The flag alias_name_used was not set on for the outer references
in subqueries. It resulted in replacement of any outer reference
resolved against an alias for a full field name when the frm
representation of a view with a subquery was generated.
If the subquery and the outer query referenced the same table in
their from lists this replacement effectively changed the meaning
of the view and led to wrong results for selects from this view.
Modified several functions to ensure setting the right value of
the alias_name_used flag for outer references resolved against
aliases.
mysql-test/r/view.result:
Added a test case for bug #26560.
mysql-test/t/view.test:
Added a test case for bug #26560.
sql/item.cc:
Fixed bug #26560.
Made the function resolve_ref_in_select_and_group analyze the return
value of the last parameter with the type of the name resolution for
the submitted reference. If the reference has been resolved against
an alias name from select list then its flag alias_name_used is set on.
Now this value is used in Item_field::fix_outer_field to initialize the flag
when the item_ref object is created for an outer reference.
Added a parameter for the second Item_ref::Item_ref constructor to initialize
properly the flag alias_name_used. The default value of the parameter is FALSE.
If this flag is set on at the creation of an object by this constructor it
will never be changed. Corrected appropriately the Item_ref::set_properties
function.
The function Item_ref::print now prints alias name for an outer reference
if the flag alias_name_used is set on.
sql/item.h:
Fixed bug #26560.
Added a parameter for the second Item_ref::Item_ref constructor to initialize
properly the flag alias_name_used. The default value of the parameter is FALSE.
A similar change has been applied to the first Item_direct_ref::Item_direct_ref
constructor.
sql/mysql_priv.h:
Fixed bug #26560.
Added an an enumeration type enum_resolution_type to return info on
how the function find_item_in_list has resolved the submitted item.
The type is used only for this function.
sql/sql_base.cc:
Fixed bug #26560.
Made the last parameter of the function find_field_in_tables return
more detailed information on how the submitted item has been resolved.
Now it says whether the item has been resolved
against an alias name,
or as a field name without alias,
or as a field name hidden by alias,
or was resolved ignoring alias.
sql/sql_select.cc:
Fixed bug #26560.
Took into account the new type of the last parameter of the function
find_item_in_list.
The flag alias_name_used was not set on for the outer references
in subqueries. It resulted in replacement of any outer reference
resolved against an alias for a full field name when the frm
representation of a view with a subquery was generated.
If the subquery and the outer query referenced the same table in
their from lists this replacement effectively changed the meaning
of the view and led to wrong results for selects from this view.
Modified several functions to ensure setting the right value of
the alias_name_used flag for outer references resolved against
aliases.
which accidentally got broken during the merge
on 16-Feb-2007.
mysql-test/r/view.result:
Updated result file.
mysql-test/t/view.test:
Fix test for views with national characters,
which accidentally got broken during the merge
on 16-Feb-2007.
Now view.test is in UTF8 character set,
so SET NAMES is required.
into olga.mysql.com:/home/igor/mysql-5.1-opt
BitKeeper/etc/gone:
auto-union
mysql-test/r/insert_select.result:
Auto merged
mysql-test/r/join_outer.result:
Auto merged
mysql-test/r/select.result:
Auto merged
mysql-test/r/view.result:
Auto merged
mysql-test/t/insert_select.test:
Auto merged
mysql-test/t/join_outer.test:
Auto merged
mysql-test/t/select.test:
Auto merged
mysys/my_getopt.c:
Auto merged
sql/field.h:
Auto merged
sql/field_conv.cc:
Auto merged
sql/item.cc:
Auto merged
sql/item_cmpfunc.cc:
Auto merged
sql/item_cmpfunc.h:
Auto merged
sql/item_func.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_show.cc:
Auto merged
sql/sql_view.cc:
Auto merged
sql/table.cc:
Auto merged
sql/table.h:
Auto merged
mysql-test/include/mix1.inc:
Manual merge
mysql-test/r/information_schema.result:
Manual merge
mysql-test/r/innodb_mysql.result:
Manual merge
mysql-test/t/information_schema.test:
Manual merge
mysql-test/t/view.test:
Manual merge
mysys/my_pthread.c:
Manual merge
sql/mysql_priv.h:
Manual merge
sql/opt_range.cc:
Manual merge
sql/sql_insert.cc:
Manual merge
sql/sql_update.cc:
Manual merge
operations)
Before this change, the boolean predicates:
- X IS TRUE,
- X IS NOT TRUE,
- X IS FALSE,
- X IS NOT FALSE
were implemented by expanding the Item tree in the parser, by using a
construct like:
Item_func_if(Item_func_ifnull(X, <value>), <value>, <value>)
Each <value> was a constant integer, either 0 or 1.
