mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-16 12:02:42 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
mariadb/mysql-test/main/ps_1general.test
Monty b6215b9b20 Update row and key fetch cost models to take into account data copy costs 
Before this patch, when calculating the cost of fetching and using a
row/key from the engine, we took into account the cost of finding a
row or key from the engine, but did not consistently take into account
index only accessed, clustered key or covered keys for all access
paths.

The cost of the WHERE clause (TIME_FOR_COMPARE) was not consistently
considered in best_access_path().  TIME_FOR_COMPARE was used in
calculation in other places, like greedy_search(), but was in some
cases (like scans) done an a different number of rows than was
accessed.

The cost calculation of row and index scans didn't take into account
the number of rows that where accessed, only the number of accepted
rows.

When using a filter, the cost of index_only_reads and cost of
accessing and disregarding 'filtered rows' where not taken into
account, which made filters cost less than there actually where.

To remedy the above, the following key & row fetch related costs
has been added:

- The cost of fetching and using a row is now split into different costs:
  - key + Row fetch cost (as before) but multiplied with the variable
  'optimizer_cache_cost' (default to 0.5). This allows the user to
  tell the optimizer the likehood of finding the key and row in the
  engine cache.
- ROW_COPY_COST, The cost copying a row from the engine to the
  sql layer or creating a row from the join_cache to the record
  buffer. Mostly affects table scan costs.
- ROW_LOOKUP_COST, the cost of fetching a row by rowid.
- KEY_COPY_COST the cost of finding the next key and copying it from
  the engine to the SQL layer. This is used when we calculate the cost
  index only reads. It makes index scans more expensive than before if
  they cover a lot of rows. (main.index_merge_myisam)
- KEY_LOOKUP_COST, the cost of finding the first key in a range.
  This replaces the old define IDX_LOOKUP_COST, but with a higher cost.
- KEY_NEXT_FIND_COST, the cost of finding the next key (and rowid).
  when doing a index scan and comparing the rowid to the filter.
  Before this cost was assumed to be 0.

All of the above constants/variables are now tuned to be somewhat in
proportion of executing complexity to each other.  There is tuning
need for these in the future, but that can wait until the above are
made user variables as that will make tuning much easier.

To make the usage of the above easy, there are new (not virtual)
cost calclation functions in handler:
- ha_read_time(), like read_time(), but take optimizer_cache_cost into
  account.
- ha_read_and_copy_time(), like ha_read_time() but take into account
  ROW_COPY_TIME
- ha_read_and_compare_time(), like ha_read_and_copy_time() but take
  TIME_FOR_COMPARE into account.
- ha_rnd_pos_time(). Read row with row id, taking ROW_COPY_COST
  into account.  This is used with filesort where we don't need
  to execute the WHERE clause again.
- ha_keyread_time(), like keyread_time() but take
  optimizer_cache_cost into account.
- ha_keyread_and_copy_time(), like ha_keyread_time(), but add
  KEY_COPY_COST.
- ha_key_scan_time(), like key_scan_time() but take
  optimizer_cache_cost nto account.
- ha_key_scan_and_compare_time(), like ha_key_scan_time(), but add
  KEY_COPY_COST & TIME_FOR_COMPARE.

I also added some setup costs for doing different types of scans and
creating temporary tables (on disk and in memory). This encourages
the optimizer to not use these for simple 'a few row' lookups if
there are adequate key lookup strategies.
- TABLE_SCAN_SETUP_COST, cost of starting a table scan.
- INDEX_SCAN_SETUP_COST, cost of starting an index scan.
- HEAP_TEMPTABLE_CREATE_COST, cost of creating in memory
  temporary table.
- DISK_TEMPTABLE_CREATE_COST, cost of creating an on disk temporary
  table.

When calculating cost of fetching ranges, we had a cost of
IDX_LOOKUP_COST (0.125) for doing a key div for a new range. This is
now replaced with 'io_cost * KEY_LOOKUP_COST (1.0) *
optimizer_cache_cost', which matches the cost we use for 'ref' and
other key lookups. The effect is that the cost is now a bit higher
when we have many ranges for a key.

Allmost all calculation with TIME_FOR_COMPARE is now done in
best_access_path(). 'JOIN::read_time' now includes the full
cost for finding the rows in the table.

