s3vectors: bound ListVectorBuckets and ListIndexes maxResults (#12218)

* s3vectors: bound ListVectorBuckets and ListIndexes maxResults

Both pass maxResults to the catalog unbounded; AWS caps these list pages
at 100, the same way ListVectors is bounded and topK is bounded.

* s3vectors: raise ListVectorBuckets/ListIndexes maxResults cap to 500

AWS documents a valid range of 1-500 for both list APIs; the shared
limit of 100 rejected legitimate page sizes.

* s3vectors: reject explicit maxResults of zero on list calls

AWS requires maxResults >= 1, but a plain int field cannot tell an
omitted field from an explicit zero. Making MaxResults a pointer keeps
the catalog default for omitted fields while a supplied zero fails
validation.

* s3vectors: test max ListIndexes page against an existing bucket

weed/s3api/s3vectors/handlers.go

@@ -301,6 +301,14 @@ func (s *Server) handleListVectorBuckets(w http.ResponseWriter, r *http.Request)
 		writeError(w, http.StatusBadRequest, "ValidationException", err.Error())
 		return
 	}
+	maxResults := 0
+	if req.MaxResults != nil {
+		if *req.MaxResults < 1 || *req.MaxResults > vectorsMaxListPage {
+			writeError(w, http.StatusBadRequest, "ValidationException", fmt.Sprintf("maxResults must be between 1 and %d", vectorsMaxListPage))
+			return
+		}
+		maxResults = *req.MaxResults
+	}
 	var resp struct {
 		TableBuckets []struct {
 			Arn       string `json:"arn"`
@@ -315,7 +323,7 @@ func (s *Server) handleListVectorBuckets(w http.ResponseWriter, r *http.Request)
 	// s3vectors actions, while showing ones they were not.
 	err := s.internalCatalog(r.Context(), "ListTableBuckets",
 		&s3tables.ListTableBucketsRequest{
-			MaxBuckets: req.MaxResults, ContinuationToken: req.NextToken, Prefix: req.Prefix,
+			MaxBuckets: maxResults, ContinuationToken: req.NextToken, Prefix: req.Prefix,
 		}, &resp)
 	if err != nil {
 		catalogErr(w, err)
@@ -1085,6 +1093,14 @@ func (s *Server) handleListIndexes(w http.ResponseWriter, r *http.Request) {
 		writeError(w, http.StatusBadRequest, "ValidationException", err.Error())
 		return
 	}
+	maxResults := 0
+	if req.MaxResults != nil {
+		if *req.MaxResults < 1 || *req.MaxResults > vectorsMaxListPage {
+			writeError(w, http.StatusBadRequest, "ValidationException", fmt.Sprintf("maxResults must be between 1 and %d", vectorsMaxListPage))
+			return
+		}
+		maxResults = *req.MaxResults
+	}
 	if _, err := s.authorizeVector(r.Context(), r, "s3vectors:ListIndexes", arns.bucket, ""); err != nil {
 		catalogErr(w, err)
 		return
@@ -1100,7 +1116,7 @@ func (s *Server) handleListIndexes(w http.ResponseWriter, r *http.Request) {
 	err = s.internalCatalog(r.Context(), "ListTables", &s3tables.ListTablesRequest{
 		TableBucketARN:    s.s3tablesArn(arns.bucket),
 		Namespace:         []string{indexNamespace},
-		MaxTables:         req.MaxResults,
+		MaxTables:         maxResults,
 		ContinuationToken: req.NextToken,
 		Prefix:            req.Prefix,
 	}, &resp)
@@ -1382,6 +1398,7 @@ const (
 	vectorsDefaultPage = 500
 	vectorsMaxPutBatch = 500
 	vectorsMaxPage     = 1000
+	vectorsMaxListPage = 500
 	vectorsMaxTopK     = 10000
 )
 

weed/s3api/s3vectors/handlers_test.go

@@ -183,6 +183,52 @@ func TestListSegmentVectorsPageLimits(t *testing.T) {
 	}
 }
 
