feat: scrape metadata after organize

2026-09-29 11:36:36 +08:00 · 2026-06-09 17:00:24 +08:00
parent 81f0cabe91
commit 76fe0dbde9
18 changed files with 470 additions and 31 deletions
@@ -250,6 +250,7 @@ func setDefaults(v *viper.Viper) {
 	v.SetDefault("downloads.smart_classify", true)
 	v.SetDefault("organizer.smart_classify", false)
 	v.SetDefault("organizer.auto_after_download", false)
+	v.SetDefault("organize.scrape_after", false)
 	v.SetDefault("organizer.categories.chinese_movie", "华语电影")
 	v.SetDefault("organizer.categories.animation_movie", "动画电影")
 	v.SetDefault("organizer.categories.foreign_movie", "外语电影")
@@ -85,6 +85,9 @@ func downloadOrganizeAllHandler(svc *service.Container) gin.HandlerFunc {
 				results = append(results, gin.H{"library": l.Name, "error": err.Error()})
 				continue
 			}
+			if svc.Scan != nil && res != nil && !res.DryRun {
+				res.Scans, res.Scrapes = scanAndScrapeAfterOrganize(c, svc, res.DestPath, l.ID, nil)
+			}
 			results = append(results, gin.H{"library": l.Name, "result": res})
 		}
 		c.JSON(http.StatusOK, gin.H{"results": results})
@@ -134,6 +134,9 @@ func organizeBulkHandler(svc *service.Container) gin.HandlerFunc {
 				out = append(out, gin.H{"library": l.Name, "error": err.Error()})
 				continue
 			}
+			if svc.Scan != nil && res != nil && !res.DryRun {
+				res.Scans, res.Scrapes = scanAndScrapeAfterOrganize(c, svc, res.DestPath, l.ID, nil)
+			}
 			out = append(out, gin.H{"library": l.Name, "result": res})
 		}
 		c.JSON(http.StatusOK, gin.H{"results": out})
@@ -21,6 +21,7 @@ type organizeReq struct {
 	TransferMode string `json:"transfer_mode"`
 	MediaType    string `json:"media_type"`
 	ScanAfter    bool   `json:"scan_after"`
+	ScrapeAfter  *bool  `json:"scrape_after"`
 	LibraryID    string `json:"library_id"`
 	DryRun       bool   `json:"dry_run"`
 }
@@ -52,24 +53,37 @@ func organizeOptionsFromReq(req organizeReq) service.OrganizeOptions {

 func organizeMediaHandler(svc *service.Container) gin.HandlerFunc {
 	return func(c *gin.Context) {
-		opts := bindOrganizeOptions(c)
+		var req organizeReq
+		_ = c.ShouldBindJSON(&req)
+		opts := organizeOptionsFromReq(req)
 		dst, err := svc.Organizer.OrganizeMediaWithOptions(c.Request.Context(), c.Param("id"), opts)
 		if err != nil {
 			c.JSON(http.StatusInternalServerError, gin.H{"error": err.Error()})
 			return
 		}
-		c.JSON(http.StatusOK, gin.H{"path": dst})
+		payload := gin.H{"path": dst}
+		if req.ScanAfter && !req.DryRun && svc.Scan != nil {
+			scans, scrapes := scanAndScrapeAfterOrganize(c, svc, dst, strings.TrimSpace(req.LibraryID), req.ScrapeAfter)
+			payload["scans"] = scans
+			payload["scrapes"] = scrapes
+		}
+		c.JSON(http.StatusOK, payload)
 	}
 }

