Skip to content

test: add test of manifestlistv1 #171

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
merged 2 commits into from
Aug 13, 2025
Merged

test: add test of manifestlistv1 #171

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fc10fd2
test: add test of manifestlistv1
Aug 11, 2025
2ef55aa
test: add & refactor manifestlistv1 test functions
Aug 13, 2025
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
221 changes: 167 additions & 54 deletions test/manifest_list_reader_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,72 +32,185 @@

namespace iceberg {

class ManifestListReaderTest : public TempFileTestBase {
class ManifestListReaderV1Test : public ::testing::Test {
protected:
static void SetUpTestSuite() { avro::AvroReader::Register(); }

void SetUp() override {
TempFileTestBase::SetUp();
local_fs_ = std::make_shared<::arrow::fs::LocalFileSystem>();
file_io_ = std::make_shared<iceberg::arrow::ArrowFileSystemFileIO>(local_fs_);
}

std::vector<ManifestFile> PrepareTestManifestList() {
std::vector<ManifestFile> manifest_files;
std::string test_dir_prefix = "/tmp/db/db/iceberg_test/metadata/";
std::vector<std::string> paths = {"2bccd69e-d642-4816-bba0-261cd9bd0d93-m0.avro",
"9b6ffacd-ef10-4abf-a89c-01c733696796-m0.avro",
"2541e6b5-4923-4bd5-886d-72c6f7228400-m0.avro",
"3118c801-d2e0-4df6-8c7a-7d4eaade32f8-m0.avro"};
std::vector<int64_t> file_size = {7433, 7431, 7433, 7431};
std::vector<int64_t> snapshot_id = {7412193043800610213, 5485972788975780755,
1679468743751242972, 1579605567338877265};
std::vector<std::vector<uint8_t>> bounds = {{'x', ';', 0x07, 0x00},
{'(', 0x19, 0x07, 0x00},
{0xd0, 0xd4, 0x06, 0x00},
{0xb8, 0xd4, 0x06, 0x00}};
for (int i = 0; i < 4; ++i) {
ManifestFile manifest_file;
manifest_file.manifest_path = test_dir_prefix + paths[i];
manifest_file.manifest_length = file_size[i];
manifest_file.partition_spec_id = 0;
manifest_file.content = ManifestFile::Content::kData;
manifest_file.sequence_number = 4 - i;
manifest_file.min_sequence_number = 4 - i;
manifest_file.added_snapshot_id = snapshot_id[i];
manifest_file.added_files_count = 1;
manifest_file.existing_files_count = 0;
manifest_file.deleted_files_count = 0;
manifest_file.added_rows_count = 1;
manifest_file.existing_rows_count = 0;
manifest_file.deleted_rows_count = 0;
PartitionFieldSummary partition;
partition.contains_null = false;
partition.contains_nan = false;
partition.lower_bound = bounds[i];
partition.upper_bound = bounds[i];
manifest_file.partitions.emplace_back(partition);
manifest_files.emplace_back(manifest_file);
}
return manifest_files;
}

std::shared_ptr<::arrow::fs::LocalFileSystem> local_fs_;
std::shared_ptr<FileIO> file_io_;

void TestManifestListReading(const std::string& resource_name,
const std::vector<ManifestFile>& expected_manifest_list) {
std::string path = GetResourcePath(resource_name);
auto manifest_reader_result = ManifestListReader::MakeReader(path, file_io_);
ASSERT_EQ(manifest_reader_result.has_value(), true);

auto manifest_reader = std::move(manifest_reader_result.value());
auto read_result = manifest_reader->Files();
ASSERT_EQ(read_result.has_value(), true);
ASSERT_EQ(read_result.value().size(), expected_manifest_list.size());
ASSERT_EQ(read_result.value(), expected_manifest_list);
}
};

TEST_F(ManifestListReaderTest, BasicTest) {
std::string path = GetResourcePath(
"snap-7412193043800610213-1-2bccd69e-d642-4816-bba0-261cd9bd0d93.avro");
auto manifest_reader_result = ManifestListReader::MakeReader(path, file_io_);
ASSERT_EQ(manifest_reader_result.has_value(), true);
auto manifest_reader = std::move(manifest_reader_result.value());
auto read_result = manifest_reader->Files();
ASSERT_EQ(read_result.has_value(), true);
ASSERT_EQ(read_result.value().size(), 4);

auto expected_manifest_list = PrepareTestManifestList();
ASSERT_EQ(read_result.value(), expected_manifest_list);
TEST_F(ManifestListReaderV1Test, PartitionTest) {
std::vector<std::string> paths = {
"iceberg-warehouse/db/v1_partition_test/metadata/"
"eafd2972-f58e-4185-9237-6378f564787e-m1.avro",
"iceberg-warehouse/db/v1_partition_test/metadata/"
"eafd2972-f58e-4185-9237-6378f564787e-m0.avro"};
std::vector<int64_t> file_size = {6185, 6113};
std::vector<int64_t> snapshot_id = {7532614258660258098, 7532614258660258098};

std::vector<std::vector<std::uint8_t>> lower_bounds = {
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x30, 0x32, 0x2D, 0x32, 0x32},
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x32}};

std::vector<std::vector<std::uint8_t>> upper_bounds = {
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x33},
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x33}};

std::vector<ManifestFile> expected_manifest_list = {
{.manifest_path = paths[0],
.manifest_length = file_size[0],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[0],
.added_files_count = 4,
.existing_files_count = 0,
.deleted_files_count = 0,
.added_rows_count = 6,
.existing_rows_count = 0,
.deleted_rows_count = 0,
.partitions = {{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[0],
.upper_bound = upper_bounds[0]}}},

