mariadb/mysql-test/suite/versioning/r/cte.result
Monty 4d987bd1ad MDEV-25292 Atomic CREATE OR REPLACE TABLE
Atomic CREATE OR REPLACE allows to keep an old table intact if the
command fails or during the crash. That is done by renaming the
original table to temporary name, as a backup and restoring it if the
CREATE fails. When the command is complete and logged the backup
table is deleted.

Atomic replace algorithm

  Two DDL chains are used for CREATE OR REPLACE:
  ddl_log_state_create (C) and ddl_log_state_rm (D).

  1. (C) Log rename of ORIG to TMP table (Rename TMP to original).
  2. Rename orignal to TMP.
  3. (C) Log CREATE_TABLE_ACTION of ORIG (drops ORIG);
  4. Do everything with ORIG (like insert data)
  5. (D) Log drop of TMP
  6. Write query to binlog (this marks (C) to be closed in
     case of failure)
  7. Execute drop of TMP through (D)
  8. Close (C) and (D)

  If there is a failure before 6) we revert the changes in (C)
  Chain (D) is only executed if 6) succeded (C is closed on
  crash recovery).

Foreign key errors will be found at the 1) stage.

Additional notes

  - CREATE TABLE without REPLACE and temporary tables is not affected
    by this commit.
    set @@drop_before_create_or_replace=1 can be used to
    get old behaviour where existing tables are dropped
    in CREATE OR REPLACE.

  - CREATE TABLE is reverted if binlogging the query fails.

  - Engines having HTON_EXPENSIVE_RENAME flag set are not affected by
    this commit. Conflicting tables marked with this flag will be
    deleted with CREATE OR REPLACE.

  - Replication execution is not affected by this commit.
    - Replication will first drop the conflicting table and then
      creating the new one.

  - CREATE TABLE .. SELECT XID usage is fixed and now there is no need
    to log DROP TABLE via DDL_CREATE_TABLE_PHASE_LOG (see comments in
    do_postlock()). XID is now correctly updated so it disables
    DDL_LOG_DROP_TABLE_ACTION. Note that binary log is flushed at the
    final stage when the table is ready. So if we have XID in the
    binary log we don't need to drop the table.

  - Three variations of CREATE OR REPLACE handled:

    1. CREATE OR REPLACE TABLE t1 (..);
    2. CREATE OR REPLACE TABLE t1 LIKE t2;
    3. CREATE OR REPLACE TABLE t1 SELECT ..;

  - Test case uses 6 combinations for engines (aria, aria_notrans,
    myisam, ib, lock_tables, expensive_rename) and 2 combinations for
    binlog types (row, stmt). Combinations help to check differences
    between the results. Error failures are tested for the above three
    variations.

  - expensive_rename tests CREATE OR REPLACE without atomic
    replace. The effect should be the same as with the old behaviour
    before this commit.

  - Triggers mechanism is unaffected by this change. This is tested in
    create_replace.test.

  - LOCK TABLES is affected. Lock restoration must be done after new
    table is created or TMP is renamed back to ORIG

  - Moved ddl_log_complete() from send_eof() to finalize_ddl(). This
    checkpoint was not executed before for normal CREATE TABLE but is
    executed now.

  - CREATE TABLE will now rollback also if writing to the binary
    logging failed. See rpl_gtid_strict.test

backup ddl log changes

- In case of a successfull CREATE OR REPLACE we only log
  the CREATE event, not the DROP TABLE event of the old table.

ddl_log.cc changes

  ddl_log_execute_action() now properly return error conditions.
  ddl_log_disable_entry() added to allow one to disable one entry.
  The entry on disk is still reserved until ddl_log_complete() is
  executed.

On XID usage

  Like with all other atomic DDL operations XID is used to avoid
  inconsistency between master and slave in the case of a crash after
  binary log is written and before ddl_log_state_create is closed. On
  recovery XIDs are taken from binary log and corresponding DDL log
  events get disabled.  That is done by
  ddl_log_close_binlogged_events().

