validate shape ids when decoding S2ShapeIndexCell

src/s2/mutable_s2shape_index_test.cc

@@ -44,6 +44,8 @@
 #include "absl/types/span.h"
 
 #include "s2/base/log_severity.h"
+#include "s2/encoded_s2cell_id_vector.h"
+#include "s2/encoded_string_vector.h"
 #include "s2/r2.h"
 #include "s2/r2rect.h"
 #include "s2/s1angle.h"
@@ -821,6 +823,32 @@ TEST_F(MutableS2ShapeIndexTest, DecoderCatchesInvalidIndex) {
   EXPECT_FALSE(ok);
 }
 
+TEST_F(MutableS2ShapeIndexTest, DecoderCatchesOutOfRangeShapeId) {
+  // An encoded index whose single cell references a shape id (5000) far beyond
+  // the number of shapes in the index (2).  Without validation, Init() accepts
+  // this and a later shape(clipped.shape_id()) lookup reads out of bounds.
+  MutableS2ShapeIndex shapes;
+  shapes.Add(
+      make_unique<S2PointVectorShape>(vector<S2Point>{S2Point(1, 0, 0)}));
+  shapes.Add(
+      make_unique<S2PointVectorShape>(vector<S2Point>{S2Point(0, 1, 0)}));
+
+  Encoder encoder;
+  encoder.Ensure(64);
+  encoder.put_varint64(10 << 2 | 0);  // max_edges_per_cell = 10, version = 0
+  s2coding::EncodeS2CellIdVector({S2CellId::FromFace(0)}, &encoder);
+  s2coding::StringVectorEncoder cells;
+  Encoder* cell = cells.AddViaEncoder();
+  cell->Ensure(8);
+  cell->put8(0x00);               // contiguous range of edges, edge id 0
+  cell->put_varint32(5000 << 4);  // shape id delta 5000, num_edges 1
+  cells.Encode(&encoder);
+
+  Decoder decoder(encoder.base(), encoder.length());
+  MutableS2ShapeIndex index;
+  EXPECT_FALSE(index.Init(&decoder, s2shapeutil::WrappedShapeFactory(&shapes)));
+}
+
 TEST(S2Shape, user_data) {
   struct MyData {
     int x, y;

src/s2/s2shape_index.cc

@@ -283,6 +283,11 @@ bool S2ShapeIndexCell::Decode(int num_shape_ids, Decoder* decoder) {
       clipped->set_contains_center((header & 4) != 0);
       if (!DecodeEdges(num_edges, clipped, decoder)) return false;
     }
+
+    // Shape ids come from the encoded data, so reject any that are out of
+    // range.  Otherwise a caller that indexes S2ShapeIndex::shape() with a
+    // decoded shape id reads out of bounds.
+    if (shape_id >= num_shape_ids) return false;
   }
   return true;
 }