Order spell and NPC extraction by document position

This commit is contained in:
2026-07-24 14:29:19 +00:00
parent 83fde83a58
commit 9bbf2535dd
8 changed files with 138 additions and 111 deletions

View File

@@ -263,7 +263,7 @@ func TestChangedSemanticSpellCatalogFingerprintCannotResumeRecordedCheckpoint(t
if _, decision := changedLoader.Normalize("spells", spellnormalize.Key, normalizeDependencies); decision.Reused { if _, decision := changedLoader.Normalize("spells", spellnormalize.Key, normalizeDependencies); decision.Reused {
t.Fatalf("changed normalize fingerprint decision = %#v, want normalize checkpoint cold miss", decision) t.Fatalf("changed normalize fingerprint decision = %#v, want normalize checkpoint cold miss", decision)
} }
changedMapping := replaceCheckpointFingerprintValue(t, fingerprints, extractSpellMappingFingerprintName(), "dnd.spells.extract_mapping.v2") changedMapping := replaceCheckpointFingerprintValue(t, fingerprints, extractSpellMappingFingerprintName(), "dnd.spells.extract_mapping.v3")
assertOnlyCheckpointFingerprintChanged(t, fingerprints, changedMapping, extractSpellMappingFingerprintName()) assertOnlyCheckpointFingerprintChanged(t, fingerprints, changedMapping, extractSpellMappingFingerprintName())
_, mappingLoader, err := checkpointHandlersForRun(settings, Options{}, materialized, changedMapping, []byte("same input"), nil, nil, "", "", true) _, mappingLoader, err := checkpointHandlersForRun(settings, Options{}, materialized, changedMapping, []byte("same input"), nil, nil, "", "", true)
if err != nil { if err != nil {

View File

@@ -6,18 +6,19 @@ import (
"gitea.maximumdirect.net/eric/notarius/internal/core/source" "gitea.maximumdirect.net/eric/notarius/internal/core/source"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd" "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity" "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
) )
func canonicalizeResponse(response *extractionResponse, doc *source.SourceDocument) { func canonicalizeResponse(response *extractionResponse, order shared.SourceRefOrder, sourceID string) {
if response == nil { if response == nil {
return return
} }
for index := range response.NPCs { for index := range response.NPCs {
canonicalizeNPC(&response.NPCs[index]) canonicalizeNPC(&response.NPCs[index], order, sourceID)
} }
sort.SliceStable(response.NPCs, func(i, j int) bool { sort.SliceStable(response.NPCs, func(i, j int) bool {
left, leftOK := earliestSourceIndex(doc, response.NPCs[i]) left, leftOK := order.EarliestValid(canonicalSourceRefs(response.NPCs[i].SourceRefs, sourceID))
right, rightOK := earliestSourceIndex(doc, response.NPCs[j]) right, rightOK := order.EarliestValid(canonicalSourceRefs(response.NPCs[j].SourceRefs, sourceID))
if leftOK != rightOK { if leftOK != rightOK {
return leftOK return leftOK
} }
@@ -28,67 +29,11 @@ func canonicalizeResponse(response *extractionResponse, doc *source.SourceDocume
}) })
} }
func canonicalizeNPC(npc *npcResponse) { func canonicalizeNPC(npc *npcResponse, order shared.SourceRefOrder, sourceID string) {
if npc == nil { if npc == nil {
return return
} }
sort.SliceStable(npc.SourceRefs, func(i, j int) bool { npc.SourceRefs = npcResponseRefs(order.Canonicalize(canonicalSourceRefs(npc.SourceRefs, sourceID)))
left := npc.SourceRefs[i]
right := npc.SourceRefs[j]
if unitSortValue(left.StartUnitID) != unitSortValue(right.StartUnitID) {
return unitSortValue(left.StartUnitID) < unitSortValue(right.StartUnitID)
}
return unitSortValue(left.EndUnitID) < unitSortValue(right.EndUnitID)
})
npc.SourceRefs = dedupeSourceRefs(npc.SourceRefs)
}
func dedupeSourceRefs(refs []npcSourceRefResponse) []npcSourceRefResponse {
if len(refs) < 2 {
return refs
}
out := refs[:0]
var previous npcSourceRefResponse
for index, ref := range refs {
if index > 0 && sameSourceRef(previous, ref) {
continue
}
out = append(out, ref)
previous = ref
}
return out
}
func sameSourceRef(left npcSourceRefResponse, right npcSourceRefResponse) bool {
return left.StartUnitID == right.StartUnitID && left.EndUnitID == right.EndUnitID
}
func earliestSourceIndex(doc *source.SourceDocument, npc npcResponse) (int, bool) {
earliest := 0
found := false
for _, ref := range npc.SourceRefs {
start := ref.StartUnitID
end := ref.EndUnitID
if start > 0 && end > 0 {
startIndex, startOK := source.UnitIndex(doc, start)
endIndex, endOK := source.UnitIndex(doc, end)
if !startOK || !endOK || startIndex > endIndex {
continue
}
if !found || startIndex < earliest {
earliest = startIndex
found = true
}
}
}
return earliest, found
}
func unitSortValue(value int) int {
if value <= 0 {
return int(^uint(0) >> 1)
}
return value
} }
func canonicalNPCList(response extractionResponse, sourceID string) dnd.NPCList { func canonicalNPCList(response extractionResponse, sourceID string) dnd.NPCList {
@@ -120,3 +65,14 @@ func canonicalSourceRefs(values []npcSourceRefResponse, sourceID string) []sourc
} }
return out return out
} }
func npcResponseRefs(values []source.SourceRef) []npcSourceRefResponse {
if values == nil {
return nil
}
out := make([]npcSourceRefResponse, len(values))
for index, value := range values {
out[index] = npcSourceRefResponse{StartUnitID: value.StartUnitID, EndUnitID: value.EndUnitID}
}
return out
}

