narratio/internal/config/validate.go

package config

import (
	"fmt"
	"net/url"
	"path/filepath"
	"regexp"
	"strings"
	"time"
)

// Validate checks resolved configuration for required fields and parseable durations.
func Validate(cfg *Config) error {
	if cfg == nil {
		return fmt.Errorf("config is nil")
	}
	if cfg.Pipeline == nil {
		return fmt.Errorf("pipeline config is required")
	}
	if cfg.Session == nil {
		return fmt.Errorf("session config is required")
	}

	if err := validatePipeline(cfg.Pipeline); err != nil {
		return fmt.Errorf("pipeline config %q invalid: %w", shortName(cfg.PipelinePath, "pipeline.yml"), err)
	}
	if err := validateSession(cfg.Session); err != nil {
		return fmt.Errorf("session config %q invalid: %w", shortName(cfg.SessionPath, "session.yml"), err)
	}
	if err := validateCrossConfig(cfg.Pipeline, cfg.Session); err != nil {
		return fmt.Errorf("pipeline/session config invalid: %w", err)
	}

	return nil
}

func validatePipeline(cfg *PipelineConfig) error {
	if strings.TrimSpace(cfg.Workspace.Root) == "" {
		return fmt.Errorf("pipeline.workspace.root is required")
	}
	if err := validateSecrets(cfg.Secrets); err != nil {
		return err
	}
	if err := validateStorage(cfg.Storage); err != nil {
		return err
	}
	if err := validateSpool(cfg.Spool); err != nil {
		return err
	}
	if err := validateArchive(cfg.Archive); err != nil {
		return err
	}
	if err := validateWhisperX(cfg.WhisperX); err != nil {
		return err
	}
	if err := validateSeriatim(cfg.Seriatim); err != nil {
		return err
	}
	if err := validateAudita(cfg.Audita); err != nil {
		return err
	}
	if err := validateNormalize(cfg.Normalize); err != nil {
		return err
	}
	if err := validateTrim(cfg.Trim); err != nil {
		return err
	}
	if err := validateScriptorium(cfg.Scriptorium); err != nil {
		return err
	}
	if err := validateDuration("pipeline.analyzer.timeout", cfg.Analyzer.Timeout); err != nil {
		return err
	}
	if err := validateDuration("pipeline.notification.timeout", cfg.Notification.Timeout); err != nil {
		return err
	}

	return nil
}

func validateStorage(cfg StorageConfig) error {
	if cfg.S3 == nil {
		return nil
	}
	if strings.TrimSpace(cfg.S3.RootPrefix) == "" {
		return fmt.Errorf("pipeline.storage.s3.root_prefix must be non-empty")
	}
	if err := validateRelativeSafePath("pipeline.storage.s3.root_prefix", cfg.S3.RootPrefix); err != nil {
		return err
	}
	if cfg.S3.Endpoint != "" && strings.TrimSpace(cfg.S3.Endpoint) == "" {
		return fmt.Errorf("pipeline.storage.s3.endpoint must be non-empty when provided")
	}
	return nil
}

func validateSpool(cfg SpoolConfig) error {
	return nil
}

func validateArchive(cfg *ArchiveConfig) error {
	if cfg == nil {
		return nil
	}
	for i, item := range cfg.PromoteArtifacts {
		prefix := fmt.Sprintf("pipeline.archive.promote_artifacts[%d]", i)
		if strings.TrimSpace(item.From) == "" {
			return fmt.Errorf("%s.from is required", prefix)
		}
		if strings.TrimSpace(item.To) == "" {
			return fmt.Errorf("%s.to is required", prefix)
		}
		if err := validateRelativeSafePath(prefix+".from", item.From); err != nil {
			return err
		}
		if err := validateRelativeSafePath(prefix+".to", item.To); err != nil {
			return err
		}
	}
	return nil
}

func validateSecrets(cfg *SecretsConfig) error {
	if cfg == nil {
		return nil
	}
	if strings.TrimSpace(cfg.EnvDir) == "" {
		return fmt.Errorf("pipeline.secrets.env_dir must be non-empty when pipeline.secrets is configured")
	}
	return nil
}

func validateNormalize(cfg *NormalizeConfig) error {
	if cfg == nil {
		return nil
	}
	if strings.TrimSpace(cfg.OutputPath) == "" {
		return fmt.Errorf("pipeline.normalize.output_path must be non-empty")
	}
	if err := validateSeriatimOutputSchema("pipeline.normalize.output_schema", cfg.OutputSchema); err != nil {
		return err
	}
	return nil
}