 func organizeLibraryHandler(svc *service.Container) gin.HandlerFunc {
 	return func(c *gin.Context) {
-		opts := bindOrganizeOptions(c)
+		var req organizeReq
+		_ = c.ShouldBindJSON(&req)
+		opts := organizeOptionsFromReq(req)
 		res, err := svc.Organizer.OrganizeLibraryWithOptions(c.Request.Context(), c.Param("id"), opts)
 		if err != nil {
 			c.JSON(http.StatusInternalServerError, gin.H{"error": err.Error()})
 			return
 		}
+		if req.ScanAfter && !req.DryRun && svc.Scan != nil {
+			res.Scans, res.Scrapes = scanAndScrapeAfterOrganize(c, svc, res.DestPath, c.Param("id"), req.ScrapeAfter)
+		}
 		c.JSON(http.StatusOK, res)
 	}
 }
@@ -95,8 +109,16 @@ func organizeDirectoryHandler(svc *service.Container) gin.HandlerFunc {
 			return
 		}
 		if req.ScanAfter && !req.DryRun && svc.Scan != nil {
-			res.Scans = svc.Scan.ScanLibrariesForPath(c.Request.Context(), res.DestPath, strings.TrimSpace(req.LibraryID))
+			res.Scans, res.Scrapes = scanAndScrapeAfterOrganize(c, svc, res.DestPath, strings.TrimSpace(req.LibraryID), req.ScrapeAfter)
 		}
 		c.JSON(http.StatusOK, res)
 	}
 }
+
+func scanAndScrapeAfterOrganize(c *gin.Context, svc *service.Container, destRoot, preferredLibraryID string, scrapeOverride *bool) ([]service.OrganizeScanSummary, []service.OrganizeScrapeSummary) {
+	scrapeAfter := service.OrganizeScrapeAfterEnabled(c.Request.Context(), svc.Repo)
+	if scrapeOverride != nil {
+		scrapeAfter = *scrapeOverride
+	}
+	return svc.Scan.ScanAndScrapeLibrariesForPath(c.Request.Context(), destRoot, preferredLibraryID, scrapeAfter)
+}
@@ -80,6 +80,7 @@ func schemaHandler(_ *service.Container) gin.HandlerFunc {
 					"items": []gin.H{
 						{"key": "organize.auto", "type": "toggle", "label": "整理源目录定时自动整理"},
 						{"key": "organizer.auto_after_download", "type": "toggle"},
+						{"key": "organize.scrape_after", "type": "toggle", "label": "整理后自动刮削"},
 						{"key": "downloads.smart_classify", "type": "toggle"},
 						{"key": "organizer.smart_classify", "type": "toggle"},
 						{"key": "organize.source_dir", "type": "text", "label": "整理源目录"},
@@ -881,7 +881,7 @@ func (d *DownloadService) onTorrentComplete(ctx context.Context, torrent QBitTor
 		return
 	}
 	if d.scanner != nil && res != nil && strings.TrimSpace(res.DestPath) != "" {
-		res.Scans = d.scanner.ScanLibrariesForPath(ctx, res.DestPath, "")
+		res.Scans, res.Scrapes = d.scanner.ScanAndScrapeLibrariesForPath(ctx, res.DestPath, "", OrganizeScrapeAfterEnabled(ctx, d.repo))
 	}
 	d.log.Info("auto organize completed torrent finished",
 		zap.String("hash", torrent.Hash),
@@ -890,6 +890,7 @@ func (d *DownloadService) onTorrentComplete(ctx context.Context, torrent QBitTor
 		zap.Int("organized", res.Organized),
 		zap.Int("replaced", res.Replaced),
 		zap.Int("skipped", res.Skipped),
+		zap.Int("scrapes", len(res.Scrapes)),
 		zap.Int("errors", len(res.Errors)))
 }

@@ -31,20 +31,30 @@ func classifyMediaCategory(input mediaClassifyInput, categories map[string]strin
 	genres := normalizeTokens(input.Genres...)
 	countries := normalizeTokens(input.Countries...)
 	languages := normalizeTokens(input.Languages...)
-	text := strings.ToLower(input.Title + " " + input.Category + " " + strings.Join(input.Genres, " "))
+	rawText := input.Title + " " + input.Category + " " + strings.Join(input.Genres, " ")
+	text := strings.ToLower(rawText)

-	isChinese := hasAny(languages, "ZH", "ZH-CN", "ZH-TW", "CN") || hasAny(countries, "CN", "TW", "HK", "MO")
-	isJapanese := hasAny(languages, "JA", "JP") || hasAny(countries, "JP") || strings.Contains(text, "日番")
-	isKorean := hasAny(languages, "KO", "KR") || hasAny(countries, "KR", "KP")
+	isChinese := hasAny(languages, "ZH", "ZH-CN", "ZH-TW", "CN") || hasAny(countries, "CN", "TW", "HK", "MO") || containsHan(rawText) || containsAnyText(text, "华语", "国产", "国剧", "国漫")
+	isJapanese := hasAny(languages, "JA", "JP") || hasAny(countries, "JP") || containsJapaneseKana(rawText) || strings.Contains(text, "日番")
+	isKorean := hasAny(languages, "KO", "KR") || hasAny(countries, "KR", "KP") || containsKoreanHangul(rawText)
 	isEastAsian := isJapanese || isKorean || hasAny(countries, "TH", "IN", "SG")
-	isWestern := hasAny(countries,
+	isWesternByMetadata := hasAny(countries,
 		"US", "GB", "UK", "FR", "DE", "CA", "AU", "NZ", "IE", "NL", "SE", "NO", "DK",
 		"FI", "ES", "IT", "PT", "AT", "CH", "BE", "RU",
 	)
+	isLatinFallback := containsLatin(rawText) && !containsHan(rawText) && !containsJapaneseKana(rawText) && !containsKoreanHangul(rawText)
+	isWestern := isWesternByMetadata || (mediaType == "tv" && isLatinFallback)
+	hasAnimeText := containsAnyText(text, "动画", "动漫", "番剧", "年番", "国漫", "日番", "bangumi", "anime")

 	hasGenre := func(values ...string) bool {
 		for _, value := range values {
-			if hasAny(genres, strings.ToUpper(value)) || strings.Contains(text, strings.ToLower(value)) {
+			if hasAny(genres, strings.ToUpper(value)) {
+				return true
+			}
+			if isDigits(value) {
+				continue
+			}
+			if strings.Contains(text, strings.ToLower(value)) {
 				return true
 			}
 		}
@@ -62,7 +72,7 @@ func classifyMediaCategory(input mediaClassifyInput, categories map[string]strin
 		if isEastAsian {
 			return categoryName(categories, "jk_movie", "日韩电影")
 		}
-		if isWestern {
+		if isWesternByMetadata {
 			return categoryName(categories, "euus_movie", "欧美电影")
 		}
 		return categoryName(categories, "foreign_movie", "外语电影")
@@ -83,7 +93,7 @@ func classifyMediaCategory(input mediaClassifyInput, categories map[string]strin
 		if hasGenre("10762", "KIDS", "儿童") {
 			return categoryName(categories, "children", "儿童")
 		}
-		if hasGenre("16", "ANIMATION", "动画", "动漫") {
+		if hasGenre("16", "ANIMATION", "动画", "动漫") || hasAnimeText {
 			if isChinese {
 				return categoryName(categories, "cn_anime", "国漫")
 			}
@@ -99,6 +109,8 @@ func classifyMediaCategory(input mediaClassifyInput, categories map[string]strin
 			return categoryName(categories, "euus_tv", "欧美剧")
 		}
 		return categoryName(categories, "uncategorized_tv", "未分类")
+	case "adult":
+		return categoryName(categories, "adult", "成人")
 	}
 	return ""
 }
@@ -158,6 +170,63 @@ func hasAny(values map[string]struct{}, needles ...string) bool {
 	return false
 }

