Fix for the bug #2976: NATURAL JOIN produces duplicate columns.

Improvement natural join code in the setup_conds(). 

mysql-test/r/join.result

@@ -277,11 +277,11 @@ CREATE TABLE t2 (ID INTEGER NOT NULL PRIMARY KEY, Value2 VARCHAR(255));
 INSERT INTO t1 VALUES (1, 'A');
 INSERT INTO t2 VALUES (1, 'B');
 SELECT * FROM t1 NATURAL JOIN t2 WHERE 1 AND (Value1 = 'A' AND Value2 <> 'B');
-ID	Value1	ID	Value2
+ID	Value1	Value2
 SELECT * FROM t1 NATURAL JOIN t2 WHERE 1 AND Value1 = 'A' AND Value2 <> 'B';
-ID	Value1	ID	Value2
+ID	Value1	Value2
 SELECT * FROM t1 NATURAL JOIN t2 WHERE (Value1 = 'A' AND Value2 <> 'B') AND 1;
-ID	Value1	ID	Value2
+ID	Value1	Value2
 drop table t1,t2;
 create table t1 (i int);
 create table t2 (i int);

mysql-test/r/join_outer.result

@@ -124,8 +124,8 @@ grp	a	c	id	a	c	d
 3	5	C	3	5	B	5
 3	6	D	3	6	C	6
 select t1.*,t2.* from t1 natural join t2;
-grp	a	c	id	a	c	d
-1	1	a	1	1	a	1
+grp	a	c	id	d
+1	1	a	1	1
 drop table t1,t2;
 CREATE TABLE t1 (
 usr_id INT unsigned NOT NULL,
@@ -467,10 +467,10 @@ count	color
 15	white
 7	green
 select * from t2 natural join t1;
-count	color	color	name
-10	green	green	lime
-7	green	green	lime
-5	black	black	grape
+count	color	name
+10	green	lime
+7	green	lime
+5	black	grape
 select t2.count, t1.name from t2 natural join t1;
 count	name
 10	lime

mysql-test/r/select.result

@@ -2279,10 +2279,10 @@ a	a	a
 2	2	2
 3	3	3
 select * from t1 natural join (t1 as t2 left join t1 as t3 using (a));
-a	a	a
-1	1	1
-2	2	2
-3	3	3
+a	a
+1	1
+2	2
+3	3
 select * from (t1 as t2 left join t1 as t3 using (a)) natural join t1;
 a	a	a
 1	1	1

sql/sql_base.cc

@@ -2293,11 +2293,20 @@ insert_fields(THD *thd,TABLE_LIST *tables, const char *db_name,
       thd->used_tables|=table->map;
       while ((field = *ptr++))
       {
-	Item_field *item= new Item_field(field);
-	if (!found++)
-	  (void) it->replace(item);		// Replace '*'
-	else
-	  it->after(item);
+        /* Skip duplicate field names if NATURAL JOIN is used */
+        if (table->outer_join ||
+            !tables->natural_join || 
+            tables->natural_join->table->outer_join ||
+            !find_field_in_table(thd, tables->natural_join->table,
+                                 field->field_name, strlen(field->field_name),
+                                 0, 0))
+        {
+          Item_field *item= new Item_field(field);
+          if (!found++)
+            (void) it->replace(item);		// Replace '*'
+          else
+            it->after(item);
+        }
 	/*
 	  Mark if field used before in this select.
 	  Used by 'insert' to verify if a field name is used twice
@@ -2377,30 +2386,26 @@ int setup_conds(THD *thd,TABLE_LIST *tables,COND **conds)
 	DBUG_RETURN(1);
       cond_and->top_level_item();
 
-      uint i,j;
-      for (i=0 ; i < t1->fields ; i++)
+      Field **t1_field, *t2_field;
+      for (t1_field= t1->field; (*t1_field); t1_field++)
       {
-	// TODO: This could be optimized to use hashed names if t2 had a hash
-	for (j=0 ; j < t2->fields ; j++)
-	{
-	  if (!my_strcasecmp(system_charset_info,
-			     t1->field[i]->field_name,
-			     t2->field[j]->field_name))
-	  {
-	    Item_func_eq *tmp=new Item_func_eq(new Item_field(t1->field[i]),
-					       new Item_field(t2->field[j]));
-	    if (!tmp)
-	      DBUG_RETURN(1);
-	    tmp->fix_length_and_dec();	// Update cmp_type
-	    tmp->const_item_cache=0;
-	    /* Mark field used for table cache */
-	    t1->field[i]->query_id=t2->field[j]->query_id=thd->query_id;
-	    cond_and->list.push_back(tmp);
-	    t1->used_keys.intersect(t1->field[i]->part_of_key);
-	    t2->used_keys.intersect(t2->field[j]->part_of_key);
-	    break;
-	  }
-	}
+        const char *t1_field_name= (*t1_field)->field_name;
+
+        if ((t2_field= find_field_in_table(thd, t2, t1_field_name,
+                                           strlen(t1_field_name), 0, 0)))
+        {
+          Item_func_eq *tmp=new Item_func_eq(new Item_field(*t1_field),
+                                             new Item_field(t2_field));
+          if (!tmp)
+            DBUG_RETURN(1);
+          tmp->fix_length_and_dec();      // Update cmp_type
+          tmp->const_item_cache= 0;
+          /* Mark field used for table cache */
+          (*t1_field)->query_id= t2_field->query_id= thd->query_id;
+          cond_and->list.push_back(tmp);
+          t1->used_keys.intersect((*t1_field)->part_of_key);
+          t2->used_keys.intersect(t2_field->part_of_key);
+        }
       }
       cond_and->used_tables_cache= t1->map | t2->map;
       thd->lex->current_select->cond_count+= cond_and->list.elements;