From 5260380cec6355416ddb316c34ef0b64f5e96eec Mon Sep 17 00:00:00 2001
From: Chris Dibak <dibak@google.com>
Date: Fri, 9 Jan 2026 03:14:23 -0800
Subject: [PATCH] Add comment why we are not using std::geometric_distribution

PiperOrigin-RevId: 854116436
---
 cc/algorithms/BUILD            | 1 +
 cc/algorithms/distributions.cc | 9 ++++++++-
 cc/algorithms/distributions.h  | 6 ++++++
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/cc/algorithms/BUILD b/cc/algorithms/BUILD
index 4a056534..450020b1 100644
--- a/cc/algorithms/BUILD
+++ b/cc/algorithms/BUILD
@@ -430,6 +430,7 @@ cc_library(
     deps = [
         ":rand",
         ":util",
+        "@com_google_absl//absl/log:check",
         "@com_google_absl//absl/memory",
         "@com_google_absl//absl/random",
         "@com_google_absl//absl/status",
diff --git a/cc/algorithms/distributions.cc b/cc/algorithms/distributions.cc
index ef42988b..8ec5c5a0 100644
--- a/cc/algorithms/distributions.cc
+++ b/cc/algorithms/distributions.cc
@@ -15,13 +15,20 @@
 //
 #include "algorithms/distributions.h"
 
+#include <algorithm>
 #include <cmath>
+#include <cstdint>
 #include <cstdlib>
+#include <limits>
+#include <memory>
+#include <utility>
 
+#include "absl/log/check.h"
 #include "absl/memory/memory.h"
 #include "absl/random/random.h"
 #include "absl/status/status.h"
 #include "absl/status/statusor.h"
+#include "absl/strings/str_cat.h"
 #include "absl/strings/string_view.h"
 #include "algorithms/rand.h"
 #include "algorithms/util.h"
@@ -145,7 +152,7 @@ double GaussianDistribution::SampleGeometric() {
 // of n must be at least 10^6. This is to ensure an accurate approximation of a
 // Gaussian distribution.
 double GaussianDistribution::SampleBinomial(double sqrt_n) {
-  long long step_size =
+  int64_t step_size =
       static_cast<long long>(std::round(std::sqrt(2.0) * sqrt_n + 1));
 
   SecureURBG& random = SecureURBG::GetInstance();
diff --git a/cc/algorithms/distributions.h b/cc/algorithms/distributions.h
index d6b1fd5b..2180c2f4 100644
--- a/cc/algorithms/distributions.h
+++ b/cc/algorithms/distributions.h
@@ -87,6 +87,12 @@ class GaussianDistribution {
 // be positive. If the result would be higher than the maximum int64_t, returns
 // the maximum int64_t, which means that users should be careful around the
 // edges of their distribution.
+//
+// Note that we are not using std::geometric_distribution, as samples returned
+// by that class are often not accurate enough for anonymization purposes. For
+// instance, when sampling with very low p, std::geometric_distribution does not
+// return odd values. In some cases, this could leak information about the
+// parity of the input.
 class GeometricDistribution {
  public:
   // Builder for GeometricDistribution.