A bug in the implementation of the function IF(a, b, c), in
Item_func_if::fix_length_and_dec(), would cause the following :
When the arguments b and c are both unsigned, the result type of the
function was signed, instead of unsigned.
When the result of the if function is signed, space for the sign could be
counted twice (in the max() expression for a signed argument, and in the
total), causing the member max_length to be too high.
An effect of this is that the final type of IF(x, int(1), int(1)) would be
int(2) instead of int(1).
With this fix, the problems found in Item_func_if::fix_length_and_dec()
have been fixed.
While it's semantically correct to represent 'X IS TRUE' with
Item_func_if(Item_func_ifnull(X, <value>), <value>, <value>),
there are however more problems with this construct.
a)
Building the parse tree involves :
- creating 5 Item instances (3 ints, 1 ifnull, 1 if),
- creating each Item calls my_pthread_getspecific_ptr() once in the operator
new(size), and a second time in the Item::Item() constructor, resulting
in a total of 10 calls to get the current thread.
Evaluating the expression involves evaluating up to 4 nodes at runtime.
This representation could be greatly simplified and improved.
b)
Transforming the parse tree internally with if(ifnull(...)) is fine as long
as this transformation is internal to the server implementation.
With views however, the result of the parse tree is later exposed by the
::print() functions, and stored as part of the view definition.
Doing this has long term consequences:
1)
The original semantic 'X IS TRUE' is lost, and replaced by the
if(ifnull(...)) expression. As a result, SHOW CREATE VIEW does not restore
the original code.
2)
Should a future version of MySQL implement the SQL BOOLEAN data type for
example, views created today using 'X IS NULL' can be exported using
mysqldump, and imported again. Such views would be converted correctly and
automatically to use a BOOLEAN column in the future version.
With 'X IS TRUE' and the current implementations, views using these
"boolean" predicates would not be converted during the export/import, and
would use integer columns instead.
The difference traces back to how SHOW CREATE VIEW preserves 'X IS NULL' but
does not preserve the 'X IS TRUE' semantic.
With this fix, internal representation of 'X IS TRUE' booleans predicates
has changed, so that:
- dedicated Item classes are created for each predicate,
- only 1 Item is created to represent 1 predicate
- my_pthread_getspecific_ptr() is invoked 1 time instead of 10
- SHOW CREATE VIEW preserves the original semantic, and prints 'X IS TRUE'.
Note that, because of the fix in Item_func_if, views created before this fix
will:
- correctly use a int(1) type instead of int(2) for boolean predicates,
- incorrectly print the if(ifnull(...), ...) expression in SHOW CREATE VIEW,
since the original semantic (X IS TRUE) has been lost.
- except for the syntax used in SHOW CREATE VIEW, these views will operate
properly, no action is needed.
Views created after this fix will operate correctly, and will preserve the
original code semantic in SHOW CREATE VIEW.
mysql-test/r/func_if.result:
IF(x, unsigned, unsigned) should be unsigned.
mysql-test/r/view.result:
Preserve the semantic of 'X IS [NOT] (TRUE|FALSE)' boolean predicates.
mysql-test/t/func_if.test:
IF(x, unsigned, unsigned) should be unsigned.
mysql-test/t/view.test:
Preserve the semantic of 'X IS [NOT] (TRUE|FALSE)' boolean predicates.
sql/item_cmpfunc.cc:
Preserve the semantic of 'X IS [NOT] (TRUE|FALSE)' boolean predicates.
IF(x, unsigned, unsigned) should be unsigned.
sql/item_cmpfunc.h:
Preserve the semantic of 'X IS [NOT] (TRUE|FALSE)' boolean predicates.
sql/sql_yacc.yy:
Preserve the semantic of 'X IS [NOT] (TRUE|FALSE)' boolean predicates.
operations)
Before this change, the boolean predicates:
- X IS TRUE,
- X IS NOT TRUE,
- X IS FALSE,
- X IS NOT FALSE
were implemented by expanding the Item tree in the parser, by using a
construct like:
Item_func_if(Item_func_ifnull(X, <value>), <value>, <value>)
Each <value> was a constant integer, either 0 or 1.