View File

@@ -13,7 +13,7 @@ import (
const Key = "dnd/npcs" const Key = "dnd/npcs"
const mappingPolicy = "dnd.npcs.extract_mapping.v1" const mappingPolicy = "dnd.npcs.extract_mapping.v2"
var requiredCapabilities = []string{ var requiredCapabilities = []string{
"chunks", "chunks",
@@ -129,6 +129,7 @@ func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRe
if err != nil { if err != nil {
return contracts.TypedExtractionResult[dnd.NPCList]{}, extractorErrorf("%w", err) return contracts.TypedExtractionResult[dnd.NPCList]{}, extractorErrorf("%w", err)
} }
order := shared.NewSourceRefOrder(req.Source)
var response extractionResponse var response extractionResponse
if _, err := e.llm.CompleteStructured(ctx, contracts.StructuredCompletionRequest{ if _, err := e.llm.CompleteStructured(ctx, contracts.StructuredCompletionRequest{
@@ -141,7 +142,7 @@ func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRe
}, &response); err != nil { }, &response); err != nil {
return contracts.TypedExtractionResult[dnd.NPCList]{}, extractorErrorf("complete structured output: %w", err) return contracts.TypedExtractionResult[dnd.NPCList]{}, extractorErrorf("complete structured output: %w", err)
} }
canonicalizeResponse(&response, req.Source) canonicalizeResponse(&response, order, req.Source.ID)
return contracts.TypedExtractionResult[dnd.NPCList]{Value: canonicalNPCList(response, req.Source.ID)}, nil return contracts.TypedExtractionResult[dnd.NPCList]{Value: canonicalNPCList(response, req.Source.ID)}, nil
} }

