From 816fe164d4a6b928287150d8b2c8a5e19b772e8e Mon Sep 17 00:00:00 2001
From: Praveen <praveen@dremio.com>
Date: Sun, 23 Jun 2019 23:51:24 +0530
Subject: [PATCH] Implement castVarcharVarchar

---
 cpp/src/gandiva/function_registry_string.cc |  4 ++
 cpp/src/gandiva/precompiled/string_ops.cc   | 12 ++++++
 cpp/src/gandiva/tests/CMakeLists.txt        |  1 +
 cpp/src/gandiva/tests/utf8_test.cc          | 48 +++++++++++++++++++++
 4 files changed, 65 insertions(+)

diff --git a/cpp/src/gandiva/function_registry_string.cc b/cpp/src/gandiva/function_registry_string.cc
index bc78ec66522..a62fde827c3 100644
--- a/cpp/src/gandiva/function_registry_string.cc
+++ b/cpp/src/gandiva/function_registry_string.cc
@@ -51,6 +51,10 @@ std::vector<NativeFunction> GetStringFunctionRegistry() {
       NativeFunction("upper", DataTypeVector{utf8()}, utf8(), kResultNullIfNull,
                      "upper_utf8", NativeFunction::kNeedsContext),
 
+      NativeFunction("castVARCHAR", DataTypeVector{utf8(), int64()}, utf8(),
+                     kResultNullIfNull, "castVARCHAR_utf8_int64",
+                     NativeFunction::kNeedsContext),
+
       NativeFunction("like", DataTypeVector{utf8(), utf8()}, boolean(), kResultNullIfNull,
                      "gdv_fn_like_utf8_utf8", NativeFunction::kNeedsFunctionHolder)};
 
diff --git a/cpp/src/gandiva/precompiled/string_ops.cc b/cpp/src/gandiva/precompiled/string_ops.cc
index 2125e18422f..545499a6a51 100644
--- a/cpp/src/gandiva/precompiled/string_ops.cc
+++ b/cpp/src/gandiva/precompiled/string_ops.cc
@@ -160,4 +160,16 @@ char* upper_utf8(int64 context, const char* data, int32 data_len, int32_t* out_l
   return ret;
 }
 
+// Truncates the string to given length
+FORCE_INLINE
+char* castVARCHAR_utf8_int64(int64 context, const char* data, int32 data_len,
+                             int64_t out_len, int32_t* out_length) {
+  // TODO: handle allocation failures
+  int32_t len = data_len <= static_cast<int32_t>(out_len) ? data_len : static_cast<int32_t>(out_len);
+  char* ret = reinterpret_cast<char*>(gdv_fn_context_arena_malloc(context, len));
+  memcpy(ret, data, len);
+  *out_length = len;
+  return ret;
+}
+
 }  // extern "C"
diff --git a/cpp/src/gandiva/tests/CMakeLists.txt b/cpp/src/gandiva/tests/CMakeLists.txt
index 0ec45fd2988..3c40b75d605 100644
--- a/cpp/src/gandiva/tests/CMakeLists.txt
+++ b/cpp/src/gandiva/tests/CMakeLists.txt
@@ -29,6 +29,7 @@ add_gandiva_test(in_expr_test)
 add_gandiva_test(null_validity_test)
 add_gandiva_test(decimal_test)
 add_gandiva_test(decimal_single_test)
+add_gandiva_test(utf8_test)
 
 add_gandiva_test(projector_test_static SOURCES projector_test.cc USE_STATIC_LINKING)
 
diff --git a/cpp/src/gandiva/tests/utf8_test.cc b/cpp/src/gandiva/tests/utf8_test.cc
index 925ceea8362..6fdc6efdf50 100644
--- a/cpp/src/gandiva/tests/utf8_test.cc
+++ b/cpp/src/gandiva/tests/utf8_test.cc
@@ -468,4 +468,52 @@ TEST_F(TestUtf8, TestToDateError) {
       << status.message();
 }
 
+TEST_F(TestUtf8, TestCastVarChar) {
+  // schema for input fields
+  auto field_a = field("a", utf8());
+  auto field_c = field("c", utf8());
+  auto schema = arrow::schema({field_a, field_c});
+
+  // output fields
+  auto res = field("res", boolean());
+
+  // build expressions.
+  auto node_a = TreeExprBuilder::MakeField(field_a);
+  auto node_c = TreeExprBuilder::MakeField(field_c);
+  // truncates the string to input length
+  auto node_b = TreeExprBuilder::MakeLiteral(static_cast<int64_t>(10));
+  auto cast_varchar =
+      TreeExprBuilder::MakeFunction("castVARCHAR", {node_a, node_b}, utf8());
+  auto equals = TreeExprBuilder::MakeFunction("equal", {cast_varchar, node_c}, boolean());
+  auto expr = TreeExprBuilder::MakeExpression(equals, res);
+
+  // Build a projector for the expressions.
+  std::shared_ptr<Projector> projector;
+  auto status = Projector::Make(schema, {expr}, TestConfiguration(), &projector);
+  EXPECT_TRUE(status.ok()) << status.message();
+
+  // Create a row-batch with some sample data
+  int num_records = 5;
+  auto array_a = MakeArrowArrayUtf8(
+      {"park", "Sparkle", "bright spark and fire", "fiery SPARK", "मदन"},
+      {true, true, false, true, true});
+
+  auto array_b =
+      MakeArrowArrayUtf8({"park", "Sparkle", "bright spar", "fiery SPAR", "मदन"},
+                         {true, true, true, true, true});
+
+  // prepare input record batch
+  auto in_batch = arrow::RecordBatch::Make(schema, num_records, {array_a, array_b});
+
+  // Evaluate expression
+  arrow::ArrayVector outputs;
+  status = projector->Evaluate(*in_batch, pool_, &outputs);
+  EXPECT_TRUE(status.ok()) << status.message();
+
+  auto exp = MakeArrowArrayBool({true, true, false, true, true},
+                                {true, true, false, true, true});
+  // Validate results
+  EXPECT_ARROW_ARRAY_EQUALS(exp, outputs[0]);
+}
+
 }  // namespace gandiva