notarius/internal/modules/dnd/locations/registry/registry_test.go

package registry

import (
	"bytes"
	"encoding/json"
	"reflect"
	"strings"
	"testing"

	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
	locationcodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/locationregistry"
	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/locations/identity"
)

func TestResolveUnboundRegistryHasEmptyIdentity(t *testing.T) {
	registry, err := Resolve(contracts.ReferenceSet{})
	if err != nil {
		t.Fatal(err)
	}
	if registry.Bound() || registry.Digest() != "" || registry.Count() != 0 || string(registry.CanonicalBytes()) != emptyRegistryContent {
		t.Fatalf("unbound registry = %#v", registry)
	}
	if registry.IdentityDigest() == "" {
		t.Fatal("unbound registry identity digest is empty")
	}
}

func TestResolveRetainsSeparateDurableAndIdentityDigests(t *testing.T) {
	registry := resolveList(t, registryFixture())
	if !registry.Bound() || registry.Count() != 2 || registry.Digest() == "" {
		t.Fatalf("registry identity = bound %t count %d digest %q", registry.Bound(), registry.Count(), registry.Digest())
	}
	if registry.IdentityDigest() == "" || registry.Digest() == registry.IdentityDigest() {
		t.Fatalf("full/identity digests = %q/%q", registry.Digest(), registry.IdentityDigest())
	}
}

func TestRegistryLookupUsesIDAndReturnsDefensiveCopies(t *testing.T) {
	registry := resolveList(t, registryFixture())
	first := registry.Locations()[0]
	if got, ok := registry.Lookup(first.ID); !ok || got.Name != first.Name || !registry.Matches(first.ID, first.Name) || registry.Matches(first.ID, "Other Tavern") {
		t.Fatalf("ID lookup/match = %#v, %t", got, ok)
	}
	if _, ok := registry.Lookup("The Tavern"); ok {
		t.Fatal("Lookup accepted a name as an ID")
	}

	locations := registry.Locations()
	locations[0].Name = "changed"
	locations[0].SourceRefs[0].SourceID = "changed"
	canonical := registry.CanonicalBytes()
	canonical[0] = '['
	if next, ok := registry.Lookup(first.ID); !ok || next.Name != first.Name || next.SourceRefs[0].SourceID != "session-alpha" {
		t.Fatalf("registry mutated through accessor: %#v, %t", next, ok)
	}
	if registry.CanonicalBytes()[0] != '{' {
		t.Fatal("registry bytes mutated through accessor")
	}
}

func TestResolveRejectsInvalidReferenceInputs(t *testing.T) {
	valid := registryFixture()
	content, err := locationcodec.New().Encode(valid)
	if err != nil {
		t.Fatal(err)
	}
	invalidIdentity := append([]byte(nil), content...)
	invalidIdentity = bytes.Replace(invalidIdentity, []byte(valid.Locations[0].ID), []byte("location:sha256:0000000000000000000000000000000000000000000000000000000000000000"), 1)
	for _, test := range []struct {
		name string
		set  contracts.ReferenceSet
		want string
	}{
		{"empty bound slot", referenceSet(), "exactly one item"},
		{"multiple items", referenceSet(item(content), item(content)), "exactly one item"},
		{"malformed JSON", referenceSet(item([]byte(`{"locations":[`))), "invalid approved"},
		{"wrong media type", referenceSet(contracts.ReferenceItem{MediaType: "text/plain", Content: content}), "media type must be"},
		{"oversized", referenceSet(item(make([]byte, MaxBytes+1))), "limit"},
		{"invalid identity", referenceSet(item(invalidIdentity)), "id_mismatch"},
	} {
		t.Run(test.name, func(t *testing.T) {
			if _, err := Resolve(test.set); err == nil || !strings.Contains(err.Error(), test.want) {
				t.Fatalf("Resolve() error = %v, want %q", err, test.want)
			}
		})
	}
}

func TestResolveRejectsReversedEvidenceInDirectRegistryReference(t *testing.T) {
	value := registryFixture()
	value.Locations[0].SourceRefs[0] = source.SourceRef{SourceID: "session-alpha", StartUnitID: 2, EndUnitID: 1}
	content, err := locationcodec.New().EncodeCandidate(value)
	if err != nil {
		t.Fatalf("EncodeCandidate() error = %v", err)
	}
	if _, err := Resolve(referenceSet(item(content))); err == nil || !strings.Contains(err.Error(), "invalid approved") {
		t.Fatalf("Resolve() error = %v, want direct-reference rejection", err)
	}
}

