audita/internal/cli/run_test.go

package cli

import (
	"bytes"
	"encoding/json"
	"io"
	"os"
	"path/filepath"
	"reflect"
	"strings"
	"testing"
)

func TestRunRootHelp(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"--help"}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d", exitCode)
	}
	if !strings.Contains(stdout.String(), "audita <command> [options]") {
		t.Fatalf("expected root usage in stdout, got %q", stdout.String())
	}
	if !strings.Contains(stdout.String(), "process") {
		t.Fatalf("expected process command in root help, got %q", stdout.String())
	}
	if stderr.Len() != 0 {
		t.Fatalf("expected empty stderr, got %q", stderr.String())
	}
}

func TestRunProcessHelpListsExpectedFlags(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"process", "--help"}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d", exitCode)
	}

	for _, expectedFlag := range []string{
		"--glossary",
		"--output",
		"--report-json",
		"--modules",
		"--llm-api-key",
		"--validation-llm-api-key",
		"--model",
		"--validation-model",
		"--base-url",
		"--validation-base-url",
		"--llm-timeout-seconds",
		"--validation-llm-timeout-seconds",
		"--validation-max-prompt-tokens",
		"--target-sections",
		"--max-retries",
		"--validation-max-retries",
		"--validation-llm-concurrency",
		"--max-section-tokens",
		"--min-section-tokens",
		"--glossary-confidence-threshold",
		"--grammar-confidence-threshold",
		"--homophones-confidence-threshold",
		"--spoken-word-confidence-threshold",
		"--normalize-max-segment-gap",
		"--normalize-ellipsis-gap",
		"--normalize-max-segment-duration",
		"--normalize-max-segment-tokens",
		"--work-dir",
		"--work-dir-retention",
	} {
		if !strings.Contains(stdout.String(), expectedFlag) {
			t.Fatalf("expected process help to include %q, got %q", expectedFlag, stdout.String())
		}
	}

	if stderr.Len() != 0 {
		t.Fatalf("expected empty stderr, got %q", stderr.String())
	}
}

func TestRunUnknownCommand(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"unknown"}, &stdout, &stderr)
	if exitCode != 2 {
		t.Fatalf("expected exit code 2, got %d", exitCode)
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout, got %q", stdout.String())
	}
	if !strings.Contains(stderr.String(), "unknown command") {
		t.Fatalf("expected unknown command error in stderr, got %q", stderr.String())
	}
}

func TestRunProcessMissingTranscriptPath(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"process", "--glossary", fixturePath("tiny_glossary.yaml")}, &stdout, &stderr)
	if exitCode == 0 {
		t.Fatalf("expected nonzero exit code")
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout, got %q", stdout.String())
	}
	if !strings.Contains(stderr.String(), "expected exactly 1 transcript JSON path argument") {
		t.Fatalf("expected missing transcript error, got %q", stderr.String())
	}
}

func TestRunProcessMissingGlossary(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"process", fixturePath("tiny_transcript.json")}, &stdout, &stderr)
	if exitCode == 0 {
		t.Fatalf("expected nonzero exit code")
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout, got %q", stdout.String())
	}
	if !strings.Contains(stderr.String(), "--glossary is required") {
		t.Fatalf("expected missing glossary error, got %q", stderr.String())
	}
}

func TestRunProcessUnreadableTranscript(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	missingTranscript := filepath.Join(t.TempDir(), "missing.json")
	exitCode := Run([]string{"process", missingTranscript, "--glossary", fixturePath("tiny_glossary.yaml")}, &stdout, &stderr)
	if exitCode == 0 {
		t.Fatalf("expected nonzero exit code")
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout, got %q", stdout.String())
	}
	if !strings.Contains(stderr.String(), "failed to read transcript file") {
		t.Fatalf("expected unreadable transcript error, got %q", stderr.String())
	}
}

func TestRunProcessMalformedTranscriptJSON(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	exitCode := Run([]string{"process", fixturePath("malformed_transcript.json"), "--glossary", fixturePath("tiny_glossary.yaml")}, &stdout, &stderr)
	if exitCode == 0 {
		t.Fatalf("expected nonzero exit code")
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout, got %q", stdout.String())
	}
	if !strings.Contains(stderr.String(), "transcript file is not valid JSON") {
		t.Fatalf("expected malformed transcript error, got %q", stderr.String())
	}
}

func TestRunProcessOutputToFile(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	transcriptPath := fixturePath("tiny_transcript.json")
	glossaryPath := fixturePath("tiny_glossary.yaml")
	outputPath := filepath.Join(t.TempDir(), "corrected.json")

