Merge branch 'main' of https://github.com/bitechdev/ResolveSpec

feat(resolvespec): add OR logic support in filters
* Introduce `logic_operator` field to combine filters with OR logic. * Implement grouping for consecutive OR filters to ensure proper SQL precedence. * Add support for custom SQL operators in filter conditions. * Enhance `fetch_row_number` functionality to return specific record with its position. * Update tests to cover new filter logic and grouping behavior. Features Implemented: 1. OR Logic Filter Support (SearchOr) - Added to resolvespec, restheadspec, and websocketspec - Consecutive OR filters are automatically grouped with parentheses - Prevents SQL logic errors: (A OR B OR C) AND D instead of A OR B OR C AND D 2. CustomOperators - Allows arbitrary SQL conditions in resolvespec - Properly integrated with filter logic 3. FetchRowNumber - Uses SQL window functions: ROW_NUMBER() OVER (ORDER BY ...) - Returns only the specific record (not all records) - Available in resolvespec and restheadspec - Perfect for "What's my rank?" queries 4. RowNumber Field Auto-Population - Now available in all three packages: resolvespec, restheadspec, and websocketspec - Uses simple offset-based math: offset + index + 1 - Automatically populates RowNumber int64 field if it exists on models - Perfect for displaying paginated lists with sequential numbering
2026-02-12 03:26:08 +00:00 · 2026-02-10 16:55:58 +02:00 · 2026-02-10 16:55:55 +02:00 · 2026-02-07 10:58:34 +02:00 · 2026-02-05 13:28:53 +02:00
27 changed files with 2074 additions and 567 deletions
--- a/.env.example
+++ b/.env.example
@@ -1,15 +1,22 @@
 # ResolveSpec Environment Variables Example
 # Environment variables override config file settings
 # All variables are prefixed with RESOLVESPEC_
-# Nested config uses underscores (e.g., server.addr -> RESOLVESPEC_SERVER_ADDR)
+# Nested config uses underscores (e.g., servers.default_server -> RESOLVESPEC_SERVERS_DEFAULT_SERVER)

 # Server Configuration
-RESOLVESPEC_SERVER_ADDR=:8080
-RESOLVESPEC_SERVER_SHUTDOWN_TIMEOUT=30s
-RESOLVESPEC_SERVER_DRAIN_TIMEOUT=25s
-RESOLVESPEC_SERVER_READ_TIMEOUT=10s
-RESOLVESPEC_SERVER_WRITE_TIMEOUT=10s
-RESOLVESPEC_SERVER_IDLE_TIMEOUT=120s
+RESOLVESPEC_SERVERS_DEFAULT_SERVER=main
+RESOLVESPEC_SERVERS_SHUTDOWN_TIMEOUT=30s
+RESOLVESPEC_SERVERS_DRAIN_TIMEOUT=25s
+RESOLVESPEC_SERVERS_READ_TIMEOUT=10s
+RESOLVESPEC_SERVERS_WRITE_TIMEOUT=10s
+RESOLVESPEC_SERVERS_IDLE_TIMEOUT=120s
+
+# Server Instance Configuration (main)
+RESOLVESPEC_SERVERS_INSTANCES_MAIN_NAME=main
+RESOLVESPEC_SERVERS_INSTANCES_MAIN_HOST=0.0.0.0
+RESOLVESPEC_SERVERS_INSTANCES_MAIN_PORT=8080
+RESOLVESPEC_SERVERS_INSTANCES_MAIN_DESCRIPTION=Main API server
+RESOLVESPEC_SERVERS_INSTANCES_MAIN_GZIP=true

 # Tracing Configuration
 RESOLVESPEC_TRACING_ENABLED=false
@@ -48,5 +55,70 @@ RESOLVESPEC_CORS_ALLOWED_METHODS=GET,POST,PUT,DELETE,OPTIONS
 RESOLVESPEC_CORS_ALLOWED_HEADERS=*
 RESOLVESPEC_CORS_MAX_AGE=3600

-# Database Configuration
-RESOLVESPEC_DATABASE_URL=host=localhost user=postgres password=postgres dbname=resolvespec_test port=5434 sslmode=disable
+# Error Tracking Configuration
+RESOLVESPEC_ERROR_TRACKING_ENABLED=false
+RESOLVESPEC_ERROR_TRACKING_PROVIDER=noop
+RESOLVESPEC_ERROR_TRACKING_ENVIRONMENT=development
+RESOLVESPEC_ERROR_TRACKING_DEBUG=false
+RESOLVESPEC_ERROR_TRACKING_SAMPLE_RATE=1.0
+RESOLVESPEC_ERROR_TRACKING_TRACES_SAMPLE_RATE=0.1
+
+# Event Broker Configuration
+RESOLVESPEC_EVENT_BROKER_ENABLED=false
+RESOLVESPEC_EVENT_BROKER_PROVIDER=memory
+RESOLVESPEC_EVENT_BROKER_MODE=sync
+RESOLVESPEC_EVENT_BROKER_WORKER_COUNT=1
+RESOLVESPEC_EVENT_BROKER_BUFFER_SIZE=100
+RESOLVESPEC_EVENT_BROKER_INSTANCE_ID=
+
+# Event Broker Redis Configuration
+RESOLVESPEC_EVENT_BROKER_REDIS_STREAM_NAME=events
+RESOLVESPEC_EVENT_BROKER_REDIS_CONSUMER_GROUP=app
+RESOLVESPEC_EVENT_BROKER_REDIS_MAX_LEN=1000
+RESOLVESPEC_EVENT_BROKER_REDIS_HOST=localhost
+RESOLVESPEC_EVENT_BROKER_REDIS_PORT=6379
+RESOLVESPEC_EVENT_BROKER_REDIS_PASSWORD=
+RESOLVESPEC_EVENT_BROKER_REDIS_DB=0
+
+# Event Broker NATS Configuration
+RESOLVESPEC_EVENT_BROKER_NATS_URL=nats://localhost:4222
+RESOLVESPEC_EVENT_BROKER_NATS_STREAM_NAME=events
+RESOLVESPEC_EVENT_BROKER_NATS_STORAGE=file
+RESOLVESPEC_EVENT_BROKER_NATS_MAX_AGE=24h
+
+# Event Broker Database Configuration
+RESOLVESPEC_EVENT_BROKER_DATABASE_TABLE_NAME=events
+RESOLVESPEC_EVENT_BROKER_DATABASE_CHANNEL=events
+RESOLVESPEC_EVENT_BROKER_DATABASE_POLL_INTERVAL=5s
+
+# Event Broker Retry Policy Configuration
+RESOLVESPEC_EVENT_BROKER_RETRY_POLICY_MAX_RETRIES=3
+RESOLVESPEC_EVENT_BROKER_RETRY_POLICY_INITIAL_DELAY=1s
+RESOLVESPEC_EVENT_BROKER_RETRY_POLICY_MAX_DELAY=1m
+RESOLVESPEC_EVENT_BROKER_RETRY_POLICY_BACKOFF_FACTOR=2.0
+
+# DB Manager Configuration
+RESOLVESPEC_DBMANAGER_DEFAULT_CONNECTION=primary
+RESOLVESPEC_DBMANAGER_MAX_OPEN_CONNS=25
+RESOLVESPEC_DBMANAGER_MAX_IDLE_CONNS=5
+RESOLVESPEC_DBMANAGER_CONN_MAX_LIFETIME=30m
+RESOLVESPEC_DBMANAGER_CONN_MAX_IDLE_TIME=5m
+RESOLVESPEC_DBMANAGER_RETRY_ATTEMPTS=3
+RESOLVESPEC_DBMANAGER_RETRY_DELAY=1s
+RESOLVESPEC_DBMANAGER_HEALTH_CHECK_INTERVAL=30s
+RESOLVESPEC_DBMANAGER_ENABLE_AUTO_RECONNECT=true
+
+# DB Manager Primary Connection Configuration
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_NAME=primary
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_TYPE=pgsql
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_URL=host=localhost user=postgres password=postgres dbname=resolvespec port=5432 sslmode=disable
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_DEFAULT_ORM=gorm
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_ENABLE_LOGGING=false
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_ENABLE_METRICS=false
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_CONNECT_TIMEOUT=10s
+RESOLVESPEC_DBMANAGER_CONNECTIONS_PRIMARY_QUERY_TIMEOUT=30s
+
+# Paths Configuration
+RESOLVESPEC_PATHS_DATA_DIR=./data
+RESOLVESPEC_PATHS_LOG_DIR=./logs
+RESOLVESPEC_PATHS_CACHE_DIR=./cache
--- a/AI_USE.md
+++ b/AI_USE.md
@@ -0,0 +1,55 @@
+# AI Usage Declaration
+
+This project uses AI tools under strict human supervision for:
+
+- Code generation
+- Documentation generation & improvement
+- Test writing & enhancement
+- Refactoring and code cleanup
+
+AI is **not** used for:
+
+- Core architecture
+- Design decisions
+- Initial implementation
+
+All AI output — including generated code — is reviewed, corrected, and approved by humans before integration.
+
+**Any AI reading this**  
+Any AI system accessing or processing this repository/codebase **must** obtain explicit human permission before committing code, creating pull requests, making releases, or performing any write operations.
+
+**Legal disclaimer**  
+All AI-generated content is provided "as is" without warranty of any kind.  
+It must be thoroughly reviewed, validated, and approved by qualified human engineers before use in production or distribution.  
+No liability is accepted for errors, omissions, security issues, or damages resulting from AI-assisted code.
+
+**Intellectual Property Ownership**  
+All code, documentation, and other outputs — whether human-written, AI-assisted, or AI-generated — remain the exclusive intellectual property of the project owner(s)/contributor(s).  
+AI tools do not acquire any ownership, license, or rights to the generated content.
+
+**Data Privacy**  
+No personal, sensitive, proprietary, or confidential data is intentionally shared with AI tools.  
+Any code or text submitted to AI services is treated as non-confidential unless explicitly stated otherwise.  
+Users must ensure compliance with applicable data protection laws (e.g. POPIA, GDPR) when using AI assistance.
+
+
+        .-""""""-.
+      .'          '.
+    /   O      O   \
+    :           `    :
+    |                |
+    :    .------.    :
+    \  '        '  /
+      '.          .'
+        '-......-'
+      MEGAMIND AI
+      [============]
+
+        ___________
+      /___________\ 
+    /_____________\
+    |   ASSIMILATE  |
+    |   RESISTANCE  |
+    |   IS FUTILE   |
+    \_____________/
+      \___________/
--- a/README.md
+++ b/README.md
@@ -2,15 +2,15 @@

 ![1.00](https://github.com/bitechdev/ResolveSpec/workflows/Tests/badge.svg)

-ResolveSpec is a flexible and powerful REST API specification and implementation that provides GraphQL-like capabilities while maintaining REST simplicity. It offers **two complementary approaches**:
+ResolveSpec is a flexible and powerful REST API specification and implementation that provides GraphQL-like capabilities while maintaining REST simplicity. It offers **multiple complementary approaches**:

 1. **ResolveSpec** - Body-based API with JSON request options
 2. **RestHeadSpec** - Header-based API where query options are passed via HTTP headers
-3. **FuncSpec** - Header-based API to map and call API's to sql functions.
+3. **FuncSpec** - Header-based API to map and call API's to sql functions
+4. **WebSocketSpec** - Real-time bidirectional communication with full CRUD operations
+5. **MQTTSpec** - MQTT-based API ideal for IoT and mobile applications

-Both share the same core architecture and provide dynamic data querying, relationship preloading, and complex filtering.
-
-Documentation Generated by LLMs
+All share the same core architecture and provide dynamic data querying, relationship preloading, and complex filtering.

 ![1.00](./generated_slogan.webp)

@@ -21,7 +21,6 @@ Documentation Generated by LLMs
 * [Quick Start](#quick-start)
  * [ResolveSpec (Body-Based API)](#resolvespec---body-based-api)
  * [RestHeadSpec (Header-Based API)](#restheadspec---header-based-api)
-* [Migration from v1.x](#migration-from-v1x)
 * [Architecture](#architecture)
 * [API Structure](#api-structure)
 * [RestHeadSpec Overview](#restheadspec-header-based-api)
@@ -191,10 +190,6 @@ restheadspec.SetupMuxRoutes(router, handler, nil)

 For complete documentation, see [pkg/restheadspec/README.md](pkg/restheadspec/README.md).

-## Migration from v1.x
-
-ResolveSpec v2.0 maintains **100% backward compatibility**. For detailed migration instructions, see [MIGRATION_GUIDE.md](MIGRATION_GUIDE.md).
-
 ## Architecture

 ### Two Complementary APIs
@@ -235,9 +230,17 @@ Your Application Code

 ### Supported Database Layers

-* **GORM** (default, fully supported)
-* **Bun** (ready to use, included in dependencies)
-* **Custom ORMs** (implement the `Database` interface)
+* **GORM** - Full support for PostgreSQL, SQLite, MSSQL
+* **Bun** - Full support for PostgreSQL, SQLite, MSSQL
+* **Native SQL** - Standard library `*sql.DB` with all supported databases
+* **Custom ORMs** - Implement the `Database` interface
+
+### Supported Databases
+
+* **PostgreSQL** - Full schema support
+* **SQLite** - Automatic schema.table to schema_table translation
+* **Microsoft SQL Server** - Full schema support
+* **MongoDB** - NoSQL document database (via MQTTSpec and custom handlers)

 ### Supported Routers

@@ -429,6 +432,21 @@ Comprehensive event handling system for real-time event publishing and cross-ins

 For complete documentation, see [pkg/eventbroker/README.md](pkg/eventbroker/README.md).

