weave_client

docs/scripts/generate_python_sdk_docs.py

@@ -263,8 +263,8 @@ def doc_module(module, root_path="./docs/reference/python-sdk", module_root_path
 def main():
     import weave
     from weave import feedback
-    from weave import weave_client as client
     from weave.trace import util
+    from weave.trace import weave_client as client
     from weave.trace_server import (
         remote_http_trace_server,
         trace_server_interface,

weave/client_context/weave_client.py

@@ -5,7 +5,7 @@
 from weave.legacy import context_state
 
 if TYPE_CHECKING:
-    from weave.weave_client import WeaveClient
+    from weave.trace.weave_client import WeaveClient
 
 _global_weave_client: Optional["WeaveClient"] = None
 lock = threading.Lock()

weave/conftest.py

@@ -22,9 +22,7 @@
     remote_http_trace_server,
     sqlite_trace_server,
 )
-from weave.trace_server import (
-    trace_server_interface as tsi,
-)
+from weave.trace_server import trace_server_interface as tsi
 
 from . import environment, logs
 from .tests import fixture_fakewandb

weave/flow/eval.py

@@ -24,7 +24,7 @@
 from weave.trace.errors import OpCallError
 from weave.trace.op import Op
 from weave.trace.vals import WeaveObject
-from weave.weave_client import get_ref
+from weave.trace.weave_client import get_ref
 
 console = Console()
 

weave/flow/obj.py

@@ -11,7 +11,7 @@
 
 from weave.trace.op import ObjectRef, Op, call
 from weave.trace.vals import WeaveObject, pydantic_getattribute
-from weave.weave_client import get_ref
+from weave.trace.weave_client import get_ref
 
 
 class Object(BaseModel):

weave/integrations/anthropic/anthropic_test.py

@@ -28,7 +28,7 @@ def _get_call_output(call: tsi.CallSchema) -> Any:
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 def test_anthropic(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("ANTHROPIC_API_KEY", "DUMMY_API_KEY")
     anthropic_client = Anthropic(
@@ -67,7 +67,7 @@ def test_anthropic(
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 def test_anthropic_stream(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("ANTHROPIC_API_KEY", "DUMMY_API_KEY")
     anthropic_client = Anthropic(
@@ -115,7 +115,7 @@ def test_anthropic_stream(
 )
 @pytest.mark.asyncio
 async def test_async_anthropic(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     anthropic_client = AsyncAnthropic(
         # This is the default and can be omitted
@@ -156,7 +156,7 @@ async def test_async_anthropic(
 )
 @pytest.mark.asyncio
 async def test_async_anthropic_stream(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     anthropic_client = AsyncAnthropic(
         # This is the default and can be omitted
@@ -204,7 +204,7 @@ async def test_async_anthropic_stream(
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 def test_tools_calling(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("ANTHROPIC_API_KEY", "DUMMY_API_KEY")
     anthropic_client = Anthropic(

weave/integrations/cohere/cohere_test.py

@@ -27,7 +27,7 @@ def op_name_from_ref(ref: str) -> str:
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 def test_cohere(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("COHERE_API_KEY", "DUMMY_API_KEY")
 
@@ -79,7 +79,7 @@ def test_cohere(
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 def test_cohere_stream(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("COHERE_API_KEY", "DUMMY_API_KEY")
     cohere_client = cohere.Client(api_key=api_key)
@@ -137,7 +137,7 @@ def test_cohere_stream(
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 async def test_cohere_async(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("COHERE_API_KEY", "DUMMY_API_KEY")
 
@@ -190,7 +190,7 @@ async def test_cohere_async(
     allowed_hosts=["api.wandb.ai", "localhost"],
 )
 async def test_cohere_async_stream(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     api_key = os.environ.get("COHERE_API_KEY", "DUMMY_API_KEY")
     cohere_client = cohere.AsyncClient(api_key=api_key)

weave/integrations/dspy/dspy_test.py

@@ -4,8 +4,8 @@
 import pytest
 
 import weave
+from weave.trace.weave_client import WeaveClient
 from weave.trace_server import trace_server_interface as tsi
-from weave.weave_client import WeaveClient
 
 
 def _get_call_output(call: tsi.CallSchema) -> Any:

weave/integrations/groq/groq_test.py

@@ -44,7 +44,7 @@ def op_name_from_ref(ref: str) -> str:
     allowed_hosts=["api.wandb.ai", "localhost", "trace.wandb.ai"],
 )
 def test_groq_quickstart(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     from groq import Groq
 
