manuzhang commented on code in PR #1066:
URL: https://github.com/apache/iceberg-cpp/pull/1066#discussion_r4389595686
##########
src/iceberg/test/snapshots_table_test.cc:
##########
@@ -21,212 +21,127 @@
#include <cstdint>
#include <memory>
-#include <string>
#include <utility>
#include <vector>
-#include <arrow/array.h>
-#include <arrow/c/bridge.h>
-#include <arrow/record_batch.h>
-#include <gmock/gmock.h>
#include <gtest/gtest.h>
-#include "iceberg/constants.h"
-#include "iceberg/inspect/metadata_table.h"
+#include "iceberg/inspect/metadata_scan_task_reader.h"
+#include "iceberg/schema.h"
+#include "iceberg/schema_field.h"
+#include "iceberg/snapshot.h"
+#include "iceberg/table_metadata.h"
+#include "iceberg/test/arrow_test_util.h"
#include "iceberg/test/matchers.h"
#include "iceberg/test/metadata_table_test_base.h"
+#include "iceberg/type.h"
+#include "iceberg/util/timepoint.h"
namespace iceberg {
-namespace {
-
-std::vector<std::pair<std::string, std::string>> GetMapEntries(
- const std::shared_ptr<::arrow::MapArray>& map_array, int64_t row) {
- auto keys =
std::static_pointer_cast<::arrow::StringArray>(map_array->keys());
- auto values =
std::static_pointer_cast<::arrow::StringArray>(map_array->items());
- std::vector<std::pair<std::string, std::string>> entries;
- entries.reserve(map_array->value_length(row));
- const auto offset = map_array->value_offset(row);
- for (int64_t index = offset; index < offset + map_array->value_length(row);
++index) {
- entries.emplace_back(keys->GetString(index), values->GetString(index));
- }
- return entries;
-}
-
-} // namespace
class SnapshotsTableTest : public MetadataTableTestBase {
protected:
void SetUp() override {
MetadataTableTestBase::SetUp();
+ metadata_->snapshots = {
+ std::make_shared<Snapshot>(
+ Snapshot{.snapshot_id = 11,
+ .parent_snapshot_id = std::nullopt,
+ .timestamp_ms = TimePointMsFromUnixMs(1234),
+ .manifest_list = "file:/manifest11.avro",
+ .summary = {{"operation", "append"}, {"total-records",
"7"}}}),
+ std::make_shared<Snapshot>(Snapshot{.snapshot_id = 12,
+ .parent_snapshot_id = 11,
+ .timestamp_ms =
TimePointMsFromUnixMs(2345),
+ .manifest_list =
"file:/manifest12.avro",
+ .summary = {{"operation",
"delete"}}}),
+ std::make_shared<Snapshot>(Snapshot{.snapshot_id = 13,
+ .parent_snapshot_id = 12,
+ .timestamp_ms =
TimePointMsFromUnixMs(3456),
+ .manifest_list =
"file:/manifest13.avro"})};
+ metadata_->current_snapshot_id = 13;
+ ICEBERG_UNWRAP_OR_FAIL(metadata_table_,
+ table_->Inspect(MetadataTableKind::kSnapshots));
+ }
- auto [snap1, snap2] = MakeTestSnapshots();
- ICEBERG_UNWRAP_OR_FAIL(
- table_, MakeTableWithSnapshots({snap1, snap2},
/*current_snapshot_id=*/2));
-
- ICEBERG_UNWRAP_OR_FAIL(snapshots_table_,
MetadataTable::Make<SnapshotsTable>(table_));
+ static std::shared_ptr<Schema> ExpectedSchema() {
+ return std::make_shared<Schema>(std::vector<SchemaField>{
+ SchemaField::MakeRequired(1, "committed_at", timestamp_tz()),
+ SchemaField::MakeRequired(2, "snapshot_id", int64()),
+ SchemaField::MakeOptional(3, "parent_id", int64()),
+ SchemaField::MakeOptional(4, "operation", string()),
+ SchemaField::MakeOptional(5, "manifest_list", string()),
+ SchemaField::MakeOptional(
+ 6, "summary",
+ std::make_shared<MapType>(SchemaField::MakeRequired(7, "key",
string()),
+ SchemaField::MakeRequired(8, "value",
string())))});
}
- std::unique_ptr<MetadataTable> snapshots_table_;
+ std::unique_ptr<MetadataTable> metadata_table_;
};
-TEST_F(SnapshotsTableTest, Construct) {
- EXPECT_EQ(snapshots_table_->kind(), MetadataTable::Kind::kSnapshots);
- EXPECT_EQ(snapshots_table_->source_table(), table_);
- EXPECT_NE(snapshots_table_->schema(), nullptr);
-}
-
-TEST_F(SnapshotsTableTest, Scan) {
- // Scan the snapshots table once and verify all columns of the result.
