🪢 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 {
removeObjectKeys,
ScoreDataTypeEnum,
type ScoreDomain,
} from "@langfuse/shared";
export {
_handleGenerateScoresForPublicApi,
_handleGetScoresCountForPublicApi,
type ScoreQueryType,
} from "@langfuse/shared/src/server";
type ScoreApiResult = Omit<ScoreDomain, "longStringValue"> & {
stringValue?: string | null;
};
type TextScoreApiResult = Omit<ScoreDomain, "longStringValue" | "value"> & {
stringValue?: string | null;
};
/**
* Converts a ScoreDomain object to API format.
* For CORRECTION scores, moves longStringValue to stringValue for API compatibility.
* For TEXT scores, removes longStringValue and value (always 0, not meaningful).
* For other score types, removes longStringValue.
*/
export function convertScoreToPublicApi(
score: ScoreDomain & { dataType: "TEXT" },
): TextScoreApiResult;
export function convertScoreToPublicApi(score: ScoreDomain): ScoreApiResult;
export function convertScoreToPublicApi(
score: ScoreDomain,
): ScoreApiResult | TextScoreApiResult {
if (score.dataType === ScoreDataTypeEnum.CORRECTION) {
const { longStringValue, ...rest } = score;
return {
...rest,
stringValue: longStringValue,
};
}
if (score.dataType === ScoreDataTypeEnum.TEXT) {
return removeObjectKeys(score, ["longStringValue", "value"]);
}
return removeObjectKeys(score, ["longStringValue"]);
}