On linking two chains together

  Chains are executed in the ascending order of entry_pos of execute
  entries. But entry_pos assignment order is undefined: it may assign
  bigger number for the first chain and then smaller number for the
  second chain. So the execution order in that case will be reverse:
  second chain will be executed first.

  To avoid that we link one chain to another. While the base chain
  (ddl_log_state_create) is active the secondary chain
  (ddl_log_state_rm) is not executed. That is: only one chain can be
  executed in two linked chains.

  The interface ddl_log_link_chains() was defined in "MDEV-22166
  ddl_log_write_execute_entry() extension".

Atomic info parameters in HA_CREATE_INFO

  Many functions in CREATE TABLE pass the same parameters. These
  parameters are part of table creation info and should be in
  HA_CREATE_INFO (or whatever). Passing parameters via single
  structure is much easier for adding new data and
  refactoring.

InnoDB changes
  Added ha_innobase::can_be_renamed_to_backup() to check if
  a table with foreign keys can be renamed.

Aria changes:
- Fixed issue in Aria engine with CREATE + locked tables
  that data was not properly commited in some cases in
  case of crashes.

Known issues:
- InnoDB tables with foreign key definitions are not fully supported
  with atomic create and replace:
  - ha_innobase::can_be_renamed_to_backup() can detect some cases
    where InnoDB does not support renaming table with foreign key
    constraints.  In this case MariaDB will drop the old table before
    creating the new one.
    The detected cases are:
    - The new and old table is using the same foreign key constraint
      name.
    - The old table has self referencing constraints.
  - If the old and new table uses the same name for a constraint the
    create of the new table will fail. The orignal table will be
    restored in this case.
  - The above issues will be fixed in a future commit.
- CREATE OR REPLACE TEMPORARY table is not full atomic. Any conflicting
  table will always be dropped before creating a new one. (Old behaviour).
2025-03-31 20:09:03 +03:00

