scriptorium/internal/usecase/runner_test.go

package usecase

import (
	"context"
	"crypto/sha256"
	"encoding/hex"
	"errors"
	"fmt"
	"regexp"
	"strings"
	"testing"

	"gitea.maximumdirect.net/eric/scriptorium/internal/defaults"
	"gitea.maximumdirect.net/eric/scriptorium/internal/domain"
	"gitea.maximumdirect.net/eric/scriptorium/internal/validate"
)

type fakePromptRepo struct {
	def         *domain.PromptDefinition
	err         error
	lastID      string
	lastVersion string
}

type fakeExecutionProfileRepo struct {
	profiles map[string]*domain.ExecutionProfile
	err      error
	lastID   string
}

func (f *fakeExecutionProfileRepo) GetProfile(ctx context.Context, id string) (*domain.ExecutionProfile, error) {
	f.lastID = id
	if f.err != nil {
		return nil, f.err
	}
	if p, ok := f.profiles[id]; ok {
		cp := *p
		return &cp, nil
	}
	return nil, errors.New("profile not found")
}

func (f *fakePromptRepo) GetPromptDefinition(ctx context.Context, id string, version string) (*domain.PromptDefinition, error) {
	f.lastID = id
	f.lastVersion = version
	if f.err != nil {
		return nil, f.err
	}
	return f.def, nil
}

type fakeArtifactReader struct {
	artifactsByURI map[string]*domain.Artifact
	errByURI       map[string]error
}

func (f *fakeArtifactReader) Read(ctx context.Context, ref domain.ArtifactRef) (*domain.Artifact, error) {
	if err, ok := f.errByURI[ref.URI]; ok {
		return nil, err
	}
	if art, ok := f.artifactsByURI[ref.URI]; ok {
		cp := *art
		return &cp, nil
	}
	return nil, errors.New("artifact not found")
}

type fakeRenderer struct {
	rendered *domain.RenderedPrompt
	err      error
}

func (f *fakeRenderer) Render(ctx context.Context, def *domain.PromptDefinition, inputs map[string]*domain.Artifact, vars map[string]string) (*domain.RenderedPrompt, error) {
	if f.err != nil {
		return nil, f.err
	}
	return f.rendered, nil
}

type fakeLLM struct {
	resp    *domain.GenerateResponse
	err     error
	lastReq domain.GenerateRequest
}

func (f *fakeLLM) Generate(ctx context.Context, req domain.GenerateRequest) (*domain.GenerateResponse, error) {
	f.lastReq = req
	if f.err != nil {
		return nil, f.err
	}
	return f.resp, nil
}

type fakeValidator struct {
	result domain.ValidationResult
	err    error
}

func (f *fakeValidator) Validate(ctx context.Context, artifact *domain.Artifact, contract domain.OutputContract) (domain.ValidationResult, error) {
	if f.err != nil {
		return domain.ValidationResult{}, f.err
	}
	return f.result, nil
}

type fakeRepairer struct {
	responses []*domain.GenerateResponse
	err       error
	calls     int
	reqs      []RepairRequest
}

func (f *fakeRepairer) Repair(ctx context.Context, req RepairRequest) (*domain.GenerateResponse, error) {
	f.calls++
	f.reqs = append(f.reqs, req)
	if f.err != nil {
		return nil, f.err
	}
	if len(f.responses) == 0 {
		return nil, errors.New("no repair response configured")
	}
	idx := f.calls - 1
	if idx >= len(f.responses) {
		idx = len(f.responses) - 1
	}
	return f.responses[idx], nil
}

func TestRunnerRunSuccessful(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatMarkdown, domain.ValidationBasic, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}}
	reader := &fakeArtifactReader{artifactsByURI: map[string]*domain.Artifact{
		"a://t": {Body: []byte("transcript"), Hash: hashString("transcript")},
		"a://g": {Body: []byte("glossary"), Hash: hashString("glossary")},
	}}
	renderer := &fakeRenderer{rendered: &domain.RenderedPrompt{Messages: []domain.RenderedMessage{{Role: "system", Content: "sys"}, {Role: "user", Content: "usr"}}}}
	llmClient := &fakeLLM{resp: &domain.GenerateResponse{Content: "# recap", Usage: domain.TokenUsage{TotalTokens: 7}}}

