From 47ca8558161a75cc8a8f0a854f576a3d8a7c3e7b Mon Sep 17 00:00:00 2001
From: Igor Nikonov <igor@clickhouse.com>
Date: Thu, 1 Dec 2022 22:14:11 +0000
Subject: [PATCH] save

---
 .../QueryPlan/Optimizations/Optimizations.h   |  3 +
 .../Optimizations/removeRedundantOrderBy.cpp  | 95 +++++++++++++++++++
 src/Processors/QueryPlan/QueryPlan.cpp        |  1 +
 3 files changed, 99 insertions(+)
 create mode 100644 src/Processors/QueryPlan/Optimizations/removeRedundantOrderBy.cpp

diff --git a/src/Processors/QueryPlan/Optimizations/Optimizations.h b/src/Processors/QueryPlan/Optimizations/Optimizations.h
index 973304b366b..ee0c0f56c0b 100644
--- a/src/Processors/QueryPlan/Optimizations/Optimizations.h
+++ b/src/Processors/QueryPlan/Optimizations/Optimizations.h
@@ -58,6 +58,9 @@ size_t tryReuseStorageOrderingForWindowFunctions(QueryPlan::Node * parent_node,
 /// Reading in order from MergeTree table if DISTINCT columns match or form a prefix of MergeTree sorting key
 size_t tryDistinctReadInOrder(QueryPlan::Node * node);
 
+/// Remove redudant ORDER BYs
+void tryRemoveRedundantOrderBy(QueryPlan::Node * parent_node);
+
 /// Put some steps under union, so that plan optimisation could be applied to union parts separately.
 /// For example, the plan can be rewritten like:
 ///                      - Something -                    - Expression - Something -
diff --git a/src/Processors/QueryPlan/Optimizations/removeRedundantOrderBy.cpp b/src/Processors/QueryPlan/Optimizations/removeRedundantOrderBy.cpp
new file mode 100644
index 00000000000..5c195bb7730
--- /dev/null
+++ b/src/Processors/QueryPlan/Optimizations/removeRedundantOrderBy.cpp
@@ -0,0 +1,95 @@
+#include <Processors/QueryPlan/AggregatingStep.h>
+#include <Processors/QueryPlan/ExpressionStep.h>
+#include <Processors/QueryPlan/FillingStep.h>
+#include <Processors/QueryPlan/ITransformingStep.h>
+#include <Processors/QueryPlan/LimitByStep.h>
+#include <Processors/QueryPlan/LimitStep.h>
+#include <Processors/QueryPlan/Optimizations/Optimizations.h>
+#include <Processors/QueryPlan/ReadFromMergeTree.h>
+#include <Processors/QueryPlan/SortingStep.h>
+#include <Common/typeid_cast.h>
+
+namespace DB::QueryPlanOptimizations
+{
+
+void tryRemoveRedundantOrderBy(QueryPlan::Node * root)
+{
+    // do top down find first order by or group by
+    Stack stack;
+    stack.push_back({.node = root});
+
+    std::vector<IQueryPlanStep *> steps_affect_order; /// aggregation or sorting
+
+    while (!stack.empty())
+    {
+        auto & frame = stack.back();
+
+        QueryPlan::Node * current_node = frame.node;
+        IQueryPlanStep * current_step = frame.node->step.get();
+        if (!steps_affect_order.empty())
+        {
+            while (true)
+            {
+                if (SortingStep * ss = typeid_cast<SortingStep *>(current_step); ss)
+                {
+                    /// if there are LIMITs on top of ORDER BY, the ORDER BY is non-removable
+                    /// if ORDER BY is with FILL WITH, it is non-removable
+                    if (typeid_cast<LimitStep *>(steps_affect_order.back()) || typeid_cast<LimitByStep *>(steps_affect_order.back())
+                        || typeid_cast<FillingStep *>(steps_affect_order.back()))
+                        break;
+
+                    bool remove_sorting = false;
+                    /// (1) aggregation
+                    if (const AggregatingStep * parent_aggr = typeid_cast<AggregatingStep *>(steps_affect_order.back()); parent_aggr)
+                    {
+                        /// check if it contains aggregation functions which depends on order
+                    }
+                    /// (2) sorting
+                    else if (SortingStep * parent_sorting = typeid_cast<SortingStep *>(steps_affect_order.back()); parent_sorting)
+                    {
+                        remove_sorting = true;
+                    }
+
+                    if (remove_sorting)
+                    {
+                        /// need to remove sorting and its expression from plan
+                        QueryPlan::Node * parent = stack.back().node;
+                        chassert(parent->children.front() == current_node);
+
+                        QueryPlan::Node * next_node = !current_node->children.empty() ? current_node->children.front() : nullptr;
+                        if (next_node && typeid_cast<ExpressionStep *>(next_node->step.get()))
+                            next_node = !current_node->children.empty() ? current_node->children.front() : nullptr;
+
+                        if (next_node)
+                            parent->children[0] = next_node;
+                    }
+                }
+            }
+        }
+
+        if (typeid_cast<LimitStep *>(current_step)
+            || typeid_cast<LimitByStep *>(current_step) /// if there are LIMITs on top of ORDER BY, the ORDER BY is non-removable
+            || typeid_cast<FillingStep *>(current_step) /// if ORDER BY is with FILL WITH, it is non-removable
+            || typeid_cast<SortingStep *>(current_step) /// ORDER BY will change order of previous sorting
+            || typeid_cast<AggregatingStep *>(current_step)) /// aggregation change order
+            steps_affect_order.push_back(current_step);
+
+        /// visit children if there are non-visited
+        if (frame.next_child < frame.node->children.size())
+        {
+            auto next_frame = Frame{.node = frame.node->children[frame.next_child]};
+            ++frame.next_child;
+            stack.push_back(next_frame);
+        }
+        /// all children are visited
+        else
+        {
+            if (!steps_affect_order.empty() && current_step == steps_affect_order.back())
+                steps_affect_order.pop_back();
+
+            stack.pop_back();
+        }
+    }
+}
+
+}
diff --git a/src/Processors/QueryPlan/QueryPlan.cpp b/src/Processors/QueryPlan/QueryPlan.cpp
index e1662d43015..d53d26ffc6a 100644
--- a/src/Processors/QueryPlan/QueryPlan.cpp
+++ b/src/Processors/QueryPlan/QueryPlan.cpp
@@ -447,6 +447,7 @@ void QueryPlan::explainPipeline(WriteBuffer & buffer, const ExplainPipelineOptio
 
 void QueryPlan::optimize(const QueryPlanOptimizationSettings & optimization_settings)
 {
+    QueryPlanOptimizations::tryRemoveRedundantOrderBy(root);
     QueryPlanOptimizations::optimizeTreeFirstPass(optimization_settings, *root, nodes);
     QueryPlanOptimizations::optimizeTreeSecondPass(optimization_settings, *root, nodes);
 }