View File

@@ -84,6 +84,43 @@ func TestExtractOrdersNPCsBySourcePositionRatherThanUnitID(t *testing.T) {
} }
} }
func TestExtractUsesDocumentOrderForNPCReferencesAndStableTies(t *testing.T) {
client := &fakeNPCsLLMClient{response: extractionResponse{NPCs: []npcResponse{
{Name: "Later", SourceRefs: responseSourceRefs(10, 10)},
{Name: "First", SourceRefs: []npcSourceRefResponse{
{StartUnitID: 10, EndUnitID: 10},
{StartUnitID: 30, EndUnitID: 30},
{StartUnitID: 30, EndUnitID: 30},
{StartUnitID: 999, EndUnitID: 0},
}},
{Name: "Second", SourceRefs: responseSourceRefs(30, 30)},
}}}
req := extractionRequest()
req.Source.Units = []source.SourceUnit{{ID: 30}, {ID: 10}}
req.Chunk.Units = append([]source.SourceUnit(nil), req.Source.Units...)
req.Chunk.Ref = source.SourceRef{SourceID: req.Source.ID, StartUnitID: 30, EndUnitID: 10}
result, err := newExtractor(t, client).Extract(context.Background(), req)
if err != nil {
t.Fatalf("Extract() error = %v", err)
}
if got := []string{result.Value.NPCs[0].Name, result.Value.NPCs[1].Name, result.Value.NPCs[2].Name}; !reflect.DeepEqual(got, []string{"First", "Second", "Later"}) {
t.Fatalf("NPC order = %#v, want document chronology with stable equal-evidence ties", got)
}
refs := result.Value.NPCs[0].SourceRefs
if got := []int{refs[0].StartUnitID, refs[1].StartUnitID, refs[2].StartUnitID}; !reflect.DeepEqual(got, []int{30, 10, 999}) {
t.Fatalf("source refs = %#v, want document order with exact duplicate removed", refs)
}
refs[0].StartUnitID = 777
for _, npc := range client.response.NPCs {
for _, ref := range npc.SourceRefs {
if ref.StartUnitID == 777 {
t.Fatal("result source references alias the model response")
}
}
}
}
func TestExtractPassesCampaignReferencesAsPromptInputs(t *testing.T) { func TestExtractPassesCampaignReferencesAsPromptInputs(t *testing.T) {
client := &fakeNPCsLLMClient{response: extractionResponse{NPCs: []npcResponse{}}} client := &fakeNPCsLLMClient{response: extractionResponse{NPCs: []npcResponse{}}}
req := extractionRequest() req := extractionRequest()

View File

@@ -5,18 +5,19 @@ import (
"gitea.maximumdirect.net/eric/notarius/internal/core/source" "gitea.maximumdirect.net/eric/notarius/internal/core/source"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd" "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
) )
func canonicalizeResponse(response *extractionResponse) { func canonicalizeResponse(response *extractionResponse, order shared.SourceRefOrder, sourceID string) {
if response == nil { if response == nil {
return return
} }
for index := range response.SpellCasts { for index := range response.SpellCasts {
canonicalizeSpellCast(&response.SpellCasts[index]) canonicalizeSpellCast(&response.SpellCasts[index], order, sourceID)
} }
sort.SliceStable(response.SpellCasts, func(i, j int) bool { sort.SliceStable(response.SpellCasts, func(i, j int) bool {
left, leftOK := earliestSourceUnit(response.SpellCasts[i]) left, leftOK := order.EarliestValid(spellSourceRefs(response.SpellCasts[i].SourceRefs, sourceID))
right, rightOK := earliestSourceUnit(response.SpellCasts[j]) right, rightOK := order.EarliestValid(spellSourceRefs(response.SpellCasts[j].SourceRefs, sourceID))
if leftOK != rightOK { if leftOK != rightOK {
return leftOK return leftOK
} }
@@ -27,53 +28,33 @@ func canonicalizeResponse(response *extractionResponse) {
}) })
} }
func canonicalizeSpellCast(spell *spellCastResponse) { func canonicalizeSpellCast(spell *spellCastResponse, order shared.SourceRefOrder, sourceID string) {
sort.SliceStable(spell.SourceRefs, func(i, j int) bool { if spell == nil {
left := spell.SourceRefs[i] return
right := spell.SourceRefs[j] }
if left.StartUnitID != right.StartUnitID { spell.SourceRefs = spellResponseRefs(order.Canonicalize(spellSourceRefs(spell.SourceRefs, sourceID)))
return unitSortValue(left.StartUnitID) < unitSortValue(right.StartUnitID)
}
return unitSortValue(left.EndUnitID) < unitSortValue(right.EndUnitID)
})
spell.SourceRefs = dedupeSourceRefs(spell.SourceRefs)
} }
func dedupeSourceRefs(refs []spellSourceRefResponse) []spellSourceRefResponse { func spellSourceRefs(refs []spellSourceRefResponse, sourceID string) []source.SourceRef {
if len(refs) < 2 { if refs == nil {
return refs return nil
} }
out := refs[:0] values := make([]source.SourceRef, len(refs))
var previous spellSourceRefResponse
for index, ref := range refs { for index, ref := range refs {
if index > 0 && sameSourceRef(previous, ref) { values[index] = source.SourceRef{SourceID: sourceID, StartUnitID: ref.StartUnitID, EndUnitID: ref.EndUnitID}
continue
}
out = append(out, ref)
previous = ref
} }
return out return values
} }
func sameSourceRef(left spellSourceRefResponse, right spellSourceRefResponse) bool { func spellResponseRefs(refs []source.SourceRef) []spellSourceRefResponse {
return left.StartUnitID == right.StartUnitID && left.EndUnitID == right.EndUnitID if refs == nil {
} return nil
func earliestSourceUnit(spell spellCastResponse) (int, bool) {
for _, ref := range spell.SourceRefs {
start := ref.StartUnitID
if start > 0 {
return start, true
}
} }
return 0, false values := make([]spellSourceRefResponse, len(refs))
} for index, ref := range refs {
values[index] = spellSourceRefResponse{StartUnitID: ref.StartUnitID, EndUnitID: ref.EndUnitID}
func unitSortValue(value int) int {
if value <= 0 {
return int(^uint(0) >> 1)
} }
return value return values
} }
func canonicalSpellList(response extractionResponse, sourceID string) dnd.SpellList { func canonicalSpellList(response extractionResponse, sourceID string) dnd.SpellList {