In the result files, many of the changes are now again close to what
they where before the "Update cost for hash and cached joins" commit,
as that commit didn't fix the filter cost (too complex to do
everything in one commit).

The above changes showed a lot of a lot of inconsistencies in
optimizer cost calculation. The main objective with the other changes
was to do calculation as similar (and accurate) as possible and to make
different plans more comparable.

Detailed list of changes:

- Calculate index_only_cost consistently and correctly for all scan
  and ref accesses. The row fetch_cost and index_only_cost now
  takes into account clustered keys, covered keys and index
  only accesses.
- cost_for_index_read now returns both full cost and index_only_cost
- Fixed cost calculation of get_sweep_read_cost() to match other
  similar costs. This is bases on the assumption that data is more
  often stored on SSD than a hard disk.
- Replaced constant 2.0 with new define TABLE_SCAN_SETUP_COST.
- Some scan cost estimates did not take into account
  TIME_FOR_COMPARE. Now all scan costs takes this into
  account. (main.show_explain)
- Added session variable optimizer_cache_hit_ratio (default 50%). By
  adjusting this on can reduce or increase the cost of index or direct
  record lookups. The effect of the default is that key lookups is now
  a bit cheaper than before. See usage of 'optimizer_cache_cost' in
  handler.h.
- JOIN_TAB::scan_time() did not take into account index only scans,
  which produced a wrong cost when index scan was used. Changed
  JOIN_TAB:::scan_time() to take into consideration clustered and
  covered keys. The values are now cached and we only have to call
  this function once. Other calls are changed to use the cached
  values.  Function renamed to JOIN_TAB::estimate_scan_time().
- Fixed that most index cost calculations are done the same way and
  more close to 'range' calculations. The cost is now lower than
  before for small data sets and higher for large data sets as we take
  into account how many keys are read (main.opt_trace_selectivity,
  main.limit_rows_examined).
- Ensured that index_scan_cost() ==
  range(scan_of_all_rows_in_table_using_one_range) +
  MULTI_RANGE_READ_INFO_CONST. One effect of this is that if there
  is choice of doing a full index scan and a range-index scan over
  almost the whole table then index scan will be preferred (no
  range-read setup cost).  (innodb.innodb, main.show_explain,
  main.range)
  - Fixed the EQ_REF and REF takes into account clustered and covered
    keys.  This changes some plans to use covered or clustered indexes
    as these are much cheaper.  (main.subselect_mat_cost,
    main.state_tables_innodb, main.limit_rows_examined)
  - Rowid filter setup cost and filter compare cost now takes into
    account fetching and checking the rowid (KEY_NEXT_FIND_COST).
    (main.partition_pruning heap.heap_btree main.log_state)
  - Added KEY_NEXT_FIND_COST to
    Range_rowid_filter_cost_info::lookup_cost to account of the time
    to find and check the next key value against the container
  - Introduced ha_keyread_time(rows) that takes into account finding
    the next row and copying the key value to 'record'
    (KEY_COPY_COST).
  - Introduced ha_key_scan_time() for calculating an index scan over
    all rows.
  - Added IDX_LOOKUP_COST to keyread_time() as a startup cost.
  - Added index_only_fetch_cost() as a convenience function to
    OPT_RANGE.
  - keyread_time() cost is slightly reduced to prefer shorter keys.
    (main.index_merge_myisam)
  - All of the above caused some index_merge combinations to be
    rejected because of cost (main.index_intersect). In some cases
    'ref' where replaced with index_merge because of the low
    cost calculation of get_sweep_read_cost().
  - Some index usage moved from PRIMARY to a covering index.
    (main.subselect_innodb)
- Changed cost calculation of filter to take KEY_LOOKUP_COST and
  TIME_FOR_COMPARE into account.  See sql_select.cc::apply_filter().
  filter parameters and costs are now written to optimizer_trace.
- Don't use matchings_records_in_range() to try to estimate the number
  of filtered rows for ranges. The reason is that we want to ensure
  that 'range' is calculated similar to 'ref'. There is also more work
  needed to calculate the selectivity when using ranges and ranges and
  filtering.  This causes filtering column in EXPLAIN EXTENDED to be
  100.00 for some cases where range cannot use filtering.
  (main.rowid_filter)
- Introduced ha_scan_time() that takes into account the CPU cost of
  finding the next row and copying the row from the engine to
  'record'. This causes costs of table scan to slightly increase and
  some test to changed their plan from ALL to RANGE or ALL to ref.
  (innodb.innodb_mysql, main.select_pkeycache)
  In a few cases where scan time of very small tables have lower cost
  than a ref or range, things changed from ref/range to ALL.
  (main.myisam, main.func_group, main.limit_rows_examined,
  main.subselect2)
- Introduced ha_scan_and_compare_time() which is like ha_scan_time()
  but also adds the cost of the where clause (TIME_FOR_COMPARE).
- Added small cost for creating temporary table for
  materialization. This causes some very small tables to use scan
  instead of materialization.
- Added checking of the WHERE clause (TIME_FOR_COMPARE) of the
  accepted rows to ROR costs in get_best_ror_intersect()
- Removed '- 0.001' from 'join->best_read' and optimize_straight_join()
  to ensure that the 'Last_query_cost' status variable contains the
  same value as the one that was calculated by the optimizer.
- Take avg_io_cost() into account in handler::keyread_time() and
  handler::read_time(). This should have no effect as it's 1.0 by
  default, except for heap that overrides these functions.
- Some 'ref_or_null' accesses changed to 'range' because of cost
  adjustments (main.order_by)
- Added scan type "scan_with_join_cache" for optimizer_trace. This is
  just to show in the trace what kind of scan was used.
- When using 'scan_with_join_cache' take into account number of
  preceding tables (as have to restore all fields for all previous
  table combination when checking the where clause)
  The new cost added is:
  (row_combinations * ROW_COPY_COST * number_of_cached_tables).
  This increases the cost of join buffering in proportion of the
  number of tables in the join buffer. One effect is that full scans
  are now done earlier as the cost is then smaller.
  (main.join_outer_innodb, main.greedy_optimizer)
- Removed the usage of 'worst_seeks' in cost_for_index_read as it
  caused wrong plans to be created; It prefered JT_EQ_REF even if it
  would be much more expensive than a full table scan. A related
  issue was that worst_seeks only applied to full lookup, not to
  clustered or index only lookups, which is not consistent. This
  caused some plans to use index scan instead of eq_ref (main.union)
- Changed federated block size from 4096 to 1500, which is the
  typical size of an IO packet.
- Added costs for reading rows to Federated. Needed as there is no
  caching of rows in the federated engine.
- Added ha_innobase::rnd_pos_time() cost function.
- A lot of extra things added to optimizer trace
  - More costs, especially for materialization and index_merge.
  - Make lables more uniform
  - Fixed a lot of minor bugs
  - Added 'trace_started()' around a lot of trace blocks.
- When calculating ORDER BY with LIMIT cost for using an index
  the cost did not take into account the number of row retrivals
  that has to be done or the cost of comparing the rows with the
  WHERE clause. The cost calculated would be just a fraction of
  the real cost. Now we calculate the cost as we do for ranges
  and 'ref'.
- 'Using index for group-by' is used a bit more than before as
  now take into account the WHERE clause cost when comparing
  with 'ref' and prefer the method with fewer row combinations.
  (main.group_min_max).

