MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
#
|
|
|
|
# Test create options with sequences
|
|
|
|
#
|
|
|
|
--source include/have_innodb.inc
|
|
|
|
--source include/have_archive.inc
|
|
|
|
|
|
|
|
drop table if exists t1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# Check some sample engines
|
|
|
|
#
|
|
|
|
|
|
|
|
create or replace sequence t1 engine=myisam;
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 engine=innodb;
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 engine=maria;
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
select * from t1;
|
|
|
|
--error ER_ILLEGAL_HA_CREATE_OPTION
|
|
|
|
create or replace sequence t1 engine=archive;
|
|
|
|
#
|
|
|
|
# The following error should be fixed. We shouldn't delete old table on errors
|
|
|
|
#
|
|
|
|
--error ER_NO_SUCH_TABLE
|
|
|
|
show create table t1;
|
|
|
|
|
|
|
|
|
|
|
|
# Check start values
|
|
|
|
create or replace sequence t1 start with 10;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 minvalue=11;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 maxvalue=13 increment by -1;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
|
|
|
|
create or replace sequence t1 increment by -1 cache 100;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 cycle;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 nocycle;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
show create sequence t1;
|
|
|
|
create or replace sequence t1 cycle minvalue= 14;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 cycle increment by -1;
|
|
|
|
show create sequence t1;
|
|
|
|
|
|
|
|
drop sequence t1;
|
|
|
|
create sequence if not exists t1;
|
|
|
|
create sequence if not exists t1 start with 10;
|
|
|
|
select * from t1;
|
|
|
|
show create sequence t1;
|
|
|
|
|
|
|
|
create or replace sequence t1 start with 10 minvalue=10 maxvalue=11 nocache cycle;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
create or replace sequence t1 start with 10 minvalue=-10 maxvalue=11 cache=10 cycle increment by 10;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
|
|
|
|
# NO MINVALUE, NO MAXVALUE
|
|
|
|
create or replace sequence t1 start with 10 NO MAXVALUE NO MINVALUE;
|
|
|
|
|
|
|
|
# Some edge cases
|
|
|
|
create or replace sequence t1 start with 10 maxvalue 10;
|
|
|
|
create or replace sequence t1 start with 10 minvalue 10;
|
|
|
|
create or replace sequence t1 start with 10 minvalue 10 maxvalue 11 cycle;
|
|
|
|
create or replace sequence t1 start with 10 maxvalue=9223372036854775806;
|
|
|
|
create or replace sequence t1 start with 10 minvalue=-9223372036854775807;
|
|
|
|
drop sequence if exists t1;
|
|
|
|
|
|
|
|
create sequence t1 increment by 0;
|
|
|
|
show create sequence t1;
|
|
|
|
select * from t1;
|
|
|
|
drop sequence t1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# Wrong usage and arguments to create sequence
|
|
|
|
#
|
|
|
|
|
|
|
|
create table t1 (a int);
|
2017-06-03 16:08:23 +03:00
|
|
|
--error ER_NOT_SEQUENCE
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
show create sequence t1;
|
|
|
|
--error ER_NOT_SEQUENCE2
|
|
|
|
drop sequence t1;
|
|
|
|
drop sequence if exists t1;
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
create sequence t1 start with 10 maxvalue=9;
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
create sequence t1 minvalue= 100 maxvalue=10;
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
