Skip to content

GH-49471, GH-49455, GH-49442, GH-49439, GH-49421 Backport several upstream fixes. #136

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
lriggs merged 5 commits into from
Apr 24, 2026
Merged

GH-49471, GH-49455, GH-49442, GH-49439, GH-49421 Backport several upstream fixes. #136

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
6fa840e
GH-49420: [C++][Gandiva] Fix castVARCHAR memory allocation and len<=0…
dmitry-chirkov-dremio Mar 13, 2026
9faa93f
GH-49438: [C++][Gandiva] Optimize LPAD/RPAD functions (#49439)
dmitry-chirkov-dremio Mar 13, 2026
87b0953
GH-49441: [C++][Gandiva] Add rand_integer function (#49442)
dmitry-chirkov-dremio Mar 6, 2026
10397c2
GH-49454: [C++][Gandiva] Fix castVARCHAR_timestamp for pre-epoch time…
dmitry-chirkov-dremio Mar 13, 2026
9aabb9e
GH-49470: [C++][Gandiva] Fix crashes in substring_index and truncate …
dmitry-chirkov-dremio Mar 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions cpp/src/gandiva/function_holder_maker_registry.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,6 +62,7 @@ FunctionHolderMakerRegistry::MakerMap FunctionHolderMakerRegistry::DefaultHolder
{"to_date", HolderMaker<ToDateHolder>},
{"random", HolderMaker<RandomGeneratorHolder>},
{"rand", HolderMaker<RandomGeneratorHolder>},
{"rand_integer", HolderMaker<RandomIntegerGeneratorHolder>},
{"regexp_replace", HolderMaker<ReplaceHolder>},
{"regexp_extract", HolderMaker<ExtractHolder>},
{"castintervalday", HolderMaker<IntervalDaysHolder>},
Expand Down
8 changes: 8 additions & 0 deletions cpp/src/gandiva/function_registry_math_ops.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,6 +103,14 @@ std::vector<NativeFunction> GetMathOpsFunctionRegistry() {
"gdv_fn_random", NativeFunction::kNeedsFunctionHolder),
NativeFunction("random", {"rand"}, DataTypeVector{int32()}, float64(),
kResultNullNever, "gdv_fn_random_with_seed",
NativeFunction::kNeedsFunctionHolder),
NativeFunction("rand_integer", {}, DataTypeVector{}, int32(), kResultNullNever,
"gdv_fn_rand_integer", NativeFunction::kNeedsFunctionHolder),
NativeFunction("rand_integer", {}, DataTypeVector{int32()}, int32(),
kResultNullNever, "gdv_fn_rand_integer_with_range",
NativeFunction::kNeedsFunctionHolder),
NativeFunction("rand_integer", {}, DataTypeVector{int32(), int32()}, int32(),
kResultNullNever, "gdv_fn_rand_integer_with_min_max",
NativeFunction::kNeedsFunctionHolder)};

return math_fn_registry_;
Expand Down
39 changes: 38 additions & 1 deletion cpp/src/gandiva/gdv_function_stubs.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,12 +70,33 @@ double gdv_fn_random(int64_t ptr) {
return (*holder)();
}

double gdv_fn_random_with_seed(int64_t ptr, int32_t seed, bool seed_validity) {
double gdv_fn_random_with_seed(int64_t ptr, int32_t /*seed*/, bool /*seed_validity*/) {
gandiva::RandomGeneratorHolder* holder =
reinterpret_cast<gandiva::RandomGeneratorHolder*>(ptr);
return (*holder)();
}

int32_t gdv_fn_rand_integer(int64_t ptr) {
gandiva::RandomIntegerGeneratorHolder* holder =
reinterpret_cast<gandiva::RandomIntegerGeneratorHolder*>(ptr);
return (*holder)();
}

int32_t gdv_fn_rand_integer_with_range(int64_t ptr, int32_t /*range*/,
bool /*range_validity*/) {
gandiva::RandomIntegerGeneratorHolder* holder =
reinterpret_cast<gandiva::RandomIntegerGeneratorHolder*>(ptr);
return (*holder)();
}

int32_t gdv_fn_rand_integer_with_min_max(int64_t ptr, int32_t /*min*/,
bool /*min_validity*/, int32_t /*max*/,
bool /*max_validity*/) {
gandiva::RandomIntegerGeneratorHolder* holder =
reinterpret_cast<gandiva::RandomIntegerGeneratorHolder*>(ptr);
return (*holder)();
}

bool gdv_fn_in_expr_lookup_int32(int64_t ptr, int32_t value, bool in_validity) {
if (!in_validity) {
return false;
Expand Down Expand Up @@ -936,6 +957,22 @@ arrow::Status ExportedStubFunctions::AddMappings(Engine* engine) const {
engine->AddGlobalMappingForFunc("gdv_fn_random_with_seed", types->double_type(), args,
reinterpret_cast<void*>(gdv_fn_random_with_seed));