Bugs fixed:
- Fixed that we don't calculate TIME_FOR_COMPARE twice for some plans,
  like in optimize_straight_join() and greedy_search()
- Fixed bug in save_explain_data where we could test for the wrong
  index when displaying 'Using index'. This caused some old plans to
  show 'Using index'.  (main.subselect_innodb, main.subselect2)
- Fixed bug in get_best_ror_intersect() where 'min_cost' was not
  updated, and the cost we compared with was not the one that was
  used.
- Fixed very wrong cost calculation for priority queues in
  check_if_pq_applicable(). (main.order_by now correctly uses priority
  queue)
- When calculating cost of EQ_REF or REF, we added the cost of
  comparing the WHERE clause with the found rows, not all row
  combinations. This made ref and eq_ref to be regarded way to cheap
  compared to other access methods.
- FORCE INDEX cost calculation didn't take into account clustered or
  covered indexes.
- JT_EQ_REF cost was estimated as avg_io_cost(), which is half the
  cost of a JT_REF key. This may be true for InnoDB primary key, but
  not for other unique keys or other engines. Now we use handler
  function to calculate the cost, which allows us to handle
  consistently clustered, covered keys and not covered keys.
- ha_start_keyread() didn't call extra_opt() if keyread was already
  enabled but still changed the 'keyread' variable (which is wrong).
  Fixed by not doing anything if keyread is already enabled.
- multi_range_read_info_cost() didn't take into account io_cost when
  calculating the cost of ranges.
- fix_semijoin_strategies_for_picked_join_order() used the wrong
  record_count when calling best_access_path() for SJ_OPT_FIRST_MATCH
  and SJ_OPT_LOOSE_SCAN.
- Hash joins didn't provide correct best_cost to the upper level, which
  means that the cost for hash_joins more expensive than calculated
  in best_access_path (a difference of 10x * TIME_OF_COMPARE).
  This is fixed in the new code thanks to that we now include
  TIME_OF_COMPARE cost in 'read_time'.

