tm_back/cmd/worker/workers/translation.go

package workers

import (
	"context"
	"strings"
	"time"
	"tm/internal/tender"
	"tm/pkg/ai_summarizer"
	"tm/pkg/logger"
	"tm/pkg/mongo"
)

// TranslationWorker backfills missing tender translations using AI summarizer API.
type TranslationWorker struct {
	Mongo          *mongo.ConnectionManager
	Logger         logger.Logger
	TenderRepo     tender.TenderRepository
	AIClient       *ai_summarizer.Client
	TargetLanguage string
	BatchSize      int
}

// NewTranslationWorker creates a translation worker for a target language.
func NewTranslationWorker(mongo *mongo.ConnectionManager, logger logger.Logger, tenderRepo tender.TenderRepository, aiClient *ai_summarizer.Client, targetLanguage string, batchSize int) *TranslationWorker {
	if strings.TrimSpace(targetLanguage) == "" {
		targetLanguage = "en"
	}
	if batchSize <= 0 {
		batchSize = 20
	}

	return &TranslationWorker{
		Mongo:          mongo,
		Logger:         logger,
		TenderRepo:     tenderRepo,
		AIClient:       aiClient,
		TargetLanguage: strings.ToLower(strings.TrimSpace(targetLanguage)),
		BatchSize:      batchSize,
	}
}

// Run starts backfilling translations for tenders missing target-language content.
func (w *TranslationWorker) Run() {
	if w.AIClient == nil {
		w.Logger.Warn("Translation worker skipped: AI client is not configured", map[string]interface{}{
			"target_language": w.TargetLanguage,
		})
		return
	}

	w.Logger.Info("Translation worker started", map[string]interface{}{
		"target_language": w.TargetLanguage,
		"batch_size":      w.BatchSize,
	})

	for {
		// Always use skip=0 because successfully translated tenders are excluded by filter.
		tenders, totalCount, err := w.TenderRepo.GetUnTranslatedTenders(context.Background(), w.TargetLanguage, w.BatchSize, 0)
		if err != nil {
			w.Logger.Error("Failed to fetch untranslated tenders", map[string]interface{}{
				"target_language": w.TargetLanguage,
				"error":           err.Error(),
			})
			break
		}

		w.Logger.Info("Fetched untranslated tenders batch", map[string]interface{}{
			"target_language": w.TargetLanguage,
			"batch_count":     len(tenders),
			"total_count":     totalCount,
		})

		if len(tenders) == 0 {
			break
		}

		for _, t := range tenders {
			w.translateTender(&t)
		}
	}

	w.Logger.Info("Translation worker completed", map[string]interface{}{
		"target_language": w.TargetLanguage,
	})
}

func (w *TranslationWorker) translateTender(t *tender.Tender) {
	if t.NoticePublicationID == "" {
		return
	}
	if existing, ok := t.Translations[w.TargetLanguage]; ok && strings.TrimSpace(existing.Title) != "" {
		w.Logger.Debug("Skipping tender translation because target language already exists", map[string]interface{}{
			"tender_id":       t.ID.Hex(),
			"notice_id":       t.NoticePublicationID,
			"target_language": w.TargetLanguage,
		})
		return
	}

	resp, err := w.AIClient.FetchTranslationOnDemand(context.Background(), ai_summarizer.TranslateRequest{
		NoticeID:    t.NoticePublicationID,
		Title:       t.Title,
		Description: t.Description,
		Language:    w.TargetLanguage,
	})
	if err != nil {
		w.Logger.Error("Failed to translate tender", map[string]interface{}{
			"tender_id":        t.ID.Hex(),
			"notice_id":        t.NoticePublicationID,
			"target_language":  w.TargetLanguage,
			"translation_error": err.Error(),
		})
		return
	}

	if t.Translations == nil {
		t.Translations = make(map[string]tender.TranslationEntry)
	}
	now := time.Now().Unix()
	// Only update the target-language slot. Existing translations in other
	// languages (e.g., "de", "fr") are preserved and never modified.
	t.Translations[w.TargetLanguage] = tender.TranslationEntry{
		Title:        resp.TranslatedTitle,
		Description:  resp.TranslatedDescription,
		Language:     w.TargetLanguage,
		TranslatedAt: now,
	}
	if t.ProcessingMetadata.TranslatedData == nil {
		t.ProcessingMetadata.TranslatedData = make(map[string]string)
	}
	t.ProcessingMetadata.TranslatedData[w.TargetLanguage] = "done"
	t.ProcessingMetadata.TranslatedAt = now

	if err := w.TenderRepo.Update(context.Background(), t); err != nil {
		w.Logger.Error("Failed to persist translated tender", map[string]interface{}{
			"tender_id":        t.ID.Hex(),
			"notice_id":        t.NoticePublicationID,
			"target_language":  w.TargetLanguage,
			"persistence_error": err.Error(),
		})
		return
	}

	w.Logger.Info("Tender translated successfully", map[string]interface{}{
		"tender_id":       t.ID.Hex(),
		"notice_id":       t.NoticePublicationID,
		"target_language": w.TargetLanguage,
	})
}