+func containsAnyText(text string, needles ...string) bool {
+	for _, needle := range needles {
+		if strings.Contains(text, strings.ToLower(needle)) {
+			return true
+		}
+	}
+	return false
+}
+
+func containsHan(text string) bool {
+	for _, r := range text {
+		if r >= '\u4e00' && r <= '\u9fff' {
+			return true
+		}
+	}
+	return false
+}
+
+func containsJapaneseKana(text string) bool {
+	for _, r := range text {
+		if (r >= '\u3040' && r <= '\u30ff') || (r >= '\u31f0' && r <= '\u31ff') {
+			return true
+		}
+	}
+	return false
+}
+
+func containsKoreanHangul(text string) bool {
+	for _, r := range text {
+		if (r >= '\uac00' && r <= '\ud7af') || (r >= '\u1100' && r <= '\u11ff') || (r >= '\u3130' && r <= '\u318f') {
+			return true
+		}
+	}
+	return false
+}
+
+func containsLatin(text string) bool {
+	for _, r := range text {
+		if (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') {
+			return true
+		}
+	}
+	return false
+}
+
+func isDigits(text string) bool {
+	if text == "" {
+		return false
+	}
+	for _, r := range text {
+		if r < '0' || r > '9' {
+			return false
+		}
+	}
+	return true
+}
+
 func categoryName(categories map[string]string, key, fallback string) string {
 	if categories != nil {
 		if name := strings.TrimSpace(categories[key]); name != "" {
@@ -57,6 +57,38 @@ func TestClassifyMediaCategoryMatchesMoviePilotStyleRules(t *testing.T) {
 			},
 			want: "纪录片",
 		},
+		{
+			name: "chinese movie title without metadata",
+			input: mediaClassifyInput{
+				MediaType: "movie",
+				Title:     "流浪地球2 2023 2160p",
+			},
+			want: "华语电影",
+		},
+		{
+			name: "latin movie title without metadata",
+			input: mediaClassifyInput{
+				MediaType: "movie",
+				Title:     "Dune 2021 2160p",
+			},
+			want: "外语电影",
+		},
+		{
+			name: "chinese tv title without metadata",
+			input: mediaClassifyInput{
+				MediaType: "tv",
+				Title:     "狂飙 S01E01 1080p",
+			},
+			want: "国产剧",
+		},
+		{
+			name: "latin tv title without metadata",
+			input: mediaClassifyInput{
+				MediaType: "tv",
+				Title:     "The Last of Us S01E01 1080p",
+			},
+			want: "欧美剧",
+		},
 	}

 	for _, tt := range tests {
@@ -49,15 +49,16 @@ func (o *OrganizerService) SetProbe(p *FFprobeService) { o.probe = p }

 // OrganizeResult reports what happened.
 type OrganizeResult struct {
-	Organized  int                   `json:"organized"`
-	Skipped    int                   `json:"skipped"`
-	Replaced   int                   `json:"replaced,omitempty"`
-	Errors     []string              `json:"errors,omitempty"`
-	SourcePath string                `json:"source_path,omitempty"`
-	DestPath   string                `json:"dest_path,omitempty"`
-	DryRun     bool                  `json:"dry_run,omitempty"`
-	Items      []OrganizePreviewItem `json:"items,omitempty"`
-	Scans      []OrganizeScanSummary `json:"scans,omitempty"`
+	Organized  int                     `json:"organized"`
+	Skipped    int                     `json:"skipped"`
+	Replaced   int                     `json:"replaced,omitempty"`
+	Errors     []string                `json:"errors,omitempty"`
+	SourcePath string                  `json:"source_path,omitempty"`
+	DestPath   string                  `json:"dest_path,omitempty"`
+	DryRun     bool                    `json:"dry_run,omitempty"`
+	Items      []OrganizePreviewItem   `json:"items,omitempty"`
+	Scans      []OrganizeScanSummary   `json:"scans,omitempty"`
+	Scrapes    []OrganizeScrapeSummary `json:"scrapes,omitempty"`
 }