View File

@@ -17,7 +17,7 @@ const Key = "dnd/spells"
const ArtifactType = "dnd.spell_cast" const ArtifactType = "dnd.spell_cast"
const SchemaVersion = "v1" const SchemaVersion = "v1"
const mappingPolicy = "dnd.spells.extract_mapping.v1" const mappingPolicy = "dnd.spells.extract_mapping.v2"
const ( const (
NPCRegistryReferenceSlot = npcregistry.ReferenceSlot NPCRegistryReferenceSlot = npcregistry.ReferenceSlot
@@ -185,6 +185,7 @@ func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRe
if err != nil { if err != nil {
return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("%w", err) return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("%w", err)
} }
order := shared.NewSourceRefOrder(req.Source)
npcRegistry, err := e.npcResolver.Resolve(req.References) npcRegistry, err := e.npcResolver.Resolve(req.References)
if err != nil { if err != nil {
return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("resolve NPC registry: %w", err) return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("resolve NPC registry: %w", err)
@@ -204,7 +205,7 @@ func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRe
}, &response); err != nil { }, &response); err != nil {
return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("complete structured output: %w", err) return contracts.TypedExtractionResult[dnd.SpellList]{}, extractorErrorf("complete structured output: %w", err)
} }
canonicalizeResponse(&response) canonicalizeResponse(&response, order, req.Source.ID)
return contracts.TypedExtractionResult[dnd.SpellList]{Value: canonicalSpellList(response, req.Source.ID)}, nil return contracts.TypedExtractionResult[dnd.SpellList]{Value: canonicalSpellList(response, req.Source.ID)}, nil
} }