	runner := NewRunner(promptRepo, execRepo, reader, renderer, llmClient, nil)
	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:      "p",
		PromptVersion: "1",
		ProfileID:     "exec",
		Inputs: map[string]domain.ArtifactRef{
			"transcript": {Type: domain.ArtifactRefFile, URI: "a://t"},
			"glossary":   {Type: domain.ArtifactRefFile, URI: "a://g"},
		},
		Execution: &domain.ExecutionTarget{Endpoint: "http://override/v1", Model: "m", Temperature: 0.3, TimeoutSeconds: 90},
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.PromptID != "p" || res.PromptVersion != "1" {
		t.Fatalf("unexpected prompt metadata: %+v", res)
	}
	if res.SelectedProfileID != "exec" {
		t.Fatalf("expected selected profile exec, got %q", res.SelectedProfileID)
	}
	if ok, _ := regexp.MatchString(`^[0-9a-f]{8}-[0-9a-f]{4}-4[0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$`, res.RunID); !ok {
		t.Fatalf("invalid run id: %q", res.RunID)
	}
	if res.PromptHash == "" || res.RenderedPromptHash == "" {
		t.Fatal("expected prompt hashes")
	}
	if res.EffectiveModelParams.Model != "m" || res.Endpoint != "http://override/v1" {
		t.Fatalf("unexpected model params: %+v", res.EffectiveModelParams)
	}
	if res.Artifact.Name != defaults.OutputArtifactName {
		t.Fatalf("expected default output artifact name %q, got %q", defaults.OutputArtifactName, res.Artifact.Name)
	}
	if res.Artifact.ContentType != defaults.ContentTypeTextMarkdown {
		t.Fatalf("expected markdown content type %q, got %q", defaults.ContentTypeTextMarkdown, res.Artifact.ContentType)
	}
	if res.RawOutput != "# recap" {
		t.Fatalf("expected raw output, got %q", res.RawOutput)
	}
	if res.Validation.Status != domain.ValidationSkipped {
		t.Fatalf("expected skipped validation, got %q", res.Validation.Status)
	}
	if llmClient.lastReq.Target.TimeoutSeconds != 90 {
		t.Fatalf("expected timeout propagation, got %d", llmClient.lastReq.Target.TimeoutSeconds)
	}
}

func TestRunnerRunExplicitProfileIDIsUsed(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	promptRepo.def.DefaultProfile = "default-prof"
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"explicit-prof": {ID: "explicit-prof", Endpoint: "http://explicit/v1", Model: "explicit"},
		"default-prof":  {ID: "default-prof", Endpoint: "http://default/v1", Model: "default"},
	}}

	runner := newMinimalRunner(promptRepo, execRepo)
	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "explicit-prof",
		Inputs:    singleInputRef(),
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if execRepo.lastID != "explicit-prof" {
		t.Fatalf("expected explicit profile lookup, got %q", execRepo.lastID)
	}
	if res.SelectedProfileID != "explicit-prof" {
		t.Fatalf("expected selected profile explicit-prof, got %q", res.SelectedProfileID)
	}
}

func TestRunnerRunPromptDefaultProfileIsUsedWhenNoExplicitProfileID(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	promptRepo.def.DefaultProfile = "from-prompt"
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"from-prompt": {ID: "from-prompt", Endpoint: "http://llm/v1", Model: "m"},
	}}

	runner := newMinimalRunner(promptRepo, execRepo)
	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID: "p",
		Inputs:   singleInputRef(),
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if execRepo.lastID != "from-prompt" {
		t.Fatalf("expected prompt default profile lookup, got %q", execRepo.lastID)
	}
	if res.SelectedProfileID != "from-prompt" {
		t.Fatalf("expected selected profile from-prompt, got %q", res.SelectedProfileID)
	}
}