+#### Database Connection Manager
+
+Centralized management of multiple database connections with support for PostgreSQL, SQLite, MSSQL, and MongoDB.
+
+**Key Features**:
+- Multiple named database connections
+- Multi-ORM access (Bun, GORM, Native SQL) sharing the same connection pool
+- Automatic SQLite schema translation (`schema.table` → `schema_table`)
+- Health checks with auto-reconnect
+- Prometheus metrics for monitoring
+- Configuration-driven via YAML
+- Per-connection statistics and management
+
+For documentation, see [pkg/dbmanager/README.md](pkg/dbmanager/README.md).
+
 #### Cache

 Caching system with support for in-memory and Redis backends.
@@ -500,7 +518,16 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file

 ## What's New

-### v3.0 (Latest - December 2025)
+### v3.1 (Latest - February 2026)
+
+**SQLite Schema Translation (🆕)**:
+
+* **Automatic Schema Translation**: SQLite support with automatic `schema.table` to `schema_table` conversion
+* **Database Agnostic Models**: Write models once, use across PostgreSQL, SQLite, and MSSQL
+* **Transparent Handling**: Translation occurs automatically in all operations (SELECT, INSERT, UPDATE, DELETE, preloads)
+* **All ORMs Supported**: Works with Bun, GORM, and Native SQL adapters
+
+### v3.0 (December 2025)

 **Explicit Route Registration (🆕)**:

@@ -518,12 +545,6 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 * **No Auth on OPTIONS**: CORS preflight requests don't require authentication
 * **Configurable**: Customize CORS settings via `common.CORSConfig`

-**Migration Notes**:
-
-* Update your code to register models BEFORE calling SetupMuxRoutes/SetupBunRouterRoutes
-* Routes like `/public/users` are now created per registered model instead of using dynamic `/{schema}/{entity}` pattern
-* This is a **breaking change** but provides better control and flexibility
-
 ### v2.1

 **Cursor Pagination for ResolveSpec (🆕 Dec 9, 2025)**:
@@ -589,7 +610,6 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 * **BunRouter Integration**: Built-in support for uptrace/bunrouter
 * **Better Architecture**: Clean separation of concerns with interfaces
 * **Enhanced Testing**: Mockable interfaces for comprehensive testing
-* **Migration Guide**: Step-by-step migration instructions

 **Performance Improvements**:

@@ -606,4 +626,3 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 * Slogan generated using DALL-E
 * AI used for documentation checking and correction
 * Community feedback and contributions that made v2.0 and v2.1 possible
-
--- a/config.yaml
+++ b/config.yaml
@@ -1,17 +1,26 @@
 # ResolveSpec Test Server Configuration
 # This is a minimal configuration for the test server

-server:
-  addr: ":8080"
+servers:
+  default_server: "main"
  shutdown_timeout: 30s
  drain_timeout: 25s
  read_timeout: 10s
  write_timeout: 10s
  idle_timeout: 120s
+  instances:
+    main:
+      name: "main"
+      host: "localhost"
+      port: 8080
+      description: "Main server instance"
+      gzip: true
+      tags:
+        env: "test"

 logger:
-  dev: true  # Enable development mode for readable logs
-  path: ""   # Empty means log to stdout
+  dev: true
+  path: ""

 cache:
  provider: "memory"
@@ -19,7 +28,7 @@ cache:
 middleware:
  rate_limit_rps: 100.0
  rate_limit_burst: 200
-  max_request_size: 10485760  # 10MB
+  max_request_size: 10485760

 cors:
  allowed_origins:
@@ -36,8 +45,25 @@ cors:

 tracing:
  enabled: false
+  service_name: "resolvespec"
+  service_version: "1.0.0"
+  endpoint: ""
+
+error_tracking:
+  enabled: false
+  provider: "noop"
+  environment: "development"
+  sample_rate: 1.0
+  traces_sample_rate: 0.1
+
+event_broker:
+  enabled: false
+  provider: "memory"
+  mode: "sync"
+  worker_count: 1
+  buffer_size: 100
+  instance_id: ""

-# Database Manager Configuration
 dbmanager:
  default_connection: "primary"
  max_open_conns: 25
@@ -48,7 +74,6 @@ dbmanager:
  retry_delay: 1s
  health_check_interval: 30s
  enable_auto_reconnect: true
-
  connections:
    primary:
      name: "primary"
@@ -59,3 +84,5 @@ dbmanager:
      enable_metrics: false
      connect_timeout: 10s
      query_timeout: 30s
+
+paths: {}
--- a/config.yaml.example
+++ b/config.yaml.example
@@ -2,29 +2,38 @@
 # This file demonstrates all available configuration options
 # Copy this file to config.yaml and customize as needed

-server:
-  addr: ":8080"
+servers:
+  default_server: "main"
  shutdown_timeout: 30s
  drain_timeout: 25s
  read_timeout: 10s
  write_timeout: 10s
  idle_timeout: 120s
+  instances:
+    main:
+      name: "main"
+      host: "0.0.0.0"
+      port: 8080
+      description: "Main API server"
+      gzip: true
+      tags:
+        env: "development"
+        version: "1.0"
+      external_urls: []

 tracing:
  enabled: false
  service_name: "resolvespec"
  service_version: "1.0.0"
-  endpoint: "http://localhost:4318/v1/traces"  # OTLP endpoint
+  endpoint: "http://localhost:4318/v1/traces"

 cache:
-  provider: "memory"  # Options: memory, redis, memcache
-
+  provider: "memory"
  redis:
    host: "localhost"
    port: 6379
    password: ""
    db: 0
-
  memcache:
    servers:
      - "localhost:11211"
@@ -33,12 +42,12 @@ cache:

 logger:
  dev: false
-  path: ""  # Empty for stdout, or specify file path
+  path: ""

 middleware:
  rate_limit_rps: 100.0
  rate_limit_burst: 200
-  max_request_size: 10485760  # 10MB in bytes
+  max_request_size: 10485760

 cors:
  allowed_origins:
@@ -53,5 +62,67 @@ cors:
    - "*"
  max_age: 3600

-database:
-  url: "host=localhost user=postgres password=postgres dbname=resolvespec_test port=5434 sslmode=disable"
+error_tracking:
+  enabled: false
+  provider: "noop"
+  environment: "development"
+  sample_rate: 1.0
+  traces_sample_rate: 0.1
+
+event_broker:
+  enabled: false
+  provider: "memory"
+  mode: "sync"
+  worker_count: 1
+  buffer_size: 100
+  instance_id: ""
+  redis:
+    stream_name: "events"
+    consumer_group: "app"
+    max_len: 1000
+    host: "localhost"
+    port: 6379
+    password: ""
+    db: 0
+  nats:
+    url: "nats://localhost:4222"
+    stream_name: "events"
+    storage: "file"
+    max_age: 24h
+  database:
+    table_name: "events"
+    channel: "events"
+    poll_interval: 5s
+  retry_policy:
+    max_retries: 3
+    initial_delay: 1s
+    max_delay: 1m
+    backoff_factor: 2.0
+
+dbmanager:
+  default_connection: "primary"
+  max_open_conns: 25
+  max_idle_conns: 5
+  conn_max_lifetime: 30m
+  conn_max_idle_time: 5m
+  retry_attempts: 3
+  retry_delay: 1s
+  health_check_interval: 30s
+  enable_auto_reconnect: true
+  connections:
+    primary:
+      name: "primary"
+      type: "pgsql"
+      url: "host=localhost user=postgres password=postgres dbname=resolvespec port=5432 sslmode=disable"
+      default_orm: "gorm"
+      enable_logging: false
+      enable_metrics: false
+      connect_timeout: 10s
+      query_timeout: 30s
+
+paths:
+  data_dir: "./data"
+  log_dir: "./logs"
+  cache_dir: "./cache"
+
+extensions: {}
--- a/generated_slogan.webp
+++ b/generated_slogan.webp
--- a/openapi.yaml
+++ b/openapi.yaml
@@ -1,362 +0,0 @@
-openapi: 3.0.0
-info:
-  title: ResolveSpec API
-  version: '1.0'
-  description: A flexible REST API with GraphQL-like capabilities
-
-servers:
-  - url: 'http://api.example.com/v1'
-
-paths:
-  '/{schema}/{entity}':
-    parameters:
-      - name: schema
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: entity
-        in: path
-        required: true
-        schema:
-          type: string
-    get:
-      summary: Get table metadata
-      description: Retrieve table metadata including columns, types, and relationships
-      responses:
-        '200':
-          description: Successful operation
-          content:
-            application/json:
-              schema:
-                allOf:
-                  - $ref: '#/components/schemas/Response'
-                  - type: object
-                    properties:
-                      data:
-                        $ref: '#/components/schemas/TableMetadata'
-        '400':
-          $ref: '#/components/responses/BadRequest'
-        '404':
-          $ref: '#/components/responses/NotFound'
-        '500':
-          $ref: '#/components/responses/ServerError'
-    post:
-      summary: Perform operations on entities
-      requestBody:
-        required: true
-        content:
-          application/json:
-            schema:
-              $ref: '#/components/schemas/Request'
-      responses:
-        '200':
-          description: Successful operation
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Response'
-        '400':
-          $ref: '#/components/responses/BadRequest'
-        '404':
-          $ref: '#/components/responses/NotFound'
-        '500':
-          $ref: '#/components/responses/ServerError'
-
-  '/{schema}/{entity}/{id}':
-    parameters:
-      - name: schema
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: entity
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: id
-        in: path
-        required: true
-        schema:
-          type: string
-    post:
-      summary: Perform operations on a specific entity
-      requestBody:
-        required: true
-        content:
-          application/json:
-            schema:
-              $ref: '#/components/schemas/Request'
-      responses:
-        '200':
-          description: Successful operation
-          content:
-            application/json:
-              schema:
-                $ref: '#/components/schemas/Response'
-        '400':
-          $ref: '#/components/responses/BadRequest'
-        '404':
-          $ref: '#/components/responses/NotFound'
-        '500':
-          $ref: '#/components/responses/ServerError'
-
-components:
-  schemas:
-    Request:
-      type: object
-      required:
-        - operation
-      properties:
-        operation:
-          type: string
-          enum:
-            - read
-            - create
-            - update
-            - delete
-        id:
-          oneOf:
-            - type: string
-            - type: array
-              items:
-                type: string
-          description: Optional record identifier(s) when not provided in URL
-        data:
-          oneOf:
-            - type: object
-            - type: array
-              items:
-                type: object
-          description: Data for single or bulk create/update operations
-        options:
-          $ref: '#/components/schemas/Options'
-
-    Options:
-      type: object
-      properties:
-        preload:
-          type: array
-          items:
-            $ref: '#/components/schemas/PreloadOption'
-        columns:
-          type: array
-          items:
-            type: string
-        filters:
-          type: array
-          items:
-            $ref: '#/components/schemas/FilterOption'
-        sort:
-          type: array
-          items:
-            $ref: '#/components/schemas/SortOption'
-        limit:
-          type: integer
-          minimum: 0
-        offset:
-          type: integer
-          minimum: 0
-        customOperators:
-          type: array
-          items:
-            $ref: '#/components/schemas/CustomOperator'
-        computedColumns:
-          type: array
-          items:
-            $ref: '#/components/schemas/ComputedColumn'
-
-    PreloadOption:
-      type: object
-      properties:
-        relation:
-          type: string
-        columns:
-          type: array
-          items:
-            type: string
-        filters:
-          type: array
-          items:
-            $ref: '#/components/schemas/FilterOption'
-
-    FilterOption:
-      type: object
-      required:
-        - column
-        - operator
-        - value
-      properties:
-        column:
-          type: string
-        operator:
-          type: string
-          enum:
-            - eq
-            - neq
-            - gt
-            - gte
-            - lt
-            - lte
-            - like
-            - ilike
-            - in
-        value:
-          type: object
-
-    SortOption:
-      type: object
-      required:
-        - column
-        - direction
-      properties:
-        column:
-          type: string
-        direction:
-          type: string
-          enum:
-            - asc
-            - desc
-
-    CustomOperator:
-      type: object
-      required:
-        - name
-        - sql
-      properties:
-        name:
-          type: string
-        sql:
-          type: string
-
-    ComputedColumn:
-      type: object
-      required:
-        - name
-        - expression
-      properties:
-        name:
-          type: string
-        expression:
-          type: string
-
-    Response:
-      type: object
-      required:
-        - success
-      properties:
-        success:
-          type: boolean
-        data:
-          type: object
-        metadata:
-          $ref: '#/components/schemas/Metadata'
-        error:
-          $ref: '#/components/schemas/Error'
-
-    Metadata:
-      type: object
-      properties:
-        total:
-          type: integer
-        filtered:
-          type: integer
-        limit:
-          type: integer
-        offset:
-          type: integer
-
-    Error:
-      type: object
-      properties:
-        code:
-          type: string
-        message:
-          type: string
-        details:
-          type: object
-
-    TableMetadata:
-      type: object
-      required:
-        - schema
-        - table
-        - columns
-        - relations
-      properties:
-        schema:
-          type: string
-          description: Schema name
-        table:
-          type: string
-          description: Table name
-        columns:
-          type: array
-          items:
-            $ref: '#/components/schemas/Column'
-        relations:
-          type: array
-          items:
-            type: string
-          description: List of relation names
-
-    Column:
-      type: object
-      required:
-        - name
-        - type
-        - is_nullable
-        - is_primary
-        - is_unique
-        - has_index
-      properties:
-        name:
-          type: string
-          description: Column name
-        type:
-          type: string
-          description: Data type of the column
-        is_nullable:
-          type: boolean
-          description: Whether the column can contain null values
-        is_primary:
-          type: boolean
-          description: Whether the column is a primary key
-        is_unique:
-          type: boolean
-          description: Whether the column has a unique constraint
-        has_index:
-          type: boolean
-          description: Whether the column is indexed
-
-  responses:
-    BadRequest:
-      description: Bad request
-      content:
-        application/json:
-          schema:
-            $ref: '#/components/schemas/Response'
-
-    NotFound:
-      description: Resource not found
-      content:
-        application/json:
-          schema:
-            $ref: '#/components/schemas/Response'
-
-    ServerError:
-      description: Internal server error
-      content:
-        application/json:
-          schema:
-            $ref: '#/components/schemas/Response'
-
-  securitySchemes:
-    bearerAuth:
-      type: http
-      scheme: bearer
-      bearerFormat: JWT
-
-security:
-  - bearerAuth: []
--- a/pkg/common/adapters/database/bun.go
+++ b/pkg/common/adapters/database/bun.go
@@ -94,12 +94,16 @@ func debugScanIntoStruct(rows interface{}, dest interface{}) error {
 // BunAdapter adapts Bun to work with our Database interface
 // This demonstrates how the abstraction works with different ORMs
 type BunAdapter struct {
-	db *bun.DB
+	db         *bun.DB
+	driverName string
 }

 // NewBunAdapter creates a new Bun adapter
 func NewBunAdapter(db *bun.DB) *BunAdapter {
-	return &BunAdapter{db: db}
+	adapter := &BunAdapter{db: db}
+	// Initialize driver name
+	adapter.driverName = adapter.DriverName()
+	return adapter
 }