@@ -97,7 +97,7 @@ def test_groq_quickstart(
     allowed_hosts=["api.wandb.ai", "localhost", "trace.wandb.ai"],
 )
 def test_groq_async_chat_completion(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     from groq import AsyncGroq
 
@@ -165,7 +165,7 @@ async def complete_chat() -> None:
     allowed_hosts=["api.wandb.ai", "localhost", "trace.wandb.ai"],
 )
 def test_groq_streaming_chat_completion(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     from groq import Groq
 
@@ -248,7 +248,7 @@ def test_groq_streaming_chat_completion(
     allowed_hosts=["api.wandb.ai", "localhost", "trace.wandb.ai"],
 )
 def test_groq_async_streaming_chat_completion(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     from groq import AsyncGroq
 
@@ -327,7 +327,7 @@ async def generate_reponse() -> str:
     allowed_hosts=["api.wandb.ai", "localhost", "trace.wandb.ai"],
 )
 def test_groq_tool_call(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     import json
 

weave/integrations/langchain/langchain.py

@@ -42,7 +42,7 @@
 )
 from weave.trace import call_context
 from weave.trace.patcher import Patcher
-from weave.weave_client import Call
+from weave.trace.weave_client import Call
 
 import_failed = False
 

weave/integrations/langchain/langchain_test.py

@@ -4,8 +4,8 @@
 import pytest
 
 import weave
+from weave.trace.weave_client import WeaveClient
 from weave.trace_server import trace_server_interface as tsi
-from weave.weave_client import WeaveClient
 
 
 def filter_body(r: Any) -> Any:

weave/integrations/litellm/litellm_test.py

@@ -60,7 +60,7 @@ def patch_litellm(request: Any) -> Generator[None, None, None]:
     filter_headers=["authorization"], allowed_hosts=["api.wandb.ai", "localhost"]
 )
 def test_litellm_quickstart(
-    client: weave.weave_client.WeaveClient, patch_litellm: None
+    client: weave.trace.weave_client.WeaveClient, patch_litellm: None
 ) -> None:
     # This is taken directly from https://docs.litellm.ai/docs/
     chat_response = litellm.completion(
@@ -101,7 +101,7 @@ def test_litellm_quickstart(
 )
 @pytest.mark.asyncio
 async def test_litellm_quickstart_async(
-    client: weave.weave_client.WeaveClient, patch_litellm: None
+    client: weave.trace.weave_client.WeaveClient, patch_litellm: None
 ) -> None:
     # This is taken directly from https://docs.litellm.ai/docs/
     chat_response = await litellm.acompletion(
@@ -144,7 +144,7 @@ async def test_litellm_quickstart_async(
     filter_headers=["authorization"], allowed_hosts=["api.wandb.ai", "localhost"]
 )
 def test_litellm_quickstart_stream(
-    client: weave.weave_client.WeaveClient, patch_litellm: None
+    client: weave.trace.weave_client.WeaveClient, patch_litellm: None
 ) -> None:
     # This is taken directly from https://docs.litellm.ai/docs/
     chat_response = litellm.completion(
@@ -187,7 +187,7 @@ def test_litellm_quickstart_stream(
 )
 @pytest.mark.asyncio
 async def test_litellm_quickstart_stream_async(
-    client: weave.weave_client.WeaveClient, patch_litellm: None
+    client: weave.trace.weave_client.WeaveClient, patch_litellm: None
 ) -> None:
     # This is taken directly from https://docs.litellm.ai/docs/
     chat_response = await litellm.acompletion(
@@ -231,7 +231,7 @@ async def test_litellm_quickstart_stream_async(
 )
 @pytest.mark.asyncio
 def test_model_predict(
-    client: weave.weave_client.WeaveClient, patch_litellm: None
+    client: weave.trace.weave_client.WeaveClient, patch_litellm: None
 ) -> None:
     class TranslatorModel(weave.Model):
         model: str

weave/integrations/llamaindex/llamaindex.py

@@ -1,6 +1,6 @@
 from weave.client_context import weave_client as weave_client_context
 from weave.trace.patcher import Patcher
-from weave.weave_client import Call
+from weave.trace.weave_client import Call
 
