Merge remote-tracking branch 'origin/better-priority-queue' into gcc-libcxx

dbms/src/Core/SortCursor.h

@@ -1,5 +1,6 @@
 #pragma once
 
+#include <cassert>
 #include <vector>
 #include <algorithm>
 
@@ -210,6 +211,8 @@ struct SortCursorWithCollation
 };
 
 
+/** Allows to fetch data from multiple sort cursors in sorted order (merging sorted data streams).
+  */
 template <typename Cursor>
 class SortingHeap
 {
@@ -232,13 +235,15 @@ public:
 
     void next()
     {
-        if (current()->isValid())
+        assert(isValid());
+
+        if (!current()->isLast())
         {
             current()->next();
-            updateTopHeap(queue.begin(), queue.end());
+            updateTop();
         }
         else
-            queue.erase(queue.begin());
+            removeTop();
     }
 
 private:
@@ -246,15 +251,17 @@ private:
     Container queue;
 
     /// This is adapted version of the function __sift_down from libc++.
-    template <typename It>
-    static void updateTopHeap(It begin, It end)
+    /// Why cannot simply use std::priority_queue?
+    /// - because it doesn't support updating the top element and requires pop and push instead.
+    void updateTop()
     {
-        size_t size = end - begin;
+        size_t size = queue.size();
         if (size < 2)
             return;
 
         size_t child_idx = 1;
-        It child_it = begin + 1;
+        auto begin = queue.begin();
+        auto child_it = begin + 1;
 
         /// Right child exists and is greater than left child.
         if (size > 2 && *child_it < *(child_it + 1))
@@ -293,6 +300,11 @@ private:
         } while (!(*child_it < top));
         *curr_it = std::move(top);
     }
+
+    void removeTop()
+    {
+        std::pop_heap(queue.begin(), queue.end());
+    }
 };
 
 }

dbms/src/DataStreams/MergeSortingBlockInputStream.cpp

@@ -178,9 +178,9 @@ Block MergeSortingBlocksBlockInputStream::readImpl()
 template <typename TSortingHeap>
 Block MergeSortingBlocksBlockInputStream::mergeImpl(TSortingHeap & queue)
 {
-    size_t num_columns = blocks[0].columns();
+    size_t num_columns = header.columns();
 
-    MutableColumns merged_columns = blocks[0].cloneEmptyColumns();
+    MutableColumns merged_columns = header.cloneEmptyColumns();
     /// TODO: reserve (in each column)
 
     /// Take rows from queue in right order and push to 'merged'.
@@ -189,31 +189,31 @@ Block MergeSortingBlocksBlockInputStream::mergeImpl(TSortingHeap & queue)
     {
         auto current = queue.current();
 
+        /// Append a row from queue.
         for (size_t i = 0; i < num_columns; ++i)
             merged_columns[i]->insertFrom(*current->all_columns[i], current->pos);
 
-        if (queue.isValid())
-            queue.next();
-        else
-            break;
-
         ++total_merged_rows;
+        ++merged_rows;
+
+        /// We don't need more rows because of limit has reached.
         if (limit && total_merged_rows == limit)
         {
-            auto res = blocks[0].cloneWithColumns(std::move(merged_columns));
             blocks.clear();
-            return res;
+            break;
         }
 
-        ++merged_rows;
+        queue.next();
+
+        /// It's enough for current output block but we will continue.
         if (merged_rows == max_merged_block_size)
-            return blocks[0].cloneWithColumns(std::move(merged_columns));
+            break;
     }
 
     if (merged_rows == 0)
         return {};
 
-    return blocks[0].cloneWithColumns(std::move(merged_columns));
+    return header.cloneWithColumns(std::move(merged_columns));
 }