create sequence t1 start with 9 minvalue=10;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create or replace sequence t1 maxvalue=13, increment by -1;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create or replace sequence t1 start with= 10 maxvalue=13;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create or replace sequence t1 maxvalue=13, increment= -1;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create or replace sequence t1 start with 10 min_value=1 NO MINVALUE;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create or replace sequence t1 start with 10 min_value=1 NO MINVALUE;
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
create sequence t1 start with 10 maxvalue=9223372036854775807;
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create sequence t1 start with 10 minvalue=-9223372036854775808;
|
2017-05-08 02:44:55 +03:00
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
create sequence t1 RESTART WITH 10;
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
|
|
|
|
# This should probably give an error
|
|
|
|
create or replace sequence t1 start with 10 NO MINVALUE minvalue=1;
|
|
|
|
drop sequence t1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# Test with LIST COLUMNS as first command
|
|
|
|
#
|
|
|
|
create sequence t1;
|
|
|
|
show fields from t1;
|
|
|
|
flush tables;
|
|
|
|
show fields from t1;
|
|
|
|
create or replace sequence t1 engine=aria;
|
|
|
|
show fields from t1;
|
|
|
|
show fields from t1;
|
|
|
|
flush tables;
|
|
|
|
|
|
|
|
#
|
|
|
|
# Test with other create options
|
|
|
|
#
|
|
|
|
|
|
|
|
create or replace sequence t1 comment= "test 1";
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
create or replace sequence t1 comment= "test 2" min_rows=1 max_rows=2;
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
create or replace sequence t1 start=1 increment= 2;
|
|
|
|
create or replace sequence t1 start 1 increment 2;
|
2017-06-03 16:08:23 +03:00
|
|
|
create or replace sequence t1 cache +1;
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
drop sequence t1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# test with create table
|
|
|
|
#
|
|
|
|
|
|
|
|
CREATE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
show create sequence t1;
|
|
|
|
show create table t1;
|
|
|
|
drop sequence t1;
|
|
|
|
|
|
|
|
# Wrong column name
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count_not_exists` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
# Wrong type
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` int(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`cycle` bigint(21) unsigned NOT NULL, /* error */
|
2017-06-03 16:08:23 +03:00
|
|
|
`cycle_count` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
|
|
|
|
# Missing NOT NULL
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21), /* error */
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
# Extra field
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
extra_field bigint(21)
|
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
# Wrong field order
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
# key
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE OR REPLACE TABLE t1 (
|
2017-06-03 16:08:23 +03:00
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
`increment` bigint(21) NOT NULL,
|
2017-06-03 16:08:23 +03:00
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL,
|
|
|
|
key key1 (next_not_cached_value)
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
) sequence=1;
|
|
|
|
|
2020-06-07 16:23:44 +03:00
|
|
|
# Check constraint
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE TABLE t1 (
|
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
|
|
|
`increment` bigint(21) NOT NULL,
|
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL,
|
|
|
|
CHECK (start_value < minimum_value)
|
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE TABLE t1 (
|