func TestRunnerRunMissingExplicitProfileAndMissingDefaultProfileFails(t *testing.T) {
	repo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	repo.def.DefaultProfile = ""
	runner := newMinimalRunner(repo, &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}})
	_, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", Inputs: singleInputRef()})
	if !errors.Is(err, ErrInvalidRequest) {
		t.Fatalf("expected ErrInvalidRequest, got %v", err)
	}
}

func TestRunnerRunInvalidDefaultProfileFails(t *testing.T) {
	repo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	repo.def.DefaultProfile = "does-not-exist"
	runner := newMinimalRunner(repo, &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{}})
	_, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", Inputs: singleInputRef()})
	if !errors.Is(err, ErrProfileLoad) {
		t.Fatalf("expected ErrProfileLoad, got %v", err)
	}
}

func TestRunnerRunExecutionProfileLoadFailure(t *testing.T) {
	repo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	runner := newMinimalRunner(repo, &fakeExecutionProfileRepo{err: errors.New("load failed")})
	_, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", ProfileID: "exec", Inputs: singleInputRef()})
	if !errors.Is(err, ErrProfileLoad) {
		t.Fatalf("expected profile load failure, got %v", err)
	}
}

func TestRunnerRunExplicitRuntimeOverrideBeatsSelectedProfileValue(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {
			ID:             "exec",
			Endpoint:       "http://profile/v1",
			Model:          "profile-model",
			Temperature:    0.2,
			MaxTokens:      500,
			TopP:           0.9,
			TimeoutSeconds: 120,
		},
	}}
	llmClient := &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), llmClient, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
		Execution: &domain.ExecutionTarget{
			Endpoint:       "http://override/v1",
			Model:          "override-model",
			Temperature:    0.7,
			TimeoutSeconds: 30,
		},
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.Endpoint != "http://override/v1" || res.ModelName != "override-model" {
		t.Fatalf("expected endpoint/model override to win, got endpoint=%q model=%q", res.Endpoint, res.ModelName)
	}
	if res.EffectiveModelParams.Temperature != 0.7 || res.EffectiveModelParams.TimeoutSeconds != 30 {
		t.Fatalf("expected numeric override to win, got %+v", res.EffectiveModelParams)
	}
	if res.EffectiveModelParams.TopP != 0.9 {
		t.Fatalf("expected non-overridden profile top_p to remain, got %v", res.EffectiveModelParams.TopP)
	}
}

func TestRunnerRunSelectedProfileBeatsBuiltInDefault(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {
			ID:             "exec",
			Endpoint:       "http://profile/v1",
			Model:          "profile-model",
			TopP:           0.8,
			TimeoutSeconds: 90,
		},
	}}
	llmClient := &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), llmClient, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.TopP != 0.8 {
		t.Fatalf("expected profile top_p to beat default, got %v", res.EffectiveModelParams.TopP)
	}
	if res.EffectiveModelParams.TimeoutSeconds != 90 {
		t.Fatalf("expected profile timeout to beat default, got %d", res.EffectiveModelParams.TimeoutSeconds)
	}
}

func TestRunnerRunBuiltInDefaultsUsedWhenProfileOmitsOptionalFields(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model"},
	}}
	llmClient := &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), llmClient, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.Temperature != defaults.ExecutionDefaultTemperature {
		t.Fatalf("expected default temperature %v, got %v", defaults.ExecutionDefaultTemperature, res.EffectiveModelParams.Temperature)
	}
	if res.EffectiveModelParams.TopP != defaults.ExecutionDefaultTopP {
		t.Fatalf("expected default top_p %v, got %v", defaults.ExecutionDefaultTopP, res.EffectiveModelParams.TopP)
	}
	if res.EffectiveModelParams.MaxTokens != defaults.ExecutionDefaultMaxTokens {
		t.Fatalf("expected default max_tokens %d, got %d", defaults.ExecutionDefaultMaxTokens, res.EffectiveModelParams.MaxTokens)
	}
	if res.EffectiveModelParams.TimeoutSeconds != defaults.ExecutionDefaultTimeoutSeconds {
		t.Fatalf("expected default timeout_seconds %d, got %d", defaults.ExecutionDefaultTimeoutSeconds, res.EffectiveModelParams.TimeoutSeconds)
	}
}