Other things:
- Added some 'if (thd->trace_started())' to speed up code
- Removed not used function Cost_estimate::is_zero()
- Simplified testing of HA_POS_ERROR in get_best_ror_intersect().
  (No cost changes)
- Moved ha_start_keyread() from join_read_const_table() to join_read_const()
  to enable keyread for all types of JT_CONST tables.
- Made a few very short functions inline in handler.h

Notes:
- In main.rowid_filter the join order of order and lineitem is swapped.
  This is because the cost of doing a range fetch of lineitem(98 rows) is
  almost as big as the whole join of order,lineitem. The filtering will
  also ensure that we only have to do very small key fetches of the rows
  in lineitem.
- main.index_merge_myisam had a few changes where we are now using
  less keys for index_merge. This is because index scans are now more
  expensive than before.
- handler->optimizer_cache_cost is updated in ha_external_lock().
  This ensures that it is up to date per statements.
  Not an optimal solution (for locked tables), but should be ok for now.
- 'DELETE FROM t1 WHERE t1.a > 0 ORDER BY t1.a' does not take cost of
  filesort into consideration when table scan is chosen.
  (main.myisam_explain_non_select_all)
- perfschema.table_aggregate_global_* has changed because an update
  on a table with 1 row will now use table scan instead of key lookup.

TODO in upcomming commits:
- Fix selectivity calculation for ranges with and without filtering and
  when there is a ref access but scan is chosen.
  For this we have to store the lowest known value for
  'accepted_records' in the OPT_RANGE structure.
- Change that records_read does not include filtered rows.
- test_if_cheaper_ordering() needs to be updated to properly calculate
  costs. This will fix tests like main.order_by_innodb,
  main.single_delete_update
- Extend get_range_limit_read_cost() to take into considering
  cost_for_index_read() if there where no quick keys. This will reduce
  the computed cost for ORDER BY with LIMIT in some cases.
  (main.innodb_ext_key)
- Fix that we take into account selectivity when counting the number
  of rows we have to read when considering using a index table scan to
  resolve ORDER BY.
- Add new calculation for rnd_pos_time() where we take into account the
  benefit of reading multiple rows from the same page.
2023-02-02 21:43:30 +03:00