func TestResolverHandlesConstructionAndOperationReferences(t *testing.T) {
	placeholder, err := NewResolver(referenceSet())
	if err != nil || placeholder.Seeded().Bound() {
		t.Fatalf("generated placeholder = %#v, %v; want unbound seed", placeholder, err)
	}

	validContent := encodeList(t, registryFixture())
	malformed := referenceSet(item([]byte(`{"locations":[`)))
	if _, err := NewResolver(malformed); err == nil {
		t.Fatal("NewResolver(malformed) error = nil")
	}
	if _, err := placeholder.Resolve(malformed); err == nil {
		t.Fatal("Resolve(malformed) error = nil")
	}

	staticContent := append([]byte(nil), validContent...)
	staticReferences := referenceSet(item(staticContent))
	seeded, err := NewResolver(staticReferences)
	if err != nil {
		t.Fatal(err)
	}
	resolved, err := seeded.Resolve(staticReferences)
	if err != nil || resolved != seeded.Seeded() {
		t.Fatalf("seed reuse = %p / %p, %v", resolved, seeded.Seeded(), err)
	}
	staticContent[0] = '['
	delete(staticReferences.Slots, ReferenceSlot)
	if seeded.Seeded().Count() != 2 || seeded.Seeded().CanonicalBytes()[0] != '{' {
		t.Fatalf("seeded registry retained construction references: %#v", seeded.Seeded())
	}
	if fallback, err := seeded.Resolve(contracts.ReferenceSet{}); err != nil || fallback != seeded.Seeded() {
		t.Fatalf("fallback = %#v, %v; want seeded registry", fallback, err)
	}
}

func TestResolverReusesEquivalentCanonicalRegistries(t *testing.T) {
	resolver, err := NewResolver(referenceSet())
	if err != nil {
		t.Fatal(err)
	}
	content, err := locationcodec.New().Encode(registryFixture())
	if err != nil {
		t.Fatal(err)
	}
	firstSet := referenceSet(item(content))
	first, err := resolver.Resolve(firstSet)
	if err != nil {
		t.Fatal(err)
	}
	spaced := append([]byte("\n  "), content...)
	spaced = append(spaced, '\n')
	second, err := resolver.Resolve(referenceSet(contracts.ReferenceItem{MediaType: "APPLICATION/JSON; charset=utf-8", Content: spaced}))
	if err != nil || second != first {
		t.Fatalf("equivalent canonical registry = %p / %p, %v", first, second, err)
	}
}

func registryFixture() dnd.LocationRegistry {
	firstRefs := []source.SourceRef{{SourceID: "session-alpha", StartUnitID: 1, EndUnitID: 1}}
	secondRefs := []source.SourceRef{{SourceID: "session-alpha", StartUnitID: 2, EndUnitID: 2}}
	return dnd.LocationRegistry{Locations: []dnd.Location{
		{ID: identity.DeriveID("The Tavern", firstRefs), Name: "The Tavern", SourceRefs: firstRefs},
		{ID: identity.DeriveID("The Tavern", secondRefs), Name: "The Tavern", SourceRefs: secondRefs},
	}}
}

func resolveList(t *testing.T, list dnd.LocationRegistry) *Registry {
	t.Helper()
	registry, err := Resolve(referenceSet(item(encodeList(t, list))))
	if err != nil {
		t.Fatalf("Resolve() error = %v", err)
	}
	return registry
}

func encodeList(t *testing.T, list dnd.LocationRegistry) []byte {
	t.Helper()
	content, err := locationcodec.New().Encode(list)
	if err != nil {
		t.Fatalf("Encode() error = %v", err)
	}
	return content
}

func item(content []byte) contracts.ReferenceItem {
	return contracts.ReferenceItem{SlotName: ReferenceSlot, MediaType: locationcodec.MediaType, Content: content}
}

func referenceSet(items ...contracts.ReferenceItem) contracts.ReferenceSet {
	return contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{ReferenceSlot: {Items: items}}}
}