319 lines
13 KiB
Text

SET @saved_stats_persistent = @@GLOBAL.innodb_stats_persistent;
SET GLOBAL innodb_stats_persistent = OFF;
set time_zone="+00:00";
set default_storage_engine=innodb;
create or replace table dept (
dept_id int(10) primary key,
name varchar(100)
) with system versioning;
create or replace table emp (
emp_id int(10) primary key,
dept_id int(10) not null,
name varchar(100) not null,
mgr int(10),
salary int(10) not null,
constraint `dept-emp-fk`
foreign key (dept_id) references dept (dept_id)
on delete cascade
on update restrict,
constraint `mgr-fk`
foreign key (mgr) references emp (emp_id)
on delete restrict
on update restrict
) with system versioning;
insert into dept (dept_id, name) values (10, "accounting");
insert into emp (emp_id, name, salary, dept_id, mgr) values
(1, "bill", 1000, 10, null),
(20, "john", 500, 10, 1),
(30, "jane", 750, 10,1 );
select row_start into @ts_1 from emp where name="jane";
update emp set mgr=30 where name ="john";
explain extended
with ancestors as (
select e.emp_id, e.name, e.mgr, e.salary from emp as e where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary from emp as e
) select * from ancestors for system_time as of @ts_1;
id select_type table type possible_keys key key_len ref rows filtered Extra
1 PRIMARY <derived2> ALL NULL NULL NULL NULL 8 100.00
2 DERIVED e ALL NULL NULL NULL NULL 4 100.00 Using where
3 UNION e ALL NULL NULL NULL NULL 4 100.00 Using where
NULL UNION RESULT <union2,3> ALL NULL NULL NULL NULL NULL NULL
Warnings:
Note 1003 with ancestors as (/* select#2 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` where `test`.`e`.`name` = 'bill' and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' union /* select#3 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` where `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu')/* select#1 */ select `ancestors`.`emp_id` AS `emp_id`,`ancestors`.`name` AS `name`,`ancestors`.`mgr` AS `mgr`,`ancestors`.`salary` AS `salary` from `ancestors`
select row_start into @ts_2 from emp where name="john";
explain extended /* All report to 'Bill' */
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp for system_time as of timestamp @ts_1 as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp for system_time as of timestamp @ts_1 as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors;
id select_type table type possible_keys key key_len ref rows filtered Extra
1 PRIMARY <derived2> ALL NULL NULL NULL NULL 4 100.00
2 DERIVED e ALL NULL NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION e ALL mgr-fk NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION <derived2> ref key0 key0 5 test.e.mgr 1 100.00
NULL UNION RESULT <union2,3> ALL NULL NULL NULL NULL NULL NULL
Warnings:
Note 1003 with recursive ancestors as (/* select#2 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` where `test`.`e`.`name` = 'bill' and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' union /* select#3 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` join `ancestors` `a` where `a`.`emp_id` = `test`.`e`.`mgr` and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu')/* select#1 */ select `ancestors`.`emp_id` AS `emp_id`,`ancestors`.`name` AS `name`,`ancestors`.`mgr` AS `mgr`,`ancestors`.`salary` AS `salary` from `ancestors`
/* All report to 'Bill' */
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp for system_time as of timestamp @ts_1 as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp for system_time as of timestamp @ts_1 as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors;
emp_id name mgr salary
1 bill NULL 1000
20 john 1 500
30 jane 1 750
explain extended with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of timestamp @ts_1;
id select_type table type possible_keys key key_len ref rows filtered Extra
1 PRIMARY <derived2> ALL NULL NULL NULL NULL 4 100.00
2 DERIVED e ALL NULL NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION e ALL mgr-fk NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION <derived2> ref key0 key0 5 test.e.mgr 1 100.00
NULL UNION RESULT <union2,3> ALL NULL NULL NULL NULL NULL NULL
Warnings:
Note 1003 with recursive ancestors as (/* select#2 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` where `test`.`e`.`name` = 'bill' and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' union /* select#3 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` join `ancestors` `a` where `a`.`emp_id` = `test`.`e`.`mgr` and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu')/* select#1 */ select `ancestors`.`emp_id` AS `emp_id`,`ancestors`.`name` AS `name`,`ancestors`.`mgr` AS `mgr`,`ancestors`.`salary` AS `salary` from `ancestors`
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of timestamp @ts_1;
emp_id name mgr salary
1 bill NULL 1000
20 john 1 500
30 jane 1 750