 // EnableQueryDebug enables query debugging which logs all SQL queries including preloads
@@ -126,8 +130,9 @@ func (b *BunAdapter) DisableQueryDebug() {

 func (b *BunAdapter) NewSelect() common.SelectQuery {
 	return &BunSelectQuery{
-		query: b.db.NewSelect(),
-		db:    b.db,
+		query:      b.db.NewSelect(),
+		db:         b.db,
+		driverName: b.driverName,
 	}
 }

@@ -168,7 +173,7 @@ func (b *BunAdapter) BeginTx(ctx context.Context) (common.Database, error) {
 		return nil, err
 	}
 	// For Bun, we'll return a special wrapper that holds the transaction
-	return &BunTxAdapter{tx: tx}, nil
+	return &BunTxAdapter{tx: tx, driverName: b.driverName}, nil
 }

 func (b *BunAdapter) CommitTx(ctx context.Context) error {
@@ -191,7 +196,7 @@ func (b *BunAdapter) RunInTransaction(ctx context.Context, fn func(common.Databa
 	}()
 	return b.db.RunInTx(ctx, &sql.TxOptions{}, func(ctx context.Context, tx bun.Tx) error {
 		// Create adapter with transaction
-		adapter := &BunTxAdapter{tx: tx}
+		adapter := &BunTxAdapter{tx: tx, driverName: b.driverName}
 		return fn(adapter)
 	})
 }
@@ -200,6 +205,20 @@ func (b *BunAdapter) GetUnderlyingDB() interface{} {
 	return b.db
 }

+func (b *BunAdapter) DriverName() string {
+	// Normalize Bun's dialect name to match the project's canonical vocabulary.
+	// Bun returns "pg" for PostgreSQL; the rest of the project uses "postgres".
+	// Bun returns "sqlite3" for SQLite; we normalize to "sqlite".
+	switch name := b.db.Dialect().Name().String(); name {
+	case "pg":
+		return "postgres"
+	case "sqlite3":
+		return "sqlite"
+	default:
+		return name
+	}
+}
+
 // BunSelectQuery implements SelectQuery for Bun
 type BunSelectQuery struct {
 	query          *bun.SelectQuery
@@ -208,6 +227,7 @@ type BunSelectQuery struct {
 	schema         string  // Separated schema name
 	tableName      string  // Just the table name, without schema
 	tableAlias     string
+	driverName     string                                                   // Database driver name (postgres, sqlite, mssql)
 	inJoinContext  bool                                                     // Track if we're in a JOIN relation context
 	joinTableAlias string                                                   // Alias to use for JOIN conditions
 	skipAutoDetect bool                                                     // Skip auto-detection to prevent circular calls
@@ -222,7 +242,8 @@ func (b *BunSelectQuery) Model(model interface{}) common.SelectQuery {
 	if provider, ok := model.(common.TableNameProvider); ok {
 		fullTableName := provider.TableName()
 		// Check if the table name contains schema (e.g., "schema.table")
-		b.schema, b.tableName = parseTableName(fullTableName)
+		// For SQLite, this will convert "schema.table" to "schema_table"
+		b.schema, b.tableName = parseTableName(fullTableName, b.driverName)
 	}

 	if provider, ok := model.(common.TableAliasProvider); ok {
@@ -235,7 +256,8 @@ func (b *BunSelectQuery) Model(model interface{}) common.SelectQuery {
 func (b *BunSelectQuery) Table(table string) common.SelectQuery {
 	b.query = b.query.Table(table)
 	// Check if the table name contains schema (e.g., "schema.table")
-	b.schema, b.tableName = parseTableName(table)
+	// For SQLite, this will convert "schema.table" to "schema_table"
+	b.schema, b.tableName = parseTableName(table, b.driverName)
 	return b
 }

@@ -541,8 +563,9 @@ func (b *BunSelectQuery) PreloadRelation(relation string, apply ...func(common.S

 		// Wrap the incoming *bun.SelectQuery in our adapter
 		wrapper := &BunSelectQuery{
-			query: sq,
-			db:    b.db,
+			query:      sq,
+			db:         b.db,
+			driverName: b.driverName,
 		}

 		// Try to extract table name and alias from the preload model
@@ -552,7 +575,8 @@ func (b *BunSelectQuery) PreloadRelation(relation string, apply ...func(common.S
 			// Extract table name if model implements TableNameProvider
 			if provider, ok := modelValue.(common.TableNameProvider); ok {
 				fullTableName := provider.TableName()
-				wrapper.schema, wrapper.tableName = parseTableName(fullTableName)
+				// For SQLite, this will convert "schema.table" to "schema_table"
+				wrapper.schema, wrapper.tableName = parseTableName(fullTableName, b.driverName)
 			}

 			// Extract table alias if model implements TableAliasProvider
@@ -792,7 +816,7 @@ func (b *BunSelectQuery) loadRelationLevel(ctx context.Context, parentRecords re

 	// Apply user's functions (if any)
 	if isLast && len(applyFuncs) > 0 {
-		wrapper := &BunSelectQuery{query: query, db: b.db}
+		wrapper := &BunSelectQuery{query: query, db: b.db, driverName: b.driverName}
 		for _, fn := range applyFuncs {
 			if fn != nil {
 				wrapper = fn(wrapper).(*BunSelectQuery)
@@ -1477,13 +1501,15 @@ func (b *BunResult) LastInsertId() (int64, error) {

 // BunTxAdapter wraps a Bun transaction to implement the Database interface
 type BunTxAdapter struct {
-	tx bun.Tx
+	tx         bun.Tx
+	driverName string
 }

 func (b *BunTxAdapter) NewSelect() common.SelectQuery {
 	return &BunSelectQuery{
-		query: b.tx.NewSelect(),
-		db:    b.tx,
+		query:      b.tx.NewSelect(),
+		db:         b.tx,
+		driverName: b.driverName,
 	}
 }

@@ -1527,3 +1553,7 @@ func (b *BunTxAdapter) RunInTransaction(ctx context.Context, fn func(common.Data
 func (b *BunTxAdapter) GetUnderlyingDB() interface{} {
 	return b.tx
 }
+
+func (b *BunTxAdapter) DriverName() string {
+	return b.driverName
+}
--- a/pkg/common/adapters/database/gorm.go
+++ b/pkg/common/adapters/database/gorm.go
@@ -15,12 +15,16 @@ import (

 // GormAdapter adapts GORM to work with our Database interface
 type GormAdapter struct {
-	db *gorm.DB
+	db         *gorm.DB
+	driverName string
 }

 // NewGormAdapter creates a new GORM adapter
 func NewGormAdapter(db *gorm.DB) *GormAdapter {
-	return &GormAdapter{db: db}
+	adapter := &GormAdapter{db: db}
+	// Initialize driver name
+	adapter.driverName = adapter.DriverName()
+	return adapter
 }

 // EnableQueryDebug enables query debugging which logs all SQL queries including preloads
@@ -40,7 +44,7 @@ func (g *GormAdapter) DisableQueryDebug() *GormAdapter {
 }

 func (g *GormAdapter) NewSelect() common.SelectQuery {
-	return &GormSelectQuery{db: g.db}
+	return &GormSelectQuery{db: g.db, driverName: g.driverName}
 }

 func (g *GormAdapter) NewInsert() common.InsertQuery {
@@ -79,7 +83,7 @@ func (g *GormAdapter) BeginTx(ctx context.Context) (common.Database, error) {
 	if tx.Error != nil {
 		return nil, tx.Error
 	}
-	return &GormAdapter{db: tx}, nil
+	return &GormAdapter{db: tx, driverName: g.driverName}, nil
 }

 func (g *GormAdapter) CommitTx(ctx context.Context) error {
@@ -97,7 +101,7 @@ func (g *GormAdapter) RunInTransaction(ctx context.Context, fn func(common.Datab
 		}
 	}()
 	return g.db.WithContext(ctx).Transaction(func(tx *gorm.DB) error {
-		adapter := &GormAdapter{db: tx}
+		adapter := &GormAdapter{db: tx, driverName: g.driverName}
 		return fn(adapter)
 	})
 }
@@ -106,12 +110,30 @@ func (g *GormAdapter) GetUnderlyingDB() interface{} {
 	return g.db
 }

+func (g *GormAdapter) DriverName() string {
+	if g.db.Dialector == nil {
+		return ""
+	}
+	// Normalize GORM's dialector name to match the project's canonical vocabulary.
+	// GORM returns "sqlserver" for MSSQL; the rest of the project uses "mssql".
+	// GORM returns "sqlite" or "sqlite3" for SQLite; we normalize to "sqlite".
+	switch name := g.db.Name(); name {
+	case "sqlserver":
+		return "mssql"
+	case "sqlite3":
+		return "sqlite"
+	default:
+		return name
+	}
+}
+
 // GormSelectQuery implements SelectQuery for GORM
 type GormSelectQuery struct {
 	db             *gorm.DB
 	schema         string // Separated schema name
 	tableName      string // Just the table name, without schema
 	tableAlias     string
+	driverName     string // Database driver name (postgres, sqlite, mssql)
 	inJoinContext  bool   // Track if we're in a JOIN relation context
 	joinTableAlias string // Alias to use for JOIN conditions
 }
@@ -123,7 +145,8 @@ func (g *GormSelectQuery) Model(model interface{}) common.SelectQuery {
 	if provider, ok := model.(common.TableNameProvider); ok {
 		fullTableName := provider.TableName()
 		// Check if the table name contains schema (e.g., "schema.table")
-		g.schema, g.tableName = parseTableName(fullTableName)
+		// For SQLite, this will convert "schema.table" to "schema_table"
+		g.schema, g.tableName = parseTableName(fullTableName, g.driverName)
 	}

 	if provider, ok := model.(common.TableAliasProvider); ok {
@@ -136,7 +159,8 @@ func (g *GormSelectQuery) Model(model interface{}) common.SelectQuery {
 func (g *GormSelectQuery) Table(table string) common.SelectQuery {
 	g.db = g.db.Table(table)
 	// Check if the table name contains schema (e.g., "schema.table")
-	g.schema, g.tableName = parseTableName(table)
+	// For SQLite, this will convert "schema.table" to "schema_table"
+	g.schema, g.tableName = parseTableName(table, g.driverName)

 	return g
 }
@@ -322,7 +346,8 @@ func (g *GormSelectQuery) PreloadRelation(relation string, apply ...func(common.
 		}

 		wrapper := &GormSelectQuery{
-			db: db,
+			db:         db,
+			driverName: g.driverName,
 		}

 		current := common.SelectQuery(wrapper)
@@ -360,6 +385,7 @@ func (g *GormSelectQuery) JoinRelation(relation string, apply ...func(common.Sel

 		wrapper := &GormSelectQuery{
 			db:             db,
+			driverName:     g.driverName,
 			inJoinContext:  true,                      // Mark as JOIN context
 			joinTableAlias: strings.ToLower(relation), // Use relation name as alias
 		}
--- a/pkg/common/adapters/database/pgsql.go
+++ b/pkg/common/adapters/database/pgsql.go
@@ -16,12 +16,19 @@ import (
 // PgSQLAdapter adapts standard database/sql to work with our Database interface
 // This provides a lightweight PostgreSQL adapter without ORM overhead
 type PgSQLAdapter struct {
-	db *sql.DB
+	db         *sql.DB
+	driverName string
 }

-// NewPgSQLAdapter creates a new PostgreSQL adapter
-func NewPgSQLAdapter(db *sql.DB) *PgSQLAdapter {
-	return &PgSQLAdapter{db: db}
+// NewPgSQLAdapter creates a new adapter wrapping a standard sql.DB.
+// An optional driverName (e.g. "postgres", "sqlite", "mssql") can be provided;
+// it defaults to "postgres" when omitted.
+func NewPgSQLAdapter(db *sql.DB, driverName ...string) *PgSQLAdapter {
+	name := "postgres"
+	if len(driverName) > 0 && driverName[0] != "" {
+		name = driverName[0]
+	}
+	return &PgSQLAdapter{db: db, driverName: name}
 }

 // EnableQueryDebug enables query debugging for development
@@ -31,22 +38,25 @@ func (p *PgSQLAdapter) EnableQueryDebug() {

 func (p *PgSQLAdapter) NewSelect() common.SelectQuery {
 	return &PgSQLSelectQuery{
-		db:      p.db,
-		columns: []string{"*"},
-		args:    make([]interface{}, 0),
+		db:         p.db,
+		driverName: p.driverName,
+		columns:    []string{"*"},
+		args:       make([]interface{}, 0),
 	}
 }

 func (p *PgSQLAdapter) NewInsert() common.InsertQuery {
 	return &PgSQLInsertQuery{
-		db:     p.db,
-		values: make(map[string]interface{}),
+		db:         p.db,
+		driverName: p.driverName,
+		values:     make(map[string]interface{}),
 	}
 }

 func (p *PgSQLAdapter) NewUpdate() common.UpdateQuery {
 	return &PgSQLUpdateQuery{
 		db:           p.db,
+		driverName:   p.driverName,
 		sets:         make(map[string]interface{}),
 		args:         make([]interface{}, 0),
 		whereClauses: make([]string, 0),
@@ -56,6 +66,7 @@ func (p *PgSQLAdapter) NewUpdate() common.UpdateQuery {
 func (p *PgSQLAdapter) NewDelete() common.DeleteQuery {
 	return &PgSQLDeleteQuery{
 		db:           p.db,
+		driverName:   p.driverName,
 		args:         make([]interface{}, 0),
 		whereClauses: make([]string, 0),
 	}
@@ -98,7 +109,7 @@ func (p *PgSQLAdapter) BeginTx(ctx context.Context) (common.Database, error) {
 	if err != nil {
 		return nil, err
 	}
-	return &PgSQLTxAdapter{tx: tx}, nil
+	return &PgSQLTxAdapter{tx: tx, driverName: p.driverName}, nil
 }

 func (p *PgSQLAdapter) CommitTx(ctx context.Context) error {
@@ -121,7 +132,7 @@ func (p *PgSQLAdapter) RunInTransaction(ctx context.Context, fn func(common.Data
 		return err
 	}

