fix dups in GLOBAL JOIN with asterisks

dbms/src/Interpreters/InterpreterSelectQuery.cpp

@@ -200,7 +200,7 @@ InterpreterSelectQuery::InterpreterSelectQuery(
     if (storage)
         table_lock = storage->lockStructureForShare(false, context.getCurrentQueryId());
 
-    syntax_analyzer_result = SyntaxAnalyzer(context, options.subquery_depth).analyze(
+    syntax_analyzer_result = SyntaxAnalyzer(context, options).analyze(
         query_ptr, source_header.getNamesAndTypesList(), required_result_column_names, storage);
     query_analyzer = std::make_unique<ExpressionAnalyzer>(
         query_ptr, syntax_analyzer_result, context, NamesAndTypesList(),

dbms/src/Interpreters/SelectQueryOptions.h

@@ -26,12 +26,14 @@ struct SelectQueryOptions
     size_t subquery_depth;
     bool only_analyze;
     bool modify_inplace;
+    bool remove_duplicates;
 
     SelectQueryOptions(QueryProcessingStage::Enum stage = QueryProcessingStage::Complete, size_t depth = 0)
         : to_stage(stage)
         , subquery_depth(depth)
         , only_analyze(false)
         , modify_inplace(false)
+        , remove_duplicates(false)
     {}
 
     SelectQueryOptions copy() const { return *this; }
@@ -58,6 +60,12 @@ struct SelectQueryOptions
 
     SelectQueryOptions & noModify() { return modify(false); }
 
+    SelectQueryOptions & removeDuplicates(bool value = true)
+    {
+        remove_duplicates = value;
+        return *this;
+    }
+
     SelectQueryOptions & noSubquery()
     {
         subquery_depth = 0;

dbms/src/Interpreters/SyntaxAnalyzer.cpp

@@ -125,22 +125,36 @@ bool hasArrayJoin(const ASTPtr & ast)
 
 /// Sometimes we have to calculate more columns in SELECT clause than will be returned from query.
 /// This is the case when we have DISTINCT or arrayJoin: we require more columns in SELECT even if we need less columns in result.
-void removeUnneededColumnsFromSelectClause(const ASTSelectQuery * select_query, const Names & required_result_columns)
+/// Also we have to remove duplicates in case of GLOBAL subqueries. Their results are placed into tables so duplicates are inpossible.
+void removeUnneededColumnsFromSelectClause(const ASTSelectQuery * select_query, const Names & required_result_columns, bool remove_dups)
 {
-    if (required_result_columns.empty())
-        return;
-
     ASTs & elements = select_query->select_expression_list->children;
 
+    std::map<String, size_t> required_columns_with_duplicate_count;
+
+    if (!required_result_columns.empty())
+    {
+        /// Some columns may be queried multiple times, like SELECT x, y, y FROM table.
+        for (const auto & name : required_result_columns)
+        {
+            if (remove_dups)
+                required_columns_with_duplicate_count[name] = 1;
+            else
+                ++required_columns_with_duplicate_count[name];
+        }
+    }
+    else if (remove_dups)
+    {
+        /// Even if we have no requirements there could be duplicates cause of asterisks. SELECT *, t.*
+        for (const auto & elem : elements)
+            required_columns_with_duplicate_count.emplace(elem->getAliasOrColumnName(), 1);
+    }
+    else
+        return;
+
     ASTs new_elements;
     new_elements.reserve(elements.size());
 
-    /// Some columns may be queried multiple times, like SELECT x, y, y FROM table.
-    /// In that case we keep them exactly same number of times.
-    std::map<String, size_t> required_columns_with_duplicate_count;
-    for (const auto & name : required_result_columns)
-        ++required_columns_with_duplicate_count[name];
-
     for (const auto & elem : elements)
     {
         String name = elem->getAliasOrColumnName();
@@ -688,7 +702,7 @@ SyntaxAnalyzerResultPtr SyntaxAnalyzer::analyze(
     /// Must be after 'normalizeTree' (after expanding aliases, for aliases not get lost)
     ///  and before 'executeScalarSubqueries', 'analyzeAggregation', etc. to avoid excessive calculations.
     if (select_query)
-        removeUnneededColumnsFromSelectClause(select_query, required_result_columns);
+        removeUnneededColumnsFromSelectClause(select_query, required_result_columns, remove_duplicates);
 
