notarius/internal/modules/dnd/normalize/locationregistry/test_helpers_test.go

package locationregistry

import (
	"context"
	"encoding/json"
	"strconv"
	"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"
	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/entityreconcile"
)

type recordingLocationNormalizerClient struct {
	response string
	err      error
	requests []contracts.StructuredCompletionRequest
}

func (c *recordingLocationNormalizerClient) CompleteStructured(_ context.Context, request contracts.StructuredCompletionRequest, output any) (contracts.StructuredCompletionResponse, error) {
	c.requests = append(c.requests, request)
	if c.err != nil {
		return contracts.StructuredCompletionResponse{}, c.err
	}
	response := c.response
	if response == "" {
		response = `{"duplicate_groups":[]}`
	}
	content, err := contextualProposalResponse(response, request.Inputs["candidates"].Content)
	if err != nil {
		return contracts.StructuredCompletionResponse{}, err
	}
	if err := json.Unmarshal(content, output); err != nil {
		return contracts.StructuredCompletionResponse{}, err
	}
	return contracts.StructuredCompletionResponse{Content: content}, nil
}

func contextualProposalResponse(response string, candidateContent []byte) ([]byte, error) {
	if !strings.Contains(response, "candidate-") {
		return []byte(response), nil
	}
	var selection struct {
		DuplicateGroups []struct {
			Members   []string `json:"members"`
			Canonical string   `json:"canonical"`
		} `json:"duplicate_groups"`
	}
	if err := json.Unmarshal([]byte(response), &selection); err != nil {
		return nil, err
	}
	var candidates struct {
		Candidates []entityreconcile.Selector `json:"candidates"`
	}
	if err := json.Unmarshal(candidateContent, &candidates); err != nil {
		return nil, err
	}
	selector := func(key string) entityreconcile.Selector {
		index, err := strconv.Atoi(strings.TrimPrefix(key, "candidate-"))
		if err != nil || index < 1 || index > len(candidates.Candidates) {
			return entityreconcile.Selector{Name: key, SourceRefs: []entityreconcile.SourceRange{}}
		}
		return candidates.Candidates[index-1].Clone()
	}
	proposal := entityreconcile.ProposalResponse{DuplicateGroups: make([]entityreconcile.DuplicateGroup, len(selection.DuplicateGroups))}
	for index, group := range selection.DuplicateGroups {
		members := make([]entityreconcile.Selector, len(group.Members))
		for memberIndex, key := range group.Members {
			members[memberIndex] = selector(key)
		}
		proposal.DuplicateGroups[index] = entityreconcile.DuplicateGroup{Members: members, Canonical: selector(group.Canonical)}
	}
	return json.Marshal(proposal)
}

func newNormalizer(t *testing.T, client contracts.StructuredLLMClient) *Normalizer {
	t.Helper()
	normalizer, err := New(client, Options{})
	if err != nil {
		t.Fatalf("New() error = %v", err)
	}
	return normalizer
}
func normalizeRequest(value dnd.LocationRegistry) contracts.TypedNormalizeRequest[dnd.LocationRegistry] {
	return contracts.TypedNormalizeRequest[dnd.LocationRegistry]{MergeOutput: contracts.MergeArtifact[dnd.LocationRegistry]{Value: value}}
}
func normalizeRequestWithSource(value dnd.LocationRegistry, doc *source.SourceDocument) contracts.TypedNormalizeRequest[dnd.LocationRegistry] {
	request := normalizeRequest(value)
	request.Source = doc
	return request
}
func semanticDocument() *source.SourceDocument {
	return &source.SourceDocument{ID: "location-session", Units: []source.SourceUnit{{ID: 10, Kind: "speech", Text: "The old mill is the Greencloak's refuge."}, {ID: 20, Kind: "speech", Text: "The mill stands on the northern road."}, {ID: 30, Kind: "speech", Text: "The tavern is beside the mill."}, {ID: 40, Kind: "speech", Text: "The mill's cellar is flooded."}}}
}
func hasWarning(warnings []contracts.Warning, reason string) bool {
	for _, warning := range warnings {
		if warning.ReasonCode == reason {
			return true
		}
	}
	return false
}