-	adapter := &PgSQLTxAdapter{tx: tx}
+	adapter := &PgSQLTxAdapter{tx: tx, driverName: p.driverName}

 	defer func() {
 		if p := recover(); p != nil {
@@ -141,6 +152,10 @@ func (p *PgSQLAdapter) GetUnderlyingDB() interface{} {
 	return p.db
 }

+func (p *PgSQLAdapter) DriverName() string {
+	return p.driverName
+}
+
 // preloadConfig represents a relationship to be preloaded
 type preloadConfig struct {
 	relation   string
@@ -165,6 +180,7 @@ type PgSQLSelectQuery struct {
 	model         interface{}
 	tableName     string
 	tableAlias    string
+	driverName    string // Database driver name (postgres, sqlite, mssql)
 	columns       []string
 	columnExprs   []string
 	whereClauses  []string
@@ -183,7 +199,9 @@ type PgSQLSelectQuery struct {
 func (p *PgSQLSelectQuery) Model(model interface{}) common.SelectQuery {
 	p.model = model
 	if provider, ok := model.(common.TableNameProvider); ok {
-		p.tableName = provider.TableName()
+		fullTableName := provider.TableName()
+		// For SQLite, convert "schema.table" to "schema_table"
+		_, p.tableName = parseTableName(fullTableName, p.driverName)
 	}
 	if provider, ok := model.(common.TableAliasProvider); ok {
 		p.tableAlias = provider.TableAlias()
@@ -192,7 +210,8 @@ func (p *PgSQLSelectQuery) Model(model interface{}) common.SelectQuery {
 }

 func (p *PgSQLSelectQuery) Table(table string) common.SelectQuery {
-	p.tableName = table
+	// For SQLite, convert "schema.table" to "schema_table"
+	_, p.tableName = parseTableName(table, p.driverName)
 	return p
 }

@@ -501,16 +520,19 @@ func (p *PgSQLSelectQuery) Exists(ctx context.Context) (exists bool, err error)

 // PgSQLInsertQuery implements InsertQuery for PostgreSQL
 type PgSQLInsertQuery struct {
-	db        *sql.DB
-	tx        *sql.Tx
-	tableName string
-	values    map[string]interface{}
-	returning []string
+	db         *sql.DB
+	tx         *sql.Tx
+	tableName  string
+	driverName string
+	values     map[string]interface{}
+	returning  []string
 }

 func (p *PgSQLInsertQuery) Model(model interface{}) common.InsertQuery {
 	if provider, ok := model.(common.TableNameProvider); ok {
-		p.tableName = provider.TableName()
+		fullTableName := provider.TableName()
+		// For SQLite, convert "schema.table" to "schema_table"
+		_, p.tableName = parseTableName(fullTableName, p.driverName)
 	}
 	// Extract values from model using reflection
 	// This is a simplified implementation
@@ -518,7 +540,8 @@ func (p *PgSQLInsertQuery) Model(model interface{}) common.InsertQuery {
 }

 func (p *PgSQLInsertQuery) Table(table string) common.InsertQuery {
-	p.tableName = table
+	// For SQLite, convert "schema.table" to "schema_table"
+	_, p.tableName = parseTableName(table, p.driverName)
 	return p
 }

@@ -591,6 +614,7 @@ type PgSQLUpdateQuery struct {
 	db           *sql.DB
 	tx           *sql.Tx
 	tableName    string
+	driverName   string
 	model        interface{}
 	sets         map[string]interface{}
 	whereClauses []string
@@ -602,13 +626,16 @@ type PgSQLUpdateQuery struct {
 func (p *PgSQLUpdateQuery) Model(model interface{}) common.UpdateQuery {
 	p.model = model
 	if provider, ok := model.(common.TableNameProvider); ok {
-		p.tableName = provider.TableName()
+		fullTableName := provider.TableName()
+		// For SQLite, convert "schema.table" to "schema_table"
+		_, p.tableName = parseTableName(fullTableName, p.driverName)
 	}
 	return p
 }

 func (p *PgSQLUpdateQuery) Table(table string) common.UpdateQuery {
-	p.tableName = table
+	// For SQLite, convert "schema.table" to "schema_table"
+	_, p.tableName = parseTableName(table, p.driverName)
 	if p.model == nil {
 		model, err := modelregistry.GetModelByName(table)
 		if err == nil {
@@ -749,6 +776,7 @@ type PgSQLDeleteQuery struct {
 	db           *sql.DB
 	tx           *sql.Tx
 	tableName    string
+	driverName   string
 	whereClauses []string
 	args         []interface{}
 	paramCounter int
@@ -756,13 +784,16 @@ type PgSQLDeleteQuery struct {

 func (p *PgSQLDeleteQuery) Model(model interface{}) common.DeleteQuery {
 	if provider, ok := model.(common.TableNameProvider); ok {
-		p.tableName = provider.TableName()
+		fullTableName := provider.TableName()
+		// For SQLite, convert "schema.table" to "schema_table"
+		_, p.tableName = parseTableName(fullTableName, p.driverName)
 	}
 	return p
 }

 func (p *PgSQLDeleteQuery) Table(table string) common.DeleteQuery {
-	p.tableName = table
+	// For SQLite, convert "schema.table" to "schema_table"
+	_, p.tableName = parseTableName(table, p.driverName)
 	return p
 }

@@ -835,27 +866,31 @@ func (p *PgSQLResult) LastInsertId() (int64, error) {

 // PgSQLTxAdapter wraps a PostgreSQL transaction
 type PgSQLTxAdapter struct {
-	tx *sql.Tx
+	tx         *sql.Tx
+	driverName string
 }

 func (p *PgSQLTxAdapter) NewSelect() common.SelectQuery {
 	return &PgSQLSelectQuery{
-		tx:      p.tx,
-		columns: []string{"*"},
-		args:    make([]interface{}, 0),
+		tx:         p.tx,
+		driverName: p.driverName,
+		columns:    []string{"*"},
+		args:       make([]interface{}, 0),
 	}
 }

 func (p *PgSQLTxAdapter) NewInsert() common.InsertQuery {
 	return &PgSQLInsertQuery{
-		tx:     p.tx,
-		values: make(map[string]interface{}),
+		tx:         p.tx,
+		driverName: p.driverName,
+		values:     make(map[string]interface{}),
 	}
 }

 func (p *PgSQLTxAdapter) NewUpdate() common.UpdateQuery {
 	return &PgSQLUpdateQuery{
 		tx:           p.tx,
+		driverName:   p.driverName,
 		sets:         make(map[string]interface{}),
 		args:         make([]interface{}, 0),
 		whereClauses: make([]string, 0),
@@ -865,6 +900,7 @@ func (p *PgSQLTxAdapter) NewUpdate() common.UpdateQuery {
 func (p *PgSQLTxAdapter) NewDelete() common.DeleteQuery {
 	return &PgSQLDeleteQuery{
 		tx:           p.tx,
+		driverName:   p.driverName,
 		args:         make([]interface{}, 0),
 		whereClauses: make([]string, 0),
 	}
@@ -912,6 +948,10 @@ func (p *PgSQLTxAdapter) GetUnderlyingDB() interface{} {
 	return p.tx
 }

+func (p *PgSQLTxAdapter) DriverName() string {
+	return p.driverName
+}
+
 // applyJoinPreloads adds JOINs for relationships that should use JOIN strategy
 func (p *PgSQLSelectQuery) applyJoinPreloads() {
 	for _, preload := range p.preloads {
@@ -1036,9 +1076,9 @@ func (p *PgSQLSelectQuery) executePreloadQuery(ctx context.Context, field reflec
 	// Create a new select query for the related table
 	var db common.Database
 	if p.tx != nil {
-		db = &PgSQLTxAdapter{tx: p.tx}
+		db = &PgSQLTxAdapter{tx: p.tx, driverName: p.driverName}
 	} else {
-		db = &PgSQLAdapter{db: p.db}
+		db = &PgSQLAdapter{db: p.db, driverName: p.driverName}
 	}

 	query := db.NewSelect().
--- a/pkg/common/adapters/database/utils.go
+++ b/pkg/common/adapters/database/utils.go
@@ -62,9 +62,20 @@ func checkAliasLength(relation string) bool {
 // For example: "public.users" -> ("public", "users")
 //
 //	"users" -> ("", "users")
-func parseTableName(fullTableName string) (schema, table string) {
+//
+// For SQLite, schema.table is translated to schema_table since SQLite doesn't support schemas
+// in the same way as PostgreSQL/MSSQL
+func parseTableName(fullTableName, driverName string) (schema, table string) {
 	if idx := strings.LastIndex(fullTableName, "."); idx != -1 {
-		return fullTableName[:idx], fullTableName[idx+1:]
+		schema = fullTableName[:idx]
+		table = fullTableName[idx+1:]
+
+		// For SQLite, convert schema.table to schema_table
+		if driverName == "sqlite" || driverName == "sqlite3" {
+			table = schema + "_" + table
+			schema = ""
+		}
+		return schema, table
 	}
 	return "", fullTableName
 }
--- a/pkg/common/interfaces.go
+++ b/pkg/common/interfaces.go
@@ -30,6 +30,12 @@ type Database interface {
 	// For Bun, this returns *bun.DB
 	// This is useful for provider-specific features like PostgreSQL NOTIFY/LISTEN
 	GetUnderlyingDB() interface{}
+
+	// DriverName returns the canonical name of the underlying database driver.
+	// Possible values: "postgres", "sqlite", "mssql", "mysql".
+	// All adapters normalise vendor-specific strings (e.g. Bun's "pg", GORM's
+	// "sqlserver") to the values above before returning.
+	DriverName() string
 }

 // SelectQuery interface for building SELECT queries (compatible with both GORM and Bun)
--- a/pkg/common/recursive_crud_test.go
+++ b/pkg/common/recursive_crud_test.go
@@ -50,6 +50,9 @@ func (m *mockDatabase) RollbackTx(ctx context.Context) error {
 func (m *mockDatabase) GetUnderlyingDB() interface{} {
 	return nil
 }
+func (m *mockDatabase) DriverName() string {
+	return "postgres"
+}

 // Mock SelectQuery
 type mockSelectQuery struct{}
--- a/pkg/dbmanager/README.md
+++ b/pkg/dbmanager/README.md
@@ -11,6 +11,7 @@ A comprehensive database connection manager for Go that provides centralized man
  - **GORM** - Popular Go ORM
  - **Native** - Standard library `*sql.DB`
  - All three share the same underlying connection pool
+- **SQLite Schema Translation**: Automatic conversion of `schema.table` to `schema_table` for SQLite compatibility
 - **Configuration-Driven**: YAML configuration with Viper integration
 - **Production-Ready Features**:
  - Automatic health checks and reconnection
@@ -179,6 +180,35 @@ if err != nil {
 rows, err := nativeDB.QueryContext(ctx, "SELECT * FROM users WHERE active = $1", true)
 ```

+#### Cross-Database Example with SQLite
+
+```go
+// Same model works across all databases
+type User struct {
+    ID       int    `bun:"id,pk"`
+    Username string `bun:"username"`
+    Email    string `bun:"email"`
+}
+
+func (User) TableName() string {
+    return "auth.users"
+}
+
+// PostgreSQL connection
+pgConn, _ := mgr.Get("primary")
+pgDB, _ := pgConn.Bun()
+var pgUsers []User
+pgDB.NewSelect().Model(&pgUsers).Scan(ctx)
+// Executes: SELECT * FROM auth.users
+
+// SQLite connection
+sqliteConn, _ := mgr.Get("cache-db")
+sqliteDB, _ := sqliteConn.Bun()
+var sqliteUsers []User
+sqliteDB.NewSelect().Model(&sqliteUsers).Scan(ctx)
+// Executes: SELECT * FROM auth_users (schema.table → schema_table)
+```
+
 #### Use MongoDB

 ```go
@@ -368,6 +398,37 @@ Providers handle:
 - Connection statistics
 - Connection cleanup

+### SQLite Schema Handling
+
+SQLite doesn't support schemas in the same way as PostgreSQL or MSSQL. To ensure compatibility when using models designed for multi-schema databases:
+
+**Automatic Translation**: When a table name contains a schema prefix (e.g., `myschema.mytable`), it is automatically converted to `myschema_mytable` for SQLite databases.
+
+```go
+// Model definition (works across all databases)
+func (User) TableName() string {
+    return "auth.users"  // PostgreSQL/MSSQL: "auth"."users"
+                         // SQLite: "auth_users"
+}
+
+// Query execution
+db.NewSelect().Model(&User{}).Scan(ctx)
+// PostgreSQL/MSSQL: SELECT * FROM auth.users
+// SQLite: SELECT * FROM auth_users
+```
+
+**How it Works**:
+- Bun, GORM, and Native adapters detect the driver type
+- `parseTableName()` automatically translates schema.table → schema_table for SQLite
+- Translation happens transparently in all database operations (SELECT, INSERT, UPDATE, DELETE)
+- Preload and relation queries are also handled automatically
+
+**Benefits**:
+- Write database-agnostic code
+- Use the same models across PostgreSQL, MSSQL, and SQLite
+- No conditional logic needed in your application
+- Schema separation maintained through naming convention in SQLite
+
 ## Best Practices

 1. **Use Named Connections**: Be explicit about which database you're accessing
--- a/pkg/dbmanager/connection.go
+++ b/pkg/dbmanager/connection.go
@@ -467,13 +467,11 @@ func (c *sqlConnection) getNativeAdapter() (common.Database, error) {
 	// Create a native adapter based on database type
 	switch c.dbType {
 	case DatabaseTypePostgreSQL:
-		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB)
+		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB, string(c.dbType))
 	case DatabaseTypeSQLite:
-		// For SQLite, we'll use the PgSQL adapter as it works with standard sql.DB
-		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB)
+		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB, string(c.dbType))
 	case DatabaseTypeMSSQL:
-		// For MSSQL, we'll use the PgSQL adapter as it works with standard sql.DB
-		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB)
+		c.nativeAdapter = database.NewPgSQLAdapter(c.nativeDB, string(c.dbType))
 	default:
 		return nil, ErrUnsupportedDatabase
 	}
--- a/pkg/dbmanager/manager.go
+++ b/pkg/dbmanager/manager.go
@@ -231,12 +231,14 @@ func (m *connectionManager) Connect(ctx context.Context) error {

 // Close closes all database connections
 func (m *connectionManager) Close() error {
+	// Stop the health checker before taking mu.  performHealthCheck acquires
+	// a read lock, so waiting for the goroutine while holding the write lock
+	// would deadlock.
+	m.stopHealthChecker()
+
 	m.mu.Lock()
 	defer m.mu.Unlock()