     /// Executing scalar subqueries - replacing them with constant values.
     executeScalarSubqueries(query, context, subquery_depth);

dbms/src/Interpreters/SyntaxAnalyzer.h

@@ -2,6 +2,7 @@
 
 #include <Interpreters/AnalyzedJoin.h>
 #include <Interpreters/Aliases.h>
+#include <Interpreters/SelectQueryOptions.h>
 
 namespace DB
 {
@@ -55,9 +56,10 @@ using SyntaxAnalyzerResultPtr = std::shared_ptr<const SyntaxAnalyzerResult>;
 class SyntaxAnalyzer
 {
 public:
-    SyntaxAnalyzer(const Context & context_, size_t subquery_depth_ = 0)
+    SyntaxAnalyzer(const Context & context_, const SelectQueryOptions & select_options = {})
         : context(context_)
-        , subquery_depth(subquery_depth_)
+        , subquery_depth(select_options.subquery_depth)
+        , remove_duplicates(select_options.remove_duplicates)
     {}
 
     SyntaxAnalyzerResultPtr analyze(
@@ -69,6 +71,7 @@ public:
 private:
     const Context & context;
     size_t subquery_depth;
+    bool remove_duplicates;
 };
 
 }

dbms/src/Interpreters/interpretSubquery.cpp

@@ -41,6 +41,8 @@ std::shared_ptr<InterpreterSelectWithUnionQuery> interpretSubquery(
     subquery_settings.extremes = 0;
     subquery_context.setSettings(subquery_settings);
 
+    auto subquery_options = SelectQueryOptions(QueryProcessingStage::Complete, subquery_depth).subquery();
+
     ASTPtr query;
     if (table || function)
     {
@@ -83,48 +85,10 @@ std::shared_ptr<InterpreterSelectWithUnionQuery> interpretSubquery(
     else
     {
         query = subquery->children.at(0);
-
-        /** Columns with the same name can be specified in a subquery. For example, SELECT x, x FROM t
-          * This is bad, because the result of such a query can not be saved to the table, because the table can not have the same name columns.
-          * Saving to the table is required for GLOBAL subqueries.
-          *
-          * To avoid this situation, we will rename the same columns.
-          */
-
-        std::set<std::string> all_column_names;
-        std::set<std::string> assigned_column_names;
-
-        if (const auto * select_with_union = query->as<ASTSelectWithUnionQuery>())
-        {
-            if (const auto * select = select_with_union->list_of_selects->children.at(0)->as<ASTSelectQuery>())
-            {
-                for (auto & expr : select->select_expression_list->children)
-                    all_column_names.insert(expr->getAliasOrColumnName());
-
-                for (auto & expr : select->select_expression_list->children)
-                {
-                    auto name = expr->getAliasOrColumnName();
-
-                    if (!assigned_column_names.insert(name).second)
-                    {
-                        size_t i = 1;
-                        while (all_column_names.end() != all_column_names.find(name + "_" + toString(i)))
-                            ++i;
-
-                        name = name + "_" + toString(i);
-                        expr = expr->clone();   /// Cancels fuse of the same expressions in the tree.
-                        expr->setAlias(name);
-
-                        all_column_names.insert(name);
-                        assigned_column_names.insert(name);
-                    }
-                }
-            }
-        }
+        subquery_options.removeDuplicates();
     }
 
-    return std::make_shared<InterpreterSelectWithUnionQuery>(
-        query, subquery_context, SelectQueryOptions(QueryProcessingStage::Complete, subquery_depth + 1), required_source_columns);
+    return std::make_shared<InterpreterSelectWithUnionQuery>(query, subquery_context, subquery_options, required_source_columns);
 }
 