func TestRunnerRunAPIKeyEnvResolvesFromEnvironment(t *testing.T) {
	t.Setenv("SCRIPTORIUM_TEST_API_KEY", "secret")
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model", APIKeyEnv: "SCRIPTORIUM_TEST_API_KEY"},
	}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", ProfileID: "exec", Inputs: singleInputRef()})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.APIKeyEnv != "SCRIPTORIUM_TEST_API_KEY" {
		t.Fatalf("expected api_key_env name in effective params, got %q", res.EffectiveModelParams.APIKeyEnv)
	}
}

func TestRunnerRunAPIKeyEnvMissingEnvironmentValueFailsClearly(t *testing.T) {
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model", APIKeyEnv: "SCRIPTORIUM_MISSING_KEY"},
	}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}, nil)

	_, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", ProfileID: "exec", Inputs: singleInputRef()})
	if !errors.Is(err, ErrInvalidRequest) {
		t.Fatalf("expected ErrInvalidRequest, got %v", err)
	}
	if !strings.Contains(err.Error(), "SCRIPTORIUM_MISSING_KEY") {
		t.Fatalf("expected missing env name in error, got %v", err)
	}
}

func TestRunnerRunRuntimeAPIKeyEnvOverrideWorks(t *testing.T) {
	const envName = "SCRIPTORIUM_RUNTIME_API_KEY"
	t.Setenv(envName, "runtime-secret")

	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model"},
	}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
		Execution: &domain.ExecutionTarget{APIKeyEnv: envName},
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.APIKeyEnv != envName {
		t.Fatalf("expected runtime api_key_env override in effective params, got %q", res.EffectiveModelParams.APIKeyEnv)
	}
}

func TestRunnerRunRuntimeAPIKeyEnvOverrideBeatsProfile(t *testing.T) {
	const profileEnv = "SCRIPTORIUM_PROFILE_API_KEY"
	const runtimeEnv = "SCRIPTORIUM_RUNTIME_API_KEY"
	t.Setenv(runtimeEnv, "runtime-secret")

	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model", APIKeyEnv: profileEnv},
	}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
		Execution: &domain.ExecutionTarget{APIKeyEnv: runtimeEnv},
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.APIKeyEnv != runtimeEnv {
		t.Fatalf("expected runtime override to beat profile api_key_env, got %q", res.EffectiveModelParams.APIKeyEnv)
	}
}

func TestRunnerRunAPIKeyValueNotPresentInMetadata(t *testing.T) {
	const envName = "SCRIPTORIUM_TEST_API_KEY"
	const secret = "top-secret-value"
	t.Setenv(envName, secret)
	promptRepo := &fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)}
	execRepo := &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
		"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model", APIKeyEnv: envName},
	}}
	runner := NewRunner(promptRepo, execRepo, defaultArtifactReader(), defaultRenderer(), &fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}}, nil)

	res, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p", ProfileID: "exec", Inputs: singleInputRef()})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.EffectiveModelParams.APIKeyEnv != envName {
		t.Fatalf("expected api key env name, got %q", res.EffectiveModelParams.APIKeyEnv)
	}
	metadataDump := fmt.Sprintf("%+v|%s|%s|%s|%s", res.EffectiveModelParams, res.Endpoint, res.ModelName, res.PromptHash, res.RenderedPromptHash)
	if strings.Contains(metadataDump, secret) {
		t.Fatalf("unexpected api key value in metadata dump: %s", metadataDump)
	}
}