- ICEBERG_UNWRAP_OR_FAIL(auto stream, snapshots_table_->Scan());
- ICEBERG_UNWRAP_OR_FAIL(auto batches, ReadAllBatches(std::move(stream)));
- ASSERT_EQ(batches.size(), 1);
- const auto& batch = batches.front();
-
- // Row and column counts.
- EXPECT_EQ(batch->num_rows(), 2);
- EXPECT_EQ(batch->num_columns(), 6);
-
- // Column 0: committed_at (timestamptz) — microseconds since epoch.
- auto committed_at =
std::static_pointer_cast<::arrow::TimestampArray>(batch->column(0));
- EXPECT_EQ(committed_at->Value(0), 1234567890000LL * 1000);
- EXPECT_EQ(committed_at->Value(1), 9876543210000LL * 1000);
-
- // Column 1: snapshot_id (long) — returned in storage order.
- auto snapshot_ids =
std::static_pointer_cast<::arrow::Int64Array>(batch->column(1));
- EXPECT_EQ(snapshot_ids->Value(0), 1);
- EXPECT_EQ(snapshot_ids->Value(1), 2);
-
- // Column 2: parent_id (long) — first snapshot has no parent.
- auto parent_ids =
std::static_pointer_cast<::arrow::Int64Array>(batch->column(2));
- EXPECT_TRUE(parent_ids->IsNull(0));
- EXPECT_FALSE(parent_ids->IsNull(1));
- EXPECT_EQ(parent_ids->Value(1), 1);
-
- // Column 3: operation (string).
- auto operations =
std::static_pointer_cast<::arrow::StringArray>(batch->column(3));
- EXPECT_EQ(operations->GetString(0), "append");
- EXPECT_EQ(operations->GetString(1), "append");
-
- // Column 4: manifest_list (string).
- auto manifest_lists =
std::static_pointer_cast<::arrow::StringArray>(batch->column(4));
- EXPECT_EQ(manifest_lists->GetString(0), "file:/tmp/manifest1.avro");
- EXPECT_EQ(manifest_lists->GetString(1), "file:/tmp/manifest2.avro");
-
- // Column 5: summary (map<string,string>) excludes the separate operation
field.
- auto summaries =
std::static_pointer_cast<::arrow::MapArray>(batch->column(5));
- EXPECT_FALSE(summaries->IsNull(0));
- EXPECT_FALSE(summaries->IsNull(1));
- EXPECT_EQ(summaries->value_length(0), 10);
- EXPECT_EQ(summaries->value_length(1), 10);
-
- auto first_summary = GetMapEntries(summaries, 0);
- EXPECT_THAT(
- first_summary,
- ::testing::Not(::testing::Contains(::testing::Pair("operation",
"append"))));
- EXPECT_THAT(first_summary,
::testing::Contains(::testing::Pair("total-records", "1")));
-
- auto second_summary = GetMapEntries(summaries, 1);
- EXPECT_THAT(
- second_summary,
- ::testing::Not(::testing::Contains(::testing::Pair("operation",
"append"))));
- EXPECT_THAT(second_summary,
::testing::Contains(::testing::Pair("total-records", "2")));
+TEST_F(SnapshotsTableTest, CanonicalSchema) {
+ const auto expected = ExpectedSchema();
+ EXPECT_EQ(*SnapshotsTable::CanonicalSchema(), *expected);
+ EXPECT_EQ(*metadata_table_->schema(), *expected);
}
-TEST_F(SnapshotsTableTest, ScanEmptySnapshotList) {
Review Comment:
Do we have the same coverage as before?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