944 lines
28 KiB
Text
Raw Blame History

###################### ps_general.test #######################
# #
# basic and miscellaneous tests for prepared statements #
# #
##############################################################
#
# NOTE: PLEASE SEE THE DETAILED DESCRIPTION AT THE BOTTOM OF THIS FILE
# BEFORE ADDING NEW TEST CASES HERE !!!
--disable_warnings
drop table if exists t5, t6, t7, t8;
drop database if exists mysqltest ;
# Cleanup from other tests
drop database if exists client_test_db;
drop database if exists testtets;
drop table if exists t1Aa,t2Aa,v1Aa,v2Aa;
drop view if exists t1Aa,t2Aa,v1Aa,v2Aa;
--enable_warnings
--disable_query_log
select '------ basic tests ------' as test_sequence ;
--enable_query_log
let $type= 'MYISAM' ;
# create the tables (t1 and t9) used in many tests
--source include/ps_create.inc
# insert data into these tables
--source include/ps_renew.inc
################ The basic functions ################
# 1. PREPARE stmt_name FROM <preparable statement>;
# <preparable statement> ::=
# 'literal_stmt' |
# @variable_ref_stmt.
# The statement may contain question marks as placeholders for parameters.
#
# Bind a statement name to a string containing a SQL statement and
# send it to the server. The server will parse the statement and
# reply with "Query Ok" or an error message.
#
PREPARE stmt FROM ' select * from t1 where a = ? ' ;
# 2. EXECUTE stmt_name [USING @var [, @var ]];
# Current values of supplied variables are used as parameters.
#
# Send the server the order to execute the statement and supply values
# for the input parameters needed.
# If no error occurs the server reply will be identical to the reply for
# the query used in PREPARE with question marks replaced with values of
# the input variables.
#
SET @var= 2 ;
EXECUTE stmt USING @var ;
# The non prepared statement with the same server reply would be:
select * from t1 where a = @var ;
# 3. DEALLOCATE PREPARE stmt_name;
#
# Send the server the order to drop the parse information.
# The server will reply with "Query Ok" or an error message.
DEALLOCATE PREPARE stmt ;
################ PREPARE ################
# prepare without parameter
prepare stmt1 from ' select 1 as my_col ' ;
# prepare with parameter
prepare stmt1 from ' select ? as my_col ' ;
# prepare must fail (incomplete statements/wrong syntax)
--error ER_PARSE_ERROR
prepare ;
--error ER_PARSE_ERROR
prepare stmt1 ;
--error ER_PARSE_ERROR
prepare stmt1 from ;
--error ER_PARSE_ERROR
prepare_garbage stmt1 from ' select 1 ' ;
--error ER_PARSE_ERROR
prepare stmt1 from_garbage ' select 1 ' ;
--error ER_PARSE_ERROR
prepare stmt1 from ' select_garbage 1 ' ;
--error ER_PARSE_ERROR
prepare from ' select 1 ' ;
--error ER_PARSE_ERROR
prepare stmt1 ' select 1 ' ;
--error ER_PARSE_ERROR
prepare ? from ' select ? as my_col ' ;
# statement in variable
set @arg00='select 1 as my_col';
prepare stmt1 from @arg00;
# prepare must fail (query variable is empty)
set @arg00='';
--error ER_EMPTY_QUERY
prepare stmt1 from @arg00;
set @arg00=NULL;
# prepare must fail (query variable is NULL)
--error ER_PARSE_ERROR
prepare stmt1 from @arg01;
prepare stmt1 from ' select * from t1 where a <= 2 ' ;
# prepare must fail (column x does not exist)
--error ER_BAD_FIELD_ERROR
prepare stmt1 from ' select * from t1 where x <= 2 ' ;
# cases derived from client_test.c: test_null()
# prepare must fail (column x does not exist)
--error ER_BAD_FIELD_ERROR
prepare stmt1 from ' insert into t1(a,x) values(?,?) ' ;
--error ER_BAD_FIELD_ERROR
prepare stmt1 from ' insert into t1(x,a) values(?,?) ' ;
--disable_warnings
drop table if exists not_exist ;
--enable_warnings
# prepare must fail (table does not exist)
--error ER_NO_SUCH_TABLE
prepare stmt1 from ' select * from not_exist where a <= 2 ' ;
# case derived from client_test.c: test_prepare_syntax()
# prepare must fail (incomplete statement)
--error ER_PARSE_ERROR
prepare stmt1 from ' insert into t1 values(? ' ;
