use settingsenum

Author: arthurpassos

src/Core/Settings.cpp

@@ -6884,11 +6884,11 @@ Maximum number of retries for exporting a merge tree part in an export partition
 Determines how long the manifest will live in ZooKeeper. It prevents the same partition from being exported twice to the same destination.
 This setting does not affect / delete in progress tasks. It'll only cleanup the completed ones.
 )", 0) \
-    DECLARE(String, export_merge_tree_part_file_already_exists_policy, "NO_OP", R"(
+    DECLARE(MergeTreePartExportFileAlreadyExistsPolicy, export_merge_tree_part_file_already_exists_policy, MergeTreePartExportFileAlreadyExistsPolicy::skip, R"(
 Possible values:
-- NO_OP - No-op if the file already exists - Default.
-- ERROR - Throw an error if the file already exists.
-- OVERWRITE - Overwrite the file
+- skip - Skip the file if it already exists.
+- error - Throw an error if the file already exists.
+- overwrite - Overwrite the file.
 )", 0) \
     DECLARE(Timezone, iceberg_timezone_for_timestamptz, "UTC", R"(
 Timezone for Iceberg timestamptz field.

src/Core/Settings.h

@@ -81,6 +81,7 @@ class WriteBuffer;
     M(CLASS_NAME, LogsLevel) \
     M(CLASS_NAME, Map) \
     M(CLASS_NAME, MaxThreads) \
+    M(CLASS_NAME, MergeTreePartExportFileAlreadyExistsPolicy) \
     M(CLASS_NAME, Milliseconds) \
     M(CLASS_NAME, MsgPackUUIDRepresentation) \
     M(CLASS_NAME, MySQLDataTypesSupport) \

src/Core/SettingsChangesHistory.cpp

@@ -50,7 +50,7 @@ const VersionToSettingsChangesMap & getSettingsChangesHistory()
             {"export_merge_tree_partition_force_export", false, false, "New setting."},
             {"export_merge_tree_partition_max_retries", 3, 3, "New setting."},
             {"export_merge_tree_partition_manifest_ttl", 180, 180, "New setting."},
-            {"export_merge_tree_part_file_already_exists_policy", "NO_OP", "NO_OP", "New setting."},
+            {"export_merge_tree_part_file_already_exists_policy", "skip", "skip", "New setting."},
             {"iceberg_timezone_for_timestamptz", "UTC", "UTC", "New setting."}
         });
         addSettingsChanges(settings_changes_history, "25.8",

src/Core/SettingsEnums.cpp

@@ -370,4 +370,6 @@ IMPLEMENT_SETTING_ENUM(
      {"manifest_list_entry", IcebergMetadataLogLevel::ManifestListEntry},
      {"manifest_file_metadata", IcebergMetadataLogLevel::ManifestFileMetadata},
      {"manifest_file_entry", IcebergMetadataLogLevel::ManifestFileEntry}})
+
+IMPLEMENT_SETTING_AUTO_ENUM(MergeTreePartExportFileAlreadyExistsPolicy, ErrorCodes::BAD_ARGUMENTS);
 }

src/Core/SettingsEnums.h

@@ -480,4 +480,14 @@ enum class IcebergMetadataLogLevel : uint8_t
 };
 
 DECLARE_SETTING_ENUM(IcebergMetadataLogLevel)
+
+enum class MergeTreePartExportFileAlreadyExistsPolicy : uint8_t
+{
+    skip,
+    error,
+    overwrite,
+};
+
+DECLARE_SETTING_ENUM(MergeTreePartExportFileAlreadyExistsPolicy)
+
 }

src/Storages/MergeTree/ExportPartTask.cpp

@@ -95,7 +95,7 @@ bool ExportPartTask::executeStep()
             manifest.data_part->name + "_" + manifest.data_part->checksums.getTotalChecksumHex(),
             block_with_partition_values,
             (*exports_list_entry)->destination_file_path,
-            manifest.file_already_exists_policy == MergeTreePartExportManifest::FileAlreadyExistsPolicy::OVERWRITE,
+            manifest.file_already_exists_policy == MergeTreePartExportManifest::FileAlreadyExistsPolicy::overwrite,
             context_copy);
     }
     catch (const Exception & e)
@@ -105,7 +105,7 @@ bool ExportPartTask::executeStep()
             ProfileEvents::increment(ProfileEvents::PartsExportDuplicated);
 
             /// File already exists and the policy is NO_OP, treat it as success.
-            if (manifest.file_already_exists_policy == MergeTreePartExportManifest::FileAlreadyExistsPolicy::NO_OP)
+            if (manifest.file_already_exists_policy == MergeTreePartExportManifest::FileAlreadyExistsPolicy::skip)
             {
                 storage.writePartLog(
                     PartLogElement::Type::EXPORT_PART,

src/Storages/MergeTree/MergeTreeData.cpp

@@ -212,7 +212,7 @@ namespace Setting
     extern const SettingsBool apply_patch_parts;
     extern const SettingsBool allow_experimental_export_merge_tree_part;
     extern const SettingsUInt64 min_bytes_to_use_direct_io;
-    extern const SettingsString export_merge_tree_part_file_already_exists_policy;
+    extern const SettingsMergeTreePartExportFileAlreadyExistsPolicy export_merge_tree_part_file_already_exists_policy;
     extern const SettingsBool output_format_parallel_formatting;
     extern const SettingsBool output_format_parquet_parallel_encoding;
 }
@@ -6253,20 +6253,11 @@ void MergeTreeData::exportPartToTable(
                         part_name, getStorageID().getFullTableName());
 