func TestRunnerRunPromptLoadFailure(t *testing.T) {
	runner := NewRunner(&fakePromptRepo{err: errors.New("boom")}, &fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}}, defaultArtifactReader(), defaultRenderer(), &fakeLLM{}, nil)
	_, err := runner.Run(context.Background(), domain.RunRequest{PromptID: "p"})
	if !errors.Is(err, ErrProfileLoad) {
		t.Fatalf("expected ErrProfileLoad, got %v", err)
	}
}

func TestRunnerRunArtifactLoadFailure(t *testing.T) {
	runner := NewRunner(
		&fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)},
		&fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}},
		&fakeArtifactReader{errByURI: map[string]error{"a://bad": errors.New("read failed")}},
		&fakeRenderer{rendered: &domain.RenderedPrompt{}},
		&fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}},
		nil,
	)

	_, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    map[string]domain.ArtifactRef{"transcript": {Type: domain.ArtifactRefFile, URI: "a://bad"}},
	})
	if !errors.Is(err, ErrArtifactLoad) {
		t.Fatalf("expected ErrArtifactLoad, got %v", err)
	}
}

func TestRunnerRunPromptRenderFailure(t *testing.T) {
	runner := NewRunner(
		&fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)},
		&fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}},
		defaultArtifactReader(),
		&fakeRenderer{err: errors.New("render failed")},
		&fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}},
		nil,
	)
	_, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
	})
	if !errors.Is(err, ErrPromptRender) {
		t.Fatalf("expected ErrPromptRender, got %v", err)
	}
}

func TestRunnerRunLLMFailure(t *testing.T) {
	runner := NewRunner(
		&fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationNone, 0)},
		&fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}},
		defaultArtifactReader(),
		defaultRenderer(),
		&fakeLLM{err: errors.New("llm failed")},
		nil,
	)
	_, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
	})
	if !errors.Is(err, ErrLLMGenerate) {
		t.Fatalf("expected ErrLLMGenerate, got %v", err)
	}
}

func TestRunnerRunValidationStillWorks(t *testing.T) {
	validator := &fakeValidator{result: domain.ValidationResult{Status: domain.ValidationFailed, Mode: domain.ValidationBasic, Errors: []string{"bad"}, IsValid: false}}
	runner := NewRunner(
		&fakePromptRepo{def: promptDef(domain.FormatText, domain.ValidationBasic, 0)},
		&fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{"exec": defaultExecutionProfile()}},
		defaultArtifactReader(),
		defaultRenderer(),
		&fakeLLM{resp: &domain.GenerateResponse{Content: "raw output"}},
		validator,
	)
	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if res.Validation.Status != domain.ValidationFailed || res.RawOutput != "raw output" {
		t.Fatalf("unexpected validation/raw output: %+v", res)
	}
}

func TestRunnerRunStructuredRepairRemainsBoundedAndUsesEffectiveModelSettings(t *testing.T) {
	repairer := &fakeRepairer{responses: []*domain.GenerateResponse{{Content: `{"broken":`}, {Content: `{"still":`}}}
	llmClient := &fakeLLM{resp: &domain.GenerateResponse{Content: `{"initial":`}}

