mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-18 05:32:52 +00:00
495c6e03aa
* Replace all Context references with std::weak_ptr * Fix shared context captured by value * Fix build * Fix Context with named sessions * Fix copy context * Fix gcc build * Merge with master and fix build * Fix gcc-9 build
82 lines
2.8 KiB
C++
82 lines
2.8 KiB
C++
#pragma once
|
|
|
|
#include <Core/Names.h>
|
|
#include <Interpreters/InDepthNodeVisitor.h>
|
|
#include <Storages/ColumnsDescription.h>
|
|
|
|
namespace DB
|
|
{
|
|
|
|
class IAST;
|
|
using ASTPtr = std::shared_ptr<IAST>;
|
|
class IDataType;
|
|
class ASTFunction;
|
|
class ASTIdentifier;
|
|
using DataTypePtr = std::shared_ptr<const IDataType>;
|
|
|
|
/// Visits AST node to rewrite alias columns in query
|
|
/// Currently works only 3 kind ways below
|
|
|
|
/// For example:
|
|
// CREATE TABLE test_table
|
|
// (
|
|
// `timestamp` DateTime,
|
|
// `value` UInt64,
|
|
// `day` Date ALIAS toDate(timestamp),
|
|
// `day1` Date ALIAS day + 1,
|
|
// `day2` Date ALIAS day1 + 1,
|
|
// `time` DateTime ALIAS timestamp
|
|
// )ENGINE = MergeTree
|
|
// PARTITION BY toYYYYMMDD(timestamp)
|
|
// ORDER BY timestamp SETTINGS index_granularity = 1;
|
|
|
|
/// 1. Rewrite the filters in query when enable optimize_respect_aliases
|
|
/// this could help with `optimize_trivial_count`, Partition Prune in `KeyCondition` and secondary indexes.
|
|
/// eg: select max(value) from test_table where day2 = today(), filters will be: ((toDate(timestamp) + 1) + 1) = today() .
|
|
|
|
/// 2. Alias on alias for `required_columns` extracted in `InterpreterSelectQuery.cpp`, it could help get all dependent physical columns for query.
|
|
/// eg: select day2 from test_table. `required_columns` can got require columns from the temporary rewritten AST `((toDate(timestamp) + 1) + 1)`.
|
|
|
|
/// 3. Help with `optimize_aggregation_in_order` and `optimize_read_in_order` in `ReadInOrderOptimizer.cpp`:
|
|
/// For queries with alias columns in `orderBy` and `groupBy`, these ASTs will not change.
|
|
/// But we generate temporary asts and generate temporary Actions to get the `InputOrderInfo`
|
|
/// eg: select day1 from test_table order by day1;
|
|
|
|
|
|
class ColumnAliasesMatcher
|
|
{
|
|
public:
|
|
using Visitor = InDepthNodeVisitor<ColumnAliasesMatcher, false>;
|
|
|
|
struct Data
|
|
{
|
|
const ColumnsDescription & columns;
|
|
|
|
/// forbidden_columns are from array join, we can't rewrite alias columns involved in array join.
|
|
/// Do not analyze joined columns.
|
|
/// They may have aliases and come to description as is.
|
|
const NameSet & forbidden_columns;
|
|
ContextPtr context;
|
|
|
|
/// private_aliases are from lambda, so these are local names.
|
|
NameSet private_aliases;
|
|
|
|
Data(const ColumnsDescription & columns_, const NameSet & forbidden_columns_, ContextPtr context_)
|
|
: columns(columns_)
|
|
, forbidden_columns(forbidden_columns_)
|
|
, context(context_)
|
|
{}
|
|
};
|
|
|
|
static void visit(ASTPtr & ast, Data & data);
|
|
static bool needChildVisit(const ASTPtr & node, const ASTPtr & child);
|
|
|
|
private:
|
|
static void visit(ASTIdentifier & node, ASTPtr & ast, Data & data);
|
|
static void visit(ASTFunction & node, ASTPtr & ast, Data & data);
|
|
};
|
|
|
|
using ColumnAliasesVisitor = ColumnAliasesMatcher::Visitor;
|
|
|
|
}
|