Merge pull request #4723 from 4ertus2/joins

Fix LEFT JOIN with const column + lots of Join.cpp refactoring

dbms/src/Interpreters/Join.h

@@ -228,6 +228,52 @@ public:
         std::unique_ptr<HashMap<UInt128, Mapped, UInt128HashCRC32>>                     keys128;
         std::unique_ptr<HashMap<UInt256, Mapped, UInt256HashCRC32>>                     keys256;
         std::unique_ptr<HashMap<UInt128, Mapped, UInt128TrivialHash>>                   hashed;
+
+        void create(Type which)
+        {
+            switch (which)
+            {
+                case Type::EMPTY:            break;
+                case Type::CROSS:            break;
+
+            #define M(NAME) \
+                case Type::NAME: NAME = std::make_unique<typename decltype(NAME)::element_type>(); break;
+                APPLY_FOR_JOIN_VARIANTS(M)
+            #undef M
+            }
+        }
+
+        size_t getTotalRowCount(Type which) const
+        {
+            switch (which)
+            {
+                case Type::EMPTY:            return 0;
+                case Type::CROSS:            return 0;
+
+            #define M(NAME) \
+                case Type::NAME: return NAME ? NAME->size() : 0;
+                APPLY_FOR_JOIN_VARIANTS(M)
+            #undef M
+            }
+
+            __builtin_unreachable();
+        }
+
+        size_t getTotalByteCountImpl(Type which) const
+        {
+            switch (which)
+            {
+                case Type::EMPTY:            return 0;
+                case Type::CROSS:            return 0;
+
+            #define M(NAME) \
+                case Type::NAME: return NAME ? NAME->getBufferSizeInBytes() : 0;
+                APPLY_FOR_JOIN_VARIANTS(M)
+            #undef M
+            }
+
+            __builtin_unreachable();
+        }
     };
 
     using MapsAny = MapsTemplate<WithFlags<false, false, RowRef>>;

dbms/tests/queries/0_stateless/00916_join_using_duplicate_columns.reference

@@ -3,3 +3,18 @@
 1
 1
 1	1
+1	0
+1	0
+1	0
+1	1
+1	0
+1	0
+1	0
+1	1
+0	2
+0	2
+0	2
+1	1
+0	2
+0	2
+0	2

dbms/tests/queries/0_stateless/00916_join_using_duplicate_columns.sql

@@ -5,4 +5,25 @@ SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 1 AS x) AS t2 USING x;
 SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 2 AS x) AS t2 USING x;
 
 SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 1 AS x) AS t2 ON t1.x = t2.x;
--- (bug) SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT materialize(1) AS x) AS t1 ALL LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT materialize(2) AS x) AS t2 ON t1.x = t2.x;
+
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY LEFT JOIN (SELECT 1 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT materialize(1) AS x) AS t1 ANY LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY LEFT JOIN (SELECT materialize(2) AS x) AS t2 ON t1.x = t2.x;
+
+SELECT * FROM (SELECT 1 AS x) AS t1 ALL RIGHT JOIN (SELECT 1 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ALL RIGHT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT materialize(1) AS x) AS t1 ALL RIGHT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ALL RIGHT JOIN (SELECT materialize(2) AS x) AS t2 ON t1.x = t2.x;
+
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY RIGHT JOIN (SELECT 1 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY RIGHT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT materialize(1) AS x) AS t1 ANY RIGHT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+SELECT * FROM (SELECT 1 AS x) AS t1 ANY RIGHT JOIN (SELECT materialize(2) AS x) AS t2 ON t1.x = t2.x;
+
+-- SET join_use_nulls = 1;
+-- SELECT * FROM (SELECT 1 AS x) AS t1 ALL LEFT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;
+-- SELECT * FROM (SELECT 1 AS x) AS t1 ALL RIGHT JOIN (SELECT 2 AS x) AS t2 ON t1.x = t2.x;