func validateTrim(cfg *TrimConfig) error {
	if cfg == nil {
		return nil
	}
	if !cfg.Enabled {
		return nil
	}

	if strings.TrimSpace(cfg.OutputPath) == "" {
		return fmt.Errorf("pipeline.trim.output_path is required when pipeline.trim.enabled is true")
	}
	if strings.TrimSpace(cfg.Bounds.PromptID) == "" {
		return fmt.Errorf("pipeline.trim.bounds.prompt_id is required when pipeline.trim.enabled is true")
	}
	if strings.TrimSpace(cfg.Bounds.TranscriptInputName) == "" {
		return fmt.Errorf("pipeline.trim.bounds.transcript_input_name is required when pipeline.trim.enabled is true")
	}
	if strings.TrimSpace(cfg.Bounds.OutputPath) == "" {
		return fmt.Errorf("pipeline.trim.bounds.output_path is required when pipeline.trim.enabled is true")
	}
	if err := validateDuration("pipeline.trim.bounds.timeout", cfg.Bounds.Timeout); err != nil {
		return err
	}
	if cfg.Bounds.RenderDebug && strings.TrimSpace(cfg.Bounds.RenderOutputPath) == "" {
		return fmt.Errorf("pipeline.trim.bounds.render_output_path is required when pipeline.trim.bounds.render_debug is true")
	}

	return nil
}

func validateWhisperX(cfg WhisperXConfig) error {
	if strings.TrimSpace(cfg.TranscribeURL) == "" {
		return fmt.Errorf("pipeline.whisperx.transcribe_url is required")
	}
	u, err := url.Parse(cfg.TranscribeURL)
	if err != nil || u.Scheme == "" || u.Host == "" {
		if err != nil {
			return fmt.Errorf("pipeline.whisperx.transcribe_url must be a valid URL: %w", err)
		}
		return fmt.Errorf("pipeline.whisperx.transcribe_url must be a valid URL")
	}
	if err := validateDuration("pipeline.whisperx.timeout", cfg.Timeout); err != nil {
		return err
	}
	if err := validateDuration("pipeline.whisperx.retry_delay", cfg.RetryDelay); err != nil {
		return err
	}
	if cfg.Retries == nil {
		return fmt.Errorf("pipeline.whisperx.retries must be set (defaults should populate this)")
	}
	if *cfg.Retries < 0 {
		return fmt.Errorf("pipeline.whisperx.retries must be >= 0")
	}
	if cfg.Concurrency == nil {
		return fmt.Errorf("pipeline.whisperx.concurrency must be set (defaults should populate this)")
	}
	if *cfg.Concurrency <= 0 {
		return fmt.Errorf("pipeline.whisperx.concurrency must be > 0")
	}
	return nil
}

func validateSeriatim(cfg SeriatimConfig) error {
	if strings.TrimSpace(cfg.Binary) == "" {
		return fmt.Errorf("pipeline.seriatim.binary is required")
	}
	if err := validateDuration("pipeline.seriatim.timeout", cfg.Timeout); err != nil {
		return err
	}
	if err := validateSeriatimOutputSchema("pipeline.seriatim.output_schema", cfg.OutputSchema); err != nil {
		return err
	}

	if cfg.CoalesceGap == nil {
		return fmt.Errorf("pipeline.seriatim.coalesce_gap must be set (defaults should populate this)")
	}
	if *cfg.CoalesceGap < 0 {
		return fmt.Errorf("pipeline.seriatim.coalesce_gap must be >= 0")
	}

	for _, item := range []struct {
		name  string
		value *float64
	}{
		{name: "pipeline.seriatim.env.overlap_word_run_gap", value: cfg.Env.OverlapWordRunGap},
		{name: "pipeline.seriatim.env.overlap_word_run_reorder_window", value: cfg.Env.OverlapWordRunReorderWindow},
		{name: "pipeline.seriatim.env.backchannel_max_duration", value: cfg.Env.BackchannelMaxDuration},
		{name: "pipeline.seriatim.env.filler_max_duration", value: cfg.Env.FillerMaxDuration},
	} {
		if item.value != nil && *item.value <= 0 {
			return fmt.Errorf("%s must be > 0 when provided", item.name)
		}
	}

	return nil
}

func validateSeriatimOutputSchema(field, value string) error {
	switch strings.TrimSpace(value) {
	case "seriatim-minimal", "seriatim-intermediate", "seriatim-full":
		return nil
	default:
		return fmt.Errorf("%s must be one of: seriatim-minimal, seriatim-intermediate, seriatim-full", field)
	}
}