A bug in the implementation of the function IF(a, b, c), in
Item_func_if::fix_length_and_dec(), would cause the following :
When the arguments b and c are both unsigned, the result type of the
function was signed, instead of unsigned.
When the result of the if function is signed, space for the sign could be
counted twice (in the max() expression for a signed argument, and in the
total), causing the member max_length to be too high.
An effect of this is that the final type of IF(x, int(1), int(1)) would be
int(2) instead of int(1).
With this fix, the problems found in Item_func_if::fix_length_and_dec()
have been fixed.
While it's semantically correct to represent 'X IS TRUE' with
Item_func_if(Item_func_ifnull(X, <value>), <value>, <value>),
there are however more problems with this construct.
a)
Building the parse tree involves :
- creating 5 Item instances (3 ints, 1 ifnull, 1 if),
- creating each Item calls my_pthread_getspecific_ptr() once in the operator
new(size), and a second time in the Item::Item() constructor, resulting
in a total of 10 calls to get the current thread.
Evaluating the expression involves evaluating up to 4 nodes at runtime.
This representation could be greatly simplified and improved.
b)
Transforming the parse tree internally with if(ifnull(...)) is fine as long
as this transformation is internal to the server implementation.
With views however, the result of the parse tree is later exposed by the
::print() functions, and stored as part of the view definition.
Doing this has long term consequences:
1)
The original semantic 'X IS TRUE' is lost, and replaced by the
if(ifnull(...)) expression. As a result, SHOW CREATE VIEW does not restore
the original code.
2)
Should a future version of MySQL implement the SQL BOOLEAN data type for
example, views created today using 'X IS NULL' can be exported using
mysqldump, and imported again. Such views would be converted correctly and
automatically to use a BOOLEAN column in the future version.
With 'X IS TRUE' and the current implementations, views using these
"boolean" predicates would not be converted during the export/import, and
would use integer columns instead.
The difference traces back to how SHOW CREATE VIEW preserves 'X IS NULL' but
does not preserve the 'X IS TRUE' semantic.
With this fix, internal representation of 'X IS TRUE' booleans predicates
has changed, so that:
- dedicated Item classes are created for each predicate,
- only 1 Item is created to represent 1 predicate
- my_pthread_getspecific_ptr() is invoked 1 time instead of 10
- SHOW CREATE VIEW preserves the original semantic, and prints 'X IS TRUE'.
Note that, because of the fix in Item_func_if, views created before this fix
will:
- correctly use a int(1) type instead of int(2) for boolean predicates,
- incorrectly print the if(ifnull(...), ...) expression in SHOW CREATE VIEW,
since the original semantic (X IS TRUE) has been lost.
- except for the syntax used in SHOW CREATE VIEW, these views will operate
properly, no action is needed.
Views created after this fix will operate correctly, and will preserve the
original code semantic in SHOW CREATE VIEW.
The function make_unireg_sortorder ignored the fact that any
view field is represented by a 'ref' object.
This could lead to wrong results for the queries containing
both GROUP BY and ORDER BY clauses.
mysql-test/r/view.result:
Added a test case for bug #26209.
mysql-test/t/view.test:
Added a test case for bug #26209.
The function make_unireg_sortorder ignored the fact that any
view field is represented by a 'ref' object.
This could lead to wrong results for the queries containing
both GROUP BY and ORDER BY clauses.
present.
A view created with CREATE VIEW ... ORDER BY ... cannot be resolved with
the MERGE algorithm, even when no other part of the CREATE VIEW statement
would require the view to be resolved using the TEMPTABLE algorithm.
The check for presence of the ORDER BY clause in the underlying select is
removed from the st_lex::can_be_merged() function.
The ORDER BY list of the underlying select is appended to the ORDER BY list
mysql-test/t/view.test:
Added a test case for bug#12122: Views with ORDER BY can't be resolved using MERGE algorithm.
mysql-test/r/view.result:
Added a test case for bug#12122: Views with ORDER BY can't be resolved using MERGE algorithm.
sql/sql_lex.cc:
Bug#12122: Views with ORDER BY can't be resolved using MERGE algorithm.