 type OrganizePreviewItem struct {
@@ -288,7 +289,7 @@ func (o *OrganizerService) OrganizeLibraryWithOptions(ctx context.Context, libra
 	sourceRoot := o.resolveSourceRoot(ctx, lib, opts.SourcePath)
 	// 目的地目录：已位于该根下的文件视为已整理；受 dest_path 覆盖与设置影响。
 	baseRoot := o.resolveBaseRoot(ctx, lib, opts.DestPath)
-	res := &OrganizeResult{}
+	res := &OrganizeResult{SourcePath: sourceRoot, DestPath: baseRoot, DryRun: opts.DryRun}
 	for i := range rows {
 		// 不在源目录内的文件跳过（不属于本次「从源目录整理」的范围）。
 		if !pathWithin(rows[i].Path, sourceRoot) {
@@ -201,6 +201,9 @@ func (o *OrganizerService) organizeSourceFile(ctx context.Context, src, sourceRo
 	if layout.MediaType == "" {
 		layout.MediaType = o.inferMediaTypeForSourceFile(src, title, season, episode)
 	}
+	if layout.Category == "" {
+		layout.Category = o.smartClassifySourceFile(ctx, src, sourceRoot, layout.MediaType, title, parsedTitle)
+	}
 	layoutRoot := destRoot
 	if layout.MediaType != "" {
 		layoutRoot = o.organizeRoot(destRoot, layout.MediaType, layout.Category)
@@ -338,6 +341,28 @@ func (o *OrganizerService) inferMediaTypeForSourceFile(src, title string, season
 	return normalizeMediaType("", title, src)
 }

+func (o *OrganizerService) smartClassifySourceFile(ctx context.Context, src, sourceRoot, mediaType, title, parsedTitle string) string {
+	if o == nil || !o.isSmartClassifyEnabled(ctx) {
+		return ""
+	}
+	seriesLike := isSeriesLibraryType(mediaType)
+	input := mediaClassifyInput{
+		MediaType: mediaType,
+		Title:     strings.Join([]string{title, parsedTitle, filepath.Base(src)}, " "),
+		Category:  strings.Join(organizeDirectoryCategoryCandidates(src, sourceRoot), " "),
+	}
+	if meta, err := ReadLocalMetadata(src, sourceRoot, seriesLike); err == nil && meta != nil && meta.HasNFO {
+		input.Title = strings.Join([]string{meta.Title, meta.OriginalName, title, parsedTitle, filepath.Base(src)}, " ")
+		input.Languages = parseCommaList(meta.Languages)
+		input.Countries = parseCommaList(meta.Countries)
+		input.Genres = parseCommaList(meta.Genres)
+		if meta.NSFW {
+			input.MediaType = "adult"
+		}
+	}
+	return sanitizeFilename(classifyMediaCategory(input, o.categoryMap()))
+}
+
 func (o *OrganizerService) inferOrganizeDirectoryLayout(src, sourceRoot string) organizeDirectoryLayout {
 	for _, name := range organizeDirectoryCategoryCandidates(src, sourceRoot) {
 		if mediaType, category := o.mediaTypeForDirectoryCategory(name); mediaType != "" && category != "" {
@@ -427,6 +427,78 @@ func TestOrganizeDirectoryUsesDownloadCategoryLayout(t *testing.T) {
 	}
 }

+func TestOrganizeDirectorySmartClassifiesUncategorizedSources(t *testing.T) {
+	root := t.TempDir()
+	src := filepath.Join(root, "downloads")
+	dest := filepath.Join(root, "media")
+	writeOrgFile(t, filepath.Join(src, "流浪地球2.2023.2160p.WEB-DL.mkv"), "cn-movie")
+	writeOrgFile(t, filepath.Join(src, "Dune.2021.2160p.WEB-DL.mkv"), "foreign-movie")
+	writeOrgFile(t, filepath.Join(src, "狂飙.S01E01.2023.1080p.WEB-DL.mkv"), "cn-tv")
+	writeOrgFile(t, filepath.Join(src, "The.Last.of.Us.S01E01.2023.1080p.WEB-DL.mkv"), "western-tv")
+
+	repos := newOrganizerTestRepo(t)
+	if err := repos.Setting.Set(t.Context(), "organizer.smart_classify", "true"); err != nil {
+		t.Fatal(err)
+	}
+	org := NewOrganizerService(&config.Config{}, zap.NewNop(), repos)
+	res, err := org.OrganizeDirectory(t.Context(), OrganizeOptions{
+		SourcePath:   src,
+		DestPath:     dest,
+		TransferMode: TransferCopy,
+	})
+	if err != nil {
+		t.Fatalf("organize directory: %v", err)
+	}
+	if res.Organized != 4 {
+		t.Fatalf("organized = %d, want 4; result=%+v", res.Organized, res)
+	}
+
+	for _, want := range []string{
+		filepath.Join(dest, "电影", "华语电影", "流浪地球2 (2023)", "流浪地球2 (2023).mkv"),
+		filepath.Join(dest, "电影", "外语电影", "Dune (2021)", "Dune (2021).mkv"),
+		filepath.Join(dest, "电视剧", "国产剧", "狂飙", "Season 01", "狂飙 - S01E01.mkv"),
+		filepath.Join(dest, "电视剧", "欧美剧", "The Last Of Us", "Season 01", "The Last Of Us - S01E01.mkv"),
+	} {
+		if _, err := os.Stat(want); err != nil {
+			t.Fatalf("expected smart classified file at %q: %v; items=%+v", want, err, res.Items)
+		}
+	}
+}
+
+func TestOrganizeDirectorySmartClassifiesWithLocalNFO(t *testing.T) {
+	root := t.TempDir()
+	src := filepath.Join(root, "downloads")
+	dest := filepath.Join(root, "media")
+	writeOrgFile(t, filepath.Join(src, "Some.Show.S01E01.2024.1080p.mkv"), "jp-anime")
+	writeOrgFile(t, filepath.Join(src, "tvshow.nfo"), `<tvshow>
+  <title>Some Show</title>
+  <genre>Animation</genre>
+  <country>JP</country>
+  <language>ja</language>
+</tvshow>`)
+
+	repos := newOrganizerTestRepo(t)
+	if err := repos.Setting.Set(t.Context(), "organizer.smart_classify", "true"); err != nil {
+		t.Fatal(err)
+	}
+	org := NewOrganizerService(&config.Config{}, zap.NewNop(), repos)
+	res, err := org.OrganizeDirectory(t.Context(), OrganizeOptions{
+		SourcePath:   src,
+		DestPath:     dest,
+		TransferMode: TransferCopy,
+	})
+	if err != nil {
+		t.Fatalf("organize directory: %v", err)
+	}
+	if res.Organized != 1 {
+		t.Fatalf("organized = %d, want 1", res.Organized)
+	}
+	want := filepath.Join(dest, "电视剧", "日番", "Some Show", "Season 01", "Some Show - S01E01.mkv")
+	if _, err := os.Stat(want); err != nil {
+		t.Fatalf("expected NFO classified episode at %q: %v", want, err)
+	}
+}
+
 func TestOrganizeDirectoryScanAfterRecursesNestedDownloadFolders(t *testing.T) {
 	root := t.TempDir()
 	src := filepath.Join(root, "downloads")
@@ -5,6 +5,7 @@ import (
 	"strings"

