Transport structured diagnostics through the pipeline

This commit is contained in:
2026-08-27 15:29:10 +00:00
parent acb04954eb
commit ba569594a1
20 changed files with 578 additions and 65 deletions

View File

@@ -25,6 +25,7 @@ type laneExtractState struct {
values []erasedExtractArtifact
serialized []CheckpointArtifact
warnings []contracts.Warning
diagnostics []contracts.DiagnosticGroup
rejected []contracts.RejectedOutput
incomplete []int
validationSummaries []artifacts.ValidationSummary
@@ -39,6 +40,7 @@ type finalizedExtractResults struct {
accepted []erasedExtractArtifact
serialized []CheckpointArtifact
warnings []contracts.Warning
diagnostics []contracts.DiagnosticGroup
rejected []contracts.RejectedOutput
incomplete []int
validationSummaries []artifacts.ValidationSummary
@@ -68,6 +70,7 @@ type extractJobResult struct {
value erasedExtractArtifact
serialized CheckpointArtifact
warnings []contracts.Warning
diagnostics []contracts.DiagnosticGroup
rejected *contracts.RejectedOutput
validationIncomplete bool
validationSummary *artifacts.ValidationSummary
@@ -458,7 +461,7 @@ func (r *Runner) runExtractJob(ctx context.Context, input RunInput, doc *source.
return producerAttemptOutput{}, terminal.record(payload, attemptErr)
}
serializedCandidate.ChunkID, serializedCandidate.ChunkIndex, serializedCandidate.ChunkRef = artifact.ChunkID, artifact.ChunkIndex, artifact.ChunkRef
return producerAttemptOutput{Value: extractAttemptValue{artifact: artifact, serialized: serializedCandidate, terminal: &terminal}, Candidate: extracted.ModelCandidate, Warnings: attemptWarnings}, nil
return producerAttemptOutput{Value: extractAttemptValue{artifact: artifact, serialized: serializedCandidate, terminal: &terminal}, Candidate: extracted.ModelCandidate, Warnings: attemptWarnings, Diagnostics: contracts.CloneProducerDiagnostics(extracted.Diagnostics)}, nil
}, func(validationCtx context.Context, output producerAttemptOutput) (validationReport, error) {
candidate, ok := output.Value.(extractAttemptValue)
if !ok {
@@ -481,6 +484,14 @@ func (r *Runner) runExtractJob(ctx context.Context, input RunInput, doc *source.
result.err = errors.Join(result.err, debugErr)
return result
}
if err == nil {
diagnostics, diagnosticErr := terminalDiagnosticGroups(terminalResult, contracts.DiagnosticOrigin{Stage: contracts.DiagnosticOriginStageExtract, StepID: input.stepID, LaneID: lane.ID, ModuleKey: lane.Extract.Module}, &chunk)
if diagnosticErr != nil {
result.err = fmt.Errorf("promote extract diagnostics: %w", diagnosticErr)
return result
}
result.diagnostics = diagnostics
}
if err == nil && terminalResult.Action == producerTerminalRejected {
result.rejected = terminalResult.Rejection
if result.rejected != nil {
@@ -532,11 +543,13 @@ func finalizeLaneExtract(checkpoints CheckpointRecorder, stepID string, state *l
if result.rejected != nil {
state.rejected = append(state.rejected, *result.rejected)
state.warnings = append(state.warnings, result.warnings...)
state.diagnostics = append(state.diagnostics, result.diagnostics...)
continue
}
state.values = append(state.values, result.value)
state.serialized = append(state.serialized, result.serialized)
state.warnings = append(state.warnings, result.warnings...)
state.diagnostics = append(state.diagnostics, result.diagnostics...)
if result.validationIncomplete {
state.incomplete = append(state.incomplete, result.chunkIndex)
}
@@ -563,6 +576,7 @@ func (r *Runner) continueLane(ctx context.Context, input RunInput, checkpoints C
accepted: state.values,
serialized: state.serialized,
warnings: state.warnings,
diagnostics: state.diagnostics,
rejected: state.rejected,
incomplete: state.incomplete,
validationSummaries: state.validationSummaries,
@@ -570,6 +584,7 @@ func (r *Runner) continueLane(ctx context.Context, input RunInput, checkpoints C
reuseEligible: state.reuseEligible,
}
local.Warnings = append(local.Warnings, cloneWarnings(results.warnings)...)
local.diagnosticGroups = append(local.diagnosticGroups, contracts.CloneDiagnosticCollection(contracts.DiagnosticCollection{Groups: results.diagnostics}).Groups...)
local.Rejected = append(local.Rejected, cloneRejectedOutputs(results.rejected)...)
local.ValidationSummaries = append(local.ValidationSummaries, cloneValidationSummaries(results.validationSummaries)...)
if err := writeDebugTimed(input.Debug, path.Join("extract", fileio.EncodePathComponent(lane.ID), "input.json"), debugTimedEnvelope{Stage: string(StageExtract), StepID: input.stepID, LaneID: lane.ID, ModuleKey: lane.Extract.Module, StartedAt: time.Now().UTC(), Payload: map[string]any{"reused": results.decision.Reused, "decision": results.decision, "source": debugSourceDocumentEnvelope(doc), "chunks": debugSourceChunkEnvelopes(chunks), "options": redactSensitiveMap(lane.Extract.Options), "metadata": redactSensitiveMap(input.Metadata)}}); err != nil {
@@ -641,6 +656,7 @@ func mergeLaneOutput(dst *RunOutput, src RunOutput) error {
dst.NormalizeOutputs = append(dst.NormalizeOutputs, cloneSerializedOutputs(src.NormalizeOutputs)...)
dst.Rejected = append(dst.Rejected, cloneRejectedOutputs(src.Rejected)...)
dst.Warnings = append(dst.Warnings, cloneWarnings(src.Warnings)...)
appendDiagnosticGroups(dst, src.diagnosticGroups)
dst.CheckpointEvents = append(dst.CheckpointEvents, src.CheckpointEvents...)
dst.ValidationSummaries = append(dst.ValidationSummaries, cloneValidationSummaries(src.ValidationSummaries)...)
if len(src.normalizeReuseEligibility) > 0 {