feat(backend): duplicate pairs, dismissals, and merge resolution

Adds the duplicate-detection backend on top of perceptual hashing:

- Two tables (edited into the original migrations): data.duplicate_pairs holds
  precomputed near-duplicate candidates (rebuilt wholesale by the rescan), and
  data.duplicate_dismissals is a global "not a duplicate" overlay that survives
  rescans. New audit actions file_merge / duplicate_dismiss.
- DuplicateService:
  - Rescan builds every pair within DUPLICATE_HASH_THRESHOLD via a BK-tree over
    the perceptual hashes and replaces the pairs table. This is the only thing
    that populates pairs, so GET never compares all-vs-all (scales to 110k+).
  - Clusters reads the precomputed pairs (ACL-filtered, non-trashed, non-
    dismissed), groups them into connected components via union-find, and
    paginates whole clusters.
  - Resolve merges a pair field-by-field: each scalar from keep or discard,
    metadata keep/discard/shallow-merge, tags/pools keep or union; then trashes
    the discarded file. Enforces edit ACL on both.
  - Dismiss records a canonical pair (view ACL on both).
- Endpoints under /files: GET /files/duplicates, POST /files/duplicates/dismiss,
  POST /files/duplicates/resolve (registered before /:id to avoid collision).
  Plain delete reuses /files/bulk/delete.
- Repo support: ListMissingPHash, ListAllPHashes, CopyPoolMemberships, plus the
  DuplicatePairRepo (ReplaceAll via COPY, ListVisible) and DismissalRepo.

Unit tests cover the BK-tree pairing, union-find clustering, metadata merge and
field validation; an integration test covers rescan -> list -> merge -> dismiss
(including that a dismissal survives a re-rescan).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/internal/handler/duplicate_handler.go

@@ -0,0 +1,122 @@
+package handler
+
+import (
+	"net/http"
+	"strconv"
+
+	"github.com/gin-gonic/gin"
+
+	"tanabata/backend/internal/domain"
+	"tanabata/backend/internal/service"
+)
+
+// DuplicateHandler handles the /files/duplicates endpoints.
+type DuplicateHandler struct {
+	dupSvc *service.DuplicateService
+}
+
+// NewDuplicateHandler creates a DuplicateHandler.
+func NewDuplicateHandler(dupSvc *service.DuplicateService) *DuplicateHandler {
+	return &DuplicateHandler{dupSvc: dupSvc}
+}
+
+// List handles GET /files/duplicates — an offset-paginated list of duplicate
+// clusters, each a group of files within the perceptual-hash threshold.
+func (h *DuplicateHandler) List(c *gin.Context) {
+	limit, offset := 20, 0
+	if n, err := strconv.Atoi(c.Query("limit")); err == nil {
+		limit = n
+	}
+	if n, err := strconv.Atoi(c.Query("offset")); err == nil {
+		offset = n
+	}
+	if limit < 1 {
+		limit = 1
+	}
+	if limit > 50 {
+		limit = 50
+	}
+	if offset < 0 {
+		offset = 0
+	}
+
+	clusters, total, err := h.dupSvc.Clusters(c.Request.Context(), limit, offset)
+	if err != nil {
+		respondError(c, err)
+		return
+	}
+
+	items := make([]gin.H, len(clusters))
+	for i, files := range clusters {
+		fs := make([]fileJSON, len(files))
+		for j, f := range files {
+			fs[j] = toFileJSON(f)
+		}
+		items[i] = gin.H{"files": fs}
+	}
+	respondJSON(c, http.StatusOK, gin.H{
+		"items":  items,
+		"total":  total,
+		"limit":  limit,
+		"offset": offset,
+	})
+}
+
+// Dismiss handles POST /files/duplicates/dismiss — mark a pair "not a duplicate".
+func (h *DuplicateHandler) Dismiss(c *gin.Context) {
+	var body struct {
+		FileIDA string `json:"file_id_a" binding:"required"`
+		FileIDB string `json:"file_id_b" binding:"required"`
+	}
+	if err := c.ShouldBindJSON(&body); err != nil {
+		respondError(c, domain.ErrValidation)
+		return
+	}
+	ids, err := parseUUIDs([]string{body.FileIDA, body.FileIDB})
+	if err != nil {
+		respondError(c, domain.ErrValidation)
+		return
+	}
+	if err := h.dupSvc.Dismiss(c.Request.Context(), ids[0], ids[1]); err != nil {
+		respondError(c, err)
+		return
+	}
+	c.Status(http.StatusNoContent)
+}
+
+// Resolve handles POST /files/duplicates/resolve — merge a duplicate pair,
+// keeping one file and folding the chosen fields in from the other. Returns the
+// updated survivor. delete_discarded defaults to true.
+func (h *DuplicateHandler) Resolve(c *gin.Context) {
+	var body struct {
+		Keep            string              `json:"keep"    binding:"required"`
+		Discard         string              `json:"discard" binding:"required"`
+		Fields          service.MergeFields `json:"fields"`
+		DeleteDiscarded *bool               `json:"delete_discarded"`
+	}
+	if err := c.ShouldBindJSON(&body); err != nil {
+		respondError(c, domain.ErrValidation)
+		return
+	}
+	ids, err := parseUUIDs([]string{body.Keep, body.Discard})
+	if err != nil {
+		respondError(c, domain.ErrValidation)
+		return
+	}
+
+	del := true
+	if body.DeleteDiscarded != nil {
+		del = *body.DeleteDiscarded
+	}
+	f, err := h.dupSvc.Resolve(c.Request.Context(), service.MergeSpec{
+		Keep:            ids[0],
+		Discard:         ids[1],
+		Fields:          body.Fields,
+		DeleteDiscarded: del,
+	})
+	if err != nil {
+		respondError(c, err)
+		return
+	}
+	respondJSON(c, http.StatusOK, toFileJSON(*f))
+}

