Distributed request to tables with Object Storage Engines

src/Storages/ObjectStorage/Azure/Configuration.cpp

@@ -459,7 +459,7 @@ void StorageAzureConfiguration::addStructureAndFormatToArgsIfNeeded(
     }
 }
 
-void StorageAzureConfiguration::setFunctionArgs(ASTs & args) const
+void StorageAzureConfiguration::getTableFunctionArguments(ASTs & args) const
 {
     if (!args.empty())
     { /// Just check

src/Storages/ObjectStorage/Azure/Configuration.h

@@ -79,7 +79,7 @@ class StorageAzureConfiguration : public StorageObjectStorage::Configuration
         ContextPtr context,
         bool with_structure) override;
 
-    void setFunctionArgs(ASTs & args) const override;
+    void getTableFunctionArguments(ASTs & args) const override;
 
 protected:
     void fromNamedCollection(const NamedCollection & collection, ContextPtr context) override;

src/Storages/ObjectStorage/HDFS/Configuration.cpp

@@ -235,7 +235,7 @@ void StorageHDFSConfiguration::addStructureAndFormatToArgsIfNeeded(
     }
 }
 
-void StorageHDFSConfiguration::setFunctionArgs(ASTs & args) const
+void StorageHDFSConfiguration::getTableFunctionArguments(ASTs & args) const
 {
     if (!args.empty())
     { /// Just check

src/Storages/ObjectStorage/HDFS/Configuration.h

@@ -65,7 +65,7 @@ class StorageHDFSConfiguration : public StorageObjectStorage::Configuration
         ContextPtr context,
         bool with_structure) override;
 
-    void setFunctionArgs(ASTs & args) const override;
+    void getTableFunctionArguments(ASTs & args) const override;
 
 private:
     void fromNamedCollection(const NamedCollection &, ContextPtr context) override;

src/Storages/ObjectStorage/S3/Configuration.cpp

@@ -585,7 +585,7 @@ void StorageS3Configuration::addStructureAndFormatToArgsIfNeeded(
     }
 }
 
-void StorageS3Configuration::setFunctionArgs(ASTs & args) const
+void StorageS3Configuration::getTableFunctionArguments(ASTs & args) const
 {
     if (!args.empty())
     { /// Just check

src/Storages/ObjectStorage/S3/Configuration.h

@@ -94,7 +94,7 @@ class StorageS3Configuration : public StorageObjectStorage::Configuration
         ContextPtr context,
         bool with_structure) override;
 
-    void setFunctionArgs(ASTs & args) const override;
+    void getTableFunctionArguments(ASTs & args) const override;
 
 private:
     void fromNamedCollection(const NamedCollection & collection, ContextPtr context) override;

src/Storages/ObjectStorage/StorageObjectStorage.h

@@ -247,9 +247,9 @@ class StorageObjectStorage::Configuration
 
     virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context);
 
-    virtual void setFunctionArgs(ASTs & /* args */) const
+    virtual void getTableFunctionArguments(ASTs & /* args */) const
     {
-        throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method setFunctionArgs is not supported by storage {}", getEngineName());
+        throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method getTableFunctionArguments is not supported by storage {}", getEngineName());
     }
 
 protected:

src/Storages/ObjectStorage/StorageObjectStorageCluster.cpp

@@ -108,6 +108,13 @@ void StorageObjectStorageCluster::updateQueryForDistributedEngineIfNeeded(ASTPtr
         return;
 
     auto * tables = select_query->tables()->as<ASTTablesInSelectQuery>();
+
+    if (tables->children.empty())
+        throw Exception(
+            ErrorCodes::LOGICAL_ERROR,
+            "Expected SELECT query from table with engine {}, got '{}'",
+            configuration->getEngineName(), queryToString(query));
+
     auto * table_expression = tables->children[0]->as<ASTTablesInSelectQueryElement>()->table_expression->as<ASTTableExpression>();
     if (!table_expression->database_and_table_name)
         return;
@@ -116,7 +123,7 @@ void StorageObjectStorageCluster::updateQueryForDistributedEngineIfNeeded(ASTPtr
 
     auto table_alias = table_identifier_typed.tryGetAlias();
 
-    std::unordered_map<std::string, std::string> engine_to_function = {
+    static std::unordered_map<std::string, std::string> engine_to_function = {
         {"S3", "s3"},
         {"Azure", "azureBlobStorage"},
         {"HDFS", "hdfs"},
@@ -153,7 +160,7 @@ void StorageObjectStorageCluster::updateQueryForDistributedEngineIfNeeded(ASTPtr
             queryToString(query));
     }
 
-    configuration->setFunctionArgs(arguments->children);
+    configuration->getTableFunctionArguments(arguments->children);
 
     function_ast->arguments = arguments;
     function_ast->children.push_back(arguments);
@@ -163,7 +170,7 @@ void StorageObjectStorageCluster::updateQueryForDistributedEngineIfNeeded(ASTPtr
 
     table_expression->database_and_table_name = nullptr;
     table_expression->table_function = function_ast_ptr;
-    table_expression->children[0].swap(function_ast_ptr);
+    table_expression->children[0] = function_ast_ptr;
 
     auto settings = select_query->settings();
     if (settings)

src/Storages/ObjectStorage/StorageObjectStorageCluster.h

@@ -37,6 +37,18 @@ class StorageObjectStorageCluster : public IStorageCluster
         const StorageSnapshotPtr & storage_snapshot,
         const ContextPtr & context) override;
 
+    /*
+    In case the table was created with `object_storage_cluster` setting,
+    modify the AST query object so that it uses the table function implementation
+    by mapping the engine name to table function name and setting `object_storage_cluster`.
+    For table like
+    CREATE TABLE table ENGINE=S3(...) SETTINGS object_storage_cluster='cluster'
+    coverts request
+    SELECT * FROM table
+    to
+    SELECT * FROM s3(...) SETTINGS object_storage_cluster='cluster'
+    to make distributed request over cluster 'cluster'.
+    */
     void updateQueryForDistributedEngineIfNeeded(ASTPtr & query);
 
     const String engine_name;