     {
-        const auto file_already_exists_policy = magic_enum::enum_cast<MergeTreePartExportManifest::FileAlreadyExistsPolicy>(Poco::toUpper(query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value));
-        if (!file_already_exists_policy)
-        {
-            throw Exception(
-                ErrorCodes::INVALID_SETTING_VALUE,
-                "Invalid value for setting 'export_merge_tree_part_file_already_exists_policy': {}",
-                query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value);
-        }
-
         MergeTreePartExportManifest manifest(
             dest_storage->getStorageID(),
             part,
             transaction_id,
-            *file_already_exists_policy,
+            query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value,
             query_context->getSettingsRef()[Setting::output_format_parallel_formatting],
             query_context->getSettingsRef()[Setting::output_format_parquet_parallel_encoding],
             query_context->getSettingsRef()[Setting::max_threads],

src/Storages/MergeTree/MergeTreePartExportManifest.h

@@ -14,12 +14,7 @@ class ExportPartTask;
 
 struct MergeTreePartExportManifest
 {
-    enum class FileAlreadyExistsPolicy
-    {
-        NO_OP,
-        ERROR,
-        OVERWRITE,
-    };
+    using FileAlreadyExistsPolicy = MergeTreePartExportFileAlreadyExistsPolicy;
 
     using DataPartPtr = std::shared_ptr<const IMergeTreeDataPart>;
 

src/Storages/StorageReplicatedMergeTree.cpp

@@ -198,7 +198,7 @@ namespace Setting
     extern const SettingsBool output_format_parallel_formatting;
     extern const SettingsBool output_format_parquet_parallel_encoding;
     extern const SettingsMaxThreads max_threads;
-    extern const SettingsString export_merge_tree_part_file_already_exists_policy;
+    extern const SettingsMergeTreePartExportFileAlreadyExistsPolicy export_merge_tree_part_file_already_exists_policy;
 }
 
 namespace MergeTreeSetting
@@ -8205,17 +8205,7 @@ void StorageReplicatedMergeTree::exportPartitionToTable(const PartitionCommand &
     manifest.parallel_formatting = query_context->getSettingsRef()[Setting::output_format_parallel_formatting];
     manifest.parquet_parallel_encoding = query_context->getSettingsRef()[Setting::output_format_parquet_parallel_encoding];
 
-    const auto file_already_exists_policy = magic_enum::enum_cast<MergeTreePartExportManifest::FileAlreadyExistsPolicy>(query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value);
-
-    if (!file_already_exists_policy)
-    {
-        throw Exception(
-            ErrorCodes::INVALID_SETTING_VALUE,
-            "Invalid value for setting 'export_merge_tree_part_file_already_exists_policy': {}",
-            query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value);
-    }
-
-    manifest.file_already_exists_policy = file_already_exists_policy.value();
+    manifest.file_already_exists_policy = query_context->getSettingsRef()[Setting::export_merge_tree_part_file_already_exists_policy].value;
 
     ops.emplace_back(zkutil::makeCreateRequest(
         fs::path(partition_exports_path) / "metadata.json", 

tests/integration/test_export_replicated_mt_partition_to_object_storage/test.py

@@ -506,7 +506,7 @@ def test_export_partition_file_already_exists_policy(cluster):
 
     # now let's try with a file exists policy that is not NO_OP
     node.query(
-        f"ALTER TABLE {mt_table} EXPORT PARTITION ID '2020' TO TABLE {s3_table} SETTINGS allow_experimental_export_merge_tree_part=1, export_merge_tree_partition_force_export=1, export_merge_tree_part_file_already_exists_policy='OVERWRITE'",
+        f"ALTER TABLE {mt_table} EXPORT PARTITION ID '2020' TO TABLE {s3_table} SETTINGS allow_experimental_export_merge_tree_part=1, export_merge_tree_partition_force_export=1, export_merge_tree_part_file_already_exists_policy='overwrite'",
         query_id=query_id_3,
     )
 
@@ -527,7 +527,7 @@ def test_export_partition_file_already_exists_policy(cluster):
     # last but not least, let's try with the error policy
     # max retries = 1 so it fails fast
     node.query(
-        f"ALTER TABLE {mt_table} EXPORT PARTITION ID '2020' TO TABLE {s3_table} SETTINGS allow_experimental_export_merge_tree_part=1, export_merge_tree_partition_force_export=1, export_merge_tree_part_file_already_exists_policy='ERROR', export_merge_tree_partition_max_retries=1",
+        f"ALTER TABLE {mt_table} EXPORT PARTITION ID '2020' TO TABLE {s3_table} SETTINGS allow_experimental_export_merge_tree_part=1, export_merge_tree_partition_force_export=1, export_merge_tree_part_file_already_exists_policy='error', export_merge_tree_partition_max_retries=1",
         query_id=query_id_4,
     )