The st_lex::can_be_merged() function now allows views with the ORDER BY
clause to be resolved using MERGE algorithm. The ORDER BY list of the view
is appended to the ORDER BY list of the embedding select.
present.
A view created with CREATE VIEW ... ORDER BY ... cannot be resolved with
the MERGE algorithm, even when no other part of the CREATE VIEW statement
would require the view to be resolved using the TEMPTABLE algorithm.
The check for presence of the ORDER BY clause in the underlying select is
removed from the st_lex::can_be_merged() function.
The ORDER BY list of the underlying select is appended to the ORDER BY list
View check option clauses were ignored for updates of multi-table
views when the updates could not be performed on fly and the rows
to update had to be put into temporary tables first.
mysql-test/r/view.result:
Added a test case for bug #25931.
mysql-test/t/view.test:
Added a test case for bug #25931.
Adjusted another existed test case to have the correct result.
sql/sql_update.cc:
Fixed bug #25931.
View check option clauses were ignored for updates of multi-table
views when the updates could not be performed on fly and the rows
to update had to be put into temporary tables first.
Added the required check to multi_update::do_updates to fix the problem.
View check option clauses were ignored for updates of multi-table
views when the updates could not be performed on fly and the rows
to update had to be put into temporary tables first.
with a column of the DATETIME type could return a wrong
result set if the WHERE clause included a BETWEEN condition
on the column.
Fixed the method Item_func_between::fix_length_and_dec
where the aggregation type for BETWEEN predicates calculated
incorrectly if the first argument was a view column of the
DATETIME type.
mysql-test/r/view.result:
Added a test case for bug #26124.
mysql-test/t/view.test:
Added a test case for bug #26124.
with a column of the DATETIME type could return a wrong
result set if the WHERE clause included a BETWEEN condition
on the column.
Fixed the method Item_func_between::fix_length_and_dec
where the aggregation type for BETWEEN predicates calculated
incorrectly if the first argument was a view column of the
DATETIME type.
into moonlight.home:/home/tomash/src/mysql_ab/mysql-5.1
mysql-test/r/view.result:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
mysql-test/t/view.test:
Manual merge.
fails
The bug was introduced with the push of the fix for bug#20953: after
the error on view creation we never reset the error state, so some
valid statements would give the same error after that.
The solution is to properly reset the error state.
mysql-test/r/view.result:
Add result for bug#25897: Some queries are no longer possible after
a CREATE VIEW fails.
mysql-test/t/view.test:
Add test case for bug#25897: Some queries are no longer possible after
a CREATE VIEW fails.
sql/sql_lex.cc:
Add st_parsing_options::reset() method, call it from lex_start().
sql/sql_lex.h:
Add st_parsing_options::reset() method, call it from constructor.
fails
The bug was introduced with the push of the fix for bug#20953: after
the error on view creation we never reset the error state, so some
valid statements would give the same error after that.
The solution is to properly reset the error state.