 	"github.com/ShukeBta/MediaStationGo/internal/model"
+	"github.com/ShukeBta/MediaStationGo/internal/repository"
 )

 // OrganizeScanSummary reports a library scan triggered after directory
@@ -20,18 +21,53 @@ type OrganizeScanSummary struct {
 	Error     string `json:"error,omitempty"`
 }

+// OrganizeScrapeSummary reports metadata enrichment triggered after organize.
+type OrganizeScrapeSummary struct {
+	LibraryID string `json:"library_id"`
+	Name      string `json:"name"`
+	Path      string `json:"path"`
+	Matched   int    `json:"matched"`
+	Skipped   bool   `json:"skipped,omitempty"`
+	Reason    string `json:"reason,omitempty"`
+	Error     string `json:"error,omitempty"`
+}
+
+// OrganizeScrapeAfterEnabled decides whether organize workflows should run
+// metadata scraping after scanning organized files into libraries.
+func OrganizeScrapeAfterEnabled(ctx context.Context, repo *repository.Container) bool {
+	if repo == nil || repo.Setting == nil {
+		return false
+	}
+	if value, err := repo.Setting.Get(ctx, "organize.scrape_after"); err == nil && strings.TrimSpace(value) != "" {
+		return parseBoolSetting(value, false)
+	}
+	if value, err := repo.Setting.Get(ctx, "scrape.auto_on_scan"); err == nil && strings.TrimSpace(value) != "" {
+		return parseBoolSetting(value, false)
+	}
+	return false
+}
+
 // ScanLibrariesForPath recursively scans libraries affected by an organize
 // destination. If preferredLibraryID is set, only that library is scanned.
 // Otherwise every enabled library whose path intersects destRoot is scanned;
 // if no path can be matched, we fall back to all enabled libraries to preserve
 // the old "scan all after ingest" UI behavior.
 func (s *ScannerService) ScanLibrariesForPath(ctx context.Context, destRoot, preferredLibraryID string) []OrganizeScanSummary {
+	scans, _ := s.ScanAndScrapeLibrariesForPath(ctx, destRoot, preferredLibraryID, false)
+	return scans
+}
+
+// ScanAndScrapeLibrariesForPath scans the affected organize target libraries,
+// then optionally runs the scraper synchronously so manual/automatic organize
+// can provide deterministic "整理 + 入库 + 刮削" behavior instead of relying on
+// a background scan hook that may or may not be enabled.
+func (s *ScannerService) ScanAndScrapeLibrariesForPath(ctx context.Context, destRoot, preferredLibraryID string, scrapeAfter bool) ([]OrganizeScanSummary, []OrganizeScrapeSummary) {
 	if s == nil || s.repo == nil || s.repo.Library == nil {
-		return nil
+		return nil, nil
 	}
 	libraries, err := s.repo.Library.List(ctx)
 	if err != nil {
-		return []OrganizeScanSummary{{Error: err.Error()}}
+		return []OrganizeScanSummary{{Error: err.Error()}}, nil
 	}
 	targets := selectOrganizeScanTargets(libraries, destRoot, preferredLibraryID)
 	out := make([]OrganizeScanSummary, 0, len(targets))
@@ -41,7 +77,7 @@ func (s *ScannerService) ScanLibrariesForPath(ctx context.Context, destRoot, pre
 			Name:      lib.Name,
 			Path:      lib.Path,
 		}
-		res, err := s.ScanLibrary(ctx, lib.ID)
+		res, err := s.scanLibrary(ctx, lib.ID, !scrapeAfter)
 		if err != nil {
 			summary.Error = err.Error()
 			out = append(out, summary)
@@ -53,6 +89,52 @@ func (s *ScannerService) ScanLibrariesForPath(ctx context.Context, destRoot, pre
 		summary.Removed = res.Removed
 		out = append(out, summary)
 	}
+	return out, s.scrapeOrganizeTargets(ctx, targets, scrapeAfter)
+}
+
+func (s *ScannerService) scrapeOrganizeTargets(ctx context.Context, targets []model.Library, scrapeAfter bool) []OrganizeScrapeSummary {
+	if len(targets) == 0 || !scrapeAfter {
+		return nil
+	}
+	out := make([]OrganizeScrapeSummary, 0, len(targets))
+	if s.scraper == nil {
+		for _, lib := range targets {
+			out = append(out, OrganizeScrapeSummary{
+				LibraryID: lib.ID,
+				Name:      lib.Name,
+				Path:      lib.Path,
+				Skipped:   true,
+				Reason:    "scraper unavailable",
+			})
+		}
+		return out
+	}
+	if !s.scraper.AnyEnabled() {
+		for _, lib := range targets {
+			out = append(out, OrganizeScrapeSummary{
+				LibraryID: lib.ID,
+				Name:      lib.Name,
+				Path:      lib.Path,
+				Skipped:   true,
+				Reason:    "no scraper provider enabled",
+			})
+		}
+		return out
+	}
+	for _, lib := range targets {
+		summary := OrganizeScrapeSummary{
+			LibraryID: lib.ID,
+			Name:      lib.Name,
+			Path:      lib.Path,
+		}
+		matched, err := s.scraper.EnrichLibrary(ctx, lib.ID)
+		if err != nil {
+			summary.Error = err.Error()
+		} else {
+			summary.Matched = matched
+		}
+		out = append(out, summary)
+	}
 	return out
 }