	exitCode := Run([]string{"process", transcriptPath, "--glossary", glossaryPath, "--output", outputPath}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d with stderr %q", exitCode, stderr.String())
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout when --output is used, got %q", stdout.String())
	}
	if stderr.Len() != 0 {
		t.Fatalf("expected empty stderr on success, got %q", stderr.String())
	}

	inputBytes := readFile(t, transcriptPath)
	outputBytes := readFile(t, outputPath)
	assertJSONSemanticallyEqual(t, inputBytes, outputBytes)
}

func TestRunProcessOutputToStdout(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	transcriptPath := fixturePath("tiny_transcript.json")
	glossaryPath := fixturePath("tiny_glossary.yaml")

	exitCode := Run([]string{"process", transcriptPath, "--glossary", glossaryPath}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d with stderr %q", exitCode, stderr.String())
	}
	if stderr.Len() != 0 {
		t.Fatalf("expected empty stderr on success, got %q", stderr.String())
	}

	inputBytes := readFile(t, transcriptPath)
	assertJSONSemanticallyEqual(t, inputBytes, stdout.Bytes())
}

func TestRunProcessReportJSONSuccess(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	transcriptPath := fixturePath("tiny_transcript.json")
	glossaryPath := fixturePath("tiny_glossary.yaml")
	outputPath := filepath.Join(t.TempDir(), "corrected.json")
	reportPath := filepath.Join(t.TempDir(), "report.json")

	exitCode := Run([]string{
		"process",
		transcriptPath,
		"--glossary",
		glossaryPath,
		"--output",
		outputPath,
		"--report-json",
		reportPath,
	}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d with stderr %q", exitCode, stderr.String())
	}
	if stdout.Len() != 0 {
		t.Fatalf("expected empty stdout when --output is used, got %q", stdout.String())
	}

	report := readProcessReport(t, reportPath)
	if report.Status != "success" {
		t.Fatalf("expected success report status, got %q", report.Status)
	}
	if report.Operation != "process" {
		t.Fatalf("expected operation process, got %q", report.Operation)
	}
	if report.TranscriptPath != transcriptPath {
		t.Fatalf("unexpected transcript path in report: %q", report.TranscriptPath)
	}
	if report.GlossaryPath != glossaryPath {
		t.Fatalf("unexpected glossary path in report: %q", report.GlossaryPath)
	}
	if report.OutputPath != outputPath {
		t.Fatalf("unexpected output path in report: %q", report.OutputPath)
	}
	if len(report.Modules) == 0 {
		t.Fatalf("expected configured module sequence in report")
	}
	if report.StartedAt == "" || report.CompletedAt == "" {
		t.Fatalf("expected started_at and completed_at timestamps in report")
	}
	if report.ErrorMessage != "" {
		t.Fatalf("did not expect error message in success report, got %q", report.ErrorMessage)
	}
}

func TestRunProcessReportJSONNoStdoutPollution(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	transcriptPath := fixturePath("tiny_transcript.json")
	glossaryPath := fixturePath("tiny_glossary.yaml")
	reportPath := filepath.Join(t.TempDir(), "report.json")

	exitCode := Run([]string{
		"process",
		transcriptPath,
		"--glossary",
		glossaryPath,
		"--report-json",
		reportPath,
	}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d with stderr %q", exitCode, stderr.String())
	}
	assertJSONSemanticallyEqual(t, readFile(t, transcriptPath), stdout.Bytes())

	reportBytes := readFile(t, reportPath)
	if bytes.Contains(stdout.Bytes(), reportBytes) {
		t.Fatalf("stdout was polluted with report JSON")
	}
}

func TestRunProcessReportJSONRedactsSecrets(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	secretPrimary := "top-secret-primary-key"
	secretValidation := "top-secret-validation-key"
	t.Setenv("AUDITA_LLM_API_KEY", secretPrimary)
	t.Setenv("AUDITA_VALIDATION_LLM_API_KEY", secretValidation)

	reportPath := filepath.Join(t.TempDir(), "report.json")
	exitCode := Run([]string{
		"process",
		fixturePath("tiny_transcript.json"),
		"--glossary",
		fixturePath("tiny_glossary.yaml"),
		"--report-json",
		reportPath,
	}, &stdout, &stderr)
	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d with stderr %q", exitCode, stderr.String())
	}