func validateAudita(cfg AuditaConfig) error {
	if strings.TrimSpace(cfg.Binary) == "" {
		return fmt.Errorf("pipeline.audita.binary is required")
	}
	if err := validateDuration("pipeline.audita.timeout", cfg.Timeout); err != nil {
		return err
	}
	for i, m := range cfg.Modules {
		module := strings.TrimSpace(m)
		if module == "" {
			return fmt.Errorf("pipeline.audita.modules[%d] must be non-empty", i)
		}
		switch module {
		case "glossary", "homophones", "spoken_word", "grammar":
		default:
			return fmt.Errorf("pipeline.audita.modules[%d] must be one of: glossary, homophones, spoken_word, grammar", i)
		}
	}
	if strings.TrimSpace(cfg.BaseURL) != "" {
		u, err := url.Parse(cfg.BaseURL)
		if err != nil || u.Scheme == "" || u.Host == "" {
			if err != nil {
				return fmt.Errorf("pipeline.audita.base_url must be a valid URL: %w", err)
			}
			return fmt.Errorf("pipeline.audita.base_url must be a valid URL")
		}
	}
	if cfg.TotalLLMConcurrency != nil && *cfg.TotalLLMConcurrency <= 0 {
		return fmt.Errorf("pipeline.audita.total_llm_concurrency must be > 0")
	}
	if cfg.ProposalLLMConcurrency != nil && *cfg.ProposalLLMConcurrency <= 0 {
		return fmt.Errorf("pipeline.audita.proposal_llm_concurrency must be > 0")
	}
	if cfg.ValidationLLMConcurrency != nil && *cfg.ValidationLLMConcurrency <= 0 {
		return fmt.Errorf("pipeline.audita.validation_llm_concurrency must be > 0")
	}
	if strings.TrimSpace(cfg.TranscriptDescription) == "" && cfg.TranscriptDescription != "" {
		return fmt.Errorf("pipeline.audita.transcript_description must be non-empty when provided")
	}
	if strings.TrimSpace(cfg.ConfigPath) == "" && cfg.ConfigPath != "" {
		return fmt.Errorf("pipeline.audita.config_path must be non-empty when provided")
	}
	switch strings.TrimSpace(cfg.OutputSchema) {
	case "", "bare-segments", "audita-v1":
	default:
		return fmt.Errorf("pipeline.audita.output_schema must be one of: bare-segments, audita-v1")
	}
	switch strings.TrimSpace(cfg.WorkDirRetention) {
	case "", "always", "auto", "never":
	default:
		return fmt.Errorf("pipeline.audita.work_dir_retention must be one of: always, auto, never")
	}
	return nil
}

func validateScriptorium(cfg *ScriptoriumConfig) error {
	if cfg == nil {
		return nil
	}
	if strings.TrimSpace(cfg.Binary) == "" {
		return fmt.Errorf("pipeline.scriptorium.binary is required when pipeline.scriptorium is configured")
	}
	if cfg.ConfigPath != "" && strings.TrimSpace(cfg.ConfigPath) == "" {
		return fmt.Errorf("pipeline.scriptorium.config_path must be non-empty when provided")
	}
	if err := validateDuration("pipeline.scriptorium.timeout", cfg.Timeout); err != nil {
		return err
	}

	for artifactName, artifactCfg := range cfg.Artifacts {
		trimmedArtifactName := strings.TrimSpace(artifactName)
		if trimmedArtifactName == "" {
			return fmt.Errorf("pipeline.scriptorium.artifacts keys must be non-empty")
		}
		if artifactCfg.Enabled && strings.TrimSpace(artifactCfg.PromptID) == "" {
			return fmt.Errorf("pipeline.scriptorium.artifacts.%s.prompt_id is required when enabled", artifactName)
		}
		if artifactCfg.Enabled && strings.TrimSpace(artifactCfg.OutputPath) == "" {
			return fmt.Errorf("pipeline.scriptorium.artifacts.%s.output_path is required when enabled", artifactName)
		}
		if err := validateDuration("pipeline.scriptorium.artifacts."+artifactName+".timeout", artifactCfg.Timeout); err != nil {
			return err
		}
		for inputName, inputCfg := range artifactCfg.Inputs {
			trimmedInputName := strings.TrimSpace(inputName)
			if trimmedInputName == "" {
				return fmt.Errorf("pipeline.scriptorium.artifacts.%s.inputs keys must be non-empty", artifactName)
			}
			if strings.TrimSpace(inputCfg.Source) == "" {
				return fmt.Errorf("pipeline.scriptorium.artifacts.%s.inputs.%s.source is required", artifactName, inputName)
			}
		}
		for varName, varValue := range artifactCfg.Vars {
			if strings.TrimSpace(varName) == "" {
				return fmt.Errorf("pipeline.scriptorium.artifacts.%s.vars keys must be non-empty", artifactName)
			}
			switch varValue.(type) {
			case bool, string:
			default:
				return fmt.Errorf("pipeline.scriptorium.artifacts.%s.vars.%s must be a string or boolean", artifactName, varName)
			}
		}
	}