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL CHECK (start_value < minimum_value),
|
|
|
|
`increment` bigint(21) NOT NULL,
|
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) NOT NULL
|
|
|
|
) sequence=1;
|
|
|
|
|
|
|
|
|
|
|
|
# Virtual field
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
CREATE TABLE t1 (
|
|
|
|
`next_not_cached_value` bigint(21) NOT NULL,
|
|
|
|
`minimum_value` bigint(21) NOT NULL,
|
|
|
|
`maximum_value` bigint(21) NOT NULL,
|
|
|
|
`start_value` bigint(21) NOT NULL,
|
|
|
|
`increment` bigint(21) NOT NULL,
|
|
|
|
`cache_size` bigint(21) unsigned NOT NULL,
|
|
|
|
`cycle_option` tinyint(1) unsigned NOT NULL,
|
|
|
|
`cycle_count` bigint(21) generated always as (1) virtual
|
|
|
|
) sequence=1;
|
|
|
|
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
drop sequence if exists t1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# DROP SEQUENCE
|
|
|
|
#
|
|
|
|
|
|
|
|
create sequence t1;
|
|
|
|
create sequence t2;
|
|
|
|
create table t3 (a int) engine=myisam;
|
|
|
|
select table_catalog, table_schema, table_name, table_type from information_schema.tables where table_catalog="test";
|
|
|
|
|
|
|
|
CREATE SEQUENCE s1;
|
|
|
|
drop sequence s1;
|
|
|
|
|
|
|
|
drop sequence if exists t1,t2,t3,t4;
|
|
|
|
drop table if exists t1,t2,t3;
|
|
|
|
|
|
|
|
CREATE TABLE t1 (a int);
|
|
|
|
CREATE TABLE t2 (a int);
|
|
|
|
CREATE SEQUENCE s1;
|
|
|
|
drop table t1,t2,s1;
|
|
|
|
|
|
|
|
CREATE TABLE t1 (a int);
|
|
|
|
CREATE TABLE t2 (a int);
|
|
|
|
CREATE SEQUENCE s1;
|
|
|
|
drop table if exists t1,t2,s1,s2;
|
|
|
|
|
|
|
|
CREATE TABLE t1 (a int);
|
|
|
|
CREATE TABLE t2 (a int);
|
|
|
|
CREATE SEQUENCE s1;
|
2020-06-05 18:55:11 +03:00
|
|
|
--error ER_NOT_SEQUENCE2
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
drop sequence t1,t2,s1,s2;
|
|
|
|
drop table if exists t1,t2;
|
|
|
|
|
|
|
|
CREATE TABLE t1 (a int);
|
|
|
|
CREATE TABLE t2 (a int);
|
|
|
|
CREATE SEQUENCE s1;
|
|
|
|
drop sequence if exists t1,t2,s1,s2;
|
|
|
|
drop table if exists t1,t2;
|
|
|
|
|
|
|
|
CREATE TEMPORARY SEQUENCE s1;
|
|
|
|
DROP SEQUENCE s1;
|
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
|
|
|
DROP TEMPORARY SEQUENCE s1;
|
|
|
|
|
|
|
|
CREATE TEMPORARY SEQUENCE s1;
|
|
|
|
CREATE SEQUENCE s2;
|
|
|
|
CREATE TEMPORARY TABLE t1 (a int);
|
|
|
|
CREATE TABLE t2 (a int);
|
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
|
|
|
DROP TEMPORARY SEQUENCE t1,t2,s1,s2;
|
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
|
|
|
DROP TEMPORARY SEQUENCE s1;
|
|
|
|
DROP TEMPORARY TABLE t1;
|
|
|
|
--error ER_BAD_TABLE_ERROR
|
|
|
|
DROP TABLE t1,t2,s1,s2;
|
|
|
|
|
|
|
|
create view v1 as (select 1);
|
|
|
|
CREATE SEQUENCE s1;
|
|
|
|
--error ER_IT_IS_A_VIEW
|
|
|
|
DROP SEQUENCE s1,v1;
|
|
|
|
drop view v1;
|
|
|
|
|
|
|
|
#
|
|
|
|
# CREATE TEMPORARY SEQUENCE
|
|
|
|
#
|
|
|
|
|
2023-05-31 11:57:45 +07:00
|
|
|
--disable_ps2_protocol
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
CREATE TEMPORARY SEQUENCE t1;
|
|
|
|
select next value for t1;
|
|
|
|
drop temporary table t1;
|
2022-03-28 15:03:09 +05:30
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
select previous value for t1;
|
|
|
|
CREATE SEQUENCE t1 start with 1 minvalue 1 maxvalue 10 increment by 1 cache 10;
|
|
|
|
select next value for t1;
|
|
|
|
select previous value for t1;
|
|
|
|
CREATE TEMPORARY SEQUENCE t1 start with 100 minvalue 100 maxvalue 200 increment by 1 cache 10;
|
|
|
|
select previous value for t1;
|
|
|
|
select next value for t1;
|
|
|
|
select previous value for t1;
|
|
|
|
drop temporary sequence t1;
|
|
|
|
select previous value for t1;
|
|
|
|
drop sequence t1;
|
|
|
|
|
|
|
|
CREATE TEMPORARY SEQUENCE t1 engine=innodb;
|
|
|
|
