🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
import { env } from "../../env";
export enum OperationType {
READ = "READ",
WRITE = "WRITE",
}
export enum Implementation {
STATEFUL = "STATEFUL",
VERSIONED = "VERSIONED",
}
type Implementations<T> = {
[Implementation.STATEFUL]: () => Promise<T>;
[Implementation.VERSIONED]: () => Promise<T>;
};
export async function executeWithDatasetServiceStrategy<T>(
operation: OperationType,
implementations: Implementations<T>,
): Promise<T> {
// Single Write Strategy: WRITE operation
if (operation === OperationType.WRITE) {
// Either write to versioned implementation OR stateful implementation
if (
env.LANGFUSE_DATASET_SERVICE_WRITE_TO_VERSIONED_IMPLEMENTATION === "true"
) {
return await implementations[Implementation.VERSIONED]();
}
return await implementations[Implementation.STATEFUL]();
}
// READ operation - use configured source
if (
env.LANGFUSE_DATASET_SERVICE_READ_FROM_VERSIONED_IMPLEMENTATION === "true"
) {
return implementations[Implementation.VERSIONED]();
}
return implementations[Implementation.STATEFUL]();
}