func TestGroundingProjectsAndResolvesUniqueAndSameNameLocations(t *testing.T) {
	doc := groundingDocument()
	sharedName := "The Tavern"
	firstRefs := []source.SourceRef{{SourceID: doc.ID, StartUnitID: 20, EndUnitID: 20}, {SourceID: doc.ID, StartUnitID: 30, EndUnitID: 30}}
	secondRefs := []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}}
	marketRefs := []source.SourceRef{{SourceID: doc.ID, StartUnitID: 40, EndUnitID: 40}}
	list := dnd.LocationRegistry{Locations: []dnd.Location{
		location(sharedName, firstRefs),
		location(sharedName, secondRefs),
		location("Market", marketRefs),
	}}
	registry := resolveList(t, list)
	grounding, err := NewGrounding(registry, doc)
	if err != nil {
		t.Fatal(err)
	}

	projection := decodeGroundingProjection(t, grounding.PromptInput().Content)
	if len(projection.Locations) != 3 {
		t.Fatalf("projection = %#v", projection)
	}
	first := projection.Locations[0]
	if first.Name != sharedName || !reflect.DeepEqual(first.RegistryRefs, []RegistryRef{{StartUnitID: 30, EndUnitID: 30}, {StartUnitID: 20, EndUnitID: 20}}) || !reflect.DeepEqual(first.Context, []ContextUnit{{UnitID: 30, Text: "Tavern entrance"}, {UnitID: 20, Text: "Tavern cellar"}}) {
		t.Fatalf("first contextual projection = %#v", first)
	}
	second := projection.Locations[1]
	if !reflect.DeepEqual(second.RegistryRefs, []RegistryRef{{StartUnitID: 10, EndUnitID: 10}}) || !reflect.DeepEqual(second.Context, []ContextUnit{{UnitID: 10, Text: "Tavern common room"}}) {
		t.Fatalf("second contextual projection = %#v", second)
	}
	market := projection.Locations[2]
	if market.Name != "Market" || len(market.RegistryRefs) != 0 || len(market.Context) != 0 {
		t.Fatalf("unique-name projection = %#v", market)
	}
	if bytes.Contains(grounding.PromptInput().Content, []byte("location:sha256:")) || bytes.Contains(grounding.PromptInput().Content, []byte(`"source_id"`)) {
		t.Fatalf("grounding projection exposed durable identity: %s", grounding.PromptInput().Content)
	}
	if grounding.PromptInput().Digest == "" || grounding.PromptInput().Digest == registry.IdentityDigest() {
		t.Fatalf("grounding/identity digests = %q/%q", grounding.PromptInput().Digest, registry.IdentityDigest())
	}

	resolved, ok := grounding.Resolve(Selector{Name: " the tavern ", RegistryRefs: []RegistryRef{{StartUnitID: 30, EndUnitID: 30}, {StartUnitID: 20, EndUnitID: 20}}})
	if !ok || resolved.ID != list.Locations[0].ID || resolved.Name != sharedName {
		t.Fatalf("Resolve(same name) = %#v, %t", resolved, ok)
	}
	if resolved, ok = grounding.Resolve(Selector{Name: "MARKET", RegistryRefs: []RegistryRef{}}); !ok || resolved.ID != list.Locations[2].ID {
		t.Fatalf("Resolve(unique name) = %#v, %t", resolved, ok)
	}
	for _, selector := range []Selector{
		{Name: "Market", RegistryRefs: []RegistryRef{{StartUnitID: 40, EndUnitID: 40}}},
		{Name: sharedName, RegistryRefs: []RegistryRef{}},
		{Name: sharedName, RegistryRefs: []RegistryRef{{StartUnitID: 30, EndUnitID: 30}}},
		{Name: sharedName, RegistryRefs: []RegistryRef{{StartUnitID: 20, EndUnitID: 20}, {StartUnitID: 30, EndUnitID: 30}}},
		{Name: "Unknown", RegistryRefs: []RegistryRef{}},
	} {
		if _, ok := grounding.Resolve(selector); ok {
			t.Fatalf("Resolve(%#v) accepted an unsupported selector", selector)
		}
	}

	input := grounding.PromptInput()
	input.Content[0] = '['
	resolved.SourceRefs[0].SourceID = "changed"
	if grounding.PromptInput().Content[0] != '{' {
		t.Fatal("grounding prompt input was mutable")
	}
	if next, ok := grounding.Resolve(Selector{Name: sharedName, RegistryRefs: []RegistryRef{{StartUnitID: 30, EndUnitID: 30}, {StartUnitID: 20, EndUnitID: 20}}}); !ok || next.SourceRefs[0].SourceID != doc.ID {
		t.Fatalf("grounding resolved location was mutable: %#v, %t", next, ok)
	}
}