// gdv_fn_rand_integer
args = {types->i64_type()};
engine->AddGlobalMappingForFunc("gdv_fn_rand_integer", types->i32_type(), args,
reinterpret_cast<void*>(gdv_fn_rand_integer));

args = {types->i64_type(), types->i32_type(), types->i1_type()};
engine->AddGlobalMappingForFunc(
"gdv_fn_rand_integer_with_range", types->i32_type(), args,
reinterpret_cast<void*>(gdv_fn_rand_integer_with_range));

args = {types->i64_type(), types->i32_type(), types->i1_type(), types->i32_type(),
types->i1_type()};
engine->AddGlobalMappingForFunc(
"gdv_fn_rand_integer_with_min_max", types->i32_type(), args,
reinterpret_cast<void*>(gdv_fn_rand_integer_with_min_max));

// gdv_fn_dec_from_string
args = {
types->i64_type(), // context
Expand Down
61 changes: 61 additions & 0 deletions cpp/src/gandiva/gdv_function_stubs_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,8 @@
#include <gmock/gmock.h>
#include <gtest/gtest.h>

#include <limits>

#include "arrow/util/logging.h"
#include "gandiva/execution_context.h"
#include "gandiva/encrypt_utils_ecb.h"
Expand Down Expand Up @@ -353,6 +355,14 @@ TEST(TestGdvFnStubs, TestCastVARCHARFromInt64) {
out_str = gdv_fn_castVARCHAR_int64_int64(ctx_ptr, 12345, 3, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "123");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_int64_int64(ctx_ptr, 347, 0, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_int64_int64(ctx_ptr, 347, -1, &out_len);
EXPECT_THAT(ctx.get_error(), ::testing::HasSubstr("Buffer length cannot be negative"));
ctx.Reset();
}

TEST(TestGdvFnStubs, TestCastVARCHARFromMilliseconds) {
Expand Down Expand Up @@ -384,6 +394,15 @@ TEST(TestGdvFnStubs, TestCastVARCHARFromMilliseconds) {
out_str = gdv_fn_castVARCHAR_date64_int64(ctx_ptr, ts, 4, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "2008");
EXPECT_FALSE(ctx.has_error());

ts = StringToTimestamp("2021-04-23 10:20:33");
out_str = gdv_fn_castVARCHAR_date64_int64(ctx_ptr, ts, 0, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_date64_int64(ctx_ptr, ts, -1, &out_len);
EXPECT_THAT(ctx.get_error(), ::testing::HasSubstr("Buffer length cannot be negative"));
ctx.Reset();
}

TEST(TestGdvFnStubs, TestCastVARCHARFromFloat) {
Expand Down Expand Up @@ -419,6 +438,14 @@ TEST(TestGdvFnStubs, TestCastVARCHARFromFloat) {
out_str = gdv_fn_castVARCHAR_float32_int64(ctx_ptr, 1.2345f, 3, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "1.2");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_float32_int64(ctx_ptr, 1.2345f, 0, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_float32_int64(ctx_ptr, 1.2345f, -1, &out_len);
EXPECT_THAT(ctx.get_error(), ::testing::HasSubstr("Buffer length cannot be negative"));
ctx.Reset();
}

TEST(TestGdvFnStubs, TestCastVARCHARFromDouble) {
Expand Down Expand Up @@ -454,6 +481,25 @@ TEST(TestGdvFnStubs, TestCastVARCHARFromDouble) {
out_str = gdv_fn_castVARCHAR_float64_int64(ctx_ptr, 1.2345, 3, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "1.2");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_float64_int64(ctx_ptr, 1.2345, 0, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_castVARCHAR_float64_int64(ctx_ptr, 1.2345, -1, &out_len);
EXPECT_THAT(ctx.get_error(), ::testing::HasSubstr("Buffer length cannot be negative"));
ctx.Reset();

// test long repeating decimal (1/3) with large buffer
out_str = gdv_fn_castVARCHAR_float64_int64(ctx_ptr, 1.0 / 3.0, 100, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "0.3333333333333333");
EXPECT_FALSE(ctx.has_error());

// test exponential notation with large negative exponent (24 chars)
out_str =
gdv_fn_castVARCHAR_float64_int64(ctx_ptr, -1.2345678901234567e-100, 100, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "-1.2345678901234567E-100");
EXPECT_FALSE(ctx.has_error());
}

TEST(TestGdvFnStubs, TestSubstringIndex) {
Expand Down Expand Up @@ -529,6 +575,21 @@ TEST(TestGdvFnStubs, TestSubstringIndex) {
out_str = gdv_fn_substring_index(ctx_ptr, "路学\\L", 8, "\\", 1, -1, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "L");
EXPECT_FALSE(ctx.has_error());

// Large counts return full string when delimiter not found enough times
out_str = gdv_fn_substring_index(ctx_ptr, "a.b.c", 5, ".", 1, -1000, &out_len);
EXPECT_EQ(std::string(out_str, out_len), "a.b.c");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_substring_index(ctx_ptr, "a.b.c", 5, ".", 1,
std::numeric_limits<int32_t>::max(), &out_len);
EXPECT_EQ(std::string(out_str, out_len), "a.b.c");
EXPECT_FALSE(ctx.has_error());

out_str = gdv_fn_substring_index(ctx_ptr, "a.b.c", 5, ".", 1,
std::numeric_limits<int32_t>::min(), &out_len);
EXPECT_EQ(std::string(out_str, out_len), "a.b.c");
EXPECT_FALSE(ctx.has_error());
}

TEST(TestGdvFnStubs, TestUpper) {
Expand Down
Loading
Loading