{.manifest_path = paths[1],
.manifest_length = file_size[1],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[1],
.added_files_count = 0,
.existing_files_count = 0,
.deleted_files_count = 2,
.added_rows_count = 0,
.existing_rows_count = 0,
.deleted_rows_count = 6,
.partitions = {{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[1],
.upper_bound = upper_bounds[1]}}}};

TestManifestListReading(
"snap-7532614258660258098-1-eafd2972-f58e-4185-9237-6378f564787e.avro",
expected_manifest_list);
}

TEST_F(ManifestListReaderV1Test, ComplexTypeTest) {
std::vector<std::string> paths = {
"iceberg-warehouse/db/v1_type_test/metadata/"
"aeffe099-3bac-4011-bc17-5875210d8dc0-m1.avro",
"iceberg-warehouse/db/v1_type_test/metadata/"
"aeffe099-3bac-4011-bc17-5875210d8dc0-m0.avro"};
std::vector<int64_t> file_size = {6498, 6513};
std::vector<int64_t> snapshot_id = {4134160420377642835, 4134160420377642835};

std::vector<ManifestFile> expected_manifest_list = {
{
.manifest_path = paths[0],
.manifest_length = file_size[0],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[0],
.added_files_count = 1,
.existing_files_count = 0,
.deleted_files_count = 0,
.added_rows_count = 2,
.existing_rows_count = 0,
.deleted_rows_count = 0,
},

{.manifest_path = paths[1],
.manifest_length = file_size[1],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[1],
.added_files_count = 0,
.existing_files_count = 0,
.deleted_files_count = 1,
.added_rows_count = 0,
.existing_rows_count = 0,
.deleted_rows_count = 3}};

TestManifestListReading(
"snap-4134160420377642835-1-aeffe099-3bac-4011-bc17-5875210d8dc0.avro",
expected_manifest_list);
}

TEST_F(ManifestListReaderV1Test, PartitionComplexTypeTest) {
std::vector<std::string> paths = {
"iceberg-warehouse/db2/v1_complex_partition_test/metadata/"
"5d690750-8fb4-4cd1-8ae7-85c7b39abe14-m0.avro",
"iceberg-warehouse/db2/v1_complex_partition_test/metadata/"
"5d690750-8fb4-4cd1-8ae7-85c7b39abe14-m1.avro"};
std::vector<int64_t> file_size = {6402, 6318};
std::vector<int64_t> snapshot_id = {7522296285847100621, 7522296285847100621};

std::vector<std::vector<std::uint8_t>> lower_bounds = {
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x32},
{0x02, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00},
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x32},
{0x02, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}};

std::vector<std::vector<std::uint8_t>> upper_bounds = {
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x34},
{0x05, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00},
{0x32, 0x30, 0x32, 0x32, 0x2D, 0x32, 0x2D, 0x32, 0x33},
{0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}};
std::vector<ManifestFile> expected_manifest_list = {
{.manifest_path = paths[0],
.manifest_length = file_size[0],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[0],
.added_files_count = 0,
.existing_files_count = 3,
.deleted_files_count = 1,
.added_rows_count = 0,
.existing_rows_count = 4,
.deleted_rows_count = 2,
.partitions = {{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[0],
.upper_bound = upper_bounds[0]},
{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[1],
.upper_bound = upper_bounds[1]}}},

{.manifest_path = paths[1],
.manifest_length = file_size[1],
.partition_spec_id = 0,
.added_snapshot_id = snapshot_id[1],
.added_files_count = 0,
.existing_files_count = 1,
.deleted_files_count = 1,
.added_rows_count = 0,
.existing_rows_count = 1,
.deleted_rows_count = 1,
.partitions = {{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[2],
.upper_bound = upper_bounds[2]},
{.contains_null = false,
.contains_nan = false,
.lower_bound = lower_bounds[3],
.upper_bound = upper_bounds[3]}}}};
TestManifestListReading(
"snap-7522296285847100621-1-5d690750-8fb4-4cd1-8ae7-85c7b39abe14.avro",
expected_manifest_list);
}

} // namespace iceberg
Binary file added test/resources/snap-4134160420377642835-1-aeffe099-3bac-4011-bc17-5875210d8dc0.avro
View file
Open in desktop
Binary file not shown.
Binary file added test/resources/snap-7522296285847100621-1-5d690750-8fb4-4cd1-8ae7-85c7b39abe14.avro
View file
Open in desktop
Binary file not shown.
Binary file added test/resources/snap-7532614258660258098-1-eafd2972-f58e-4185-9237-6378f564787e.avro
View file
Open in desktop
Binary file not shown.
Loading