Refactor the single-run HTTP contract
This commit is contained in:
@@ -106,38 +106,36 @@ func (h *Handler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
|
||||
model = executionTargetOverrideFromModelOverrideDTO(req.Model)
|
||||
}
|
||||
|
||||
runContext := r.Context()
|
||||
var operationTimeout time.Duration
|
||||
if req.RunTimeoutSeconds != nil && *req.RunTimeoutSeconds > 0 {
|
||||
var cancel context.CancelFunc
|
||||
runContext, cancel = context.WithTimeout(runContext, time.Duration(*req.RunTimeoutSeconds)*time.Second)
|
||||
defer cancel()
|
||||
operationTimeout = time.Duration(*req.RunTimeoutSeconds) * time.Second
|
||||
}
|
||||
mappedRequest := promptkit.RunRequest{
|
||||
PromptID: req.PromptID,
|
||||
PromptVersion: req.PromptVersion,
|
||||
ProfileID: req.ProfileID,
|
||||
SessionID: req.SessionID,
|
||||
Inputs: mappedInputs,
|
||||
Vars: req.Vars,
|
||||
Execution: model,
|
||||
TransientRetries: req.TransientRetries,
|
||||
OperationTimeout: operationTimeout,
|
||||
}
|
||||
var collector *executionreport.Collector
|
||||
if req.IncludeExecutionReport {
|
||||
collector = executionreport.NewCollector()
|
||||
mappedRequest.Observer = collector.Observe
|
||||
}
|
||||
res, err := h.runner.Run(runContext, mappedRequest)
|
||||
res, err := h.runner.Run(r.Context(), mappedRequest)
|
||||
var execution *executionreport.Execution
|
||||
if collector != nil {
|
||||
execution = collector.Snapshot()
|
||||
}
|
||||
if err != nil {
|
||||
switch runContext.Err() {
|
||||
case context.DeadlineExceeded:
|
||||
switch {
|
||||
case errors.Is(err, promptkit.ErrOperationTimeout), errors.Is(err, context.DeadlineExceeded) && errors.Is(r.Context().Err(), context.DeadlineExceeded):
|
||||
writeRunFailure(w, http.StatusGatewayTimeout, "deadline_exceeded", "run execution deadline exceeded", execution, h.options.MaxResponseBytes)
|
||||
return
|
||||
case context.Canceled:
|
||||
case errors.Is(err, context.Canceled) && errors.Is(r.Context().Err(), context.Canceled):
|
||||
writeRunFailure(w, http.StatusRequestTimeout, "request_canceled", "run request canceled", execution, h.options.MaxResponseBytes)
|
||||
return
|
||||
}
|
||||
@@ -159,14 +157,11 @@ func (h *Handler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
|
||||
Metadata: metadataDTO{
|
||||
RunID: res.RunID,
|
||||
PromptID: res.PromptID,
|
||||
PromptVersion: res.PromptVersion,
|
||||
PromptHash: res.PromptHash,
|
||||
RenderedPromptHash: res.RenderedPromptHash,
|
||||
SelectedProfileID: res.SelectedProfileID,
|
||||
SelectedBackendID: res.SelectedBackendID,
|
||||
SessionID: res.SessionID,
|
||||
ModelName: res.ModelName,
|
||||
Endpoint: res.Endpoint,
|
||||
ModelParams: modelParamsDTOFromExecutionTarget(res.EffectiveModelParams),
|
||||
InputHashes: res.InputHashes,
|
||||
Usage: tokenUsageDTO{
|
||||
@@ -176,12 +171,9 @@ func (h *Handler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
|
||||
CachedTokens: res.Usage.CachedTokens,
|
||||
CacheWriteTokens: res.Usage.CacheWriteTokens,
|
||||
},
|
||||
StartTime: res.StartTime,
|
||||
EndTime: res.EndTime,
|
||||
DurationMS: res.Duration.Milliseconds(),
|
||||
ValidationMode: string(res.Validation.Mode),
|
||||
ValidationStatus: string(res.Validation.Status),
|
||||
RepairAttemptsUsed: res.Validation.RepairAttempts,
|
||||
StartTime: res.StartTime,
|
||||
EndTime: res.EndTime,
|
||||
DurationMS: res.Duration.Milliseconds(),
|
||||
},
|
||||
Execution: execution,
|
||||
}
|
||||
@@ -221,7 +213,6 @@ func executionTargetOverrideFromModelOverrideDTO(dto *modelOverrideRequestDTO) *
|
||||
func modelParamsDTOFromExecutionTarget(target promptkit.ExecutionTarget) modelParamsDTO {
|
||||
return modelParamsDTO{
|
||||
Endpoint: target.Endpoint,
|
||||
BackendID: target.BackendID,
|
||||
Model: target.Model,
|
||||
Temperature: target.Temperature,
|
||||
MaxTokens: target.MaxTokens,
|
||||
@@ -255,16 +246,16 @@ func mapRunError(err error) (int, string, string) {
|
||||
return http.StatusBadRequest, "profile_required", "profile_id is required when prompt default_profile is not set"
|
||||
case errors.Is(err, promptkit.ErrAPIKeyEnvMissing):
|
||||
return http.StatusBadRequest, "api_key_env_missing", "api_key_env is set but the environment variable is missing"
|
||||
case errors.Is(err, ErrFileNotAllowed), errors.Is(err, ErrFileOutsideRoot):
|
||||
return http.StatusBadRequest, "artifact_not_allowed", "file input artifact is not allowed"
|
||||
case errors.Is(err, ErrFileTooLarge):
|
||||
return http.StatusRequestEntityTooLarge, "artifact_too_large", "file input artifact is too large"
|
||||
case errors.Is(err, promptkit.ErrPromptLoad):
|
||||
return http.StatusBadRequest, "prompt_load_failed", "failed to load prompt definition"
|
||||
case errors.Is(err, promptkit.ErrProfileLoad):
|
||||
return http.StatusBadRequest, "profile_load_failed", "failed to load execution profile"
|
||||
case errors.Is(err, promptkit.ErrInvalidRequest):
|
||||
return http.StatusBadRequest, "invalid_request", "invalid run request"
|
||||
case errors.Is(err, ErrFileNotAllowed), errors.Is(err, ErrFileOutsideRoot):
|
||||
return http.StatusBadRequest, "artifact_not_allowed", "file input artifact is not allowed"
|
||||
case errors.Is(err, ErrFileTooLarge):
|
||||
return http.StatusRequestEntityTooLarge, "artifact_too_large", "file input artifact is too large"
|
||||
case errors.Is(err, promptkit.ErrArtifactLoad):
|
||||
return http.StatusBadRequest, "artifact_read_failed", "failed to read input artifact"
|
||||
case errors.Is(err, promptkit.ErrPromptRender):
|
||||
|
||||
Reference in New Issue
Block a user