Copilot commented on code in PR #293:
URL: https://github.com/apache/iceberg-cpp/pull/293#discussion_r2504393925
##########
src/iceberg/v3_metadata.cc:
##########
@@ -23,47 +23,68 @@
#include "iceberg/manifest_entry.h"
#include "iceberg/manifest_list.h"
#include "iceberg/schema.h"
+#include "iceberg/schema_internal.h"
#include "iceberg/util/macros.h"
namespace iceberg {
+std::shared_ptr<Schema> ManifestEntryAdapterV3::EntrySchema(
+ std::shared_ptr<StructType> partition_type) {
+ return WrapFileSchema(FileType(std::move(partition_type)));
+}
+
+std::shared_ptr<Schema> ManifestEntryAdapterV3::WrapFileSchema(
+ std::shared_ptr<StructType> file_schema) {
+ return std::make_shared<Schema>(std::vector<SchemaField>{
+ ManifestEntry::kStatus,
+ ManifestEntry::kSnapshotId,
+ ManifestEntry::kSequenceNumber,
+ ManifestEntry::kFileSequenceNumber,
+ SchemaField::MakeRequired(ManifestEntry::kDataFileFieldId,
+ ManifestEntry::kDataFileField,
std::move(file_schema)),
+ });
+}
+
+std::shared_ptr<StructType> ManifestEntryAdapterV3::FileType(
+ std::shared_ptr<StructType> partition_type) {
+ return std::make_shared<StructType>(std::vector<SchemaField>{
+ DataFile::kContent.AsRequired(),
+ DataFile::kFilePath,
+ DataFile::kFileFormat,
+ SchemaField::MakeRequired(DataFile::kPartitionFieldId,
DataFile::kPartitionField,
+ std::move(partition_type),
DataFile::kPartitionDoc),
+ DataFile::kRecordCount,
+ DataFile::kFileSize,
+ DataFile::kColumnSizes,
+ DataFile::kValueCounts,
+ DataFile::kNullValueCounts,
+ DataFile::kNanValueCounts,
+ DataFile::kLowerBounds,
+ DataFile::kUpperBounds,
+ DataFile::kKeyMetadata,
+ DataFile::kSplitOffsets,
+ DataFile::kEqualityIds,
+ DataFile::kSortOrderId,
+ DataFile::kFirstRowId,
+ DataFile::kReferencedDataFile,
+ DataFile::kContentOffset,
+ DataFile::kContentSize,
+ });
+}
+
Status ManifestEntryAdapterV3::Init() {
- static std::unordered_set<int32_t> kManifestEntryFieldIds{
- ManifestEntry::kStatus.field_id(),
- ManifestEntry::kSnapshotId.field_id(),
- ManifestEntry::kDataFileFieldId,
- ManifestEntry::kSequenceNumber.field_id(),
- ManifestEntry::kFileSequenceNumber.field_id(),
- DataFile::kContent.field_id(),
- DataFile::kFilePath.field_id(),
- DataFile::kFileFormat.field_id(),
- DataFile::kPartitionFieldId,
- DataFile::kRecordCount.field_id(),
- DataFile::kFileSize.field_id(),
- DataFile::kColumnSizes.field_id(),
- DataFile::kValueCounts.field_id(),
- DataFile::kNullValueCounts.field_id(),
- DataFile::kNanValueCounts.field_id(),
- DataFile::kLowerBounds.field_id(),
- DataFile::kUpperBounds.field_id(),
- DataFile::kKeyMetadata.field_id(),
- DataFile::kSplitOffsets.field_id(),
- DataFile::kEqualityIds.field_id(),
- DataFile::kSortOrderId.field_id(),
- DataFile::kFirstRowId.field_id(),
- DataFile::kReferencedDataFile.field_id(),
- DataFile::kContentOffset.field_id(),
- DataFile::kContentSize.field_id(),
- };
- ICEBERG_RETURN_UNEXPECTED(InitSchema(kManifestEntryFieldIds));
- ICEBERG_ASSIGN_OR_RAISE(metadata_["schema"], ToJsonString(*manifest_schema_))
- if (partition_spec_ != nullptr) {
- ICEBERG_ASSIGN_OR_RAISE(metadata_["partition-spec"],
ToJsonString(*partition_spec_));
- metadata_["partition-spec-id"] =
std::to_string(partition_spec_->spec_id());
- }
+ // ICEBERG_ASSIGN_OR_RAISE(metadata_["schema"],
ToJsonString(*manifest_schema_))
Review Comment:
Commented-out code should be removed or explained with a TODO comment. If
this assignment will be needed in the future, add a TODO comment explaining
when and why it should be re-enabled.
