eric/notarius
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add canonical chunk plan source model

Browse Source
This commit is contained in:
Eric Rakestraw
2026-07-17 23:47:22 +00:00
parent 1c13e1d64a
commit 3bfac14397
16 changed files with 755 additions and 73 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
internal/framework/checkpoint/loader.go
View File

@@ -242,15 +242,25 @@ func sourceChunksFromEnvelope(values []chunkEnvelope) ([]source.Chunk, error) {
if err != nil {
return nil, err
}
annotations, err := source.CanonicalizeChunkAnnotations(value.Annotations)
if err != nil {
return nil, fmt.Errorf("canonicalize chunk %q annotations: %w", value.ID, err)
}
planAnnotations, err := source.CanonicalizeChunkAnnotations(value.PlanAnnotations)
if err != nil {
return nil, fmt.Errorf("canonicalize chunk %q plan_annotations: %w", value.ID, err)
}
out = append(out, source.Chunk{
ID: value.ID,
SourceID: value.SourceID,
Index: value.Index,
Ref: value.Ref,
Content: content,
MediaType: value.Content.MediaType,
Units: cloneSourceUnits(value.Units),
Metadata: cloneMetadata(value.Metadata),
ID: value.ID,
SourceID: value.SourceID,
Index: value.Index,
Ref: value.Ref,
Content: content,
MediaType: value.Content.MediaType,
Units: cloneSourceUnits(value.Units),
Metadata: cloneMetadata(value.Metadata),
Annotations: annotations,
PlanAnnotations: planAnnotations,
})
}
return out, nil

32
internal/framework/checkpoint/recorder.go
View File

@@ -251,13 +251,15 @@ type chunksEnvelope struct {
}
type chunkEnvelope struct {
ID string `json:"id"`
SourceID string `json:"source_id"`
Index int `json:"index"`
Ref source.SourceRef `json:"ref"`
Content binaryEnvelope `json:"content"`
Units []source.SourceUnit `json:"units,omitempty"`
Metadata map[string]any `json:"metadata,omitempty"`
ID string `json:"id"`
SourceID string `json:"source_id"`
Index int `json:"index"`
Ref source.SourceRef `json:"ref"`
Content binaryEnvelope `json:"content"`
Units []source.SourceUnit `json:"units,omitempty"`
Metadata map[string]any `json:"metadata,omitempty"`
Annotations source.ChunkAnnotations `json:"annotations,omitempty"`
PlanAnnotations source.ChunkAnnotations `json:"plan_annotations,omitempty"`
}
type binaryEnvelope struct {
@@ -320,13 +322,15 @@ func chunkEnvelopes(chunks []source.Chunk) []chunkEnvelope {
out := make([]chunkEnvelope, 0, len(chunks))
for _, chunk := range chunks {
out = append(out, chunkEnvelope{
ID: chunk.ID,
SourceID: chunk.SourceID,
Index: chunk.Index,
Ref: chunk.Ref,
Content: binaryEnvelopeFromContent(chunk.Content, chunk.MediaType, chunk.Metadata, nil),
Units: cloneSourceUnits(chunk.Units),
Metadata: cloneMetadata(chunk.Metadata),
ID: chunk.ID,
SourceID: chunk.SourceID,
Index: chunk.Index,
Ref: chunk.Ref,
Content: binaryEnvelopeFromContent(chunk.Content, chunk.MediaType, chunk.Metadata, nil),
Units: cloneSourceUnits(chunk.Units),
Metadata: cloneMetadata(chunk.Metadata),
Annotations: source.CloneChunkAnnotations(chunk.Annotations),
PlanAnnotations: source.CloneChunkAnnotations(chunk.PlanAnnotations),
})
}
return out

33
internal/framework/checkpoint/recorder_test.go
View File

@@ -5,6 +5,7 @@ import (
"encoding/json"
"os"
"path/filepath"
"reflect"
"strings"
"testing"
@@ -33,6 +34,12 @@ func TestWorkspaceRecorderWritesSuccessfulCheckpointFiles(t *testing.T) {
Content: []byte("chunk content"),
MediaType: "text/plain",
Units: doc.Units,
Annotations: source.ChunkAnnotations{
"same": json.RawMessage(`{"range":1}`),
},
PlanAnnotations: source.ChunkAnnotations{
"same": json.RawMessage(`{"plan":2}`),
},
},
}
@@ -57,7 +64,9 @@ func TestWorkspaceRecorderWritesSuccessfulCheckpointFiles(t *testing.T) {
assertManifestStatus(t, filepath.Join(root, "chunk", "manifest.json"), coreworkspace.StatusSucceeded)
var chunkPayload struct {
Chunks []struct {
Content struct {
Annotations source.ChunkAnnotations `json:"annotations"`
PlanAnnotations source.ChunkAnnotations `json:"plan_annotations"`
Content struct {
ContentBase64 string `json:"content_base64"`
ContentDigest string `json:"content_digest"`
} `json:"content"`
@@ -77,6 +86,28 @@ func TestWorkspaceRecorderWritesSuccessfulCheckpointFiles(t *testing.T) {
if got, want := chunkPayload.Chunks[0].Content.ContentDigest, contentDigest([]byte("chunk content")); got != want {
t.Fatalf("content digest = %q, want %q", got, want)
}
if !jsonEqual(t, chunkPayload.Chunks[0].Annotations["same"], json.RawMessage(`{"range":1}`)) || !jsonEqual(t, chunkPayload.Chunks[0].PlanAnnotations["same"], json.RawMessage(`{"plan":2}`)) {
t.Fatalf("checkpoint annotations = %s / %s", chunkPayload.Chunks[0].Annotations["same"], chunkPayload.Chunks[0].PlanAnnotations["same"])
}
loaded, decision := (&WorkspaceLoader{root: root}).Chunk("generic", doc.Digest)
if !decision.Reused || len(loaded.Chunks) != 1 {
t.Fatalf("loaded chunk checkpoint = %#v, decision = %#v", loaded, decision)
}
if string(loaded.Chunks[0].Annotations["same"]) != `{"range":1}` || string(loaded.Chunks[0].PlanAnnotations["same"]) != `{"plan":2}` {
t.Fatalf("loaded canonical annotations = %s / %s", loaded.Chunks[0].Annotations["same"], loaded.Chunks[0].PlanAnnotations["same"])
}
}
func jsonEqual(t *testing.T, left, right []byte) bool {
t.Helper()
var leftValue, rightValue any
if err := json.Unmarshal(left, &leftValue); err != nil {
t.Fatalf("decode left JSON: %v", err)
}
if err := json.Unmarshal(right, &rightValue); err != nil {
t.Fatalf("decode right JSON: %v", err)
}
return reflect.DeepEqual(leftValue, rightValue)
}
func TestWorkspaceArtifactCheckpointsRoundTripCodecIdentityAndBytes(t *testing.T) {