+// An explicit maxResults of zero is rejected like any out-of-range value;
+// an omitted field still reaches the catalog as 0 and pages by default.
+func TestListRequestMaxResultsBounds(t *testing.T) {
+	fs := s3tablestest.Start(t)
+	s := NewServer(s3tables.NewManagerClient(fs.Client), allowAllAuth{}, newFakeVectors(0))
+
+	zero, over, max := 0, vectorsMaxListPage+1, vectorsMaxListPage
+	for name, req := range map[string]*ListVectorBucketsRequest{
+		"zero": {MaxResults: &zero},
+		"over": {MaxResults: &over},
+	} {
+		if w := postJSONAs(t, "", s.handleListVectorBuckets, req); w.Code != http.StatusBadRequest {
+			t.Fatalf("ListVectorBuckets %s: %d %s, want 400", name, w.Code, w.Body.String())
+		}
+	}
+	for name, req := range map[string]*ListIndexesRequest{
+		"zero": {VectorBucketName: "vec-bucket", MaxResults: &zero},
+		"over": {VectorBucketName: "vec-bucket", MaxResults: &over},
+	} {
+		if w := postJSONAs(t, "", s.handleListIndexes, req); w.Code != http.StatusBadRequest {
+			t.Fatalf("ListIndexes %s: %d %s, want 400", name, w.Code, w.Body.String())
+		}
+	}
+	if w := postJSONAs(t, "", s.handleListVectorBuckets, &ListVectorBucketsRequest{}); w.Code != http.StatusOK {
+		t.Fatalf("ListVectorBuckets omitted maxResults: %d %s", w.Code, w.Body.String())
+	}
+	if w := postJSONAs(t, "", s.handleListVectorBuckets, &ListVectorBucketsRequest{MaxResults: &max}); w.Code != http.StatusOK {
+		t.Fatalf("ListVectorBuckets maxResults=%d: %d %s", max, w.Code, w.Body.String())
+	}
+	if w := postJSONAs(t, "", s.handleCreateVectorBucket, &CreateVectorBucketRequest{VectorBucketName: "vec-bucket"}); w.Code != http.StatusOK {
+		t.Fatalf("create bucket: %d %s", w.Code, w.Body.String())
+	}
+	if w := postJSONAs(t, "", s.handleCreateIndex, &CreateIndexRequest{
+		VectorBucketName: "vec-bucket",
+		IndexName:        "idx",
+		DataType:         "float32",
+		Dimension:        3,
+		DistanceMetric:   "euclidean",
+	}); w.Code != http.StatusOK {
+		t.Fatalf("create index: %d %s", w.Code, w.Body.String())
+	}
+	if w := postJSONAs(t, "", s.handleListIndexes, &ListIndexesRequest{VectorBucketName: "vec-bucket", MaxResults: &max}); w.Code != http.StatusOK {
+		t.Fatalf("ListIndexes maxResults=%d: %d %s", max, w.Code, w.Body.String())
+	}
+}
+
 func TestArnParts(t *testing.T) {
 	var v vectorBucketARNs
 

weed/s3api/s3vectors/types.go

@@ -75,7 +75,7 @@ type VectorBucket struct {
 }
 
 type ListVectorBucketsRequest struct {
-	MaxResults int    `json:"maxResults"`
+	MaxResults *int   `json:"maxResults"`
 	NextToken  string `json:"nextToken"`
 	Prefix     string `json:"prefix"`
 }
@@ -158,7 +158,7 @@ type GetIndexResponse struct {
 }
 
 type ListIndexesRequest struct {
-	MaxResults       int    `json:"maxResults"`
+	MaxResults       *int   `json:"maxResults"`
 	NextToken        string `json:"nextToken"`
 	Prefix           string `json:"prefix"`
 	VectorBucketName string `json:"vectorBucketName"`