```suggestion
```
##########
src/iceberg/v2_metadata.cc:
##########
@@ -23,44 +23,63 @@
#include "iceberg/manifest_entry.h"
#include "iceberg/manifest_list.h"
#include "iceberg/schema.h"
+#include "iceberg/schema_internal.h"
#include "iceberg/util/macros.h"
namespace iceberg {
+std::shared_ptr<Schema> ManifestEntryAdapterV2::EntrySchema(
+ std::shared_ptr<StructType> partition_type) {
+ return WrapFileSchema(FileType(std::move(partition_type)));
+}
+std::shared_ptr<Schema> ManifestEntryAdapterV2::WrapFileSchema(
+ std::shared_ptr<StructType> file_schema) {
+ return std::make_shared<Schema>(std::vector<SchemaField>{
+ ManifestEntry::kStatus,
+ ManifestEntry::kSnapshotId,
+ ManifestEntry::kSequenceNumber,
+ ManifestEntry::kFileSequenceNumber,
+ SchemaField::MakeRequired(ManifestEntry::kDataFileFieldId,
+ ManifestEntry::kDataFileField,
std::move(file_schema)),
+ });
+}
+std::shared_ptr<StructType> ManifestEntryAdapterV2::FileType(
+ std::shared_ptr<StructType> partition_type) {
+ return std::make_shared<StructType>(std::vector<SchemaField>{
+ DataFile::kContent.AsRequired(),
+ DataFile::kFilePath,
+ DataFile::kFileFormat,
+ SchemaField::MakeRequired(DataFile::kPartitionFieldId,
DataFile::kPartitionField,
+ std::move(partition_type),
DataFile::kPartitionDoc),
+ DataFile::kRecordCount,
+ DataFile::kFileSize,
+ DataFile::kColumnSizes,
+ DataFile::kValueCounts,
+ DataFile::kNullValueCounts,
+ DataFile::kNanValueCounts,
+ DataFile::kLowerBounds,
+ DataFile::kUpperBounds,
+ DataFile::kKeyMetadata,
+ DataFile::kSplitOffsets,
+ DataFile::kEqualityIds,
+ DataFile::kSortOrderId,
+ DataFile::kReferencedDataFile,
+ });
+}
+
Status ManifestEntryAdapterV2::Init() {
- static std::unordered_set<int32_t> kManifestEntryFieldIds{
- ManifestEntry::kStatus.field_id(),
- ManifestEntry::kSnapshotId.field_id(),
- ManifestEntry::kSequenceNumber.field_id(),
- ManifestEntry::kFileSequenceNumber.field_id(),
- ManifestEntry::kDataFileFieldId,
- DataFile::kContent.field_id(),
- DataFile::kFilePath.field_id(),
- DataFile::kFileFormat.field_id(),
- DataFile::kPartitionFieldId,
- DataFile::kRecordCount.field_id(),
- DataFile::kFileSize.field_id(),
- DataFile::kColumnSizes.field_id(),
- DataFile::kValueCounts.field_id(),
- DataFile::kNullValueCounts.field_id(),
- DataFile::kNanValueCounts.field_id(),
- DataFile::kLowerBounds.field_id(),
- DataFile::kUpperBounds.field_id(),
- DataFile::kKeyMetadata.field_id(),
- DataFile::kSplitOffsets.field_id(),
- DataFile::kEqualityIds.field_id(),
- DataFile::kSortOrderId.field_id(),
- DataFile::kReferencedDataFile.field_id(),
- };
- ICEBERG_RETURN_UNEXPECTED(InitSchema(kManifestEntryFieldIds));
- ICEBERG_ASSIGN_OR_RAISE(metadata_["schema"], ToJsonString(*manifest_schema_))
- if (partition_spec_ != nullptr) {
- ICEBERG_ASSIGN_OR_RAISE(metadata_["partition-spec"],
ToJsonString(*partition_spec_));
- metadata_["partition-spec-id"] =
std::to_string(partition_spec_->spec_id());
- }
+ // ICEBERG_ASSIGN_OR_RAISE(metadata_["schema"],
ToJsonString(*manifest_schema_))
Review Comment:
Commented-out code should be removed or explained with a TODO comment. If
this assignment will be needed in the future, add a TODO comment explaining
when and why it should be re-enabled.
```suggestion
```
--
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]
