narratio/internal/config/validate_adapters.go

package config

import (
	"fmt"
	"net/url"
	"strings"
)

func validateNotification(cfg NotificationConfig) error {
	if strings.EqualFold(strings.TrimSpace(cfg.Mode), DefaultNotificationMode) {
		return nil
	}
	return fmt.Errorf("pipeline.notification.mode must be %q until a notification provider is configured", DefaultNotificationMode)
}

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 == nil {
		return fmt.Errorf("pipeline.trim.enabled must be set (defaults should populate this)")
	}
	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 validateRender(cfg *RenderConfig) error {
	if cfg == nil {
		return nil
	}
	if cfg.Enabled == nil {
		return fmt.Errorf("pipeline.render.enabled must be set (defaults should populate this)")
	}
	if cfg.IncludeTimestamps == nil {
		return fmt.Errorf("pipeline.render.include_timestamps must be set (defaults should populate this)")
	}
	if cfg.IncludeSegmentIDs == nil {
		return fmt.Errorf("pipeline.render.include_segment_ids must be set (defaults should populate this)")
	}
	format := strings.TrimSpace(cfg.Format)
	if format != "markdown" {
		return fmt.Errorf("pipeline.render.format must be markdown")
	}
	if cfg.Title != "" && strings.TrimSpace(cfg.Title) == "" {
		return fmt.Errorf("pipeline.render.title must be non-empty when provided")
	}
	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.IsAbs() || u.Host == "" || (strings.ToLower(u.Scheme) != "http" && strings.ToLower(u.Scheme) != "https") {
		if err != nil {
			return fmt.Errorf("pipeline.whisperx.transcribe_url must be an absolute http or https URL: %w", err)
		}
		return fmt.Errorf("pipeline.whisperx.transcribe_url must be an absolute http or https 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.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")
	}
	return nil
}