	reportBytes := readFile(t, reportPath)
	if strings.Contains(string(reportBytes), secretPrimary) || strings.Contains(string(reportBytes), secretValidation) {
		t.Fatalf("report JSON leaked API key material: %q", string(reportBytes))
	}
}

func TestRunProcessReportJSONBestEffortOnFailure(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	reportPath := filepath.Join(t.TempDir(), "report.json")
	exitCode := Run([]string{
		"process",
		fixturePath("malformed_transcript.json"),
		"--glossary",
		fixturePath("tiny_glossary.yaml"),
		"--report-json",
		reportPath,
	}, &stdout, &stderr)
	if exitCode == 0 {
		t.Fatalf("expected nonzero exit code for malformed transcript")
	}

	report := readProcessReport(t, reportPath)
	if report.Status != "failed" {
		t.Fatalf("expected failed report status, got %q", report.Status)
	}
	if report.ErrorMessage == "" {
		t.Fatalf("expected error message in failed report")
	}
	if !strings.Contains(report.ErrorMessage, "not valid JSON") {
		t.Fatalf("expected malformed JSON failure in report error, got %q", report.ErrorMessage)
	}
}

func TestRunProcessCLIOverridesEnvironment(t *testing.T) {
	var stdout bytes.Buffer
	var stderr bytes.Buffer

	t.Setenv("AUDITA_MODEL", "env-model")
	t.Setenv("AUDITA_VALIDATION_LLM_CONCURRENCY", "2")

	transcriptPath := fixturePath("tiny_transcript.json")
	glossaryPath := fixturePath("tiny_glossary.yaml")

	var captured processInvocation
	originalRunner := processRunner
	processRunner = func(inv processInvocation, output io.Writer) error {
		_ = output
		captured = inv
		return nil
	}
	t.Cleanup(func() {
		processRunner = originalRunner
	})

	exitCode := Run([]string{
		"process",
		transcriptPath,
		"--glossary",
		glossaryPath,
		"--model",
		"cli-model",
		"--validation-llm-concurrency",
		"5",
	}, &stdout, &stderr)

	if exitCode != 0 {
		t.Fatalf("expected exit code 0, got %d", exitCode)
	}
	if captured.Config.PrimaryLLM.Model != "cli-model" {
		t.Fatalf("expected CLI model override, got %q", captured.Config.PrimaryLLM.Model)
	}
	if captured.Config.ValidationLLM.Concurrency == nil || *captured.Config.ValidationLLM.Concurrency != 5 {
		t.Fatalf("expected CLI validation concurrency override, got %#v", captured.Config.ValidationLLM.Concurrency)
	}
	if captured.GlossaryPath != glossaryPath {
		t.Fatalf("unexpected glossary path: %q", captured.GlossaryPath)
	}
}

func fixturePath(name string) string {
	return filepath.Join("testdata", name)
}

func readFile(t *testing.T, path string) []byte {
	t.Helper()
	data, err := os.ReadFile(path)
	if err != nil {
		t.Fatalf("failed to read file %q: %v", path, err)
	}
	return data
}

func assertJSONSemanticallyEqual(t *testing.T, expected []byte, actual []byte) {
	t.Helper()
	if !json.Valid(actual) {
		t.Fatalf("actual output is not valid JSON: %q", string(actual))
	}

	var expectedValue any
	var actualValue any
	if err := json.Unmarshal(expected, &expectedValue); err != nil {
		t.Fatalf("failed to unmarshal expected JSON: %v", err)
	}
	if err := json.Unmarshal(actual, &actualValue); err != nil {
		t.Fatalf("failed to unmarshal actual JSON: %v", err)
	}
	if !reflect.DeepEqual(expectedValue, actualValue) {
		t.Fatalf("JSON content mismatch: expected %q got %q", string(expected), string(actual))
	}
}

type minimalProcessReport struct {
	Status         string   `json:"status"`
	Operation      string   `json:"operation"`
	TranscriptPath string   `json:"transcript_path"`
	GlossaryPath   string   `json:"glossary_path"`
	OutputPath     string   `json:"output_path"`
	Modules        []string `json:"modules"`
	StartedAt      string   `json:"started_at"`
	CompletedAt    string   `json:"completed_at"`
	ErrorMessage   string   `json:"error_message"`
}

func readProcessReport(t *testing.T, path string) minimalProcessReport {
	t.Helper()
	raw := readFile(t, path)
	var report minimalProcessReport
	if err := json.Unmarshal(raw, &report); err != nil {
		t.Fatalf("failed to unmarshal process report JSON: %v (raw: %q)", err, string(raw))
	}
	return report
}