 TRANSFORM_EMBEDDINGS = False
 ALLOWED_ROOT_EVENT_TYPES = ("query",)

weave/integrations/llamaindex/llamaindex_test.py

@@ -81,7 +81,7 @@ def fake_api_key() -> Generator[None, None, None]:
     before_record_request=filter_body,
 )
 def test_llamaindex_quickstart(
-    client: weave.weave_client.WeaveClient, fake_api_key: None
+    client: weave.trace.weave_client.WeaveClient, fake_api_key: None
 ) -> None:
     # This is taken directly from  https://docs.llamaindex.ai/en/stable/getting_started/starter_example/
     from llama_index.core import SimpleDirectoryReader, VectorStoreIndex
@@ -105,7 +105,7 @@ def test_llamaindex_quickstart(
 )
 @pytest.mark.asyncio
 async def test_llamaindex_quickstart_async(
-    client: weave.weave_client.WeaveClient, fake_api_key: None
+    client: weave.trace.weave_client.WeaveClient, fake_api_key: None
 ) -> None:
     # This is taken directly from  https://docs.llamaindex.ai/en/stable/getting_started/starter_example/
     from llama_index.core import SimpleDirectoryReader, VectorStoreIndex

weave/integrations/mistral/mistral_test.py

@@ -24,7 +24,7 @@ def _get_call_output(call: tsi.CallSchema) -> Any:
 @pytest.mark.vcr(
     filter_headers=["authorization"], allowed_hosts=["api.wandb.ai", "localhost"]
 )
-def test_mistral_quickstart(client: weave.weave_client.WeaveClient) -> None:
+def test_mistral_quickstart(client: weave.trace.weave_client.WeaveClient) -> None:
     # This is taken directly from https://docs.mistral.ai/getting-started/quickstart/
     api_key = os.environ.get("MISTRAL_API_KEY", "DUMMY_API_KEY")
     model = "mistral-large-latest"
@@ -77,7 +77,9 @@ def test_mistral_quickstart(client: weave.weave_client.WeaveClient) -> None:
     filter_headers=["authorization"], allowed_hosts=["api.wandb.ai", "localhost"]
 )
 @pytest.mark.asyncio
-async def test_mistral_quickstart_async(client: weave.weave_client.WeaveClient) -> None:
+async def test_mistral_quickstart_async(
+    client: weave.trace.weave_client.WeaveClient,
+) -> None:
     # This is taken directly from https://docs.mistral.ai/getting-started/quickstart/
     api_key = os.environ.get("MISTRAL_API_KEY", "DUMMY_API_KEY")
     model = "mistral-large-latest"
@@ -125,7 +127,9 @@ async def test_mistral_quickstart_async(client: weave.weave_client.WeaveClient)
 @pytest.mark.vcr(
     filter_headers=["authorization"], allowed_hosts=["api.wandb.ai", "localhost"]
 )
-def test_mistral_quickstart_with_stream(client: weave.weave_client.WeaveClient) -> None:
+def test_mistral_quickstart_with_stream(
+    client: weave.trace.weave_client.WeaveClient,
+) -> None:
     # This is taken directly from https://docs.mistral.ai/getting-started/quickstart/
     api_key = os.environ.get("MISTRAL_API_KEY", "DUMMY_API_KEY")
     model = "mistral-large-latest"
@@ -182,7 +186,7 @@ def test_mistral_quickstart_with_stream(client: weave.weave_client.WeaveClient)
 )
 @pytest.mark.asyncio
 async def test_mistral_quickstart_with_stream_async(
-    client: weave.weave_client.WeaveClient,
+    client: weave.trace.weave_client.WeaveClient,
 ) -> None:
     # This is taken directly from https://docs.mistral.ai/getting-started/quickstart/
     api_key = os.environ.get("MISTRAL_API_KEY", "DUMMY_API_KEY")