descrybe/apps/api/internal/processing/categorize_ai_test.go

package processing

import (
	"context"
	"strings"
	"testing"
)

func TestTryAICategorize_picksTaxonomyUniqueID(t *testing.T) {
	e := &Engine{
		Completer: stubCompleter{fn: func(system, user string) (Completion, error) {
			if strings.Contains(strings.ToLower(system), "categoryid") {
				if !strings.Contains(user, "Available Categories:") {
					t.Fatalf("expected Available Categories in user prompt")
				}
				if !strings.Contains(user, "(ID: 50)") {
					t.Fatalf("expected ID 50 in list: %s", user)
				}
				return Completion{
					Text:         `{"categoryId":"50","confidence":0.91}`,
					TotalTokens:  12,
					PromptTokens: 8,
					OutputTokens: 4,
				}, nil
			}
			return Completion{Text: `{"name":"Gorenje Cooker","description":"Freestanding cooker for the kitchen."}`, TotalTokens: 5}, nil
		}},
		Vector: NoopVectorCategorizer{},
	}
	out, err := e.RunSteps(context.Background(), "co", ProductInput{
		Name:        "Gorenje stove",
		Description: "Freestanding cooker",
		CategoryNamesByUID: map[string]string{
			"28": "TV",
			"50": "Štedilniki",
		},
	}, "full", nil, StepPolicy{AllowAI: true, AllowEPREL: true})
	if err != nil {
		t.Fatal(err)
	}
	if out.Category != "50" {
		t.Fatalf("Category=%q want 50", out.Category)
	}
	if out.CategoryName != "Štedilniki" {
		t.Fatalf("CategoryName=%q want Štedilniki", out.CategoryName)
	}
	if src, _ := out.FieldSources["category"].(string); src != "llm" {
		t.Fatalf("field_sources.category=%v want llm", out.FieldSources["category"])
	}
	if out.TotalTokens < 12 {
		t.Fatalf("TotalTokens=%d want >=12 from categorize", out.TotalTokens)
	}
}

func TestTryAICategorize_rejectsInventedID(t *testing.T) {
	e := &Engine{
		Completer: stubCompleter{fn: func(system, user string) (Completion, error) {
			if strings.Contains(strings.ToLower(system), "categoryid") {
				return Completion{Text: `{"categoryId":"99999","confidence":0.9}`, TotalTokens: 3}, nil
			}
			return Completion{Text: `{"name":"X","description":"Y"}`, TotalTokens: 5}, nil
		}},
		Vector: NoopVectorCategorizer{},
	}
	out, err := e.RunSteps(context.Background(), "co", ProductInput{
		Name: "Widget",
		CategoryNamesByUID: map[string]string{
			"50": "Štedilniki",
		},
	}, "full", nil, StepPolicy{AllowAI: true})
	if err != nil {
		t.Fatal(err)
	}
	if out.Category != "" {
		t.Fatalf("Category=%q want empty (invented id rejected)", out.Category)
	}
	found := false
	for _, n := range out.Notes {
		if strings.Contains(n, "ai_categorize: ignored") {
			found = true
			break
		}
	}
	if !found {
		t.Fatalf("expected reject note, got %v", out.Notes)
	}
}

func TestTryAICategorize_resolvesDisplayName(t *testing.T) {
	e := &Engine{
		Completer: stubCompleter{fn: func(system, user string) (Completion, error) {
			if strings.Contains(strings.ToLower(system), "categoryid") {
				return Completion{Text: `{"categoryId":"Štedilniki","confidence":0.8}`, TotalTokens: 3}, nil
			}
			return Completion{Text: `{"name":"X","description":"Y"}`, TotalTokens: 5}, nil
		}},
		Vector: NoopVectorCategorizer{},
	}
	out, err := e.RunSteps(context.Background(), "co", ProductInput{
		Name: "Cooker",
		CategoryNamesByUID: map[string]string{
			"50": "Štedilniki",
		},
	}, "full", nil, StepPolicy{AllowAI: true})
	if err != nil {
		t.Fatal(err)
	}
	if out.Category != "50" {
		t.Fatalf("Category=%q want 50 (name coerced)", out.Category)
	}
}

func TestTryAICategorize_skippedWithoutTaxonomy(t *testing.T) {
	calls := 0
	e := &Engine{
		Completer: stubCompleter{fn: func(system, user string) (Completion, error) {
			calls++
			return Completion{Text: `{"name":"A","description":"B"}`, TotalTokens: 2}, nil
		}},
		Vector: NoopVectorCategorizer{},
	}
	out, err := e.RunSteps(context.Background(), "co", ProductInput{
		Name: "Lone product",
	}, "full", nil, StepPolicy{AllowAI: true})
	if err != nil {
		t.Fatal(err)
	}
	if out.Category != "" {
		t.Fatalf("Category=%q want empty", out.Category)
	}
	// Only enhance should call Completer (categorize skips with no taxonomy).
	if calls != 1 {
		t.Fatalf("completer calls=%d want 1 (enhance only)", calls)
	}
}

func TestHeuristicCompleter_categorizeReturnsJSON(t *testing.T) {
	h := HeuristicCompleter{}
	user := ProductCategorizeUser("Stove", "Cooker", []categoryOption{
		{UniqueID: "50", Name: "Štedilniki"},
		{UniqueID: "28", Name: "TV"},
	})
	comp, err := h.Complete(context.Background(), ProductCategorizeSystem, user)
	if err != nil {
		t.Fatal(err)
	}
	obj, err := ParseJSONObject(comp.Text)
	if err != nil {
		t.Fatal(err)
	}
	got := categoryIDFromCategorizeJSON(obj)
	if got != "50" && got != "28" {
		t.Fatalf("categoryId=%q want 50 or 28 from Available Categories", got)
	}
	opts := categoryOptionsFromNames(map[string]string{"50": "Štedilniki", "28": "TV"})
	found := false
	for _, o := range opts {
		if o.UniqueID == got {
			found = true
			break
		}
	}
	if !found {
		t.Fatalf("categoryId=%q not in options", got)
	}
}

func TestCategoryOptionsFromNames_sortAndCap(t *testing.T) {
	names := map[string]string{"2": "Beta", "1": "Alpha", "3": "Gamma"}
	opts := categoryOptionsFromNames(names)
	if len(opts) != 3 || opts[0].UniqueID != "1" || opts[1].UniqueID != "2" {
		t.Fatalf("opts=%v", opts)
	}
}