From 5d311ea76ddc8bdbb357b6afdf64dfce6ece39a7 Mon Sep 17 00:00:00 2001
From: David Wendt <45795991+davidwendt@users.noreply.github.com>
Date: Thu, 5 Oct 2023 13:21:31 -0400
Subject: [PATCH] Fix strings replace for adjacent, identical multi-byte UTF-8
 character targets (#14235)

Fixes bug that can occur when replacing all occurrences in a string using a multi-byte UTF-8 target when the target matches sequentially in the same string -- some characters were missed.
Specialized gtest is also added.

Found while working on #13891

Authors:
  - David Wendt (https://github.com/davidwendt)

Approvers:
  - Bradley Dice (https://github.com/bdice)
  - Nghia Truong (https://github.com/ttnghia)

URL: https://github.com/rapidsai/cudf/pull/14235
---
 cpp/src/strings/replace/replace.cu  |  2 +-
 cpp/tests/strings/replace_tests.cpp | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/cpp/src/strings/replace/replace.cu b/cpp/src/strings/replace/replace.cu
index a622d1a742d..acc1502f4d6 100644
--- a/cpp/src/strings/replace/replace.cu
+++ b/cpp/src/strings/replace/replace.cu
@@ -97,7 +97,7 @@ struct replace_row_parallel_fn {
       } else {
         bytes += d_repl.size_bytes() - d_target.size_bytes();
       }
-      position = d_str.find(d_target, position + d_target.size_bytes());
+      position = d_str.find(d_target, position + d_target.length());
       --max_n;
     }
     if (out_ptr)  // copy whats left (or right depending on your point of view)
diff --git a/cpp/tests/strings/replace_tests.cpp b/cpp/tests/strings/replace_tests.cpp
index f143983aded..f04bb832f09 100644
--- a/cpp/tests/strings/replace_tests.cpp
+++ b/cpp/tests/strings/replace_tests.cpp
@@ -246,6 +246,28 @@ TEST_F(StringsReplaceTest, ReplaceEndOfString)
   CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
 }
 
+TEST_F(StringsReplaceTest, ReplaceAdjacentMultiByteTarget)
+{
+  auto input = cudf::test::strings_column_wrapper({"ééééééé", "eéeéeée", "eeeeeee"});
+  auto strings_view = cudf::strings_column_view(input);
+  // replace all occurrences of 'é' with 'e'
+  cudf::test::strings_column_wrapper expected({"eeeeeee", "eeeeeee", "eeeeeee"});
+
+  auto stream = cudf::get_default_stream();
+  auto mr     = rmm::mr::get_current_device_resource();
+
+  auto target  = cudf::string_scalar("é", true, stream);
+  auto repl    = cudf::string_scalar("e", true, stream);
+  auto results = cudf::strings::replace(strings_view, target, repl);
+  CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
+  results = cudf::strings::detail::replace<algorithm::CHAR_PARALLEL>(
+    strings_view, target, repl, -1, stream, mr);
+  CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
+  results = cudf::strings::detail::replace<algorithm::ROW_PARALLEL>(
+    strings_view, target, repl, -1, stream, mr);
+  CUDF_TEST_EXPECT_COLUMNS_EQUAL(*results, expected);
+}
+
 TEST_F(StringsReplaceTest, ReplaceSlice)
 {
   std::vector<char const*> h_strings{"Héllo", "thesé", nullptr, "ARE THE", "tést strings", ""};