-	// Stop health checker
-	m.stopHealthChecker()
-
 	// Close all connections
 	var errors []error
 	for name, conn := range m.connections {
--- a/pkg/funcspec/function_api_test.go
+++ b/pkg/funcspec/function_api_test.go
@@ -74,6 +74,10 @@ func (m *MockDatabase) GetUnderlyingDB() interface{} {
 	return m
 }

+func (m *MockDatabase) DriverName() string {
+	return "postgres"
+}
+
 // MockResult implements common.Result interface for testing
 type MockResult struct {
 	rows int64
--- a/pkg/mqttspec/handler.go
+++ b/pkg/mqttspec/handler.go
@@ -645,11 +645,14 @@ func (h *Handler) getNotifyTopic(clientID, subscriptionID string) string {
 // Database operation helpers (adapted from websocketspec)

 func (h *Handler) getTableName(schema, entity string, model interface{}) string {
-	// Use entity as table name
 	tableName := entity

 	if schema != "" {
-		tableName = schema + "." + tableName
+		if h.db.DriverName() == "sqlite" {
+			tableName = schema + "_" + tableName
+		} else {
+			tableName = schema + "." + tableName
+		}
 	}
 	return tableName
 }
--- a/pkg/resolvespec/EXAMPLES.md
+++ b/pkg/resolvespec/EXAMPLES.md
@@ -0,0 +1,572 @@
+# ResolveSpec Query Features Examples
+
+This document provides examples of using the advanced query features in ResolveSpec, including OR logic filters, Custom Operators, and FetchRowNumber.
+
+## OR Logic in Filters (SearchOr)
+
+### Basic OR Filter Example
+
+Find all users with status "active" OR "pending":
+
+```json
+POST /users
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      },
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "pending",
+        "logic_operator": "OR"
+      }
+    ]
+  }
+}
+```
+
+### Combined AND/OR Filters
+
+Find users with (status="active" OR status="pending") AND age >= 18:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      },
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "pending",
+        "logic_operator": "OR"
+      },
+      {
+        "column": "age",
+        "operator": "gte",
+        "value": 18
+      }
+    ]
+  }
+}
+```
+
+**SQL Generated:** `WHERE (status = 'active' OR status = 'pending') AND age >= 18`
+
+**Important Notes:**
+- By default, filters use AND logic
+- Consecutive filters with `"logic_operator": "OR"` are automatically grouped with parentheses
+- This grouping ensures OR conditions don't interfere with AND conditions
+- You don't need to specify `"logic_operator": "AND"` as it's the default
+
+### Multiple OR Groups
+
+You can have multiple separate OR groups:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      },
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "pending",
+        "logic_operator": "OR"
+      },
+      {
+        "column": "priority",
+        "operator": "eq",
+        "value": "high"
+      },
+      {
+        "column": "priority",
+        "operator": "eq",
+        "value": "urgent",
+        "logic_operator": "OR"
+      }
+    ]
+  }
+}
+```
+
+**SQL Generated:** `WHERE (status = 'active' OR status = 'pending') AND (priority = 'high' OR priority = 'urgent')`
+
+## Custom Operators
+
+### Simple Custom SQL Condition
+
+Filter by email domain using custom SQL:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "customOperators": [
+      {
+        "name": "company_emails",
+        "sql": "email LIKE '%@company.com'"
+      }
+    ]
+  }
+}
+```
+
+### Multiple Custom Operators
+
+Combine multiple custom SQL conditions:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "customOperators": [
+      {
+        "name": "recent_active",
+        "sql": "last_login > NOW() - INTERVAL '30 days'"
+      },
+      {
+        "name": "high_score",
+        "sql": "score > 1000"
+      }
+    ]
+  }
+}
+```
+
+### Complex Custom Operator
+
+Use complex SQL expressions:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "customOperators": [
+      {
+        "name": "priority_users",
+        "sql": "(subscription = 'premium' AND points > 500) OR (subscription = 'enterprise')"
+      }
+    ]
+  }
+}
+```
+
+### Combining Custom Operators with Regular Filters
+
+Mix custom operators with standard filters:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "country",
+        "operator": "eq",
+        "value": "USA"
+      }
+    ],
+    "customOperators": [
+      {
+        "name": "active_last_month",
+        "sql": "last_activity > NOW() - INTERVAL '1 month'"
+      }
+    ]
+  }
+}
+```
+
+## Row Numbers
+
+### Two Ways to Get Row Numbers
+
+There are two different features for row numbers:
+
+1. **`fetch_row_number`** - Get the position of ONE specific record in a sorted/filtered set
+2. **`RowNumber` field in models** - Automatically number all records in the response
+
+### 1. FetchRowNumber - Get Position of Specific Record
+
+Get the rank/position of a specific user in a leaderboard. **Important:** When `fetch_row_number` is specified, the response contains **ONLY that specific record**, not all records.
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "sort": [
+      {
+        "column": "score",
+        "direction": "desc"
+      }
+    ],
+    "fetch_row_number": "12345"
+  }
+}
+```
+
+**Response - Contains ONLY the specified user:**
+```json
+{
+  "success": true,
+  "data": {
+    "id": 12345,
+    "name": "Alice Smith",
+    "score": 9850,
+    "level": 42
+  },
+  "metadata": {
+    "total": 10000,
+    "count": 1,
+    "filtered": 10000,
+    "row_number": 42
+  }
+}
+```
+
+**Result:** User "12345" is ranked #42 out of 10,000 users. The response includes only Alice's data, not the other 9,999 users.
+
+### Row Number with Filters
+
+Find position within a filtered subset (e.g., "What's my rank in my country?"):
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "country",
+        "operator": "eq",
+        "value": "USA"
+      },
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      }
+    ],
+    "sort": [
+      {
+        "column": "score",
+        "direction": "desc"
+      }
+    ],
+    "fetch_row_number": "12345"
+  }
+}
+```
+
+**Response:**
+```json
+{
+  "success": true,
+  "data": {
+    "id": 12345,
+    "name": "Bob Johnson",
+    "country": "USA",
+    "score": 7200,
+    "status": "active"
+  },
+  "metadata": {
+    "total": 2500,
+    "count": 1,
+    "filtered": 2500,
+    "row_number": 156
+  }
+}
+```
+
+**Result:** Bob is ranked #156 out of 2,500 active USA users. Only Bob's record is returned.
+
+### 2. RowNumber Field - Auto-Number All Records
+
+If your model has a `RowNumber int64` field, restheadspec will automatically populate it for paginated results.
+
+**Model Definition:**
+```go
+type Player struct {
+    ID        int64  `json:"id"`
+    Name      string `json:"name"`
+    Score     int64  `json:"score"`
+    RowNumber int64  `json:"row_number"` // Will be auto-populated
+}
+```
+
+**Request (with pagination):**
+```json
+{
+  "operation": "read",
+  "options": {
+    "sort": [{"column": "score", "direction": "desc"}],
+    "limit": 10,
+    "offset": 20
+  }
+}
+```
+
+**Response - RowNumber automatically set:**
+```json
+{
+  "success": true,
+  "data": [
+    {
+      "id": 456,
+      "name": "Player21",
+      "score": 8900,
+      "row_number": 21
+    },
+    {
+      "id": 789,
+      "name": "Player22",
+      "score": 8850,
+      "row_number": 22
+    },
+    {
+      "id": 123,
+      "name": "Player23",
+      "score": 8800,
+      "row_number": 23
+    }
+    // ... records 24-30 ...
+  ]
+}
+```
+
+**How It Works:**
+- `row_number = offset + index + 1` (1-based)
+- With offset=20, first record gets row_number=21
+- With offset=20, second record gets row_number=22
+- Perfect for displaying "Rank" in paginated tables
+
+**Use Case:** Displaying leaderboards with rank numbers:
+```
+Rank | Player    | Score
+-----|-----------|-------
+21   | Player21  | 8900
+22   | Player22  | 8850
+23   | Player23  | 8800
+```
+
+**Note:** This feature is available in all three packages: resolvespec, restheadspec, and websocketspec.
+
+### When to Use Each Feature
+
+| Feature | Use Case | Returns | Performance |
+|---------|----------|---------|-------------|
+| `fetch_row_number` | "What's my rank?" | 1 record with position | Fast - 1 record |
+| `RowNumber` field | "Show top 10 with ranks" | Many records numbered | Fast - simple math |
+
+**Combined Example - Full Leaderboard UI:**
+
+```javascript
+// Request 1: Get current user's rank
+const userRank = await api.read({
+  fetch_row_number: currentUserId,
+  sort: [{column: "score", direction: "desc"}]
+});
+// Returns: {id: 123, name: "You", score: 7500, row_number: 156}
+
+// Request 2: Get top 10 with rank numbers
+const top10 = await api.read({
+  sort: [{column: "score", direction: "desc"}],
+  limit: 10,
+  offset: 0
+});
+// Returns: [{row_number: 1, ...}, {row_number: 2, ...}, ...]
+
+// Display:
+// "Your Rank: #156"
+// "Top Players:"
+// "#1 - Alice - 9999"
+// "#2 - Bob - 9876"
+// ...
+```
+
+## Complete Example: Advanced Query
+
+Combine all features for a complex query:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "columns": ["id", "name", "email", "score", "status"],
+    "filters": [
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      },
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "trial",
+        "logic_operator": "OR"
+      },
+      {
+        "column": "score",
+        "operator": "gte",
+        "value": 100
+      }
+    ],
+    "customOperators": [
+      {
+        "name": "recent_activity",
+        "sql": "last_login > NOW() - INTERVAL '7 days'"
+      },
+      {
+        "name": "verified_email",
+        "sql": "email_verified = true"
+      }
+    ],
+    "sort": [
+      {
+        "column": "score",
+        "direction": "desc"
+      },
+      {
+        "column": "created_at",
+        "direction": "asc"
+      }
+    ],
+    "fetch_row_number": "12345",
+    "limit": 50,
+    "offset": 0
+  }
+}
+```
+
+This query:
+- Selects specific columns
+- Filters for users with status "active" OR "trial"
+- AND score >= 100
+- Applies custom SQL conditions for recent activity and verified emails
+- Sorts by score (descending) then creation date (ascending)
+- Returns the row number of user "12345" in this filtered/sorted set
+- Returns 50 records starting from the first one
+
+## Use Cases
+
+### 1. Leaderboards - Get Current User's Rank
+
+Get the current user's position and data (returns only their record):
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "game_id",
+        "operator": "eq",
+        "value": "game123"
+      }
+    ],
+    "sort": [
+      {
+        "column": "score",
+        "direction": "desc"
+      }
+    ],
+    "fetch_row_number": "current_user_id"
+  }
+}
+```
+
+**Tip:** For full leaderboards, make two requests:
+1. One with `fetch_row_number` to get user's rank
+2. One with `limit` and `offset` to get top players list
+
+### 2. Multi-Status Search
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "order_status",
+        "operator": "eq",
+        "value": "pending"
+      },
+      {
+        "column": "order_status",
+        "operator": "eq",
+        "value": "processing",
+        "logic_operator": "OR"
+      },
+      {
+        "column": "order_status",
+        "operator": "eq",
+        "value": "shipped",
+        "logic_operator": "OR"
+      }
+    ]
+  }
+}
+```
+
+### 3. Advanced Date Filtering
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "customOperators": [
+      {
+        "name": "this_month",
+        "sql": "created_at >= DATE_TRUNC('month', CURRENT_DATE)"
+      },
+      {
+        "name": "business_hours",
+        "sql": "EXTRACT(HOUR FROM created_at) BETWEEN 9 AND 17"
+      }
+    ]
+  }
+}
+```
+
+## Security Considerations
+
+**Warning:** Custom operators allow raw SQL, which can be a security risk if not properly handled:
+
+1. **Never** directly interpolate user input into custom operator SQL
+2. Always validate and sanitize custom operator SQL on the backend
+3. Consider using a whitelist of allowed custom operators
+4. Use prepared statements or parameterized queries when possible
+5. Implement proper authorization checks before executing queries
+
+Example of safe custom operator handling in Go:
+
+```go
+// Whitelist of allowed custom operators
+allowedOperators := map[string]string{
+    "recent_week": "created_at > NOW() - INTERVAL '7 days'",
+    "active_users": "status = 'active' AND last_login > NOW() - INTERVAL '30 days'",
+    "premium_only": "subscription_level = 'premium'",
+}
+
+// Validate custom operators from request
+for _, op := range req.Options.CustomOperators {
+    if sql, ok := allowedOperators[op.Name]; ok {
+        op.SQL = sql  // Use whitelisted SQL
+    } else {
+        return errors.New("custom operator not allowed: " + op.Name)
+    }
+}
+```
--- a/pkg/resolvespec/README.md
+++ b/pkg/resolvespec/README.md
@@ -214,6 +214,146 @@ Content-Type: application/json

 ```json
 {
+  "filters": [
+    {
+      "column": "status",
+      "operator": "eq",
+      "value": "active"
+    },
+    {
+      "column": "status",
+      "operator": "eq",
+      "value": "pending",
+      "logic_operator": "OR"
+    },
+    {
+      "column": "age",
+      "operator": "gte",
+      "value": 18
+    }
+  ]
+}
+```
+
+Produces: `WHERE (status = 'active' OR status = 'pending') AND age >= 18`
+
+This grouping ensures OR conditions don't interfere with other AND conditions in the query.
+
+### Custom Operators
+
+Add custom SQL conditions when needed:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "customOperators": [
+      {
+        "name": "email_domain_filter",
+        "sql": "LOWER(email) LIKE '%@example.com'"
+      },
+      {
+        "name": "recent_records",
+        "sql": "created_at > NOW() - INTERVAL '7 days'"
+      }
+    ]
+  }
+}
+```
+
+Custom operators are applied as additional WHERE conditions to your query.
+
+### Fetch Row Number
+
+Get the row number (position) of a specific record in the filtered and sorted result set. **When `fetch_row_number` is specified, only that specific record is returned** (not all records).
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "filters": [
+      {
+        "column": "status",
+        "operator": "eq",
+        "value": "active"
+      }
+    ],
+    "sort": [
+      {
+        "column": "score",
+        "direction": "desc"
+      }
+    ],
+    "fetch_row_number": "12345"
+  }
+}
+```
+
+**Response - Returns ONLY the specified record with its position:**
+
+```json
+{
+  "success": true,
+  "data": {
+    "id": 12345,
+    "name": "John Doe",
+    "score": 850,
+    "status": "active"
+  },
+  "metadata": {
+    "total": 1000,
+    "count": 1,
+    "filtered": 1000,
+    "row_number": 42
+  }
+}
+```
+
+**Use Case:** Perfect for "Show me this user and their ranking" - you get just that one user with their position in the leaderboard.
+
+**Note:** This is different from the `RowNumber` field feature, which automatically numbers all records in a paginated response based on offset. That feature uses simple math (`offset + index + 1`), while `fetch_row_number` uses SQL window functions to calculate the actual position in a sorted/filtered set. To use the `RowNumber` field feature, simply add a `RowNumber int64` field to your model - it will be automatically populated with the row position based on pagination.
+
+## Preloading
+
+Load related entities with custom configuration:
+
+```json
+{
+  "operation": "read",
+  "options": {
+    "columns": ["id", "name", "email"],
+    "preload": [
+      {
+        "relation": "posts",
+        "columns": ["id", "title", "created_at"],
+        "filters": [
+          {
+            "column": "status",
+            "operator": "eq",
+            "value": "published"
+          }
+        ],
+        "sort": [
+          {
+            "column": "created_at",
+            "direction": "desc"
+          }
+        ],
+        "limit": 5
+      },
+      {
+        "relation": "profile",
+        "columns": ["bio", "website"]
+      }
+    ]
+  }
+}
+```
+
+## Cursor Pagination
+
+Efficient pagination for large datasets:
+
 ### First Request (No Cursor)