select next value for t1;
|
|
|
|
drop temporary table t1;
|
2022-03-28 15:03:09 +05:30
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
select previous value for t1;
|
|
|
|
CREATE SEQUENCE t1 start with 1 minvalue 1 maxvalue 10 increment by 1 cache 10 engine=innodb;
|
|
|
|
select next value for t1;
|
|
|
|
select previous value for t1;
|
|
|
|
CREATE TEMPORARY SEQUENCE t1 start with 100 minvalue 100 maxvalue 200 increment by 1 cache 10 engine=innodb;
|
2018-01-11 15:39:36 +02:00
|
|
|
START TRANSACTION WITH CONSISTENT SNAPSHOT;
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
select previous value for t1;
|
|
|
|
select next value for t1;
|
|
|
|
select previous value for t1;
|
|
|
|
drop temporary sequence t1;
|
|
|
|
select previous value for t1;
|
|
|
|
drop sequence t1;
|
2023-05-31 11:57:45 +07:00
|
|
|
--enable_ps2_protocol
|
MDEV-10139 Support for SEQUENCE objects
Working features:
CREATE OR REPLACE [TEMPORARY] SEQUENCE [IF NOT EXISTS] name
[ INCREMENT [ BY | = ] increment ]
[ MINVALUE [=] minvalue | NO MINVALUE ]
[ MAXVALUE [=] maxvalue | NO MAXVALUE ]
[ START [ WITH | = ] start ] [ CACHE [=] cache ] [ [ NO ] CYCLE ]
ENGINE=xxx COMMENT=".."
SELECT NEXT VALUE FOR sequence_name;
SELECT NEXTVAL(sequence_name);
SELECT PREVIOUS VALUE FOR sequence_name;
SELECT LASTVAL(sequence_name);
SHOW CREATE SEQUENCE sequence_name;
SHOW CREATE TABLE sequence_name;
CREATE TABLE sequence-structure ... SEQUENCE=1
ALTER TABLE sequence RENAME TO sequence2;
RENAME TABLE sequence TO sequence2;
DROP [TEMPORARY] SEQUENCE [IF EXISTS] sequence_names
Missing features
- SETVAL(value,sequence_name), to be used with replication.
- Check replication, including checking that sequence tables are marked
not transactional.
- Check that a commit happens for NEXT VALUE that changes table data (may
already work)
- ALTER SEQUENCE. ANSI SQL version of setval.
- Share identical sequence entries to not add things twice to table list.
- testing insert/delete/update/truncate/load data
- Run and fix Alibaba sequence tests (part of mysql-test/suite/sql_sequence)
- Write documentation for NEXT VALUE / PREVIOUS_VALUE
- NEXTVAL in DEFAULT
- Ensure that NEXTVAL in DEFAULT uses database from base table
- Two NEXTVAL for same row should give same answer.
- Oracle syntax sequence_table.nextval, without any FOR or FROM.
- Sequence tables are treated as 'not read constant tables' by SELECT; Would
be better if we would have a separate list for sequence tables so that
select doesn't know about them, except if refereed to with FROM.
Other things done:
- Improved output for safemalloc backtrack
- frm_type_enum changed to Table_type
- Removed lex->is_view and replaced with lex->table_type. This allows
use to more easy check if item is view, sequence or table.
- Added table flag HA_CAN_TABLES_WITHOUT_ROLLBACK, needed for handlers
that want's to support sequences
- Added handler calls:
- engine_name(), to simplify getting engine name for partition and sequences
- update_first_row(), to be able to do efficient sequence implementations.
- Made binlog_log_row() global to be able to call it from ha_sequence.cc
- Added handler variable: row_already_logged, to be able to flag that the
changed row is already logging to replication log.
- Added CF_DB_CHANGE and CF_SCHEMA_CHANGE flags to simplify
deny_updates_if_read_only_option()
- Added sp_add_cfetch() to avoid new conflicts in sql_yacc.yy
- Moved code for add_table_options() out from sql_show.cc::show_create_table()
- Added String::append_longlong() and used it in sql_show.cc to simplify code.
- Added extra option to dd_frm_type() and ha_table_exists to indicate if
the table is a sequence. Needed by DROP SQUENCE to not drop a table.