	return nil
}

func validateSession(cfg *SessionConfig) error {
	if strings.TrimSpace(cfg.SessionID) == "" {
		return fmt.Errorf("session.session_id is required")
	}
	if strings.TrimSpace(cfg.Campaign) == "" {
		return fmt.Errorf("session.campaign is required")
	}

	if strings.TrimSpace(cfg.Inputs.SpeakersFile) == "" {
		return fmt.Errorf("session.inputs.speakers_file is required")
	}
	if strings.TrimSpace(cfg.Inputs.AutocorrectFile) == "" {
		return fmt.Errorf("session.inputs.autocorrect_file is required")
	}
	if strings.TrimSpace(cfg.Inputs.GlossaryFile) == "" {
		return fmt.Errorf("session.inputs.glossary_file is required")
	}

	hasAudioDir := strings.TrimSpace(cfg.Inputs.AudioDir) != ""
	hasAudioFiles := len(cfg.Inputs.AudioFiles) > 0
	hasAudioS3 := cfg.Inputs.AudioS3 != nil
	if hasAudioS3 {
		if strings.TrimSpace(cfg.Inputs.AudioS3.Prefix) == "" {
			return fmt.Errorf("session.inputs.audio_s3.prefix is required when session.inputs.audio_s3 is configured")
		}
		if err := validateRelativeSafePath("session.inputs.audio_s3.prefix", cfg.Inputs.AudioS3.Prefix); err != nil {
			return err
		}
	}
	if hasAudioS3 && (hasAudioDir || hasAudioFiles) {
		return fmt.Errorf("session.inputs.audio_dir/audio_files and session.inputs.audio_s3 are mutually exclusive")
	}
	if !hasAudioDir && !hasAudioFiles && !hasAudioS3 {
		return fmt.Errorf("session.inputs requires audio_dir, at least one audio_files entry, or audio_s3")
	}

	return nil
}

func validateCrossConfig(pipeline *PipelineConfig, session *SessionConfig) error {
	if pipeline == nil || session == nil {
		return nil
	}
	if pipeline.Storage.S3 == nil {
		return nil
	}

	audioS3Enabled := session.Inputs.AudioS3 != nil
	archiveUploadEnabled := archiveUploadConfiguredForS3(pipeline)
	if (audioS3Enabled || archiveUploadEnabled) && strings.TrimSpace(pipeline.Storage.S3.Bucket) == "" {
		return fmt.Errorf("pipeline.storage.s3.bucket is required when S3 session audio or archive upload is enabled")
	}
	return nil
}

func archiveUploadConfiguredForS3(pipeline *PipelineConfig) bool {
	if pipeline == nil || pipeline.Archive == nil {
		return false
	}
	if !strings.EqualFold(strings.TrimSpace(pipeline.Storage.Backend), "s3") {
		return false
	}
	enabled := true
	if pipeline.Archive.Enabled != nil {
		enabled = *pipeline.Archive.Enabled
	}
	upload := true
	if pipeline.Archive.UploadRun != nil {
		upload = *pipeline.Archive.UploadRun
	}
	return enabled && upload
}

var windowsAbsPathRE = regexp.MustCompile(`^[A-Za-z]:[\\/].*`)

func validateRelativeSafePath(fieldName, value string) error {
	trimmed := strings.TrimSpace(value)
	if trimmed == "" {
		return fmt.Errorf("%s must be non-empty", fieldName)
	}
	if filepath.IsAbs(trimmed) || strings.HasPrefix(trimmed, "/") || strings.HasPrefix(trimmed, "\\") || windowsAbsPathRE.MatchString(trimmed) {
		return fmt.Errorf("%s must be a relative path", fieldName)
	}

	normalized := strings.ReplaceAll(trimmed, "\\", "/")
	for _, segment := range strings.Split(normalized, "/") {
		if segment == ".." {
			return fmt.Errorf("%s must not contain path traversal", fieldName)
		}
	}
	return nil
}

func validateDuration(fieldName, value string) error {
	trimmed := strings.TrimSpace(value)
	if trimmed == "" {
		return nil
	}

	if _, err := time.ParseDuration(trimmed); err != nil {
		return fmt.Errorf("%s must be a valid duration: %w", fieldName, err)
	}

	return nil
}