freeplay 0.5.0a2__tar.gz → 0.5.0a4__tar.gz

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: freeplay
-Version: 0.5.0a2
+Version: 0.5.0a4
 Summary: 
 License: MIT
 Author: FreePlay Engineering

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "freeplay"
-version = "0.5.0a2"
+version = "0.5.0a4"
 description = ""
 authors = ["FreePlay Engineering <engineering@freeplay.ai>"]
 license = "MIT"

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/src/freeplay/resources/recordings.py

@@ -21,7 +21,7 @@ from freeplay.resources.prompts import (
     PromptInfo,
 )
 from freeplay.resources.sessions import SessionInfo, TraceInfo
-from freeplay.support import CallSupport
+from freeplay.support import CallSupport, media_inputs_to_json
 from freeplay.utils import convert_provider_message_to_dict
 
 logger = logging.getLogger(__name__)
@@ -109,18 +109,7 @@ class RecordResponse:
     completion_id: str
 
 
-def media_inputs_to_json(media_input: MediaInput) -> Dict[str, Any]:
-    if isinstance(media_input, MediaInputUrl):
-        return {
-            "type": media_input.type,
-            "url": media_input.url
-        }
-    else:
-        return {
-            "type": media_input.type,
-            "data": media_input.data,
-            "content_type": media_input.content_type
-        }
+
 
 class Recordings:
     def __init__(self, call_support: CallSupport):

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/src/freeplay/resources/test_cases.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from typing import List, Optional, Dict, Any
 
-from freeplay.model import InputVariables, NormalizedMessage
+from freeplay.model import InputVariables, NormalizedMessage, MediaInputMap
 from freeplay.support import CallSupport, DatasetTestCaseRequest, DatasetTestCasesRetrievalResponse
 
 
@@ -13,12 +13,14 @@ class DatasetTestCase:
             output: Optional[str],
             history: Optional[List[NormalizedMessage]] = None,
             metadata: Optional[Dict[str, str]] = None,
+            media_inputs: Optional[MediaInputMap] = None,
             id: Optional[str] = None, # Only set on retrieval
     ):
         self.inputs = inputs
         self.output = output
         self.history = history
         self.metadata = metadata
+        self.media_inputs = media_inputs
         self.id = id
 
 
@@ -44,7 +46,7 @@ class TestCases:
         return self.create_many(project_id, dataset_id, [test_case])
 
     def create_many(self, project_id: str, dataset_id: str, test_cases: List[DatasetTestCase]) -> Dataset:
-        dataset_test_cases = [DatasetTestCaseRequest(test_case.history, test_case.inputs, test_case.metadata, test_case.output) for test_case in test_cases]
+        dataset_test_cases = [DatasetTestCaseRequest(test_case.history, test_case.inputs, test_case.metadata, test_case.output, test_case.media_inputs) for test_case in test_cases]
         self.call_support.create_test_cases(project_id, dataset_id, dataset_test_cases)
         return Dataset(dataset_id, test_cases)
 

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/src/freeplay/resources/test_runs.py

@@ -1,5 +1,6 @@
 import warnings
 from dataclasses import dataclass
+from uuid import UUID
 from typing import Any, Dict, List, Optional, Union
 
 from freeplay.model import InputVariables, MediaInputBase64, MediaInputUrl, TestRunInfo
@@ -117,10 +118,11 @@ class TestRuns:
             include_outputs: bool = False,
             name: Optional[str] = None,
             description: Optional[str] = None,
-            flavor_name: Optional[str] = None
+            flavor_name: Optional[str] = None,
+            target_evaluation_ids: Optional[List[UUID]] = None,
     ) -> TestRun:
         test_run = self.call_support.create_test_run(
-            project_id, testlist, include_outputs, name, description, flavor_name)
+            project_id, testlist, include_outputs, name, description, flavor_name, target_evaluation_ids)
         test_cases = [
             CompletionTestCase(
                 test_case_id=test_case.id,

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/src/freeplay/support.py

@@ -1,6 +1,7 @@
 from dataclasses import asdict, dataclass, field
 from json import JSONEncoder
 from typing import Any, Dict, List, Literal, Optional, Union
+from uuid import UUID
 
 from freeplay import api_support
 from freeplay.api_support import try_decode
@@ -11,7 +12,7 @@ from freeplay.model import (
     MediaInputBase64,
     MediaInputUrl,
     NormalizedMessage,
-    TestRunInfo,
+    TestRunInfo, MediaInputMap, MediaInput,
 )
 
 CustomMetadata = Optional[Dict[str, Union[str, int, float, bool]]]
@@ -35,7 +36,6 @@ class ToolSchema:
 
 Role = Literal['system', 'user', 'assistant']
 
-
 MediaType = Literal["image", "audio", "video", "file"]
 
 
@@ -56,6 +56,7 @@ class TemplateChatMessage:
 class HistoryTemplateMessage:
     kind: Literal["history"]
 
+
 TemplateMessage = Union[HistoryTemplateMessage, TemplateChatMessage]
 
 
@@ -94,6 +95,20 @@ class ProjectInfos:
     projects: List[ProjectInfo]
 