View File

@@ -284,6 +284,57 @@ func TestExtractOrdersAndDeduplicatesEvidence(t *testing.T) {
} }
} }
func TestExtractUsesDocumentOrderForReferencesAndSpellCasts(t *testing.T) {
client := &fakeSpellsLLMClient{response: extractionResponse{SpellCasts: []spellCastResponse{
{Caster: "Later", Spell: "Fire Bolt", SourceRefs: responseSourceRefs(10, 10)},
{Caster: "Earlier", Spell: "Cure Wounds", SourceRefs: []spellSourceRefResponse{
{StartUnitID: 10, EndUnitID: 10},
{StartUnitID: 30, EndUnitID: 30},
{StartUnitID: 30, EndUnitID: 30},
{StartUnitID: 999, EndUnitID: 0},
}},
{Caster: "Unavailable", Spell: "Healing Word", SourceRefs: []spellSourceRefResponse{{StartUnitID: 999, EndUnitID: 0}}},
}}}
req := extractionRequest()
req.Source.Units = []source.SourceUnit{{ID: 30}, {ID: 10}}
req.Chunk.Units = append([]source.SourceUnit(nil), req.Source.Units...)
req.Chunk.Ref = source.SourceRef{SourceID: req.Source.ID, StartUnitID: 30, EndUnitID: 10}
result, err := newExtractor(t, client).Extract(context.Background(), req)
if err != nil {
t.Fatalf("Extract() error = %v", err)
}
if got := []string{result.Value.SpellCasts[0].Spell, result.Value.SpellCasts[1].Spell, result.Value.SpellCasts[2].Spell}; !reflect.DeepEqual(got, []string{"Cure Wounds", "Fire Bolt", "Healing Word"}) {
t.Fatalf("spell order = %#v, want document chronology followed by invalid evidence", got)
}
refs := result.Value.SpellCasts[0].SourceRefs
if got := []int{refs[0].StartUnitID, refs[1].StartUnitID, refs[2].StartUnitID}; !reflect.DeepEqual(got, []int{30, 10, 999}) {
t.Fatalf("source refs = %#v, want document order with exact duplicate removed", refs)
}
refs[0].StartUnitID = 777
for _, spell := range client.response.SpellCasts {
for _, ref := range spell.SourceRefs {
if ref.StartUnitID == 777 {
t.Fatal("result source references alias the model response")
}
}
}
}
func TestExtractPreservesStableSpellOrderForEqualEvidence(t *testing.T) {
client := &fakeSpellsLLMClient{response: extractionResponse{SpellCasts: []spellCastResponse{
{Caster: "First", Spell: "Cure Wounds", SourceRefs: responseSourceRefs(2, 2)},
{Caster: "Second", Spell: "Fire Bolt", SourceRefs: responseSourceRefs(2, 2)},
}}}
result, err := newExtractor(t, client).Extract(context.Background(), extractionRequest())
if err != nil {
t.Fatalf("Extract() error = %v", err)
}
if got := []string{result.Value.SpellCasts[0].Caster, result.Value.SpellCasts[1].Caster}; !reflect.DeepEqual(got, []string{"First", "Second"}) {
t.Fatalf("equal-evidence order = %#v, want stable response order", got)
}
}
func TestExtractPreservesInvalidEvidenceForValidators(t *testing.T) { func TestExtractPreservesInvalidEvidenceForValidators(t *testing.T) {
client := &fakeSpellsLLMClient{response: extractionResponse{SpellCasts: []spellCastResponse{{ client := &fakeSpellsLLMClient{response: extractionResponse{SpellCasts: []spellCastResponse{{
Caster: "Aria", Spell: "Cure Wounds", Caster: "Aria", Spell: "Cure Wounds",

View File

@@ -41,8 +41,8 @@ func TestNPCOutputGroundsSpellAndCombatConsumersThroughOneOperation(t *testing.T
t.Fatalf("Prepare() error = %v", err) t.Fatalf("Prepare() error = %v", err)
} }
for name, value := range map[string]string{ for name, value := range map[string]string{
"extract:npcs:dnd/npcs:mapping_policy": "dnd.npcs.extract_mapping.v1", "extract:npcs:dnd/npcs:mapping_policy": "dnd.npcs.extract_mapping.v2",
"extract:spells:dnd/spells:mapping_policy": "dnd.spells.extract_mapping.v1", "extract:spells:dnd/spells:mapping_policy": "dnd.spells.extract_mapping.v2",
} { } {
assertFingerprintValue(t, prepared.CheckpointFingerprints(), name, value) assertFingerprintValue(t, prepared.CheckpointFingerprints(), name, value)
} }