--error ER_PARSE_ERROR
prepare stmt1 from ' select a, b from t1
where a=? and where ' ;
################ EXECUTE ################
# execute must fail (statement never_prepared never prepared)
--error ER_UNKNOWN_STMT_HANDLER
execute never_prepared ;
# execute must fail (prepare stmt1 just failed,
# but there was a successful prepare of stmt1 before)
prepare stmt1 from ' select * from t1 where a <= 2 ' ;
--error ER_NO_SUCH_TABLE
prepare stmt1 from ' select * from not_exist where a <= 2 ' ;
--error ER_UNKNOWN_STMT_HANDLER
execute stmt1 ;
# drop the table between prepare and execute
create table t5
(
a int primary key,
b char(30),
c int
);
insert into t5( a, b, c) values( 1, 'original table', 1);
prepare stmt2 from ' select * from t5 ' ;
execute stmt2 ;
drop table t5 ;
# execute must fail (table was dropped after prepare)
--error ER_NO_SUCH_TABLE
execute stmt2 ;
# cases derived from client_test.c: test_select_prepare()
# 1. drop + create table (same column names/types/order)
# between prepare and execute
create table t5
(
a int primary key,
b char(30),
c int
);
insert into t5( a, b, c) values( 9, 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# 2. drop + create table (same column names/types but different order)
# between prepare and execute
create table t5
(
a int primary key,
c int,
b char(30)
);
insert into t5( a, b, c) values( 9, 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# 3. drop + create table (same column names/types/order+extra column)
# between prepare and execute
create table t5
(
a int primary key,
b char(30),
c int,
d timestamp default '2008-02-23 09:23:45'
);
insert into t5( a, b, c) values( 9, 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# 4. drop + create table (same column names/types, different order +
# additional column) between prepare and execute
create table t5
(
a int primary key,
d timestamp default '2008-02-23 09:23:45',
b char(30),
c int
);
insert into t5( a, b, c) values( 9, 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# 5. drop + create table (same column names/order, different types)
# between prepare and execute
create table t5
(
a timestamp default '2004-02-29 18:01:59',
b char(30),
c int
);
insert into t5( b, c) values( 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# 6. drop + create table (same column types/order, different names)
# between prepare and execute
create table t5
(
f1 int primary key,
f2 char(30),
f3 int
);
insert into t5( f1, f2, f3) values( 9, 'recreated table', 9);
execute stmt2 ;
drop table t5 ;
# execute without parameter
prepare stmt1 from ' select * from t1 where a <= 2 ' ;
execute stmt1 ;
# execute with parameter
set @arg00=1 ;
set @arg01='two' ;
prepare stmt1 from ' select * from t1 where a <= ? ' ;
execute stmt1 using @arg00;
# execute must fail (too small number of parameters)
--error ER_WRONG_ARGUMENTS
execute stmt1 ;
# execute must fail (too big number of parameters)
--error ER_WRONG_ARGUMENTS
execute stmt1 using @arg00, @arg01;
# execute must fail (parameter is not set)
execute stmt1 using @not_set;
################ DEALLOCATE ################
# deallocate must fail (the statement 'never_prepared' was never prepared)
--error ER_UNKNOWN_STMT_HANDLER
deallocate prepare never_prepared ;
# deallocate must fail (prepare stmt1 just failed,
# but there was a successful prepare before)
prepare stmt1 from ' select * from t1 where a <= 2 ' ;
--error ER_NO_SUCH_TABLE
prepare stmt1 from ' select * from not_exist where a <= 2 ' ;
--error ER_UNKNOWN_STMT_HANDLER
deallocate prepare stmt1;
create table t5
(
a int primary key,
b char(10)
);
prepare stmt2 from ' select a,b from t5 where a <= 2 ' ;
drop table t5 ;
# deallocate prepared statement where the table was dropped after prepare
deallocate prepare stmt2;
## parallel use of more than one prepared statement handlers
# switch between different queries
prepare stmt1 from ' select a from t1 where a <= 2 ' ;
