thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-12-02 20:42:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

slight changes

Browse Source
This commit is contained in:
Igor Nikonov 2022-09-07 22:02:50 +00:00
parent 30860290de
commit c23412ae77
1 changed files with 25 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
src/Processors/QueryPlan/Optimizations/distinctReadInOrder.cpp
View File

@ -1,7 +1,10 @@
#include <Interpreters/ExpressionActions.h>
#include <Parsers/ASTSelectQuery.h>
#include <Processors/QueryPlan/DistinctStep.h>
#include <Processors/QueryPlan/ITransformingStep.h>
#include <Processors/QueryPlan/Optimizations/Optimizations.h>
#include <Processors/QueryPlan/ReadFromMergeTree.h>
#include <Storages/ReadInOrderOptimizer.h>
#include <Common/typeid_cast.h>
namespace DB::QueryPlanOptimizations
@ -10,7 +13,9 @@ namespace DB::QueryPlanOptimizations
///
size_t tryDistinctReadInOrder(QueryPlan::Node * parent_node, QueryPlan::Nodes &)
{
/// check if storage already read in order
/// walk through the plan
/// (1) check if there is preliminary distinct node
/// (2) check if nodes below preliminary distinct preserve sorting
QueryPlan::Node * node = parent_node;
DistinctStep const * pre_distinct = nullptr;
while (!node->children.empty())
@ -36,31 +41,36 @@ size_t tryDistinctReadInOrder(QueryPlan::Node * parent_node, QueryPlan::Nodes &)
if (!pre_distinct)
return 0;
auto const * storage = typeid_cast<ReadFromMergeTree const *>(node->step.get());
if (!storage)
auto * read_from_merge_tree = typeid_cast<ReadFromMergeTree *>(node->step.get());
if (!read_from_merge_tree)
return 0;
const DataStream & output = storage->getOutputStream();
/// check if reading in order is already there
const DataStream & output = read_from_merge_tree->getOutputStream();
if (output.sort_mode != DataStream::SortMode::Chunk)
return 0;
/// check if DISTINCT has the same columns as sorting key
const SortDescription & sort_desc = output.sort_description;
const DataStream & distinct_output = pre_distinct->getOutputStream();
const auto & distinct_columns = distinct_output.distinct_columns;
auto it = sort_desc.begin();
for (; it != sort_desc.end(); ++it)
{
if (distinct_columns.end() == std::find(begin(distinct_columns), end(distinct_columns), it->column_name))
break;
}
const auto & distinct_columns = pre_distinct->getOutputStream().distinct_columns;
/// apply optimization only when distinct columns match or form prefix of sorting key
/// todo: check if reading in order optimization would be benefitial sorting key is prefix of columns in DISTINCT
if (it != sort_desc.end())
if (sort_desc.size() < distinct_columns.size())
return 0;
/// if so, set storage to read in order
/// check if DISTINCT has the same columns as sorting key
SortDescription distinct_sort_desc;
distinct_sort_desc.reserve(sort_desc.size());
for (const auto & column_desc : sort_desc)
{
if (distinct_columns.end() == std::find(begin(distinct_columns), end(distinct_columns), column_desc.column_name))
break;
distinct_sort_desc.push_back(column_desc);
}
if (!distinct_sort_desc.empty())
return 0;
// TODO: provide input order info to read_from_merge_tree
return 0;
}