Merge pull request #1002 from guardrails-ai/dtam/add_support_for_outp…

…ut_format_json_schema

add experimental helper to get us a much cleaner interface to openai …

guardrails/decorators/experimental.py

@@ -0,0 +1,15 @@
+import functools
+from guardrails.logger import logger
+
+
+def experimental(func):
+    """Decorator to mark a function as experimental."""
+
+    @functools.wraps(func)
+    def wrapper(*args, **kwargs):
+        logger.warn(
+            f"The function '{func.__name__}' is experimental and subject to change."
+        )
+        return func(*args, **kwargs)
+
+    return wrapper

guardrails/guard.py

@@ -86,10 +86,13 @@
     ValidatorMap,
 )
 
-from guardrails.utils.tools_utils import (
+from guardrails.utils.structured_data_utils import (
     # Prevent duplicate declaration in the docs
     json_function_calling_tool as json_function_calling_tool_util,
+    output_format_json_schema as output_format_json_schema,
 )
+from guardrails.decorators.experimental import experimental
+
 from guardrails.settings import settings
 
 
@@ -1331,6 +1334,10 @@ def to_dict(self) -> Dict[str, Any]:
 
         return i_guard.to_dict()
 
+    @experimental
+    def response_format_json_schema(self) -> Dict[str, Any]:
+        return output_format_json_schema(schema=self._base_model)  # type: ignore
+
     def json_function_calling_tool(
         self,
         tools: Optional[list] = None,

guardrails/utils/structured_data_utils.py

@@ -0,0 +1,75 @@
+from typing import List, Optional
+from guardrails.logger import logger
+from guardrails.classes.schema.processed_schema import ProcessedSchema
+from guardrails.types.pydantic import ModelOrListOfModels
+
+
+# takes processed schema and converts it to a openai tool object
+def schema_to_tool(schema) -> dict:
+    tool = {
+        "type": "function",
+        "function": {
+            "name": "gd_response_tool",
+            "description": "A tool for generating responses to guardrails."
+            " It must be called last in every response.",
+            "parameters": schema,
+            "required": schema["required"] or [],
+        },
+    }
+    return tool
+
+
+def set_additional_properties_false_iteratively(schema):
+    stack = [schema]
+    while stack:
+        current = stack.pop()
+        if isinstance(current, dict):
+            if "properties" in current:
+                current["required"] = list(
+                    current["properties"].keys()
+                )  # this has to be set
+            if "maximum" in current:
+                logger.warn("Property maximum is not supported." " Dropping")
+                current.pop("maximum")  # the api does not like these set
+            if "minimum" in current:
+                logger.warn("Property maximum is not supported." " Dropping")
+                current.pop("minimum")  # the api does not like these set
+            if "default" in current:
+                logger.warn("Property default is not supported. Marking field Required")
+                current.pop("default")  # the api does not like these set
+            for prop in current.values():
+                stack.append(prop)
+        elif isinstance(current, list):
+            for prop in current:
+                stack.append(prop)
+        if (
+            isinstance(current, dict)
+            and "additionalProperties" not in current
+            and "type" in current
+            and current["type"] == "object"
+        ):
+            current["additionalProperties"] = False  # the api needs these set
+
+
+def json_function_calling_tool(
+    schema: ProcessedSchema,
+    tools: Optional[List] = None,
+) -> List:
+    tools = tools or []
+    tools.append(schema_to_tool(schema))  # type: ignore
+    return tools
+
+
+def output_format_json_schema(schema: ModelOrListOfModels) -> dict:
+    parsed_schema = schema.model_json_schema()  # type: ignore
+
+    set_additional_properties_false_iteratively(parsed_schema)
+
+    return {
+        "type": "json_schema",
+        "json_schema": {
+            "name": parsed_schema["title"],
+            "schema": parsed_schema,
+            "strict": True,
+        },  # type: ignore
+    }

tests/unit_tests/utils/test_tools_utils.py → tests/unit_tests/utils/test_structured_data_utils.py

@@ -3,7 +3,11 @@
 
 from guardrails.schema.pydantic_schema import pydantic_model_to_schema
 
-from guardrails.utils.tools_utils import json_function_calling_tool, schema_to_tool
+from guardrails.utils.structured_data_utils import (
+    json_function_calling_tool,
+    schema_to_tool,
+    output_format_json_schema,
+)
 
 
 class Delivery(BaseModel):
@@ -141,3 +145,89 @@ def test_json_function_calling_tool():
             },
         }
     ]
+
+
+def test_output_format_json_schema():
+    schema = output_format_json_schema(Schedule)
+    assert schema == {
+        "type": "json_schema",
+        "json_schema": {
+            "name": "Schedule",
+            "schema": {
+                "additionalProperties": False,
+                "$defs": {
+                    "Delivery": {
+                        "additionalProperties": False,
+                        "properties": {
+                            "customer": {
+                                "description": "customer name",
+                                "title": "Customer",
+                                "type": "string",
+                            },
+                            "pickup_time": {
+                                "description": "date and time of pickup",
+                                "title": "Pickup Time",
+                                "type": "string",
+                            },
+                            "pickup_location": {
+                                "description": "address of pickup",
+                                "title": "Pickup Location",
+                                "type": "string",
+                            },
+                            "dropoff_time": {
+                                "description": "date and time of dropoff",
+                                "title": "Dropoff Time",
+                                "type": "string",
+                            },
+                            "dropoff_location": {
+                                "description": "address of dropoff",
+                                "title": "Dropoff Location",
+                                "type": "string",
+                            },
+                            "price": {
+                                "description": "price of delivery with"
+                                " currency symbol included",
+                                "title": "Price",
+                                "type": "string",
+                            },
+                            "items": {
+                                "description": "items for pickup/delivery typically"
+                                " something a single person can carry on a bike",
+                                "title": "Items",
+                                "type": "string",
+                            },
+                            "number_items": {
+                                "description": "number of items",
+                                "title": "Number Items",
+                                "type": "integer",
+                            },
+                        },
+                        "required": [
+                            "customer",
+                            "pickup_time",
+                            "pickup_location",
+                            "dropoff_time",
+                            "dropoff_location",
+                            "price",
+                            "items",
+                            "number_items",
+                        ],
+                        "title": "Delivery",
+                        "type": "object",
+                    }
+                },
+                "properties": {
+                    "deliveries": {
+                        "description": "deliveries for messenger",
+                        "items": {"$ref": "#/$defs/Delivery"},
+                        "title": "Deliveries",
+                        "type": "array",
+                    }
+                },
+                "required": ["deliveries"],
+                "title": "Schedule",
+                "type": "object",
+            },
+            "strict": True,
+        },
+    }