prepare stmt2 from ' select b from t1 where a <= 2 ' ;
execute stmt2 ;
execute stmt1 ;
# switch between statement handlers of the same query
prepare stmt1 from ' select a from t1 where a <= 2 ' ;
prepare stmt2 from ' select a from t1 where a <= 2 ' ;
execute stmt2 ;
execute stmt1 ;
deallocate prepare stmt1 ;
# Will the deallocate of stmt1 with the same query affect stmt2 ?
execute stmt2 ;
--disable_query_log
select '------ show and misc tests ------' as test_sequence ;
--enable_query_log
--disable_warnings
drop table if exists t2;
--enable_warnings
create table t2
(
a int primary key, b char(10)
);
################ SHOW COMMANDS ################
prepare stmt4 from ' show databases ';
execute stmt4;
prepare stmt4 from ' show tables from test like ''t2%'' ';
execute stmt4;
prepare stmt4 from ' show columns from t2 where field in (select ?) ';
SET @arg00="a";
execute stmt4 using @arg00;
SET @arg00="b";
execute stmt4 using @arg00;
SET @arg00=1;
execute stmt4 using @arg00;
prepare stmt4 from ' show columns from t2 from test like ''a%'' ';
execute stmt4;
create index t2_idx on t2(b);
prepare stmt4 from ' show index from t2 from test ';
execute stmt4;
prepare stmt4 from ' show table status from test like ''t2%'' ';
# egalize date and time values
--replace_column 8 # 12 # 13 # 14 # 19 #
# Bug#4288 : prepared statement 'show table status ..', wrong output on execute
execute stmt4;
# try the same with the big table
prepare stmt4 from ' show table status from test like ''t9%'' ';
# egalize date and time values
--replace_column 8 # 12 # 13 # 14 # 19 #
# Bug#4288
execute stmt4;
prepare stmt4 from ' show status like ''Threads_running'' ';
--replace_column 2 #
execute stmt4;
prepare stmt4 from ' show variables like ''sql_mode'' ';
execute stmt4;
prepare stmt4 from ' show engine myisam logs ';
execute stmt4;
prepare stmt4 from ' show grants for user ';
prepare stmt4 from ' show create table t2 ';
prepare stmt4 from ' show master status ';
prepare stmt4 from ' show master logs ';
prepare stmt4 from ' show slave status ';
prepare stmt4 from ' show warnings limit 20 ';
prepare stmt4 from ' show errors limit 20 ';
prepare stmt4 from ' show storage engines ';
# The output depends upon the precise order in which
# storage engines are registered, so we switch off the output.
--disable_result_log
execute stmt4;
--enable_result_log
################ MISC STUFF ################
## get a warning and an error
# cases derived from client_test.c: test_warnings(), test_errors()
--disable_warnings
drop table if exists t5;
--enable_warnings
prepare stmt1 from ' drop table if exists t5 ' ;
execute stmt1 ;
prepare stmt1 from ' drop table t5 ' ;
--error ER_BAD_TABLE_ERROR
execute stmt1 ;
## SELECT @@version
# cases derived from client_test.c: test_select_version()
#
# TODO: Metadata check is temporary disabled here, because metadata of
# this statement also depends on @@version contents and you can't apply
# replace_column and replace_result to it. It will be enabled again when
# support of replace_column and replace_result on metadata will be
# implemented.
#
#--enable_metadata
prepare stmt1 from ' SELECT @@version ' ;
# egalize the version
--replace_column 1 <version>
execute stmt1 ;
#--disable_metadata
## do @var:= and set @var=
# cases derived from client_test.c: test_do_set()
prepare stmt_do from ' do @var:= (1 in (select a from t1)) ' ;
prepare stmt_set from ' set @var= (1 in (select a from t1)) ' ;
let $1= 3 ;
while ($1)
{
execute stmt_do ;
--disable_query_log
select @var as 'content of @var is:' ;
--enable_query_log
execute stmt_set ;
--disable_query_log
select @var as 'content of @var is:' ;
--enable_query_log
dec $1 ;
}
# the same test with a table containing one column and 'select *'
--disable_warnings
drop table if exists t5 ;
--enable_warnings
create table t5 (a int) ;
prepare stmt_do from ' do @var:= (1 in (select a from t5)) ' ;