@@ -0,0 +1,70 @@
+package service
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+
+	"go.uber.org/zap"
+
+	"github.com/ShukeBta/MediaStationGo/internal/config"
+	"github.com/ShukeBta/MediaStationGo/internal/model"
+)
+
+func TestOrganizeDirectoryScanAndScrapeAfter(t *testing.T) {
+	scraper, repos, closeServer := newTestScraper(t)
+	defer closeServer()
+	if err := repos.DB.AutoMigrate(&model.Setting{}); err != nil {
+		t.Fatal(err)
+	}
+
+	root := t.TempDir()
+	src := filepath.Join(root, "downloads")
+	dest := filepath.Join(root, "media")
+	sourceFile := filepath.Join(src, "Spy.x.Family.S01E01.2022.1080p.mkv")
+	writeOrgFile(t, sourceFile, "episode")
+
+	lib := model.Library{
+		Name:    "剧集",
+		Path:    filepath.Join(dest, "电视剧"),
+		Type:    "tv",
+		Enabled: true,
+	}
+	if err := repos.Library.Create(t.Context(), &lib); err != nil {
+		t.Fatal(err)
+	}
+
+	organizer := NewOrganizerService(&config.Config{}, zap.NewNop(), repos)
+	res, err := organizer.OrganizeDirectory(t.Context(), OrganizeOptions{
+		SourcePath:   src,
+		DestPath:     dest,
+		TransferMode: TransferCopy,
+		MediaType:    "tv",
+	})
+	if err != nil {
+		t.Fatalf("organize directory: %v", err)
+	}
+	if res.Organized != 1 {
+		t.Fatalf("organized = %d, want 1", res.Organized)
+	}
+
+	scanner := NewScannerService(&config.Config{}, zap.NewNop(), repos, NewHub(zap.NewNop()), nil, scraper)
+	scans, scrapes := scanner.ScanAndScrapeLibrariesForPath(t.Context(), res.DestPath, "", true)
+	if len(scans) != 1 || scans[0].Added != 1 {
+		t.Fatalf("scans = %#v, want one scan with added=1", scans)
+	}
+	if len(scrapes) != 1 || scrapes[0].Matched != 1 || scrapes[0].Error != "" || scrapes[0].Skipped {
+		t.Fatalf("scrapes = %#v, want one successful matched scrape", scrapes)
+	}
+
+	var media model.Media
+	if err := repos.DB.Where("path LIKE ?", "%Spy Family - S01E01.mkv").First(&media).Error; err != nil {
+		t.Fatal(err)
+	}
+	if media.ScrapeStatus != "matched" || media.TMDbID != 12345 {
+		t.Fatalf("media scrape status=%q tmdb=%d, want matched/12345", media.ScrapeStatus, media.TMDbID)
+	}
+	if _, err := os.Stat(media.Path); err != nil {
+		t.Fatalf("organized file missing at %q: %v", media.Path, err)
+	}
+}
@@ -80,6 +80,10 @@ type ScanResult struct {

 // ScanLibrary walks the library root and persists discovered media files.
 func (s *ScannerService) ScanLibrary(ctx context.Context, libraryID string) (*ScanResult, error) {
+	return s.scanLibrary(ctx, libraryID, true)
+}
+
+func (s *ScannerService) scanLibrary(ctx context.Context, libraryID string, autoScrape bool) (*ScanResult, error) {
 	lib, err := s.repo.Library.FindByID(ctx, libraryID)
 	if err != nil || lib == nil {
 		return nil, err
@@ -124,7 +128,7 @@ func (s *ScannerService) ScanLibrary(ctx context.Context, libraryID string) (*Sc