2017-03-25 23:36:56 +02:00
|
|
|
|
|
|
|
#
|
|
|
|
# Check that we can't create anything with the sequence engine
|
|
|
|
#
|
|
|
|
|
|
|
|
--error ER_UNKNOWN_STORAGE_ENGINE
|
|
|
|
create table t1 (a int) engine=sql_sequence;
|
2017-10-22 12:26:32 +03:00
|
|
|
|
2017-10-22 13:01:33 +03:00
|
|
|
--echo #
|
|
|
|
--echo # MDEV-13711 assertion on CREATE LIKE fix
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
create sequence s;
|
|
|
|
create table t like s;
|
|
|
|
show create table t;
|
|
|
|
show create sequence t;
|
|
|
|
drop tables t, s;
|
|
|
|
|
2017-10-22 12:26:32 +03:00
|
|
|
--echo #
|
|
|
|
--echo # MDEV-13714 SEQUENCE option fix
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
create or replace table s (
|
|
|
|
`next_value` bigint(21) not null,
|
|
|
|
`min_value` bigint(21) not null,
|
|
|
|
`max_value` bigint(21) not null,
|
|
|
|
`start` bigint(21) not null,
|
|
|
|
`increment` bigint(21) not null,
|
|
|
|
`cache` bigint(21) not null,
|
|
|
|
`cycle` tinyint(1) unsigned not null,
|
|
|
|
`round` bigint(21) not null)
|
|
|
|
sequence=0;
|
|
|
|
|
|
|
|
create or replace table s2 (
|
|
|
|
`next_value` bigint(21) not null,
|
|
|
|
`min_value` bigint(21) not null,
|
|
|
|
`max_value` bigint(21) not null,
|
|
|
|
`start` bigint(21) not null,
|
|
|
|
`increment` bigint(21) not null,
|
|
|
|
`cache` bigint(21) not null,
|
|
|
|
`cycle` tinyint(1) unsigned not null,
|
|
|
|
`round` bigint(21) not null)
|
|
|
|
sequence=default;
|
|
|
|
|
|
|
|
show create table s;
|
|
|
|
show create table s2;
|
|
|
|
--error ER_NOT_SEQUENCE
|
|
|
|
show create sequence s;
|
|
|
|
--error ER_NOT_SEQUENCE
|
|
|
|
show create sequence s2;
|
|
|
|
drop table s,s2;
|
2017-10-22 20:14:52 +03:00
|
|
|
|
|
|
|
--echo #
|
|
|
|
--echo # MDEV-13721 Assertion is_lock_owner() failed in mysql_rm_table_no_locks
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
create or replace sequence s;
|
|
|
|
create temporary table s (i int);
|
|
|
|
drop sequence s;
|
|
|
|
show create table s;
|
|
|
|
drop table s;
|
|
|
|
|
|
|
|
create or replace sequence s;
|
|
|
|
create temporary sequence s;
|
|
|
|
show create table s;
|
|
|
|
drop sequence s;
|
|
|
|
show create table s;
|
|
|
|
drop table s;
|
|
|
|
|
|
|
|
create or replace sequence s;
|
|
|
|
create temporary sequence s;
|
|
|
|
drop temporary sequence s;
|
|
|
|
show create table s;
|
|
|
|
drop table s;
|
|
|
|
|
|
|
|
create temporary sequence s;
|
|
|
|
drop temporary table s;
|
|
|
|
create temporary table s (i int);
|
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
|
|
|
drop temporary sequence s;
|
|
|
|
drop table s;
|
2018-02-06 17:09:26 +02:00
|
|
|
|
|
|
|
--echo #
|
|
|
|
--echo # MDEV-15115 Assertion failure in CREATE SEQUENCE...ROW_FORMAT=REDUNDANT
