Deduplicate organize by multi-source metadata

2026-10-07 22:06:38 +08:00 · 2026-06-14 01:25:44 +08:00
parent 4a3ed77395
commit a623702276
14 changed files with 424 additions and 49 deletions
@@ -1,6 +1,9 @@
 package service

 import (
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
 	"os"
 	"path/filepath"
 	"testing"
@@ -102,6 +105,107 @@ func TestOrganizeDirectoryUsesScraperMatchBeforeRename(t *testing.T) {
 	}
 }

+func TestOrganizeDirectoryDedupsByExternalIDBeforeRename(t *testing.T) {
+	scraper, repos, closeServer := newTestScraper(t)
+	defer closeServer()
+
+	root := t.TempDir()
+	src := filepath.Join(root, "downloads")
+	dest := filepath.Join(root, "media")
+	sourceFile := filepath.Join(src, "Spy.x.Family.S01E01.2022.2160p.mkv")
+	writeOrgFile(t, sourceFile, "episode")
+
+	existingPath := filepath.Join(dest, "电视剧", "旧错误名", "Season 01", "旧错误名 - S01E01.mkv")
+	writeOrgFile(t, existingPath, "existing")
+	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)
+	}
+	if err := repos.DB.Create(&model.Media{
+		LibraryID:    lib.ID,
+		Title:        "旧错误名",
+		Path:         existingPath,
+		SeasonNum:    1,
+		EpisodeNum:   1,
+		TMDbID:       12345,
+		ScrapeStatus: "matched",
+	}).Error; err != nil {
+		t.Fatal(err)
+	}
+
+	organizer := NewOrganizerService(&config.Config{}, zap.NewNop(), repos)
+	organizer.SetScraper(scraper)
+	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 != 0 || res.Skipped != 1 {
+		t.Fatalf("organize result = organized %d skipped %d, want 0/1; items=%#v errors=%#v", res.Organized, res.Skipped, res.Items, res.Errors)
+	}
+	if len(res.Items) != 1 || res.Items[0].Reason != organizeSkipDuplicateLibrary {
+		t.Fatalf("source should be skipped as external-id duplicate: %#v", res.Items)
+	}
+	if _, err := os.Stat(sourceFile); err != nil {
+		t.Fatalf("duplicate source should remain untouched: %v", err)
+	}
+}
+
+func TestOrganizeDirectoryUsesBangumiForAnimeRename(t *testing.T) {
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if r.URL.Path != "/search/subject/frieren" {
+			http.NotFound(w, r)
+			return
+		}
+		w.Header().Set("Content-Type", "application/json")
+		_ = json.NewEncoder(w).Encode(map[string]any{
+			"results": 1,
+			"list": []map[string]any{{
+				"id":       889,
+				"name":     "Frieren",
+				"name_cn":  "葬送的芙莉莲",
+				"air_date": "2023-09-29",
+			}},
+		})
+	}))
+	defer upstream.Close()
+
+	repos := newOrganizerTestRepo(t)
+	cfg := &config.Config{}
+	bangumi := NewBangumiProvider(cfg, zap.NewNop())
+	bangumi.base = upstream.URL
+	scraper := NewScraperService(cfg, zap.NewNop(), repos, nil, bangumi, nil, nil, NewHub(zap.NewNop()))
+
+	root := t.TempDir()
+	src := filepath.Join(root, "downloads")
+	dest := filepath.Join(root, "media")
+	sourceFile := filepath.Join(src, "Frieren.S01E01.1080p.mkv")
+	writeOrgFile(t, sourceFile, "episode")
+
+	organizer := NewOrganizerService(cfg, zap.NewNop(), repos)
+	organizer.SetScraper(scraper)
+	res, err := organizer.OrganizeDirectory(t.Context(), OrganizeOptions{
+		SourcePath:   src,
+		DestPath:     dest,
+		TransferMode: TransferCopy,
+		MediaType:    "anime",
+	})
+	if err != nil {
+		t.Fatalf("organize directory: %v", err)
+	}
+	want := filepath.Join(dest, "电视剧", "葬送的芙莉莲", "Season 01", "葬送的芙莉莲 - S01E01.mkv")
+	if res.Organized != 1 {
+		t.Fatalf("organized = %d, want 1; items=%#v errors=%#v", res.Organized, res.Items, res.Errors)
+	}
+	if _, err := os.Stat(want); err != nil {
+		t.Fatalf("organized file should use Bangumi metadata path %q: %v", want, err)
+	}
+}
+
 func TestOrganizeScanAndScrapeRetriesNoMatchRows(t *testing.T) {
 	scraper, repos, closeServer := newTestScraper(t)
 	defer closeServer()