	runner := NewRunnerWithRepairer(
		&fakePromptRepo{def: promptDef(domain.FormatJSON, domain.ValidationJSON, 1)},
		&fakeExecutionProfileRepo{profiles: map[string]*domain.ExecutionProfile{
			"exec": {ID: "exec", Endpoint: "http://profile/v1", Model: "profile-model", TimeoutSeconds: 55},
		}},
		defaultArtifactReader(),
		defaultRenderer(),
		llmClient,
		validate.NewStandardValidator("."),
		repairer,
	)
	res, err := runner.Run(context.Background(), domain.RunRequest{
		PromptID:  "p",
		ProfileID: "exec",
		Inputs:    singleInputRef(),
		Execution: &domain.ExecutionTarget{Endpoint: "http://override/v1", Model: "override-model", TimeoutSeconds: 22},
	})
	if err != nil {
		t.Fatalf("expected no error, got %v", err)
	}
	if repairer.calls != 1 || res.Validation.RepairAttempts != 1 {
		t.Fatalf("expected one bounded repair, calls=%d attempts=%d", repairer.calls, res.Validation.RepairAttempts)
	}
	if len(repairer.reqs) != 1 {
		t.Fatalf("expected one repair request, got %d", len(repairer.reqs))
	}
	if repairer.reqs[0].Target.Endpoint != "http://override/v1" || repairer.reqs[0].Target.Model != "override-model" {
		t.Fatalf("expected repair to use effective target, got %+v", repairer.reqs[0].Target)
	}
	if repairer.reqs[0].Target.TimeoutSeconds != 22 {
		t.Fatalf("expected repair to use effective timeout, got %d", repairer.reqs[0].Target.TimeoutSeconds)
	}
}

func TestBuildOutputArtifactDefaults(t *testing.T) {
	tests := []struct {
		name        string
		format      domain.OutputFormat
		contentType string
	}{
		{name: "text", format: domain.FormatText, contentType: defaults.ContentTypeTextPlain},
		{name: "markdown", format: domain.FormatMarkdown, contentType: defaults.ContentTypeTextMarkdown},
		{name: "json", format: domain.FormatJSON, contentType: defaults.ContentTypeApplicationJSON},
	}

	for _, tc := range tests {
		t.Run(tc.name, func(t *testing.T) {
			art := buildOutputArtifact("body", tc.format)
			if art.Name != defaults.OutputArtifactName {
				t.Fatalf("expected artifact name %q, got %q", defaults.OutputArtifactName, art.Name)
			}
			if art.ContentType != tc.contentType {
				t.Fatalf("expected content type %q, got %q", tc.contentType, art.ContentType)
			}
		})
	}
}

func promptDef(format domain.OutputFormat, mode domain.ValidationMode, attempts int) *domain.PromptDefinition {
	return &domain.PromptDefinition{
		ID:             "p",
		Version:        "1",
		DefaultProfile: "exec",
		Inputs:         []domain.PromptInput{{Name: "transcript", Required: true}},
		Templates:      []domain.PromptMessageTemplate{{Role: "user", Content: "x"}},
		OutputFormat:   format,
		Validation: domain.OutputContract{
			ValidationMode: mode,
			RepairAttempts: attempts,
			Format:         format,
		},
	}
}

func hashString(s string) string {
	sum := sha256.Sum256([]byte(s))
	return hex.EncodeToString(sum[:])
}

func defaultExecutionProfile() *domain.ExecutionProfile {
	return &domain.ExecutionProfile{
		ID:       "exec",
		Endpoint: "http://llm/v1",
		Model:    "model-from-profile",
	}
}

func defaultArtifactReader() *fakeArtifactReader {
	return &fakeArtifactReader{artifactsByURI: map[string]*domain.Artifact{
		"a://ok": {Body: []byte("x"), Hash: hashString("x")},
	}}
}

func defaultRenderer() *fakeRenderer {
	return &fakeRenderer{rendered: &domain.RenderedPrompt{Messages: []domain.RenderedMessage{{Role: "user", Content: "hello"}}}}
}

func singleInputRef() map[string]domain.ArtifactRef {
	return map[string]domain.ArtifactRef{"transcript": {Type: domain.ArtifactRefFile, URI: "a://ok"}}
}

func newMinimalRunner(promptRepo *fakePromptRepo, execRepo *fakeExecutionProfileRepo) *Runner {
	return NewRunner(
		promptRepo,
		execRepo,
		defaultArtifactReader(),
		defaultRenderer(),
		&fakeLLM{resp: &domain.GenerateResponse{Content: "ok"}},
		nil,
	)
}