 }

dbms/tests/queries/0_stateless/00850_global_join_dups.reference

@@ -1,2 +1,7 @@
 1
 0
+0
+0	0
+0
+0	0
+0	0

dbms/tests/queries/0_stateless/00850_global_join_dups.sql

@@ -35,49 +35,49 @@ GLOBAL INNER JOIN
 ) USING dummy;
 
 
--- SET asterisk_left_columns_only = 0;
---
--- SELECT * FROM remote('127.0.0.2', system.one)
--- GLOBAL INNER JOIN
--- (
---    SELECT *, dummy
---    FROM ( SELECT dummy FROM remote('127.0.0.2', system.one) ) t1
---    GLOBAL INNER JOIN ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t2
---    USING dummy
--- ) USING dummy;
---
--- SELECT * FROM remote('127.0.0.2', system.one)
--- GLOBAL INNER JOIN
--- (
---   SELECT *, t1.*, t2.*
---    FROM ( SELECT toUInt8(1) AS dummy ) t1
---    INNER JOIN ( SELECT toUInt8(1) AS dummy ) t2
---    USING dummy
--- ) USING dummy;
---
--- SELECT * FROM remote('127.0.0.2', system.one)
--- GLOBAL INNER JOIN
--- (
---    SELECT *, dummy
---    FROM ( SELECT toUInt8(1) AS dummy ) t1
---    INNER JOIN ( SELECT toUInt8(1) AS dummy ) t2
---    USING dummy
--- ) USING dummy;
---
--- SELECT * FROM remote('127.0.0.2', system.one)
--- GLOBAL INNER JOIN
--- (
---     SELECT *
---     FROM ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t1
---     GLOBAL INNER JOIN ( SELECT toUInt8(1) AS dummy ) t2
---     USING dummy
--- ) USING dummy;
---
--- SELECT * FROM remote('127.0.0.2', system.one)
--- GLOBAL INNER JOIN
--- (
---     SELECT *
---     FROM ( SELECT toUInt8(1) AS dummy ) t1
---     GLOBAL INNER JOIN ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t2
---     USING dummy
--- ) USING dummy;
+SET asterisk_left_columns_only = 0;
+
+SELECT * FROM remote('127.0.0.2', system.one)
+GLOBAL INNER JOIN
+(
+    SELECT *, dummy
+    FROM ( SELECT dummy FROM remote('127.0.0.2', system.one) ) t1
+    GLOBAL INNER JOIN ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t2
+    USING dummy
+) USING dummy;
+
+SELECT * FROM remote('127.0.0.2', system.one)
+GLOBAL INNER JOIN
+(
+    SELECT *, t1.*, t2.*
+    FROM ( SELECT toUInt8(0) AS dummy ) t1
+    INNER JOIN ( SELECT toUInt8(0) AS dummy ) t2
+    USING dummy
+) USING dummy;
+
+SELECT * FROM remote('127.0.0.2', system.one)
+GLOBAL INNER JOIN
+(
+    SELECT *, dummy
+    FROM ( SELECT toUInt8(0) AS dummy ) t1
+    INNER JOIN ( SELECT toUInt8(0) AS dummy ) t2
+    USING dummy
+) USING dummy;
+
+SELECT * FROM remote('127.0.0.2', system.one)
+GLOBAL INNER JOIN
+(
+    SELECT *, dummy as other
+    FROM ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t1
+    GLOBAL INNER JOIN ( SELECT toUInt8(0) AS dummy ) t2
+    USING dummy
+) USING dummy;
+
+SELECT * FROM remote('127.0.0.2', system.one)
+GLOBAL INNER JOIN
+(
+    SELECT *, dummy, dummy as other
+    FROM ( SELECT toUInt8(0) AS dummy ) t1
+    GLOBAL INNER JOIN ( SELECT dummy FROM remote('127.0.0.3', system.one) ) t2
+    USING dummy
+) USING dummy;