into mysql.com:/home/gluh/MySQL/Merge/5.1
mysql-test/r/view.result:
Auto merged
mysql-test/t/innodb.test:
Auto merged
mysql-test/t/view.test:
Auto merged
sql/item.h:
Auto merged
sql/mysqld.cc:
Auto merged
sql/sql_class.cc:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/sql_update.cc:
Auto merged
into mysql.com:/home/gluh/MySQL/Merge/5.1-opt
BitKeeper/etc/ignore:
auto-union
mysql-test/r/func_in.result:
Auto merged
mysql-test/r/range.result:
Auto merged
mysql-test/r/sp-code.result:
Auto merged
mysql-test/r/trigger.result:
Auto merged
mysql-test/r/udf.result:
Auto merged
mysql-test/t/func_in.test:
Auto merged
mysql-test/t/range.test:
Auto merged
mysql-test/t/trigger.test:
Auto merged
sql/item.cc:
Auto merged
sql/item.h:
Auto merged
sql/item_func.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_show.cc:
Auto merged
sql/sql_table.cc:
Auto merged
mysql-test/r/select.result:
manual merge
mysql-test/r/view.result:
manual merge
mysql-test/t/select.test:
manual merge
mysql-test/t/view.test:
manual merge
into macbook.gmz:/Users/kgeorge/mysql/work/merge-5.1-opt
mysql-test/r/view.result:
Auto merged
mysql-test/t/view.test:
Auto merged
sql/item_func.cc:
Auto merged
geometry dependent parts moved to proper .test files
mysql-test/r/gis.result:
result fixed
mysql-test/r/innodb.result:
result fixed
mysql-test/r/innodb_gis.result:
result fixed
mysql-test/r/view.result:
result fixed
mysql-test/t/gis.test:
HAVE_GEOMETRY dependent part moved here from view.test
mysql-test/t/innodb.test:
HAVE_GEOMETRY dependent part moved to innodb_gis.test
BitKeeper/etc/ignore:
Added libmysqld/sql_servers.cc to the ignore list
mysql-test/t/innodb_gis.test:
HAVE_GEOMETRY dependent part moved here from innodb.test
mysql-test/t/view.test:
HAVE_GEOMETRY dependent part moved to gis.test
into macbook.gmz:/Users/kgeorge/mysql/work/merge-5.1-opt
mysql-test/r/func_in.result:
Auto merged
mysql-test/r/range.result:
Auto merged
mysql-test/r/select.result:
Auto merged
mysql-test/r/sp-code.result:
Auto merged
mysql-test/r/trigger.result:
Auto merged
mysql-test/r/udf.result:
Auto merged
mysql-test/r/view.result:
Auto merged
mysql-test/t/func_in.test:
Auto merged
mysql-test/t/func_str.test:
Auto merged
mysql-test/t/range.test:
Auto merged
mysql-test/t/select.test:
Auto merged
mysql-test/t/trigger.test:
Auto merged
mysql-test/t/udf.test:
Auto merged
sql/item.cc:
Auto merged
sql/item.h:
Auto merged
sql/item_cmpfunc.cc:
Auto merged
sql/item_func.cc:
Auto merged
sql/item_strfunc.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/opt_range.h:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/sql_union.cc:
Auto merged
sql/sql_update.cc:
Auto merged
mysql-test/r/func_str.result:
merge 5.0-opt -> 5.1-opt
mysql-test/r/group_by.result:
merge 5.0-opt -> 5.1-opt
mysql-test/r/insert.result:
merge 5.0-opt -> 5.1-opt
mysql-test/r/subselect.result:
merge 5.0-opt -> 5.1-opt
mysql-test/t/group_by.test:
merge 5.0-opt -> 5.1-opt
mysql-test/t/insert.test:
merge 5.0-opt -> 5.1-opt
mysql-test/t/subselect.test:
merge 5.0-opt -> 5.1-opt
mysql-test/t/view.test:
merge 5.0-opt -> 5.1-opt
sql/item_cmpfunc.h:
merge 5.0-opt -> 5.1-opt
sql/item_strfunc.h:
merge 5.0-opt -> 5.1-opt
into mysql.com:/home/gluh/MySQL/Merge/5.0-opt
mysql-test/r/func_in.result:
Auto merged
mysql-test/r/range.result:
Auto merged
mysql-test/r/sp-code.result:
Auto merged
mysql-test/t/func_in.test:
Auto merged
mysql-test/t/range.test:
Auto merged
mysql-test/t/trigger.test:
Auto merged
mysql-test/t/view.test:
Auto merged
sql/item.cc:
Auto merged
sql/item_func.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_table.cc:
Auto merged
mysql-test/r/select.result:
manual merge
mysql-test/r/view.result:
manual merge
mysql-test/t/select.test:
manual merge
mysql-test/r/create.result:
Fixed error number
mysql-test/r/rpl_sp.result:
Fixed error number
mysql-test/r/sp.result:
Fixed error number
mysql-test/r/view.result:
Better fix for now() handling
mysql-test/t/ndb_basic.test:
Use error names instead of numbers
mysql-test/t/ndb_index_unique.test:
Use error names instead of numbers
mysql-test/t/rpl_skip_error-slave.opt:
Change to new error number
mysql-test/t/sp.test:
Change to new error number
mysql-test/t/view.test:
Better fix for now() handling
