bug处理优化

bug处理优化

internal/service/task_tracker_organize_items.go

@@ -0,0 +1,146 @@
+package service
+
+import "strings"
+
+// Item-level task status. A single BackgroundTask (one organize → scan →
+// scrape run) is broken down into per-file / per-library TaskItemRecord rows so
+// the operator can watch exactly which file is being organized, renamed, ingested
+// or scraped, and retry only the failed ones.
+const (
+	ItemStatusPending   = "pending"   // 待进行
+	ItemStatusRunning   = "running"   // 进行中
+	ItemStatusSucceeded = "succeeded" // 成功
+	ItemStatusFailed    = "failed"    // 失败
+)
+
+// Item kind / phase. 整理与重命名 share the "organize" phase (per file);
+// scan（入库）与 scrape（刮削）are reported per library.
+const (
+	ItemKindOrganize = "organize"
+	ItemKindScan     = "scan"
+	ItemKindScrape   = "scrape"
+)
+
+// TaskItemRecord is a single row on the live tasks board. Each record maps to
+// exactly one file (organize) or one library (ingest / scrape).
+type TaskItemRecord struct {
+	ID        string `json:"id"`
+	Kind      string `json:"kind"`              // organize / scan / scrape
+	Status    string `json:"status"`            // pending / running / succeeded / failed
+	Name      string `json:"name"`              // display name (file base name or library name)
+	Source    string `json:"source,omitempty"`  // source file path (organize)
+	DestPath  string `json:"dest_path,omitempty"`
+	LibraryID string `json:"library_id,omitempty"`
+	Error     string `json:"error,omitempty"`
+}
+
+// organizeItemsFromResult converts the final organize result items into
+// per-file task records. The organize result is only complete after the whole
+// directory walk finishes, so this is called once at stage boundaries rather
+// than per file. Each item is keyed by its source path for stable identity.
+func organizeItemsFromResult(res *OrganizeResult) []TaskItemRecord {
+	if res == nil || len(res.Items) == 0 {
+		return nil
+	}
+	out := make([]TaskItemRecord, 0, len(res.Items))
+	for _, item := range res.Items {
+		rec := TaskItemRecord{
+			ID:       "organize:" + item.Source,
+			Kind:     ItemKindOrganize,
+			Name:     itemBaseName(item.Source, item.Title),
+			Source:   item.Source,
+			DestPath: item.Target,
+		}
+		switch item.Action {
+		case "error":
+			rec.Status = ItemStatusFailed
+			rec.Error = strings.TrimSpace(item.Reason)
+		default:
+			// organize / replace / reclassify / cleanup / skip all count as done.
+			rec.Status = ItemStatusSucceeded
+		}
+		out = append(out, rec)
+	}
+	return out
+}
+
+// scanItemsFromResult converts per-library scan summaries into task records.
+func scanItemsFromResult(res *OrganizeResult) []TaskItemRecord {
+	if res == nil || len(res.Scans) == 0 {
+		return nil
+	}
+	out := make([]TaskItemRecord, 0, len(res.Scans))
+	for _, scan := range res.Scans {
+		rec := TaskItemRecord{
+			ID:        "scan:" + scan.LibraryID,
+			Kind:      ItemKindScan,
+			Name:      scan.Name,
+			LibraryID: scan.LibraryID,
+			Status:    ItemStatusSucceeded,
+		}
+		if scan.Error != "" {
+			rec.Status = ItemStatusFailed
+			rec.Error = scan.Error
+		}
+		out = append(out, rec)
+	}
+	return out
+}
+
+// scrapeItemsFromResult converts per-library scrape summaries into task records.
+func scrapeItemsFromResult(res *OrganizeResult) []TaskItemRecord {
+	if res == nil || len(res.Scrapes) == 0 {
+		return nil
+	}
+	out := make([]TaskItemRecord, 0, len(res.Scrapes))
+	for _, scrape := range res.Scrapes {
+		rec := TaskItemRecord{
+			ID:        "scrape:" + scrape.LibraryID,
+			Kind:      ItemKindScrape,
+			Name:      scrape.Name,
+			LibraryID: scrape.LibraryID,
+			Status:    ItemStatusSucceeded,
+		}
+		if scrape.Error != "" {
+			rec.Status = ItemStatusFailed
+			rec.Error = scrape.Error
+			out = append(out, rec)
+			continue
+		}
+		if scrape.Skipped {
+			rec.Status = ItemStatusSucceeded
+		}
+		out = append(out, rec)
+	}
+	return out
+}
+
+// combineOrganizeItems merges organize + scan + scrape item rows into one flat
+// list ordered by phase, deduping by ID (scan/scrape share the library ID and
+// there is at most one of each per run).
+func combineOrganizeItems(res *OrganizeResult) []TaskItemRecord {
+	items := organizeItemsFromResult(res)
+	items = append(items, scanItemsFromResult(res)...)
+	items = append(items, scrapeItemsFromResult(res)...)
+	return items
+}
+
+func itemBaseName(source, title string) string {
+	if strings.TrimSpace(title) != "" && !strings.Contains(title, ".") {
+		return strings.TrimSpace(title)
+	}
+	if source == "" {
+		return ""
+	}
+	base := source
+	if idx := strings.LastIndexAny(base, "/\\"); idx >= 0 {
+		base = base[idx+1:]
+	}
+	if idx := strings.LastIndex(base, "."); idx > 0 {
+		base = base[:idx]
+	}
+	if base == "" {
+		return source
+	}
+	return base
+}