 	// Online enrichment is opt-in. Local NFO is always consumed first during
 	// the scan, and matched rows are excluded from EnrichLibrary's pending set.
-	if s.scraper != nil && s.scraper.AnyEnabled() && s.autoScrapeEnabled(ctx) {
+	if autoScrape && s.scraper != nil && s.scraper.AnyEnabled() && s.autoScrapeEnabled(ctx) {
 		go func(libID string) {
 			if _, err := s.scraper.EnrichLibrary(context.Background(), libID); err != nil {
 				s.log.Warn("scraper enrich failed", zap.Error(err))
@@ -255,7 +255,7 @@ func (s *SchedulerService) jobOrganizeSource(ctx context.Context) error {
 		return err
 	}
 	if s.scanner != nil && res != nil && strings.TrimSpace(res.DestPath) != "" {
-		res.Scans = s.scanner.ScanLibrariesForPath(ctx, res.DestPath, "")
+		res.Scans, res.Scrapes = s.scanner.ScanAndScrapeLibrariesForPath(ctx, res.DestPath, "", OrganizeScrapeAfterEnabled(ctx, s.repo))
 	}
 	if s.log != nil && res != nil {
 		s.log.Info("scheduled source organize finished",
@@ -264,6 +264,7 @@ func (s *SchedulerService) jobOrganizeSource(ctx context.Context) error {
 			zap.Int("organized", res.Organized),
 			zap.Int("replaced", res.Replaced),
 			zap.Int("skipped", res.Skipped),
+			zap.Int("scrapes", len(res.Scrapes)),
 			zap.Int("errors", len(res.Errors)),
 		)
 	}
@@ -12,6 +12,7 @@ export interface OrganizeOverrides {
  transfer_mode?: string
  media_type?: string
  scan_after?: boolean
+  scrape_after?: boolean
  library_id?: string
  dry_run?: boolean
 }
@@ -76,6 +77,15 @@ export const toolsAPI = {
          removed: number
          error?: string
        }>
+        scrapes?: Array<{
+          library_id: string
+          name: string
+          path: string
+          matched: number
+          skipped?: boolean
+          reason?: string
+          error?: string
+        }>
      }>(
        '/admin/organize/source',
        opts,
@@ -45,9 +45,19 @@ function formatScanSummary(scans: Array<{ name: string; added: number; updated:
  return ` · 扫描 ${ok.length}/${scans.length} 个库 · 新入库 ${added} · 更新 ${updated} · 访问 ${visited}`
 }

