fix(datasets): add schema arguments to create_dataset (#1457)

* fix(datasets): add schema arguments to create_dataset

* fix tests

Author: hassiebp

langfuse/_client/client.py

@@ -3250,20 +3250,28 @@ def create_dataset(
         name: str,
         description: Optional[str] = None,
         metadata: Optional[Any] = None,
+        input_schema: Optional[Any] = None,
+        expected_output_schema: Optional[Any] = None,
     ) -> Dataset:
         """Create a dataset with the given name on Langfuse.
 
         Args:
             name: Name of the dataset to create.
             description: Description of the dataset. Defaults to None.
             metadata: Additional metadata. Defaults to None.
+            input_schema: JSON Schema for validating dataset item inputs. When set, all new items will be validated against this schema.
+            expected_output_schema: JSON Schema for validating dataset item expected outputs. When set, all new items will be validated against this schema.
 
         Returns:
             Dataset: The created dataset as returned by the Langfuse API.
         """
         try:
             body = CreateDatasetRequest(
-                name=name, description=description, metadata=metadata
+                name=name,
+                description=description,
+                metadata=metadata,
+                inputSchema=input_schema,
+                expectedOutputSchema=expected_output_schema,
             )
             langfuse_logger.debug(f"Creating datasets {body}")
 

tests/test_datasets.py

@@ -146,7 +146,7 @@ def test_dataset_run_with_metadata_and_description():
     dataset_name = create_uuid()
     langfuse.create_dataset(name=dataset_name)
 
-    input = json.dumps({"input": "Hello World"})
+    input = {"input": "Hello World"}
     langfuse.create_dataset_item(dataset_name=dataset_name, input=input)
 
     dataset = langfuse.get_dataset(dataset_name)
@@ -187,7 +187,7 @@ def test_get_dataset_runs():
     dataset_name = create_uuid()
     langfuse.create_dataset(name=dataset_name)
 
-    input = json.dumps({"input": "Hello World"})
+    input = {"input": "Hello World"}
     langfuse.create_dataset_item(dataset_name=dataset_name, input=input)
 
     dataset = langfuse.get_dataset(dataset_name)