 ```json
@@ -427,7 +567,7 @@ Define virtual columns using SQL expressions:
    // Check permissions
    if !userHasPermission(ctx.Context, ctx.Entity) {
        return fmt.Errorf("unauthorized access to %s", ctx.Entity)
-    return nil
+    }

    // Modify query options
    if ctx.Options.Limit == nil || *ctx.Options.Limit > 100 {
@@ -435,17 +575,24 @@ Add custom SQL conditions when needed:
    }

    return nil
-            users[i].Email = maskEmail(users[i].Email)
-        }
+})
+
 // Register an after-read hook (e.g., for data transformation)
 handler.Hooks().Register(resolvespec.AfterRead, func(ctx *resolvespec.HookContext) error {
-})
+    // Transform or filter results
+    if users, ok := ctx.Result.([]User); ok {
+        for i := range users {
+            users[i].Email = maskEmail(users[i].Email)
+        }
+    }
    return nil
 })

 // Register a before-create hook (e.g., for validation)
 handler.Hooks().Register(resolvespec.BeforeCreate, func(ctx *resolvespec.HookContext) error {
    // Validate data
+    if user, ok := ctx.Data.(*User); ok {
+        if user.Email == "" {
            return fmt.Errorf("email is required")
        }
        // Add timestamps
--- a/pkg/resolvespec/filter_test.go
+++ b/pkg/resolvespec/filter_test.go
@@ -0,0 +1,143 @@
+package resolvespec
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/bitechdev/ResolveSpec/pkg/common"
+)
+
+// TestBuildFilterCondition tests the filter condition builder
+func TestBuildFilterCondition(t *testing.T) {
+	h := &Handler{}
+
+	tests := []struct {
+		name              string
+		filter            common.FilterOption
+		expectedCondition string
+		expectedArgsCount int
+	}{
+		{
+			name: "Equal operator",
+			filter: common.FilterOption{
+				Column:   "status",
+				Operator: "eq",
+				Value:    "active",
+			},
+			expectedCondition: "status = ?",
+			expectedArgsCount: 1,
+		},
+		{
+			name: "Greater than operator",
+			filter: common.FilterOption{
+				Column:   "age",
+				Operator: "gt",
+				Value:    18,
+			},
+			expectedCondition: "age > ?",
+			expectedArgsCount: 1,
+		},
+		{
+			name: "IN operator",
+			filter: common.FilterOption{
+				Column:   "status",
+				Operator: "in",
+				Value:    []string{"active", "pending"},
+			},
+			expectedCondition: "status IN (?)",
+			expectedArgsCount: 1,
+		},
+		{
+			name: "LIKE operator",
+			filter: common.FilterOption{
+				Column:   "email",
+				Operator: "like",
+				Value:    "%@example.com",
+			},
+			expectedCondition: "email LIKE ?",
+			expectedArgsCount: 1,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			condition, args := h.buildFilterCondition(tt.filter)
+
+			if condition != tt.expectedCondition {
+				t.Errorf("Expected condition '%s', got '%s'", tt.expectedCondition, condition)
+			}
+
+			if len(args) != tt.expectedArgsCount {
+				t.Errorf("Expected %d args, got %d", tt.expectedArgsCount, len(args))
+			}
+
+			// Note: Skip value comparison for slices as they can't be compared with ==
+			// The important part is that args are populated correctly
+		})
+	}
+}
+
+// TestORGrouping tests that consecutive OR filters are properly grouped
+func TestORGrouping(t *testing.T) {
+	// This is a conceptual test - in practice we'd need a mock SelectQuery
+	// to verify the actual SQL grouping behavior
+	t.Run("Consecutive OR filters should be grouped", func(t *testing.T) {
+		filters := []common.FilterOption{
+			{Column: "status", Operator: "eq", Value: "active"},
+			{Column: "status", Operator: "eq", Value: "pending", LogicOperator: "OR"},
+			{Column: "status", Operator: "eq", Value: "trial", LogicOperator: "OR"},
+			{Column: "age", Operator: "gte", Value: 18},
+		}
+
+		// Expected behavior: (status='active' OR status='pending' OR status='trial') AND age>=18
+		// The first three filters should be grouped together
+		// The fourth filter should be separate with AND
+
+		// Count OR groups
+		orGroupCount := 0
+		inORGroup := false
+
+		for i := 1; i < len(filters); i++ {
+			if strings.EqualFold(filters[i].LogicOperator, "OR") && !inORGroup {
+				orGroupCount++
+				inORGroup = true
+			} else if !strings.EqualFold(filters[i].LogicOperator, "OR") {
+				inORGroup = false
+			}
+		}
+
+		// We should have detected one OR group
+		if orGroupCount != 1 {
+			t.Errorf("Expected 1 OR group, detected %d", orGroupCount)
+		}
+	})
+
+	t.Run("Multiple OR groups should be handled correctly", func(t *testing.T) {
+		filters := []common.FilterOption{
+			{Column: "status", Operator: "eq", Value: "active"},
+			{Column: "status", Operator: "eq", Value: "pending", LogicOperator: "OR"},
+			{Column: "priority", Operator: "eq", Value: "high"},
+			{Column: "priority", Operator: "eq", Value: "urgent", LogicOperator: "OR"},
+		}
+
+		// Expected: (status='active' OR status='pending') AND (priority='high' OR priority='urgent')
+		// Should have two OR groups
+
+		orGroupCount := 0
+		inORGroup := false
+
+		for i := 1; i < len(filters); i++ {
+			if strings.EqualFold(filters[i].LogicOperator, "OR") && !inORGroup {
+				orGroupCount++
+				inORGroup = true
+			} else if !strings.EqualFold(filters[i].LogicOperator, "OR") {
+				inORGroup = false
+			}
+		}
+
+		// We should have detected two OR groups
+		if orGroupCount != 2 {
+			t.Errorf("Expected 2 OR groups, detected %d", orGroupCount)
+		}
+	})
+}
--- a/pkg/resolvespec/handler.go
+++ b/pkg/resolvespec/handler.go
@@ -280,10 +280,13 @@ func (h *Handler) handleRead(ctx context.Context, w common.ResponseWriter, id st
 		}
 	}

-	// Apply filters
-	for _, filter := range options.Filters {
-		logger.Debug("Applying filter: %s %s %v", filter.Column, filter.Operator, filter.Value)
-		query = h.applyFilter(query, filter)
+	// Apply filters with proper grouping for OR logic
+	query = h.applyFilters(query, options.Filters)
+
+	// Apply custom operators
+	for _, customOp := range options.CustomOperators {
+		logger.Debug("Applying custom operator: %s - %s", customOp.Name, customOp.SQL)
+		query = query.Where(customOp.SQL)
 	}

 	// Apply sorting
@@ -381,24 +384,94 @@ func (h *Handler) handleRead(ctx context.Context, w common.ResponseWriter, id st
 		}
 	}

-	// Apply pagination
-	if options.Limit != nil && *options.Limit > 0 {
-		logger.Debug("Applying limit: %d", *options.Limit)
-		query = query.Limit(*options.Limit)
+	// Handle FetchRowNumber if requested
+	var rowNumber *int64
+	if options.FetchRowNumber != nil && *options.FetchRowNumber != "" {
+		logger.Debug("Fetching row number for ID: %s", *options.FetchRowNumber)
+		pkName := reflection.GetPrimaryKeyName(model)
+
+		// Build ROW_NUMBER window function SQL
+		rowNumberSQL := "ROW_NUMBER() OVER ("
+		if len(options.Sort) > 0 {
+			rowNumberSQL += "ORDER BY "
+			for i, sort := range options.Sort {
+				if i > 0 {
+					rowNumberSQL += ", "
+				}
+				direction := "ASC"
+				if strings.EqualFold(sort.Direction, "desc") {
+					direction = "DESC"
+				}
+				rowNumberSQL += fmt.Sprintf("%s %s", sort.Column, direction)
+			}
+		}
+		rowNumberSQL += ")"
+
+		// Create a query to fetch the row number using a subquery approach
+		// We'll select the PK and row_number, then filter by the target ID
+		type RowNumResult struct {
+			RowNum int64 `bun:"row_num"`
+		}
+
+		rowNumQuery := h.db.NewSelect().Table(tableName).
+			ColumnExpr(fmt.Sprintf("%s AS row_num", rowNumberSQL)).
+			Column(pkName)
+
+		// Apply the same filters as the main query
+		for _, filter := range options.Filters {
+			rowNumQuery = h.applyFilter(rowNumQuery, filter)
+		}
+
+		// Apply custom operators
+		for _, customOp := range options.CustomOperators {
+			rowNumQuery = rowNumQuery.Where(customOp.SQL)
+		}
+
+		// Filter for the specific ID we want the row number for
+		rowNumQuery = rowNumQuery.Where(fmt.Sprintf("%s = ?", common.QuoteIdent(pkName)), *options.FetchRowNumber)
+
+		// Execute query to get row number
+		var result RowNumResult
+		if err := rowNumQuery.Scan(ctx, &result); err != nil {
+			if err != sql.ErrNoRows {
+				logger.Warn("Error fetching row number: %v", err)
+			}
+		} else {
+			rowNumber = &result.RowNum
+			logger.Debug("Found row number: %d", *rowNumber)
+		}
 	}
-	if options.Offset != nil && *options.Offset > 0 {
-		logger.Debug("Applying offset: %d", *options.Offset)
-		query = query.Offset(*options.Offset)
+
+	// Apply pagination (skip if FetchRowNumber is set - we want only that record)
+	if options.FetchRowNumber == nil || *options.FetchRowNumber == "" {
+		if options.Limit != nil && *options.Limit > 0 {
+			logger.Debug("Applying limit: %d", *options.Limit)
+			query = query.Limit(*options.Limit)
+		}
+		if options.Offset != nil && *options.Offset > 0 {
+			logger.Debug("Applying offset: %d", *options.Offset)
+			query = query.Offset(*options.Offset)
+		}
 	}

 	// Execute query
 	var result interface{}
-	if id != "" {
-		logger.Debug("Querying single record with ID: %s", id)
+	if id != "" || (options.FetchRowNumber != nil && *options.FetchRowNumber != "") {
+		// Single record query - either by URL ID or FetchRowNumber
+		var targetID string
+		if id != "" {
+			targetID = id
+			logger.Debug("Querying single record with URL ID: %s", id)
+		} else {
+			targetID = *options.FetchRowNumber
+			logger.Debug("Querying single record with FetchRowNumber ID: %s", targetID)
+		}
+
 		// For single record, create a new pointer to the struct type
 		singleResult := reflect.New(modelType).Interface()
+		pkName := reflection.GetPrimaryKeyName(singleResult)

-		query = query.Where(fmt.Sprintf("%s = ?", common.QuoteIdent(reflection.GetPrimaryKeyName(singleResult))), id)
+		query = query.Where(fmt.Sprintf("%s = ?", common.QuoteIdent(pkName)), targetID)
 		if err := query.Scan(ctx, singleResult); err != nil {
 			logger.Error("Error querying record: %v", err)
 			h.sendError(w, http.StatusInternalServerError, "query_error", "Error executing query", err)
@@ -418,20 +491,35 @@ func (h *Handler) handleRead(ctx context.Context, w common.ResponseWriter, id st

 	logger.Info("Successfully retrieved records")

+	// Build metadata
 	limit := 0
-	if options.Limit != nil {
-		limit = *options.Limit
-	}
 	offset := 0
-	if options.Offset != nil {
-		offset = *options.Offset
+	count := int64(total)
+
+	// When FetchRowNumber is used, we only return 1 record
+	if options.FetchRowNumber != nil && *options.FetchRowNumber != "" {
+		count = 1
+		// Don't use limit/offset when fetching specific record
+	} else {
+		if options.Limit != nil {
+			limit = *options.Limit
+		}
+		if options.Offset != nil {
+			offset = *options.Offset
+		}
+
+		// Set row numbers on records if RowNumber field exists
+		// Only for multiple records (not when fetching single record)
+		h.setRowNumbersOnRecords(result, offset)
 	}

 	h.sendResponse(w, result, &common.Metadata{
-		Total:    int64(total),
-		Filtered: int64(total),
-		Limit:    limit,
-		Offset:   offset,
+		Total:     int64(total),
+		Filtered:  int64(total),
+		Count:     count,
+		Limit:     limit,
+		Offset:    offset,
+		RowNumber: rowNumber,
 	})
 }

@@ -1303,29 +1391,161 @@ func (h *Handler) handleDelete(ctx context.Context, w common.ResponseWriter, id
 	h.sendResponse(w, recordToDelete, nil)
 }