 
+def media_inputs_to_json(media_input: MediaInput) -> Dict[str, Any]:
+    if isinstance(media_input, MediaInputUrl):
+        return {
+            "type": media_input.type,
+            "url": media_input.url
+        }
+    else:
+        return {
+            "type": media_input.type,
+            "data": media_input.data,
+            "content_type": media_input.content_type
+        }
+
+
 class PromptTemplateEncoder(JSONEncoder):
     def default(self, prompt_template: PromptTemplate) -> Dict[str, Any]:
         return prompt_template.__dict__
@@ -176,12 +191,19 @@ class TestRunRetrievalResponse:
 
 
 class DatasetTestCaseRequest:
-    def __init__(self, history: Optional[List[NormalizedMessage]], inputs: InputVariables,
-                 metadata: Optional[Dict[str, str]], output: Optional[str]) -> None:
+    def __init__(
+            self,
+            history: Optional[List[NormalizedMessage]],
+            inputs: InputVariables,
+            metadata: Optional[Dict[str, str]],
+            output: Optional[str],
+            media_inputs: Optional[MediaInputMap] = None,
+    ) -> None:
         self.history: Optional[List[NormalizedMessage]] = history
         self.inputs: InputVariables = inputs
         self.metadata: Optional[Dict[str, str]] = metadata
         self.output: Optional[str] = output
+        self.media_inputs = media_inputs
 
 
 class DatasetTestCaseResponse:
@@ -325,7 +347,8 @@ class CallSupport:
             include_outputs: bool = False,
             name: Optional[str] = None,
             description: Optional[str] = None,
-            flavor_name: Optional[str] = None
+            flavor_name: Optional[str] = None,
+            target_evaluation_ids: Optional[List[UUID]] = None
     ) -> TestRunResponse:
         response = api_support.post_raw(
             api_key=self.freeplay_api_key,
@@ -335,7 +358,10 @@ class CallSupport:
                 'include_outputs': include_outputs,
                 'test_run_name': name,
                 'test_run_description': description,
-                'flavor_name': flavor_name
+                'flavor_name': flavor_name,
+                'target_evaluation_ids': [
+                    str(id) for id in target_evaluation_ids
+                ] if target_evaluation_ids is not None else None
             },
         )
 
@@ -403,13 +429,22 @@ class CallSupport:
         if response.status_code != 201:
             raise freeplay_response_error('Error while deleting session.', response)
 
-    def create_test_cases(self, project_id: str, dataset_id: str, test_cases: List[DatasetTestCaseRequest]) -> None:
+    def create_test_cases(
+            self,
+            project_id: str,
+            dataset_id: str,
+            test_cases: List[DatasetTestCaseRequest]
+    ) -> None:
         examples = [
             {
                 "history": test_case.history,
                 "output": test_case.output,
                 "metadata": test_case.metadata,
-                "inputs": test_case.inputs
+                "inputs": test_case.inputs,
+                "media_inputs": {
+                    name: media_inputs_to_json(media_input)
+                    for name, media_input in test_case.media_inputs.items()
+                } if test_case.media_inputs is not None else None
             } for test_case in test_cases]
         payload: Dict[str, Any] = {"examples": examples}
         url = f'{self.api_base}/v2/projects/{project_id}/datasets/id/{dataset_id}/test-cases'

{freeplay-0.5.0a2 → freeplay-0.5.0a4}/src/freeplay/utils.py

@@ -75,8 +75,42 @@ def get_user_agent() -> str:
 # Recursively convert Pydantic models, lists, and dicts to dict compatible format -- used to allow us to accept
 # provider message shapes (usually generated types) or the default {'content': ..., 'role': ...} shape.
 def convert_provider_message_to_dict(obj: Any) -> Any:
+    """
+    Convert provider message objects to dictionaries.
+    For Vertex AI objects, automatically converts to camelCase.
+    """
+    # List of possible raw attribute names in Vertex AI objects
+    vertex_raw_attrs = [
+        '_raw_content',      # For Content objects
+        '_raw_tool',         # For Tool objects  
+        '_raw_message',      # For message objects
+        '_raw_candidate',    # For Candidate objects
+        '_raw_response',     # For response objects
+        '_raw_function_declaration',  # For FunctionDeclaration
+        '_raw_generation_config',     # For GenerationConfig
+        '_pb',              # Generic protobuf attribute
+    ]
+    
+    # Check for Vertex AI objects with raw protobuf attributes
+    for attr_name in vertex_raw_attrs:
+        if hasattr(obj, attr_name):
+            raw_obj = getattr(obj, attr_name)
+            if raw_obj is not None:
+                try:
+                    # Use the metaclass to_dict with camelCase conversion
+                    return type(raw_obj).to_dict(
+                        raw_obj,
+                        preserving_proto_field_name=False,  # camelCase
+                        use_integers_for_enums=False,  # Keep as strings (we'll lowercase them)
+                        including_default_value_fields=False  # Exclude defaults
+                    )
+                except:  # noqa: E722
+                    # If we can't convert, continue to the next attribute
+                    pass
+    
+    # For non-Vertex AI objects, use their standard to_dict methods
     if hasattr(obj, 'to_dict') and callable(getattr(obj, 'to_dict')):
-        # Vertex AI has a to_dict method
+        # Regular to_dict (for Vertex AI wrappers without _raw_* attributes)
         return obj.to_dict()
     elif hasattr(obj, 'model_dump'):
         # Pydantic v2
@@ -85,7 +119,11 @@ def convert_provider_message_to_dict(obj: Any) -> Any:
         # Pydantic v1
         return obj.dict(encode_json=True)
     elif isinstance(obj, dict):
+        # Handle dictionaries recursively
         return {k: convert_provider_message_to_dict(v) for k, v in obj.items()}
     elif isinstance(obj, list):
+        # Handle lists recursively
         return [convert_provider_message_to_dict(item) for item in obj]
+    
+    # Return as-is for primitive types
     return obj