crate) fn execute_polymorph(
&self,
disc: &str,
options: &std::collections::BTreeMap<String, String>,
result: &mut ValidationResult,
) -> Result<bool, ValidationError> {
// 1. O(1) Fast-Path Router & Extractor
let instance_val = self.instance.as_object().and_then(|o| o.get(disc)).and_then(|t| t.as_str());
if let Some(val) = instance_val {
result.evaluated_keys.insert(disc.to_string());
if let Some(target_id) = options.get(val) {
if let Some(target_schema) = self.db.schemas.get(target_id) {
let derived = self.derive_for_schema(target_schema.as_ref(), false);
let sub_res = derived.validate()?;
let is_valid = sub_res.is_valid();
result.merge(sub_res);
return Ok(is_valid);
} else {
result.errors.push(ValidationError {
code: "MISSING_COMPILED_SCHEMA".to_string(),
message: format!("Polymorphic router target '{}' does not exist in the database schemas map", target_id),
path: self.path.to_string(),
});
return Ok(false);
}
} else {
result.errors.push(ValidationError {
code: if self.schema.family.is_some() { "NO_FAMILY_MATCH".to_string() } else { "NO_ONEOF_MATCH".to_string() },
message: format!("Payload provided discriminator {}='{}' which matches none of the required candidate sub-schemas", disc, val),
path: self.path.to_string(),
});
return Ok(false);
}
} else {
result.errors.push(ValidationError {
code: "MISSING_TYPE".to_string(),
message: format!("Missing '{}' discriminator. Unable to resolve polymorphic boundaries", disc),
path: self.path.to_string(),
});
return Ok(false);
}
}
pub(crate) fn validate_type_inheritance(
&self,
result: &mut ValidationResult,
@ -323,17 +173,17 @@ impl<'a> ValidationContext<'a> {
let mut custom_types = Vec::new();
match &self.schema.type_ {
Some(crate::database::schema::SchemaTypeOrArray::Single(t)) => {
if !crate::database::schema::is_primitive_type(t) {
Some(crate::database::object::SchemaTypeOrArray::Single(t)) => {
if !crate::database::object::is_primitive_type(t) {
custom_types.push(t.clone());
}
}
Some(crate::database::schema::SchemaTypeOrArray::Multiple(arr)) => {
Some(crate::database::object::SchemaTypeOrArray::Multiple(arr)) => {
if arr.contains(&payload_primitive.to_string()) || (payload_primitive == "integer" && arr.contains(&"number".to_string())) {
// It natively matched a primitive in the array options, skip forcing custom proxy fallback
} else {
for t in arr {
if !crate::database::schema::is_primitive_type(t) {
if !crate::database::object::is_primitive_type(t) {
custom_types.push(t.clone());
}
}

View File

@ -1 +1 @@
1.0.109
1.0.110