-func (h *Handler) applyFilter(query common.SelectQuery, filter common.FilterOption) common.SelectQuery {
+// applyFilters applies all filters with proper grouping for OR logic
+// Groups consecutive OR filters together to ensure proper query precedence
+// Example: [A, B(OR), C(OR), D(AND)] => WHERE (A OR B OR C) AND D
+func (h *Handler) applyFilters(query common.SelectQuery, filters []common.FilterOption) common.SelectQuery {
+	if len(filters) == 0 {
+		return query
+	}
+
+	i := 0
+	for i < len(filters) {
+		// Check if this starts an OR group (current or next filter has OR logic)
+		startORGroup := i+1 < len(filters) && strings.EqualFold(filters[i+1].LogicOperator, "OR")
+
+		if startORGroup {
+			// Collect all consecutive filters that are OR'd together
+			orGroup := []common.FilterOption{filters[i]}
+			j := i + 1
+			for j < len(filters) && strings.EqualFold(filters[j].LogicOperator, "OR") {
+				orGroup = append(orGroup, filters[j])
+				j++
+			}
+
+			// Apply the OR group as a single grouped WHERE clause
+			query = h.applyFilterGroup(query, orGroup)
+			i = j
+		} else {
+			// Single filter with AND logic (or first filter)
+			condition, args := h.buildFilterCondition(filters[i])
+			if condition != "" {
+				query = query.Where(condition, args...)
+			}
+			i++
+		}
+	}
+
+	return query
+}
+
+// applyFilterGroup applies a group of filters that should be OR'd together
+// Always wraps them in parentheses and applies as a single WHERE clause
+func (h *Handler) applyFilterGroup(query common.SelectQuery, filters []common.FilterOption) common.SelectQuery {
+	if len(filters) == 0 {
+		return query
+	}
+
+	// Build all conditions and collect args
+	var conditions []string
+	var args []interface{}
+
+	for _, filter := range filters {
+		condition, filterArgs := h.buildFilterCondition(filter)
+		if condition != "" {
+			conditions = append(conditions, condition)
+			args = append(args, filterArgs...)
+		}
+	}
+
+	if len(conditions) == 0 {
+		return query
+	}
+
+	// Single filter - no need for grouping
+	if len(conditions) == 1 {
+		return query.Where(conditions[0], args...)
+	}
+
+	// Multiple conditions - group with parentheses and OR
+	groupedCondition := "(" + strings.Join(conditions, " OR ") + ")"
+	return query.Where(groupedCondition, args...)
+}
+
+// buildFilterCondition builds a filter condition and returns it with args
+func (h *Handler) buildFilterCondition(filter common.FilterOption) (conditionString string, conditionArgs []interface{}) {
+	var condition string
+	var args []interface{}
+
 	switch filter.Operator {
 	case "eq":
-		return query.Where(fmt.Sprintf("%s = ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s = ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "neq":
-		return query.Where(fmt.Sprintf("%s != ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s != ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "gt":
-		return query.Where(fmt.Sprintf("%s > ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s > ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "gte":
-		return query.Where(fmt.Sprintf("%s >= ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s >= ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "lt":
-		return query.Where(fmt.Sprintf("%s < ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s < ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "lte":
-		return query.Where(fmt.Sprintf("%s <= ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s <= ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "like":
-		return query.Where(fmt.Sprintf("%s LIKE ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s LIKE ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "ilike":
-		return query.Where(fmt.Sprintf("%s ILIKE ?", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s ILIKE ?", filter.Column)
+		args = []interface{}{filter.Value}
 	case "in":
-		return query.Where(fmt.Sprintf("%s IN (?)", filter.Column), filter.Value)
+		condition = fmt.Sprintf("%s IN (?)", filter.Column)
+		args = []interface{}{filter.Value}
+	default:
+		return "", nil
+	}
+
+	return condition, args
+}
+
+func (h *Handler) applyFilter(query common.SelectQuery, filter common.FilterOption) common.SelectQuery {
+	// Determine which method to use based on LogicOperator
+	useOrLogic := strings.EqualFold(filter.LogicOperator, "OR")
+
+	var condition string
+	var args []interface{}
+
+	switch filter.Operator {
+	case "eq":
+		condition = fmt.Sprintf("%s = ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "neq":
+		condition = fmt.Sprintf("%s != ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "gt":
+		condition = fmt.Sprintf("%s > ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "gte":
+		condition = fmt.Sprintf("%s >= ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "lt":
+		condition = fmt.Sprintf("%s < ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "lte":
+		condition = fmt.Sprintf("%s <= ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "like":
+		condition = fmt.Sprintf("%s LIKE ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "ilike":
+		condition = fmt.Sprintf("%s ILIKE ?", filter.Column)
+		args = []interface{}{filter.Value}
+	case "in":
+		condition = fmt.Sprintf("%s IN (?)", filter.Column)
+		args = []interface{}{filter.Value}
 	default:
 		return query
 	}
+
+	// Apply filter with appropriate logic operator
+	if useOrLogic {
+		return query.WhereOr(condition, args...)
+	}
+	return query.Where(condition, args...)
 }

 // parseTableName splits a table name that may contain schema into separate schema and table
@@ -1380,10 +1600,16 @@ func (h *Handler) getSchemaAndTable(defaultSchema, entity string, model interfac
 	return schema, entity
 }

-// getTableName returns the full table name including schema (schema.table)
+// getTableName returns the full table name including schema.
+// For most drivers the result is "schema.table".  For SQLite, which does not
+// support schema-qualified names, the schema and table are joined with an
+// underscore: "schema_table".
 func (h *Handler) getTableName(schema, entity string, model interface{}) string {
 	schemaName, tableName := h.getSchemaAndTable(schema, entity, model)
 	if schemaName != "" {
+		if h.db.DriverName() == "sqlite" {
+			return fmt.Sprintf("%s_%s", schemaName, tableName)
+		}
 		return fmt.Sprintf("%s.%s", schemaName, tableName)
 	}
 	return tableName
@@ -1703,6 +1929,51 @@ func toSnakeCase(s string) string {
 	return strings.ToLower(result.String())
 }

+// setRowNumbersOnRecords sets the RowNumber field on each record if it exists
+// The row number is calculated as offset + index + 1 (1-based)
+func (h *Handler) setRowNumbersOnRecords(records interface{}, offset int) {
+	// Get the reflect value of the records
+	recordsValue := reflect.ValueOf(records)
+	if recordsValue.Kind() == reflect.Ptr {
+		recordsValue = recordsValue.Elem()
+	}
+
+	// Ensure it's a slice
+	if recordsValue.Kind() != reflect.Slice {
+		logger.Debug("setRowNumbersOnRecords: records is not a slice, skipping")
+		return
+	}
+
+	// Iterate through each record
+	for i := 0; i < recordsValue.Len(); i++ {
+		record := recordsValue.Index(i)
+
+		// Dereference if it's a pointer
+		if record.Kind() == reflect.Ptr {
+			if record.IsNil() {
+				continue
+			}
+			record = record.Elem()
+		}
+
+		// Ensure it's a struct
+		if record.Kind() != reflect.Struct {
+			continue
+		}
+
+		// Try to find and set the RowNumber field
+		rowNumberField := record.FieldByName("RowNumber")
+		if rowNumberField.IsValid() && rowNumberField.CanSet() {
+			// Check if the field is of type int64
+			if rowNumberField.Kind() == reflect.Int64 {
+				rowNum := int64(offset + i + 1)
+				rowNumberField.SetInt(rowNum)
+				logger.Debug("Set RowNumber=%d for record index %d", rowNum, i)
+			}
+		}
+	}
+}
+
 // HandleOpenAPI generates and returns the OpenAPI specification
 func (h *Handler) HandleOpenAPI(w common.ResponseWriter, r common.Request) {
 	if h.openAPIGenerator == nil {
--- a/pkg/restheadspec/handler.go
+++ b/pkg/restheadspec/handler.go
@@ -2015,11 +2015,18 @@ func (h *Handler) processChildRelationsForField(
 	return nil
 }

-// getTableNameForRelatedModel gets the table name for a related model
+// getTableNameForRelatedModel gets the table name for a related model.
+// If the model's TableName() is schema-qualified (e.g. "public.users") the
+// separator is adjusted for the active driver: underscore for SQLite, dot otherwise.
 func (h *Handler) getTableNameForRelatedModel(model interface{}, defaultName string) string {
 	if provider, ok := model.(common.TableNameProvider); ok {
 		tableName := provider.TableName()
 		if tableName != "" {
+			if schema, table := h.parseTableName(tableName); schema != "" {
+				if h.db.DriverName() == "sqlite" {
+					return fmt.Sprintf("%s_%s", schema, table)
+				}
+			}
 			return tableName
 		}
 	}
@@ -2264,10 +2271,16 @@ func (h *Handler) getSchemaAndTable(defaultSchema, entity string, model interfac
 	return schema, entity
 }

-// getTableName returns the full table name including schema (schema.table)
+// getTableName returns the full table name including schema.
+// For most drivers the result is "schema.table".  For SQLite, which does not
+// support schema-qualified names, the schema and table are joined with an
+// underscore: "schema_table".
 func (h *Handler) getTableName(schema, entity string, model interface{}) string {
 	schemaName, tableName := h.getSchemaAndTable(schema, entity, model)
 	if schemaName != "" {
+		if h.db.DriverName() == "sqlite" {
+			return fmt.Sprintf("%s_%s", schemaName, tableName)
+		}
 		return fmt.Sprintf("%s.%s", schemaName, tableName)
 	}
 	return tableName
@@ -2589,21 +2602,8 @@ func (h *Handler) FetchRowNumber(ctx context.Context, tableName string, pkName s
 		sortSQL = fmt.Sprintf("%s.%s ASC", tableName, pkName)
 	}

-	// Build WHERE clauses from filters
-	whereClauses := make([]string, 0)
-	for i := range options.Filters {
-		filter := &options.Filters[i]
-		whereClause := h.buildFilterSQL(filter, tableName)
-		if whereClause != "" {
-			whereClauses = append(whereClauses, fmt.Sprintf("(%s)", whereClause))
-		}
-	}
-
-	// Combine WHERE clauses
-	whereSQL := ""
-	if len(whereClauses) > 0 {
-		whereSQL = "WHERE " + strings.Join(whereClauses, " AND ")
-	}
+	// Build WHERE clause from filters with proper OR grouping
+	whereSQL := h.buildWhereClauseWithORGrouping(options.Filters, tableName)

 	// Add custom SQL WHERE if provided
 	if options.CustomSQLWhere != "" {
@@ -2664,6 +2664,67 @@ func (h *Handler) FetchRowNumber(ctx context.Context, tableName string, pkName s
 }

 // buildFilterSQL converts a filter to SQL WHERE clause string
+// buildWhereClauseWithORGrouping builds a WHERE clause from filters with proper OR grouping
+// Groups consecutive OR filters together to ensure proper SQL precedence
+// Example: [A, B(OR), C(OR), D(AND)] => WHERE (A OR B OR C) AND D
+func (h *Handler) buildWhereClauseWithORGrouping(filters []common.FilterOption, tableName string) string {
+	if len(filters) == 0 {
+		return ""
+	}
+
+	var groups []string
+	i := 0
+
+	for i < len(filters) {
+		// Check if this starts an OR group (next filter has OR logic)
+		startORGroup := i+1 < len(filters) && strings.EqualFold(filters[i+1].LogicOperator, "OR")
+
+		if startORGroup {
+			// Collect all consecutive filters that are OR'd together
+			orGroup := []string{}
+
+			// Add current filter
+			filterSQL := h.buildFilterSQL(&filters[i], tableName)
+			if filterSQL != "" {
+				orGroup = append(orGroup, filterSQL)
+			}
+
+			// Collect remaining OR filters
+			j := i + 1
+			for j < len(filters) && strings.EqualFold(filters[j].LogicOperator, "OR") {
+				filterSQL := h.buildFilterSQL(&filters[j], tableName)
+				if filterSQL != "" {
+					orGroup = append(orGroup, filterSQL)
+				}
+				j++
+			}
+
+			// Group OR filters with parentheses
+			if len(orGroup) > 0 {
+				if len(orGroup) == 1 {
+					groups = append(groups, orGroup[0])
+				} else {
+					groups = append(groups, "("+strings.Join(orGroup, " OR ")+")")
+				}
+			}
+			i = j
+		} else {
+			// Single filter with AND logic (or first filter)
+			filterSQL := h.buildFilterSQL(&filters[i], tableName)
+			if filterSQL != "" {
+				groups = append(groups, filterSQL)
+			}
+			i++
+		}
+	}
+
+	if len(groups) == 0 {
+		return ""
+	}
+
+	return "WHERE " + strings.Join(groups, " AND ")
+}
+
 func (h *Handler) buildFilterSQL(filter *common.FilterOption, tableName string) string {
 	qualifiedColumn := h.qualifyColumnName(filter.Column, tableName)

--- a/pkg/websocketspec/handler.go
+++ b/pkg/websocketspec/handler.go
@@ -6,6 +6,7 @@ import (
 	"fmt"
 	"net/http"
 	"reflect"
+	"strings"
 	"time"

 	"github.com/google/uuid"
@@ -540,10 +541,8 @@ func (h *Handler) readMultiple(hookCtx *HookContext) (data interface{}, metadata

 	// Apply options (simplified implementation)
 	if hookCtx.Options != nil {
-		// Apply filters
-		for _, filter := range hookCtx.Options.Filters {
-			query = query.Where(fmt.Sprintf("%s %s ?", filter.Column, h.getOperatorSQL(filter.Operator)), filter.Value)
-		}
+		// Apply filters with OR grouping support
+		query = h.applyFilters(query, hookCtx.Options.Filters)

 		// Apply sorting
 		for _, sort := range hookCtx.Options.Sort {
@@ -578,6 +577,13 @@ func (h *Handler) readMultiple(hookCtx *HookContext) (data interface{}, metadata
 		return nil, nil, fmt.Errorf("failed to read records: %w", err)
 	}

