chore: add example for langserve (#264)

python/examples/langserve/per_request_metadata.py

@@ -0,0 +1,42 @@
+"""
+This example demonstrates how to add metadata on a per-request basis. Here we record the
+referer from request headers, but a more common use case to record the user_id as shown in
+the following example:
+https://github.com/langchain-ai/langserve/blob/4bf7b0f4386f9bab0026d28e0003584f654ddbbe/examples/chat_with_persistence_and_user/server.py#L107
+"""  # noqa: E501
+
+from typing import Any, Dict
+
+import uvicorn
+from fastapi import FastAPI, Request
+from langchain_core.runnables import ConfigurableField
+from langchain_openai import ChatOpenAI
+from langserve import add_routes
+from openinference.instrumentation.langchain import LangChainInstrumentor
+from opentelemetry import trace as trace_api
+from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
+from opentelemetry.sdk import trace as trace_sdk
+from opentelemetry.sdk.trace.export import SimpleSpanProcessor
+
+endpoint = "http://127.0.0.1:6006/v1/traces"
+tracer_provider = trace_sdk.TracerProvider()
+trace_api.set_tracer_provider(tracer_provider)
+tracer_provider.add_span_processor(SimpleSpanProcessor(OTLPSpanExporter(endpoint)))
+
+LangChainInstrumentor().instrument()
+
+
+async def per_req_config_modifier(config: Dict[str, Any], request: Request) -> Dict[str, Any]:
+    config["configurable"] = {"metadata": {"referer": request.headers["referer"]}}
+    return config
+
+
+app = FastAPI()
+add_routes(
+    app,
+    ChatOpenAI().configurable_fields(metadata=ConfigurableField(id="metadata")),
+    per_req_config_modifier=per_req_config_modifier,
+    path="/openai",
+)
+if __name__ == "__main__":
+    uvicorn.run(app, host="localhost", port=8000)