fix(sdk): remove API key

.github/actions/js-integration-tests/action.yml

@@ -7,6 +7,9 @@ inputs:
   langchain-api-key:
     description: "Langchain"
     required: true
+  langchain-endpoint:
+    description: "LangSmith Endpoint"
+    required: true
   openai-api-key:
     description: "OpenAI API key"
     required: false
@@ -34,5 +37,6 @@ runs:
       working-directory: js
       env:
         LANGCHAIN_TRACING_V2: "true"
+        LANGCHAIN_ENDPOINT: ${{ inputs.langchain-endpoint }}
         LANGCHAIN_API_KEY: ${{ inputs.langchain-api-key }}
         OPENAI_API_KEY: ${{ inputs.openai-api-key }}

.github/actions/python-integration-tests/action.yml

@@ -7,6 +7,9 @@ inputs:
   langchain-api-key:
     description: "Langchain"
     required: true
+  langchain-endpoint:
+    description: "LangSmith Endpoint"
+    required: true
   openai-api-key:
     description: "OpenAI API key"
     required: false
@@ -40,6 +43,7 @@ runs:
     - name: Run integration tests
       env:
         LANGCHAIN_TRACING_V2: "true"
+        LANGCHAIN_ENDPOINT: ${{ inputs.langchain-endpoint }}
         LANGCHAIN_API_KEY: ${{ inputs.langchain-api-key }}
         OPENAI_API_KEY: ${{ inputs.openai-api-key }}
       run: make integration_tests_fast
@@ -49,6 +53,7 @@ runs:
     - name: Run doctest
       env:
         LANGCHAIN_TRACING_V2: "true"
+        LANGCHAIN_ENDPOINT: ${{ inputs.langchain-endpoint }}
         LANGCHAIN_API_KEY: ${{ inputs.langchain-api-key }}
         OPENAI_API_KEY: ${{ inputs.openai-api-key }}
         ANTHROPIC_API_KEY: ${{ inputs.anthropic-api-key }}
@@ -60,6 +65,7 @@ runs:
     - name: Run Evaluation
       env:
           LANGCHAIN_TRACING_V2: "true"
+          LANGCHAIN_ENDPOINT: ${{ inputs.langchain-endpoint }}
           LANGCHAIN_API_KEY: ${{ inputs.langchain-api-key }}
           OPENAI_API_KEY: ${{ inputs.openai-api-key }}
           ANTHROPIC_API_KEY: ${{ inputs.anthropic-api-key }}

.github/workflows/integration_tests.yml

@@ -47,6 +47,7 @@ jobs:
         uses: ./.github/actions/python-integration-tests
         with:
           python-version: 3.11
+          langchain-endpoint: https://api.smith.langchain.com
           langchain-api-key: ${{ secrets.LANGSMITH_API_KEY }}
           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
           anthropic-api-key: ${{ secrets.ANTHROPIC_API_KEY }}
@@ -75,5 +76,6 @@ jobs:
         uses: ./.github/actions/js-integration-tests
         with:
           node-version: 20.x
+          langchain-endpoint: https://api.smith.langchain.com
           langchain-api-key: ${{ secrets.LANGSMITH_API_KEY }}
           openai-api-key: ${{ secrets.OPENAI_API_KEY }}

python/langsmith/evaluation/_runner.py

@@ -579,6 +579,7 @@ def evaluate_comparative(
         ...     evaluators=[score_preferences],
         ...     client=client,
         ... )  # doctest: +ELLIPSIS
+        View the pairwise evaluation results at:...
     """  # noqa: E501
     if len(experiments) < 2:
         raise ValueError("Comparative evaluation requires at least 2 experiments.")

python/tests/integration_tests/test_client.py

@@ -35,10 +35,7 @@ def wait_for(
 
 
 @pytest.fixture
-def langchain_client(monkeypatch: pytest.MonkeyPatch) -> Client:
-    # monkeypatch.setenv("LANGCHAIN_ENDPOINT", "https://api.smith.langchain.com")
-    monkeypatch.setenv("LANGCHAIN_ENDPOINT", "https://dev.api.smith.langchain.com")
-    monkeypatch.setenv("LANGCHAIN_API_KEY", "ls__29e47afc8ad24323aef27444617fb8db")
+def langchain_client() -> Client:
     return Client()
 
 
@@ -217,7 +214,6 @@ def test_create_dataset(
 ) -> None:
     """Test persisting runs and adding feedback."""
     monkeypatch.setenv("LANGCHAIN_ENDPOINT", "https://dev.api.smith.langchain.com")
-    monkeypatch.setenv("LANGCHAIN_API_KEY", "ls__29e47afc8ad24323aef27444617fb8db")
     dataset_name = "__test_create_dataset" + uuid4().hex[:4]
     if langchain_client.has_dataset(dataset_name=dataset_name):
         langchain_client.delete_dataset(dataset_name=dataset_name)

python/tests/integration_tests/test_runs.py

@@ -1,11 +1,10 @@
 import asyncio
-import os
 import time
 from collections import defaultdict
 from concurrent.futures import ThreadPoolExecutor
 from typing import AsyncGenerator, Generator, Optional
 
-import pytest
+import pytest  # type: ignore
 
 from langsmith import utils as ls_utils
 from langsmith.client import Client
@@ -15,13 +14,7 @@
 
 @pytest.fixture
 def langchain_client() -> Generator[Client, None, None]:
-    original = os.environ.get("LANGCHAIN_ENDPOINT")
-    os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"
     yield Client()
-    if original is None:
-        os.environ.pop("LANGCHAIN_ENDPOINT")
-    else:
-        os.environ["LANGCHAIN_ENDPOINT"] = original
 
 
 def poll_runs_until_count(