+function formatScrapeSummary(scrapes: Array<{ name: string; matched: number; skipped?: boolean; reason?: string; error?: string }>): string {
+  if (scrapes.length === 0) return ''
+  const ok = scrapes.filter((scrape) => !scrape.error && !scrape.skipped)
+  const skipped = scrapes.filter((scrape) => scrape.skipped).length
+  const matched = ok.reduce((sum, scrape) => sum + (scrape.matched ?? 0), 0)
+  if (ok.length === 0 && skipped > 0) return ` · 刮削跳过 ${skipped} 个库`
+  return ` · 刮削 ${ok.length}/${scrapes.length} 个库 · 匹配 ${matched}${skipped ? ` · 跳过 ${skipped}` : ''}`
+}
+
 type AutoOrganizeConfig = {
  enabled: string
  afterDownload: string
+  scrapeAfter: string
  sourceDir: string
  targetDir: string
  transferMode: string
@@ -57,6 +67,7 @@ type AutoOrganizeConfig = {
 const AUTO_ORGANIZE_DEFAULTS: AutoOrganizeConfig = {
  enabled: 'false',
  afterDownload: 'false',
+  scrapeAfter: 'false',
  sourceDir: '',
  targetDir: '',
  transferMode: 'hardlink',
@@ -66,6 +77,7 @@ const AUTO_ORGANIZE_DEFAULTS: AutoOrganizeConfig = {
 const AUTO_ORGANIZE_KEYS: Record<keyof AutoOrganizeConfig, string> = {
  enabled: 'organize.auto',
  afterDownload: 'organizer.auto_after_download',
+  scrapeAfter: 'organize.scrape_after',
  sourceDir: 'organize.source_dir',
  targetDir: 'organize.target_dir',
  transferMode: 'organize.transfer_mode',
@@ -83,6 +95,7 @@ function mergeAutoOrganizeSettings(rows: Setting[]): AutoOrganizeConfig {
  return {
    enabled: idx[AUTO_ORGANIZE_KEYS.enabled] ?? AUTO_ORGANIZE_DEFAULTS.enabled,
    afterDownload: idx[AUTO_ORGANIZE_KEYS.afterDownload] ?? AUTO_ORGANIZE_DEFAULTS.afterDownload,
+    scrapeAfter: idx[AUTO_ORGANIZE_KEYS.scrapeAfter] ?? AUTO_ORGANIZE_DEFAULTS.scrapeAfter,
    sourceDir: idx[AUTO_ORGANIZE_KEYS.sourceDir] ?? AUTO_ORGANIZE_DEFAULTS.sourceDir,
    targetDir: idx[AUTO_ORGANIZE_KEYS.targetDir] ?? AUTO_ORGANIZE_DEFAULTS.targetDir,
    transferMode: idx[AUTO_ORGANIZE_KEYS.transferMode] ?? AUTO_ORGANIZE_DEFAULTS.transferMode,
@@ -114,6 +127,7 @@ export function FileManagerPage() {
  const [organizeTransferMode, setOrganizeTransferMode] = useState('hardlink')
  const [organizeMediaType, setOrganizeMediaType] = useState('auto')
  const [scanAfter, setScanAfter] = useState(true)
+  const [scrapeAfter, setScrapeAfter] = useState(false)
  const [organizeBusy, setOrganizeBusy] = useState('')
  const [previewItems, setPreviewItems] = useState<Array<{
    source: string
@@ -160,7 +174,9 @@ export function FileManagerPage() {
    adminAPI
      .listSettings()
      .then((rows) => {
-        setAutoConfig(mergeAutoOrganizeSettings(rows))
+        const nextConfig = mergeAutoOrganizeSettings(rows)
+        setAutoConfig(nextConfig)
+        setScrapeAfter(settingOn(nextConfig.scrapeAfter))
        setAutoDirty(false)
      })
      .catch(() => undefined)
@@ -309,7 +325,7 @@ export function FileManagerPage() {
    if (!dryRun) {
      const ok = await confirmAction({
        title: '确认整理入库',
-        message: `来源：${organizeSource}\n目标：${organizeDestPath}\n方式：${organizeTransferMode}${scanAfter ? '\n整理完成后会扫描入库。' : ''}`,
+        message: `来源：${organizeSource}\n目标：${organizeDestPath}\n方式：${organizeTransferMode}${scanAfter ? '\n整理完成后会扫描入库。' : ''}${scanAfter && scrapeAfter ? '\n扫描后会自动刮削。' : ''}`,
        confirmText: '开始整理',
      })
      if (!ok) return
@@ -322,6 +338,7 @@ export function FileManagerPage() {
        transfer_mode: organizeTransferMode,
        media_type: organizeMediaType === 'auto' ? undefined : organizeMediaType,
        scan_after: !dryRun && scanAfter,
+        scrape_after: !dryRun && scanAfter && scrapeAfter,
        library_id: !dryRun && scanAfter && organizeLibraryID ? organizeLibraryID : undefined,
        dry_run: dryRun,
      })
@@ -340,7 +357,8 @@ export function FileManagerPage() {
        return
      }
      const scanText = scanAfter ? formatScanSummary(result.scans ?? []) : ''
-      toast.success(`整理完成：新增 ${result.organized} · 替换 ${replaced} · 跳过 ${result.skipped}${scanText}`)
+      const scrapeText = scanAfter && scrapeAfter ? formatScrapeSummary(result.scrapes ?? []) : ''
+      toast.success(`整理完成：新增 ${result.organized} · 替换 ${replaced} · 跳过 ${result.skipped}${scanText}${scrapeText}`)
      refresh()
    } catch (err: unknown) {
      toast.error((err as { response?: { data?: { error?: string } } })?.response?.data?.error ?? '整理失败')
@@ -498,6 +516,14 @@ export function FileManagerPage() {
            />
            qB 下载完成后自动整理
          </label>
+          <label className="flex items-center gap-2 rounded-lg border border-gray-200 bg-gray-50 px-2 py-1 text-xs text-ink-100">
+            <input
+              type="checkbox"
+              checked={settingOn(autoConfig.scrapeAfter)}
+              onChange={(event) => changeAutoConfig('scrapeAfter', event.target.checked ? 'true' : 'false')}
+            />
+            整理后自动刮削
+          </label>
          <span className="text-xs text-sand-500">
            {autoDirty ? '有未保存设置' : '设置已同步'} · 定时任务名：organize_source
          </span>
@@ -568,6 +594,15 @@ export function FileManagerPage() {
              <input type="checkbox" checked={scanAfter} onChange={(event) => setScanAfter(event.target.checked)} />
              整理后扫描入库
            </label>
+            <label className="flex items-center gap-2 rounded-lg border border-gray-200 bg-gray-50 px-2 py-1 text-xs text-ink-100">
+              <input
+                type="checkbox"
+                checked={scanAfter && scrapeAfter}
+                disabled={!scanAfter}
+                onChange={(event) => setScrapeAfter(event.target.checked)}
+              />
+              整理后自动刮削
+            </label>
            <button
              type="button"
              className="neon-button !border-primary-400/30 !bg-white !text-brand-500"
@@ -160,6 +160,13 @@ const GROUPS: SettingGroup[] = [
        type: 'toggle',
        hint: '开启后 qB 下载完成时，系统会优先使用种子的 content_path 整理该文件/目录，并在整理完成后扫描目标媒体库。',
      },
+      {
+        key: 'organize.scrape_after',
+        label: '整理后自动刮削',
+        type: 'toggle',
+        hint: '开启后，手动/自动整理完成并扫描入库后，会立即触发 TMDb/豆瓣/Bangumi/JavBus/JavDB 等元数据刮削。需要先配置可用刮削源。',
+        defaultValue: 'false',
+      },
      {
        key: 'downloads.smart_classify',
        label: '下载器智能分类',