func TestGroundingHandlesEmptyRegistriesAndIdentityFingerprints(t *testing.T) {
	doc := groundingDocument()
	empty := resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{}})
	grounding, err := NewGrounding(empty, doc)
	if err != nil {
		t.Fatal(err)
	}
	if projection := decodeGroundingProjection(t, grounding.PromptInput().Content); projection.Locations == nil || len(projection.Locations) != 0 {
		t.Fatalf("empty projection = %#v", projection)
	}
	if _, ok := grounding.Resolve(Selector{Name: "Market", RegistryRefs: []RegistryRef{}}); ok {
		t.Fatal("empty grounding resolved a location")
	}

	baseRefs := []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}}
	base := resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{location("Market", baseRefs)}})
	expandedRefs := append(append([]source.SourceRef(nil), baseRefs...), source.SourceRef{SourceID: doc.ID, StartUnitID: 40, EndUnitID: 40})
	expanded := resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{location("Market", expandedRefs)}})
	baseGrounding, err := NewGrounding(base, doc)
	if err != nil {
		t.Fatal(err)
	}
	expandedGrounding, err := NewGrounding(expanded, doc)
	if err != nil {
		t.Fatal(err)
	}
	if base.IdentityDigest() != expanded.IdentityDigest() || baseGrounding.PromptInput().Digest != expandedGrounding.PromptInput().Digest {
		t.Fatalf("identity/model fingerprints = %q/%q and %q/%q", base.IdentityDigest(), expanded.IdentityDigest(), baseGrounding.PromptInput().Digest, expandedGrounding.PromptInput().Digest)
	}
}

func TestGroundingRejectsUnsafeContextualConstruction(t *testing.T) {
	doc := groundingDocument()
	validRefs := []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}}
	valid := resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{location("Market", validRefs)}})
	if _, err := NewGrounding(valid, nil); err == nil {
		t.Fatal("NewGrounding(nil source) error = nil")
	}

	for _, test := range []struct {
		name     string
		registry *Registry
	}{
		{
			name: "foreign reference",
			registry: resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{
				location("The Tavern", []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}}),
				location("The Tavern", []source.SourceRef{{SourceID: "other-source", StartUnitID: 10, EndUnitID: 10}}),
			}}),
		},
		{
			name: "invalid reference",
			registry: resolveList(t, dnd.LocationRegistry{Locations: []dnd.Location{
				location("The Tavern", []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}}),
				location("The Tavern", []source.SourceRef{{SourceID: doc.ID, StartUnitID: 999, EndUnitID: 999}}),
			}}),
		},
		{
			name: "selector collision",
			registry: &Registry{list: dnd.LocationRegistry{Locations: []dnd.Location{
				{ID: "first", Name: "The Tavern", SourceRefs: validRefs},
				{ID: "second", Name: "The Tavern", SourceRefs: validRefs},
			}}},
		},
		{
			name:     "empty comparison name",
			registry: &Registry{list: dnd.LocationRegistry{Locations: []dnd.Location{{ID: "first", Name: "  ", SourceRefs: validRefs}}}},
		},
	} {
		t.Run(test.name, func(t *testing.T) {
			if _, err := NewGrounding(test.registry, doc); err == nil || strings.Contains(err.Error(), "Tavern entrance") {
				t.Fatalf("NewGrounding() error = %v, want content-safe failure", err)
			}
		})
	}
}

type decodedGroundingProjection struct {
	Locations []struct {
		Name         string        `json:"name"`
		RegistryRefs []RegistryRef `json:"registry_refs"`
		Context      []ContextUnit `json:"context"`
	} `json:"locations"`
}

func decodeGroundingProjection(t *testing.T, content []byte) decodedGroundingProjection {
	t.Helper()
	var projection decodedGroundingProjection
	if err := json.Unmarshal(content, &projection); err != nil {
		t.Fatal(err)
	}
	return projection
}

func groundingDocument() *source.SourceDocument {
	return &source.SourceDocument{ID: "session-alpha", Units: []source.SourceUnit{
		{ID: 30, Text: "Tavern entrance"},
		{ID: 10, Text: "Tavern common room"},
		{ID: 20, Text: "Tavern cellar"},
		{ID: 40, Text: "Market square"},
	}}
}

func location(name string, refs []source.SourceRef) dnd.Location {
	return dnd.Location{ID: identity.DeriveID(name, refs), Name: name, SourceRefs: refs}
}