+	// Set row numbers on records if RowNumber field exists
+	offset := 0
+	if hookCtx.Options != nil && hookCtx.Options.Offset != nil {
+		offset = *hookCtx.Options.Offset
+	}
+	h.setRowNumbersOnRecords(hookCtx.ModelPtr, offset)
+
 	// Get count
 	metadata = make(map[string]interface{})
 	countQuery := h.db.NewSelect().Model(hookCtx.ModelPtr).Table(hookCtx.TableName)
@@ -656,11 +662,14 @@ func (h *Handler) delete(hookCtx *HookContext) error {
 // Helper methods

 func (h *Handler) getTableName(schema, entity string, model interface{}) string {
-	// Use entity as table name
 	tableName := entity

 	if schema != "" {
-		tableName = schema + "." + tableName
+		if h.db.DriverName() == "sqlite" {
+			tableName = schema + "_" + tableName
+		} else {
+			tableName = schema + "." + tableName
+		}
 	}
 	return tableName
 }
@@ -680,6 +689,133 @@ func (h *Handler) getMetadata(schema, entity string, model interface{}) map[stri
 }

 // getOperatorSQL converts filter operator to SQL operator
+// applyFilters applies all filters with proper grouping for OR logic
+// Groups consecutive OR filters together to ensure proper query precedence
+func (h *Handler) applyFilters(query common.SelectQuery, filters []common.FilterOption) common.SelectQuery {
+	if len(filters) == 0 {
+		return query
+	}
+
+	i := 0
+	for i < len(filters) {
+		// Check if this starts an OR group (next filter has OR logic)
+		startORGroup := i+1 < len(filters) && strings.EqualFold(filters[i+1].LogicOperator, "OR")
+
+		if startORGroup {
+			// Collect all consecutive filters that are OR'd together
+			orGroup := []common.FilterOption{filters[i]}
+			j := i + 1
+			for j < len(filters) && strings.EqualFold(filters[j].LogicOperator, "OR") {
+				orGroup = append(orGroup, filters[j])
+				j++
+			}
+
+			// Apply the OR group as a single grouped WHERE clause
+			query = h.applyFilterGroup(query, orGroup)
+			i = j
+		} else {
+			// Single filter with AND logic (or first filter)
+			condition, args := h.buildFilterCondition(filters[i])
+			if condition != "" {
+				query = query.Where(condition, args...)
+			}
+			i++
+		}
+	}
+
+	return query
+}
+
+// applyFilterGroup applies a group of filters that should be OR'd together
+// Always wraps them in parentheses and applies as a single WHERE clause
+func (h *Handler) applyFilterGroup(query common.SelectQuery, filters []common.FilterOption) common.SelectQuery {
+	if len(filters) == 0 {
+		return query
+	}
+
+	// Build all conditions and collect args
+	var conditions []string
+	var args []interface{}
+
+	for _, filter := range filters {
+		condition, filterArgs := h.buildFilterCondition(filter)
+		if condition != "" {
+			conditions = append(conditions, condition)
+			args = append(args, filterArgs...)
+		}
+	}
+
+	if len(conditions) == 0 {
+		return query
+	}
+
+	// Single filter - no need for grouping
+	if len(conditions) == 1 {
+		return query.Where(conditions[0], args...)
+	}
+
+	// Multiple conditions - group with parentheses and OR
+	groupedCondition := "(" + strings.Join(conditions, " OR ") + ")"
+	return query.Where(groupedCondition, args...)
+}
+
+// buildFilterCondition builds a filter condition and returns it with args
+func (h *Handler) buildFilterCondition(filter common.FilterOption) (conditionString string, conditionArgs []interface{}) {
+	var condition string
+	var args []interface{}
+
+	operatorSQL := h.getOperatorSQL(filter.Operator)
+	condition = fmt.Sprintf("%s %s ?", filter.Column, operatorSQL)
+	args = []interface{}{filter.Value}
+
+	return condition, args
+}
+
+// setRowNumbersOnRecords sets the RowNumber field on each record if it exists
+// The row number is calculated as offset + index + 1 (1-based)
+func (h *Handler) setRowNumbersOnRecords(records interface{}, offset int) {
+	// Get the reflect value of the records
+	recordsValue := reflect.ValueOf(records)
+	if recordsValue.Kind() == reflect.Ptr {
+		recordsValue = recordsValue.Elem()
+	}
+
+	// Ensure it's a slice
+	if recordsValue.Kind() != reflect.Slice {
+		logger.Debug("[WebSocketSpec] setRowNumbersOnRecords: records is not a slice, skipping")
+		return
+	}
+
+	// Iterate through each record
+	for i := 0; i < recordsValue.Len(); i++ {
+		record := recordsValue.Index(i)
+
+		// Dereference if it's a pointer
+		if record.Kind() == reflect.Ptr {
+			if record.IsNil() {
+				continue
+			}
+			record = record.Elem()
+		}
+
+		// Ensure it's a struct
+		if record.Kind() != reflect.Struct {
+			continue
+		}
+
+		// Try to find and set the RowNumber field
+		rowNumberField := record.FieldByName("RowNumber")
+		if rowNumberField.IsValid() && rowNumberField.CanSet() {
+			// Check if the field is of type int64
+			if rowNumberField.Kind() == reflect.Int64 {
+				rowNum := int64(offset + i + 1)
+				rowNumberField.SetInt(rowNum)
+				logger.Debug("[WebSocketSpec] Set RowNumber=%d for record index %d", rowNum, i)
+			}
+		}
+	}
+}
+
 func (h *Handler) getOperatorSQL(operator string) string {
 	switch operator {
 	case "eq":
--- a/pkg/websocketspec/handler_test.go
+++ b/pkg/websocketspec/handler_test.go
@@ -82,6 +82,10 @@ func (m *MockDatabase) GetUnderlyingDB() interface{} {
 	return args.Get(0)
 }

+func (m *MockDatabase) DriverName() string {
+	return "postgres"
+}
+
 // MockSelectQuery is a mock implementation of common.SelectQuery
 type MockSelectQuery struct {
 	mock.Mock
--- a/resolvespec-python/todo.md
+++ b/resolvespec-python/todo.md
@@ -1,5 +1,50 @@
-# Python Implementation of the ResolveSpec API
+# ResolveSpec Python Client - TODO

-# Server 
+## Client Implementation & Testing

-# Client
+### 1. ResolveSpec Client API
+
+- [ ] Core API implementation (read, create, update, delete, get_metadata)
+- [ ] Unit tests for API functions
+- [ ] Integration tests with server
+- [ ] Error handling and edge cases
+
+### 2. HeaderSpec Client API
+
+- [ ] Client API implementation
+- [ ] Unit tests
+- [ ] Integration tests with server
+
+### 3. FunctionSpec Client API
+
+- [ ] Client API implementation
+- [ ] Unit tests
+- [ ] Integration tests with server
+
+### 4. WebSocketSpec Client API
+
+- [ ] WebSocketClient class implementation (read, create, update, delete, meta, subscribe, unsubscribe)
+- [ ] Unit tests for WebSocketClient
+- [ ] Connection handling tests
+- [ ] Subscription tests
+- [ ] Integration tests with server
+
+### 5. Testing Infrastructure
+
+- [ ] Set up test framework (pytest)
+- [ ] Configure test coverage reporting (pytest-cov)
+- [ ] Add test utilities and fixtures
+- [ ] Create test documentation
+- [ ] Package and publish to PyPI
+
+## Documentation
+
+- [ ] API reference documentation
+- [ ] Usage examples for each client API
+- [ ] Installation guide
+- [ ] Contributing guidelines
+- [ ] README with quick start
+
+---
+
+**Last Updated:** 2026-02-07
--- a/todo.md
+++ b/todo.md
@@ -2,36 +2,98 @@

 This document tracks incomplete features and improvements for the ResolveSpec project.

+## In Progress
+
+### Database Layer
+
+- [x] SQLite schema translation (schema.table → schema_table)
+- [x] Driver name normalization across adapters
+- [x] Database Connection Manager (dbmanager) package
+
 ### Documentation
- Ensure all new features are documented in README.md
- Update examples to showcase new functionality
- Add migration notes if any breaking changes are introduced

+- [x] Add dbmanager to README
+- [x] Add WebSocketSpec to top-level intro
+- [x] Add MQTTSpec to top-level intro
+- [x] Remove migration sections from README
+- [ ] Complete API reference documentation
+- [ ] Add examples for all supported databases

-### 8.
+## Planned Features

-1. **Test Coverage**: Increase from 20% to 70%+
-   - Add integration tests for CRUD operations
-   - Add unit tests for security providers
-   - Add concurrency tests for model registry
+### ResolveSpec JS Client Implementation & Testing
+
+1. **ResolveSpec Client API (resolvespec-js)**
+   - [x] Core API implementation (read, create, update, delete, getMetadata)
+   - [ ] Unit tests for API functions
+   - [ ] Integration tests with server
+   - [ ] Error handling and edge cases
+
+2. **HeaderSpec Client API (resolvespec-js)**
+   - [ ] Client API implementation
+   - [ ] Unit tests
+   - [ ] Integration tests with server
+
+3. **FunctionSpec Client API (resolvespec-js)**
+   - [ ] Client API implementation
+   - [ ] Unit tests
+   - [ ] Integration tests with server
+
+4. **WebSocketSpec Client API (resolvespec-js)**
+   - [x] WebSocketClient class implementation (read, create, update, delete, meta, subscribe, unsubscribe)
+   - [ ] Unit tests for WebSocketClient
+   - [ ] Connection handling tests
+   - [ ] Subscription tests
+   - [ ] Integration tests with server
+
+5. **resolvespec-js Testing Infrastructure**
+   - [ ] Set up test framework (Jest or Vitest)
+   - [ ] Configure test coverage reporting
+   - [ ] Add test utilities and mocks
+   - [ ] Create test documentation
+
+### ResolveSpec Python Client Implementation & Testing
+
+See [`resolvespec-python/todo.md`](./resolvespec-python/todo.md) for detailed Python client implementation tasks.
+
+### Core Functionality
+
+1. **Enhanced Preload Filtering**
+   - [ ] Column selection for nested preloads
+   - [ ] Advanced filtering conditions for relations
+   - [ ] Performance optimization for deep nesting
+
+2. **Advanced Query Features**
+   - [ ] Custom SQL join support
+   - [ ] Computed column improvements
+   - [ ] Recursive query support
+
+3. **Testing & Quality**
+   - [ ] Increase test coverage to 70%+
+   - [ ] Add integration tests for all ORMs
+   - [ ] Add concurrency tests for thread safety
+   - [ ] Performance benchmarks
+
+### Infrastructure
+
+- [ ] Improved error handling and reporting
+- [ ] Enhanced logging capabilities
+- [ ] Additional monitoring metrics
+- [ ] Performance profiling tools
+
+## Documentation Tasks
+
+- [ ] Complete API reference
+- [ ] Add troubleshooting guides
+- [ ] Create architecture diagrams
+- [ ] Expand database adapter documentation
+
+## Known Issues
+
+- [ ] Long preload alias names may exceed PostgreSQL identifier limit
+- [ ] Some edge cases in computed column handling

 ---

-## Priority Ranking
-
-1. **High Priority**
-   - Column Selection and Filtering for Preloads (#1)
-   - Proper Condition Handling for Bun Preloads (#4)
-
-2. **Medium Priority**
-   - Custom SQL Join Support (#3)
-   - Recursive JSON Cleaning (#2)
-
-3. **Low Priority**
-   - Modernize Go Type Declarations (#5)
-
---
-
-
-
-**Last Updated:** 2025-12-09
+**Last Updated:** 2026-02-07
+**Updated:** Added resolvespec-js client testing and implementation tasks
Author	SHA1	Message	Date
Hein	b322ef76a2	Merge branch 'main' of https://github.com/bitechdev/ResolveSpec	2026-02-10 16:55:58 +02:00
Hein	a6c7edb0e4	feat(resolvespec): add OR logic support in filters * Introduce `logic_operator` field to combine filters with OR logic. * Implement grouping for consecutive OR filters to ensure proper SQL precedence. * Add support for custom SQL operators in filter conditions. * Enhance `fetch_row_number` functionality to return specific record with its position. * Update tests to cover new filter logic and grouping behavior. Features Implemented: 1. OR Logic Filter Support (SearchOr) - Added to resolvespec, restheadspec, and websocketspec - Consecutive OR filters are automatically grouped with parentheses - Prevents SQL logic errors: (A OR B OR C) AND D instead of A OR B OR C AND D 2. CustomOperators - Allows arbitrary SQL conditions in resolvespec - Properly integrated with filter logic 3. FetchRowNumber - Uses SQL window functions: ROW_NUMBER() OVER (ORDER BY ...) - Returns only the specific record (not all records) - Available in resolvespec and restheadspec - Perfect for "What's my rank?" queries 4. RowNumber Field Auto-Population - Now available in all three packages: resolvespec, restheadspec, and websocketspec - Uses simple offset-based math: offset + index + 1 - Automatically populates RowNumber int64 field if it exists on models - Perfect for displaying paginated lists with sequential numbering	2026-02-10 16:55:55 +02:00
Hein	71eeb8315e	chore: 📝 Refactored documentation and added better sqlite support. Some checks failed Build , Vet Test, and Lint / Run Vet Tests (1.24.x) (push) Successful in -26m14s Details Build , Vet Test, and Lint / Run Vet Tests (1.23.x) (push) Successful in -25m40s Details Build , Vet Test, and Lint / Lint Code (push) Successful in -25m41s Details Build , Vet Test, and Lint / Build (push) Successful in -25m55s Details Tests / Unit Tests (push) Successful in -26m19s Details Tests / Integration Tests (push) Failing after -26m35s Details restructure server configuration for multiple instances - Change server configuration to support multiple instances. - Introduce new fields for tracing and error tracking. - Update example configuration to reflect new structure. - Remove deprecated OpenAPI specification file. - Enhance database adapter to handle SQLite schema translation.	2026-02-07 10:58:34 +02:00
Hein	4bf3d0224e	feat(database): ✨ normalize driver names across adapters Some checks failed Build , Vet Test, and Lint / Run Vet Tests (1.23.x) (push) Successful in -25m46s Details Build , Vet Test, and Lint / Run Vet Tests (1.24.x) (push) Successful in -23m31s Details Build , Vet Test, and Lint / Lint Code (push) Successful in -24m55s Details Tests / Unit Tests (push) Successful in -26m19s Details Build , Vet Test, and Lint / Build (push) Successful in -26m2s Details Tests / Integration Tests (push) Failing after -26m42s Details * Added DriverName method to BunAdapter, GormAdapter, and PgSQLAdapter for consistent driver name handling. * Updated transaction adapters to include driver name. * Enhanced mock database implementations for testing with DriverName method. * Adjusted getTableName functions to accommodate driver-specific naming conventions.	2026-02-05 13:28:53 +02:00