Add opt-in JSON chunk map export

This commit is contained in:
2026-07-23 15:02:01 +00:00
parent 97c9a8e5ce
commit 16a998055c
2 changed files with 261 additions and 5 deletions

View File

@@ -9,6 +9,7 @@ import (
"sort"
"strings"
"gitea.maximumdirect.net/eric/notarius/internal/framework/chunkmap"
"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
)
@@ -17,18 +18,25 @@ const Key = "json"
const contentTypeJSON = "application/json"
const chunkMapFileName = "chunk-map.json"
var safeOutputFileChar = regexp.MustCompile(`[^A-Za-z0-9._-]`)
var _ contracts.OutputEncoder = (*Encoder)(nil)
type Options struct{}
type Options struct {
IncludeChunkMap bool
}
type Encoder struct {
options Options
}
func New() *Encoder {
options, _ := DecodeOptions(nil)
return NewWithOptions(Options{})
}
func NewWithOptions(options Options) *Encoder {
return &Encoder{options: options}
}
@@ -47,7 +55,7 @@ func (e *Encoder) Encode(ctx context.Context, req contracts.OutputRequest) (cont
return contracts.OutputResult{}, encoderErrorf("context error before encoding: %w", err)
}
files, err := logicalFiles(req)
files, err := logicalFiles(req, e.options)
if err != nil {
return contracts.OutputResult{}, err
}
@@ -79,9 +87,16 @@ func validateOptions(options map[string]any) error {
}
func DecodeOptions(options map[string]any) (Options, error) {
if err := pipeline.RejectUnknownOptions(options); err != nil {
if err := pipeline.RejectUnknownOptions(options, "include_chunk_map"); err != nil {
return Options{}, encoderErrorf("%w", err)
}
if value, ok := options["include_chunk_map"]; ok {
enabled, ok := value.(bool)
if !ok {
return Options{}, encoderErrorf("option %q must be a boolean", "include_chunk_map")
}
return Options{IncludeChunkMap: enabled}, nil
}
return Options{}, nil
}
@@ -90,6 +105,16 @@ type indexFile struct {
OutputFiles []outputFileIndex `json:"output_files"`
RejectedFile string `json:"rejected_file"`
WarningsFile string `json:"warnings_file"`
ChunkMap *chunkMapIndex `json:"chunk_map,omitempty"`
}
type chunkMapIndex struct {
ArtifactKind contracts.ArtifactKind `json:"artifact_kind"`
File string `json:"file"`
MediaType string `json:"media_type"`
SchemaID string `json:"schema_id"`
SchemaName string `json:"schema_name"`
SchemaVersion string `json:"schema_version"`
}
type outputFileIndex struct {
@@ -110,7 +135,7 @@ type warningsFile struct {
Warnings []contracts.Warning `json:"warnings"`
}
func logicalFiles(req contracts.OutputRequest) ([]contracts.OutputFile, error) {
func logicalFiles(req contracts.OutputRequest, options Options) ([]contracts.OutputFile, error) {
outputs := cloneNormalizeOutputs(req.NormalizeOutputs)
sort.SliceStable(outputs, func(i, j int) bool {
return outputs[i].LaneID < outputs[j].LaneID
@@ -156,6 +181,14 @@ func logicalFiles(req contracts.OutputRequest) ([]contracts.OutputFile, error) {
RejectedFile: "rejected.json",
WarningsFile: "warnings.json",
}
if options.IncludeChunkMap && req.ChunkMap != nil {
chunkMapOutput, chunkMapDescriptor, err := serializedChunkMapFile(*req.ChunkMap)
if err != nil {
return nil, err
}
files = append(files, chunkMapOutput)
index.ChunkMap = &chunkMapDescriptor
}
indexOutput, err := jsonFile("index.json", index)
if err != nil {
return nil, err
@@ -175,6 +208,33 @@ func logicalFiles(req contracts.OutputRequest) ([]contracts.OutputFile, error) {
return files, nil
}
func serializedChunkMapFile(artifact contracts.SerializedArtifact) (contracts.OutputFile, chunkMapIndex, error) {
if artifact.Kind != chunkmap.ArtifactKind {
return contracts.OutputFile{}, chunkMapIndex{}, encoderErrorf("chunk map has unexpected artifact kind %q", artifact.Kind)
}
if artifact.Schema.ID != chunkmap.SchemaID || artifact.Schema.Name != chunkmap.SchemaName || artifact.Schema.Version != chunkmap.SchemaVersion {
return contracts.OutputFile{}, chunkMapIndex{}, encoderErrorf("chunk map has unexpected schema identity")
}
if strings.TrimSpace(artifact.MediaType) != chunkmap.MediaType {
return contracts.OutputFile{}, chunkMapIndex{}, encoderErrorf("chunk map has unsupported media type %q", artifact.MediaType)
}
if _, err := chunkmap.New().Decode(artifact.Content); err != nil {
return contracts.OutputFile{}, chunkMapIndex{}, encoderErrorf("decode chunk map: %w", err)
}
file, err := serializedOutputFile(chunkMapFileName, artifact)
if err != nil {
return contracts.OutputFile{}, chunkMapIndex{}, err
}
return file, chunkMapIndex{
ArtifactKind: artifact.Kind,
File: chunkMapFileName,
MediaType: chunkmap.MediaType,
SchemaID: artifact.Schema.ID,
SchemaName: artifact.Schema.Name,
SchemaVersion: artifact.Schema.Version,
}, nil
}
func serializedOutputFile(name string, artifact contracts.SerializedArtifact) (contracts.OutputFile, error) {
content := append([]byte(nil), artifact.Content...)
if len(content) == 0 {