backend/internal/handler/router.go

@@ -26,6 +26,7 @@ func NewRouter(
 	auth *AuthMiddleware,
 	authHandler *AuthHandler,
 	fileHandler *FileHandler,
+	duplicateHandler *DuplicateHandler,
 	tagHandler *TagHandler,
 	categoryHandler *CategoryHandler,
 	poolHandler *PoolHandler,
@@ -80,7 +81,11 @@ func NewRouter(
 		files.GET("", fileHandler.List)
 		files.POST("", fileHandler.Upload)
 
-		// Bulk + import routes registered before /:id to prevent param collision.
+		// Bulk + import + duplicates routes registered before /:id to prevent
+		// param collision (e.g. "duplicates" being captured as :id).
+		files.GET("/duplicates", duplicateHandler.List)
+		files.POST("/duplicates/dismiss", duplicateHandler.Dismiss)
+		files.POST("/duplicates/resolve", duplicateHandler.Resolve)
 		files.POST("/bulk/tags", fileHandler.BulkSetTags)
 		files.POST("/bulk/delete", fileHandler.BulkDelete)
 		files.POST("/bulk/review", fileHandler.BulkReview)

backend/internal/handler/router_test.go

@@ -10,7 +10,7 @@ import "testing"
 func TestNewRouterRegisters(t *testing.T) {
 	r, err := NewRouter(
 		(*AuthMiddleware)(nil), (*AuthHandler)(nil),
-		(*FileHandler)(nil), (*TagHandler)(nil), (*CategoryHandler)(nil), (*PoolHandler)(nil),
+		(*FileHandler)(nil), (*DuplicateHandler)(nil), (*TagHandler)(nil), (*CategoryHandler)(nil), (*PoolHandler)(nil),
 		(*UserHandler)(nil), (*ACLHandler)(nil), (*AuditHandler)(nil),
 		"", nil,
 	)