|
|
|
|
--echo #
|
|
|
|
CREATE SEQUENCE seq1 ENGINE=InnoDB ROW_FORMAT=REDUNDANT;
|
2018-04-06 18:54:35 +03:00
|
|
|
|
|
|
|
--replace_column 1 # 5 #
|
|
|
|
SELECT * FROM INFORMATION_SCHEMA.INNODB_SYS_TABLES WHERE NAME='test/seq1';
|
|
|
|
|
2018-02-06 17:09:26 +02:00
|
|
|
DROP SEQUENCE seq1;
|
|
|
|
CREATE TEMPORARY SEQUENCE seq1 ENGINE=InnoDB ROW_FORMAT=REDUNDANT;
|
|
|
|
DROP TEMPORARY SEQUENCE seq1;
|
2018-10-29 16:12:52 +02:00
|
|
|
|
|
|
|
--echo #
|
|
|
|
--echo # MDEV-17503 CREATE SEQUENCE failed with innodb_force_primary_key =1
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
set global innodb_force_primary_key =1;
|
|
|
|
CREATE SEQUENCE s1 START WITH 100 INCREMENT BY 10 ENGINE=innodb;
|
|
|
|
set global innodb_force_primary_key=default;
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
ALTER TABLE s1 ADD PRIMARY KEY (next_not_cached_value);
|
|
|
|
DROP SEQUENCE s1;
|
2022-03-28 15:03:09 +05:30
|
|
|
|
|
|
|
--echo #
|
|
|
|
--echo # Beginning of 10.4 Test
|
|
|
|
--echo #
|
|
|
|
--echo # MDEV-13005: Fixing bugs in SEQUENCE, part 3
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
--echo # Task 1:
|
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
|
|
|
SET @x = PREVIOUS VALUE FOR x;
|
2022-03-28 15:08:25 +05:30
|
|
|
|
|
|
|
--echo # Task 2:
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
CREATE SEQUENCE x START WITH 1 INCREMENT BY 123456789012345678;
|
2022-03-28 15:12:33 +05:30
|
|
|
|
|
|
|
--echo # Task 3:
|
|
|
|
--error ER_SEQUENCE_INVALID_DATA
|
|
|
|
CREATE SEQUENCE seq1 START WITH 1 cache -1;
|
2022-03-28 15:18:42 +05:30
|
|
|
|
|
|
|
--echo # Task 4:
|
|
|
|
CREATE TEMPORARY TABLE s1 (s1 INT);
|
2022-04-06 11:51:36 +03:00
|
|
|
--error ER_UNKNOWN_SEQUENCES
|
2022-03-28 15:18:42 +05:30
|
|
|
DROP TEMPORARY SEQUENCE s1;
|
|
|
|
DROP TEMPORARY TABLE s1;
|
|
|
|
|
|
|
|
--echo # Task 5:
|
|
|
|
CREATE TEMPORARY TABLE s1 (s1 INT);
|
|
|
|
--error ER_PARSE_ERROR
|
|
|
|
CREATE TEMPORARY SEQUENCE s1 (s1 INT);
|
|
|
|
DROP TEMPORARY TABLE s1;
|
2022-03-28 15:45:29 +05:30
|
|
|
|
|
|
|
--echo # Task 6:
|
|
|
|
CREATE SEQUENCE seq1 START WITH 2;
|
|
|
|
--error ER_TRG_ON_VIEW_OR_TEMP_TABLE
|
|
|
|
CREATE TRIGGER s1 BEFORE UPDATE ON seq1 FOR EACH ROW SET @a= 5;
|
|
|
|
DROP SEQUENCE seq1;
|
|
|
|
|
2023-09-26 18:17:04 +02:00
|
|
|
--echo #
|
|
|
|
--echo # MDEV-29771: Server crashes in check_sequence_fields upon
|
|
|
|
--echo # CREATE TABLE .. SEQUENCE=1 AS SELECT ..
|
|
|
|
--echo #
|
|
|
|
|
|
|
|
--error ER_SEQUENCE_INVALID_TABLE_STRUCTURE
|
|
|
|
create table s sequence=1 as select 1;
|
|
|
|
|
|
|
|
--echo #
|
2022-03-28 15:45:29 +05:30
|
|
|
--echo # End of 10.4 test
|
2023-09-26 18:17:04 +02:00
|
|
|
--echo #
|