#include #include #include #include #include #include #include #include #include #include namespace DB { namespace ErrorCodes { extern const int ILLEGAL_TYPE_OF_ARGUMENT; } const char * toString(ColumnTransfomerType type) { switch (type) { case ColumnTransfomerType::APPLY: return "APPLY"; case ColumnTransfomerType::EXCEPT: return "EXCEPT"; case ColumnTransfomerType::REPLACE: return "REPLACE"; } } /// ApplyColumnTransformerNode implementation const char * toString(ApplyColumnTransformerType type) { switch (type) { case ApplyColumnTransformerType::LAMBDA: return "LAMBDA"; case ApplyColumnTransformerType::FUNCTION: return "FUNCTION"; } } ApplyColumnTransformerNode::ApplyColumnTransformerNode(QueryTreeNodePtr expression_node_) { if (expression_node_->getNodeType() == QueryTreeNodeType::LAMBDA) apply_transformer_type = ApplyColumnTransformerType::LAMBDA; else if (expression_node_->getNodeType() == QueryTreeNodeType::FUNCTION) apply_transformer_type = ApplyColumnTransformerType::FUNCTION; else throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Apply column transformer expression must be lambda or function. Actual {}", expression_node_->getNodeTypeName()); children.resize(1); children[expression_child_index] = std::move(expression_node_); } void ApplyColumnTransformerNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, size_t indent) const { buffer << std::string(indent, ' ') << "APPLY COLUMN TRANSFORMER id: " << format_state.getNodeId(this); buffer << ", apply_transformer_type: " << toString(apply_transformer_type); buffer << '\n' << std::string(indent + 2, ' ') << "EXPRESSION" << '\n'; const auto & expression_node = getExpressionNode(); expression_node->dumpTreeImpl(buffer, format_state, indent + 4); } bool ApplyColumnTransformerNode::isEqualImpl(const IQueryTreeNode & rhs) const { const auto & rhs_typed = assert_cast(rhs); return apply_transformer_type == rhs_typed.apply_transformer_type; } void ApplyColumnTransformerNode::updateTreeHashImpl(IQueryTreeNode::HashState & hash_state) const { hash_state.update(static_cast(getTransformerType())); hash_state.update(static_cast(getApplyTransformerType())); getExpressionNode()->updateTreeHash(hash_state); } ASTPtr ApplyColumnTransformerNode::toASTImpl() const { auto ast_apply_transformer = std::make_shared(); const auto & expression_node = getExpressionNode(); if (apply_transformer_type == ApplyColumnTransformerType::FUNCTION) { auto & function_expression = expression_node->as(); ast_apply_transformer->func_name = function_expression.getFunctionName(); ast_apply_transformer->parameters = function_expression.getParametersNode()->toAST(); } else { auto & lambda_expression = expression_node->as(); if (!lambda_expression.getArgumentNames().empty()) ast_apply_transformer->lambda_arg = lambda_expression.getArgumentNames()[0]; ast_apply_transformer->lambda = lambda_expression.toAST(); } return ast_apply_transformer; } QueryTreeNodePtr ApplyColumnTransformerNode::cloneImpl() const { return std::make_shared(getExpressionNode()); } /// ExceptColumnTransformerNode implementation bool ExceptColumnTransformerNode::isColumnMatching(const std::string & column_name) const { if (column_matcher) return RE2::PartialMatch(column_name, *column_matcher); for (const auto & name : except_column_names) if (column_name == name) return true; return false; } const char * toString(ExceptColumnTransformerType type) { switch (type) { case ExceptColumnTransformerType::REGEXP: return "REGEXP"; case ExceptColumnTransformerType::COLUMN_LIST: return "COLUMN_LIST"; } } void ExceptColumnTransformerNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, size_t indent) const { buffer << std::string(indent, ' ') << "EXCEPT COLUMN TRANSFORMER id: " << format_state.getNodeId(this); buffer << ", except_transformer_type: " << toString(except_transformer_type); if (column_matcher) { buffer << ", pattern: " << column_matcher->pattern(); return; } else { buffer << ", identifiers: "; size_t except_column_names_size = except_column_names.size(); for (size_t i = 0; i < except_column_names_size; ++i) { buffer << except_column_names[i]; if (i + 1 != except_column_names_size) buffer << ", "; } } } bool ExceptColumnTransformerNode::isEqualImpl(const IQueryTreeNode & rhs) const { const auto & rhs_typed = assert_cast(rhs); if (except_transformer_type != rhs_typed.except_transformer_type || is_strict != rhs_typed.is_strict || except_column_names != rhs_typed.except_column_names) return false; const auto & rhs_column_matcher = rhs_typed.column_matcher; if (!column_matcher && !rhs_column_matcher) return true; else if (column_matcher && !rhs_column_matcher) return false; else if (!column_matcher && rhs_column_matcher) return false; return column_matcher->pattern() == rhs_column_matcher->pattern(); } void ExceptColumnTransformerNode::updateTreeHashImpl(IQueryTreeNode::HashState & hash_state) const { hash_state.update(static_cast(getTransformerType())); hash_state.update(static_cast(getExceptTransformerType())); for (const auto & column_name : except_column_names) { hash_state.update(column_name.size()); hash_state.update(column_name); } if (column_matcher) { const auto & pattern = column_matcher->pattern(); hash_state.update(pattern.size()); hash_state.update(pattern); } } ASTPtr ExceptColumnTransformerNode::toASTImpl() const { auto ast_except_transformer = std::make_shared(); if (column_matcher) { ast_except_transformer->setPattern(column_matcher->pattern()); return ast_except_transformer; } ast_except_transformer->children.reserve(except_column_names.size()); for (const auto & name : except_column_names) ast_except_transformer->children.push_back(std::make_shared(name)); return ast_except_transformer; } QueryTreeNodePtr ExceptColumnTransformerNode::cloneImpl() const { if (except_transformer_type == ExceptColumnTransformerType::REGEXP) return std::make_shared(column_matcher); return std::make_shared(except_column_names, is_strict); } /// ReplaceColumnTransformerNode implementation ReplaceColumnTransformerNode::ReplaceColumnTransformerNode(const std::vector & replacements_, bool is_strict_) : is_strict(is_strict_) { children.resize(1); children[replacements_child_index] = std::make_shared(); auto & replacement_expressions_nodes = getReplacements().getNodes(); std::unordered_set replacement_names_set; for (const auto & replacement : replacements_) { auto [_, inserted] = replacement_names_set.emplace(replacement.column_name); if (!inserted) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Expressions in column transformer replace should not contain same replacement {} more than once", replacement.column_name); replacements_names.push_back(replacement.column_name); replacement_expressions_nodes.push_back(replacement.expression_node); } } QueryTreeNodePtr ReplaceColumnTransformerNode::findReplacementExpression(const std::string & expression_name) { auto it = std::find(replacements_names.begin(), replacements_names.end(), expression_name); if (it == replacements_names.end()) return {}; size_t replacement_index = it - replacements_names.begin(); auto & replacement_expressions_nodes = getReplacements().getNodes(); return replacement_expressions_nodes[replacement_index]; } void ReplaceColumnTransformerNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, size_t indent) const { buffer << std::string(indent, ' ') << "REPLACE COLUMN TRANSFORMER id: " << format_state.getNodeId(this); auto & replacements_nodes = getReplacements().getNodes(); size_t replacements_size = replacements_nodes.size(); buffer << '\n' << std::string(indent + 2, ' ') << "REPLACEMENTS " << replacements_size << '\n'; for (size_t i = 0; i < replacements_size; ++i) { const auto & replacement_name = replacements_names[i]; buffer << std::string(indent + 4, ' ') << "REPLACEMENT NAME " << replacement_name; buffer << " EXPRESSION" << '\n'; const auto & expression_node = replacements_nodes[i]; expression_node->dumpTreeImpl(buffer, format_state, indent + 6); if (i + 1 != replacements_size) buffer << '\n'; } } bool ReplaceColumnTransformerNode::isEqualImpl(const IQueryTreeNode & rhs) const { const auto & rhs_typed = assert_cast(rhs); return is_strict == rhs_typed.is_strict && replacements_names == rhs_typed.replacements_names; } void ReplaceColumnTransformerNode::updateTreeHashImpl(IQueryTreeNode::HashState & hash_state) const { hash_state.update(static_cast(getTransformerType())); auto & replacement_expressions_nodes = getReplacements().getNodes(); size_t replacements_size = replacement_expressions_nodes.size(); hash_state.update(replacements_size); for (size_t i = 0; i < replacements_size; ++i) { const auto & replacement_name = replacements_names[i]; hash_state.update(replacement_name.size()); hash_state.update(replacement_name); replacement_expressions_nodes[i]->updateTreeHash(hash_state); } } ASTPtr ReplaceColumnTransformerNode::toASTImpl() const { auto ast_replace_transformer = std::make_shared(); auto & replacement_expressions_nodes = getReplacements().getNodes(); size_t replacements_size = replacement_expressions_nodes.size(); ast_replace_transformer->children.reserve(replacements_size); for (size_t i = 0; i < replacements_size; ++i) { auto replacement_ast = std::make_shared(); replacement_ast->name = replacements_names[i]; replacement_ast->expr = replacement_expressions_nodes[i]->toAST(); ast_replace_transformer->children.push_back(replacement_ast); } return ast_replace_transformer; } QueryTreeNodePtr ReplaceColumnTransformerNode::cloneImpl() const { ReplaceColumnTransformerNodePtr result_replace_transformers(new ReplaceColumnTransformerNode()); result_replace_transformers->is_strict = is_strict; result_replace_transformers->replacements_names = replacements_names; return result_replace_transformers; } }