explain extended with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select name from emp where emp_id in (select emp_id from ancestors for system_time as of timestamp @ts_1);
id select_type table type possible_keys key key_len ref rows filtered Extra
1 PRIMARY emp ALL PRIMARY NULL NULL NULL 4 100.00 Using where
1 PRIMARY <derived2> ref key0 key0 5 test.emp.emp_id 1 100.00 FirstMatch(emp)
2 DERIVED e ALL NULL NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION e ALL mgr-fk NULL NULL NULL 4 100.00 Using where
3 RECURSIVE UNION <derived2> ref key0 key0 5 test.e.mgr 1 100.00
NULL UNION RESULT <union2,3> ALL NULL NULL NULL NULL NULL NULL
Warnings:
Note 1003 with recursive ancestors as (/* select#2 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` where `test`.`e`.`name` = 'bill' and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' union /* select#3 */ select `test`.`e`.`emp_id` AS `emp_id`,`test`.`e`.`name` AS `name`,`test`.`e`.`mgr` AS `mgr`,`test`.`e`.`salary` AS `salary` from `test`.`emp` FOR SYSTEM_TIME AS OF TIMESTAMP @`ts_1` `e` join `ancestors` `a` where `a`.`emp_id` = `test`.`e`.`mgr` and `test`.`e`.`row_end` > TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu' and `test`.`e`.`row_start` <= TIMESTAMP/*WITH LOCAL TIME ZONE*/'YYYY-MM-DD hh:ss:mm:.uuuuuu')/* select#1 */ select `test`.`emp`.`name` AS `name` from `test`.`emp` semi join (`ancestors`) where `ancestors`.`emp_id` = `test`.`emp`.`emp_id` and `test`.`emp`.`row_end` = TIMESTAMP'2106-02-07 06:28:15.999999'
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select name from emp where emp_id in (select emp_id from ancestors for system_time as of timestamp @ts_1);
name
bill
john
jane
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of @ts_2,
ancestors for system_time as of @ts_2 a2;
emp_id name mgr salary emp_id name mgr salary
1 bill NULL 1000 1 bill NULL 1000
30 jane 1 750 1 bill NULL 1000
20 john 30 500 1 bill NULL 1000
1 bill NULL 1000 30 jane 1 750
30 jane 1 750 30 jane 1 750
20 john 30 500 30 jane 1 750
1 bill NULL 1000 20 john 30 500
30 jane 1 750 20 john 30 500
20 john 30 500 20 john 30 500
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of @ts_2,
ancestors for system_time as of now() a2;
ERROR HY000: Conflicting FOR SYSTEM_TIME clauses in WITH RECURSIVE
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors,
ancestors for system_time as of @ts_2 a2;
ERROR HY000: Conflicting FOR SYSTEM_TIME clauses in WITH RECURSIVE
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of @ts_2,
ancestors a2;
ERROR HY000: Conflicting FOR SYSTEM_TIME clauses in WITH RECURSIVE
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr, e.salary
from emp as e
where name = 'bill'
union
select e.emp_id, e.name, e.mgr, e.salary
from emp as e,
ancestors as a
where e.mgr = a.emp_id
)
select * from ancestors for system_time as of @ts_2
where emp_id in (select * from ancestors);
ERROR HY000: Conflicting FOR SYSTEM_TIME clauses in WITH RECURSIVE
# SYSTEM_TIME to internal recursive instance is prohibited
with recursive cte as
(
select * from emp
union all
select * from cte for system_time as of @ts_1
)
select * from cte;
ERROR HY000: Table `cte` is not system-versioned
create or replace table emp ( emp_id int, name varchar(127), mgr int) with system versioning;
Warnings:
Warning 122 Engine does not support atomic create or replace for table 'emp'. Original table will be deleted
Warning 1215 Cannot add foreign key constraint for `emp`
create or replace table addr ( emp_id int, address varchar(100)) with system versioning;
insert emp values (1, 'bill', 0), (2, 'bill', 1), (3, 'kate', 1);
insert addr values (1, 'Moscow'), (2, 'New York'), (3, 'London');
set @ts=now(6);
delete from emp;
delete from addr;
with recursive
ancestors
as
(
select e.emp_id, e.name, e.mgr
from emp for system_time as of timestamp @ts as e
where name = 'bill'
union
select ee.emp_id, ee.name, ee.mgr
from emp for system_time as of timestamp @ts as ee, ancestors as a
where ee.mgr = a.emp_id
)
select * from ancestors;
emp_id name mgr
1 bill 0
2 bill 1
3 kate 1
insert emp values (4, 'john', 1);
insert addr values (4, 'Paris');
with ancestors as (select * from emp natural join addr) select * from ancestors;
emp_id name mgr address
4 john 1 Paris
with ancestors as (select * from emp natural join addr) select * from ancestors for system_time all;
emp_id name mgr address
1 bill 0 Moscow
2 bill 1 New York
3 kate 1 London
4 john 1 Paris
with ancestors as (select * from (select * from emp natural join addr) for system_time all as t) select * from ancestors;
emp_id name mgr address
1 bill 0 Moscow
2 bill 1 New York
3 kate 1 London
4 john 1 Paris
select * from (select * from emp natural join addr) for system_time all as t;
emp_id name mgr address
1 bill 0 Moscow
2 bill 1 New York
3 kate 1 London
4 john 1 Paris
drop table emp;
drop table dept;
drop table addr;
SET GLOBAL innodb_stats_persistent = @saved_stats_persistent;