Skip to content

SCHEMA_VALIDATION.md

Latest commit

 

History

History
230 lines (171 loc) · 5.25 KB

SCHEMA_VALIDATION.md

File metadata and controls

230 lines (171 loc) · 5.25 KB

EPR MCP Server Input Validation Schemas

This document describes the input validation schemas implemented for the EPR MCP Server tools.

Overview

The schema validation system uses Pydantic models to ensure that all input data to the MCP tools is properly validated before processing. This helps prevent errors, improves security, and provides clear error messages when invalid data is provided.

Schema Components

Base Models

  • BaseModel: All schemas inherit from Pydantic's BaseModel
  • Field: Used for field definitions with validation rules and descriptions
  • validator: Custom validation decorators for complex field validation

Search Schemas

EventSearchInput

Validates input for searching events:

{
    "name": "optional string",
    "version": "optional string",
    "release": "optional string",
    "platform_id": "optional string",
    "package": "optional string",
    "description": "optional string",
    "success": "optional boolean",
    "event_receiver_id": "optional string"
}

EventReceiverSearchInput

Validates input for searching event receivers:

{
    "name": "optional string",
    "type": "optional string",
    "version": "optional string",
    "description": "optional string"
}

EventReceiverGroupSearchInput

Validates input for searching event receiver groups:

{
    "name": "optional string",
    "type": "optional string",
    "version": "optional string",
    "description": "optional string"
}

Creation Schemas

EventCreateInput

Validates input for creating events (all fields required):

{
    "name": "string (min_length=1)",
    "version": "string (min_length=1)",
    "release": "string (min_length=1)",
    "platform_id": "string (min_length=1)",
    "package": "string (min_length=1)",
    "description": "string (min_length=1)",
    "event_receiver_id": "string (min_length=1)",
    "success": "boolean",
    "payload": "dict (required, must be dictionary)"
}

EventReceiverCreateInput

Validates input for creating event receivers (all fields required):

{
    "name": "string (min_length=1)",
    "type": "string (min_length=1)",
    "version": "string (min_length=1)",
    "description": "string (min_length=1)"
}

EventReceiverGroupCreateInput

Validates input for creating event receiver groups (all fields required):

{
    "name": "string (min_length=1)",
    "type": "string (min_length=1)",
    "version": "string (min_length=1)",
    "description": "string (min_length=1)",
    "event_receiver_ids": "list of strings (min_items=1, all non-empty)"
}

Fetch Schema

FetchInput

Validates input for fetch operations:

{
    "id": "string (min_length=1, trimmed)"
}

Data Wrappers

SearchDataWrapper

Wraps search inputs in a data field:

{
    "data": EventSearchInput | EventReceiverSearchInput | EventReceiverGroupSearchInput
}

CreateDataWrapper

Wraps creation inputs in a data field:

{
    "data": EventCreateInput | EventReceiverCreateInput | EventReceiverGroupCreateInput
}

Usage

Validation Function

The main validation function is validate_input(operation, input_data):

from schemas import validate_input
from pydantic import ValidationError

try:
    validated_data = validate_input("search_events", {
        "data": {
            "name": "foo",
            "version": "1.0.0"
        }
    })
    # Use validated_data...
except (ValidationError, ValueError) as e:
    # Handle validation error
    print(f"Validation error: {e}")

Operation Mapping

The system supports these operations:

  • search_events → SearchDataWrapper
  • search_receivers → SearchDataWrapper
  • search_groups → SearchDataWrapper
  • create_event → CreateDataWrapper
  • create_receiver → CreateDataWrapper
  • create_group → CreateDataWrapper
  • fetch_event → FetchInput
  • fetch_receiver → FetchInput
  • fetch_group → FetchInput

Integration with MCP Tools

Each MCP tool now includes validation:

@mcp.tool(title="Search Events", description="Search for events in EPR")
async def search_events(data: dict) -> str:
    try:
        validated_data = validate_input("search_events", data)
    except (ValidationError, ValueError) as e:
        return f"Validation error: {str(e)}"

    # Process validated_data...

Benefits

  1. Type Safety: Ensures correct data types for all fields
  2. Required Field Validation: Prevents missing required fields
  3. Format Validation: Validates string lengths, list contents, etc.
  4. Clear Error Messages: Pydantic provides detailed validation errors
  5. Documentation: Schema serves as documentation for expected input formats
  6. Security: Prevents injection attacks and malformed data

Testing

Run the validation tests:

cd /home/bsmith/go/src/github.com/xbcsmith/epr-workshop/11-epr-mcp-server/src
python test_schemas.py

Error Handling

Validation errors are caught and returned as user-friendly error messages:

  • Missing required fields
  • Invalid data types
  • Empty strings where content is required
  • Invalid list contents
  • Malformed dictionaries

Dependencies

  • pydantic>=2.0.0: Core validation library
  • typing: Type hints for Python < 3.9 compatibility