prepare stmt_set from ' set @var= (1 in (select a from t5)) ' ;
let $1= 3 ;
while ($1)
{
execute stmt_do ;
--disable_query_log
select @var as 'content of @var is:' ;
--enable_query_log
execute stmt_set ;
--disable_query_log
select @var as 'content of @var is:' ;
--enable_query_log
dec $1 ;
}
drop table t5 ;
deallocate prepare stmt_do ;
deallocate prepare stmt_set ;
## nonsense like prepare of prepare,execute or deallocate
--error ER_UNSUPPORTED_PS
prepare stmt1 from ' prepare stmt2 from '' select 1 '' ' ;
--error ER_UNSUPPORTED_PS
prepare stmt1 from ' execute stmt2 ' ;
--error ER_UNSUPPORTED_PS
prepare stmt1 from ' deallocate prepare never_prepared ' ;
prepare stmt1 from 'alter view v1 as select 2';
## switch the database connection
prepare stmt4 from ' use test ' ;
## create/drop database
prepare stmt3 from ' create database mysqltest ';
create database mysqltest ;
prepare stmt3 from ' drop database mysqltest ';
drop database mysqltest ;
#### table related commands
## describe
prepare stmt3 from ' describe t2 ';
execute stmt3;
drop table t2 ;
--error ER_NO_SUCH_TABLE
execute stmt3;
## lock/unlock
prepare stmt3 from ' lock tables t1 read ' ;
prepare stmt3 from ' unlock tables ' ;
## Load/Unload table contents
--let $datafile = $MYSQLTEST_VARDIR/tmp/data.txt
--disable_warnings
--error 0,1
--remove_file $datafile
--enable_warnings
--replace_result $MYSQLTEST_VARDIR <MYSQLTEST_VARDIR>
eval prepare stmt1 from ' load data infile ''$datafile''
into table t1 fields terminated by ''\t'' ';
--replace_result $MYSQLTEST_VARDIR <MYSQLTEST_VARDIR>
eval prepare stmt1 from ' select * into outfile ''$datafile'' from t1 ';
execute stmt1 ;
##
prepare stmt1 from ' optimize table t1 ' ;
prepare stmt1 from ' analyze table t1 ' ;
prepare stmt1 from ' checksum table t1 ' ;
prepare stmt1 from ' repair table t1 ' ;
--remove_file $datafile
## handler
prepare stmt1 from ' handler t1 open ';
## commit/rollback
prepare stmt3 from ' commit ' ;
prepare stmt3 from ' rollback ' ;
# the view protocol has to use the same connection,
# not util connection
--disable_service_connection
## switch the sql_mode
prepare stmt4 from ' SET sql_mode=ansi ';
execute stmt4;
# check if the sql_mode is now ansi
select 'a' || 'b' ;
prepare stmt4 from ' SET sql_mode="" ';
execute stmt4;
# check if the sql_mode is not ansi
select '2' || '3' ;
# Will a switch of the sqlmode affect the execution of already prepared
# statements ?
prepare stmt5 from ' select ''2'' || ''3'' ' ;
execute stmt5;
SET sql_mode=ansi;
execute stmt5;
SET sql_mode="";
--enable_service_connection
prepare stmt1 from ' flush local privileges ' ;
prepare stmt1 from ' reset query cache ' ;
prepare stmt1 from ' KILL 0 ';
## simple explain
# cases derived from client_test.c: test_explain_bug()
prepare stmt1 from ' explain select a from t1 order by b ';
# PS protocol gives slightly different metadata
--disable_ps_protocol
--enable_metadata
--replace_result 4096 4_OR_8_K 8192 4_OR_8_K
execute stmt1;
--disable_metadata
SET @arg00=1 ;
prepare stmt1 from ' explain select a from t1 force index (primary) where a > ? order by b ';
--enable_metadata
--replace_result 4096 4_OR_8_K 8192 4_OR_8_K
execute stmt1 using @arg00;
--disable_metadata
--enable_ps_protocol
## parameters with probably problematic characters (quote, double quote)
# cases derived from client_test.c: test_logs()
# try if
--disable_warnings
drop table if exists t2;
--enable_warnings
create table t2 (id smallint, name varchar(20)) ;
prepare stmt1 from ' insert into t2 values(?, ?) ' ;
set @id= 9876 ;
set @arg00= 'MySQL - Open Source Database' ;
set @arg01= "'" ;
set @arg02= '"' ;
set @arg03= "my'sql'" ;
set @arg04= 'my"sql"' ;
insert into t2 values ( @id , @arg00 );
insert into t2 values ( @id , @arg01 );
insert into t2 values ( @id , @arg02 );
insert into t2 values ( @id , @arg03 );
insert into t2 values ( @id , @arg04 );
prepare stmt1 from ' select * from t2 where id= ? and name= ? ';
execute stmt1 using @id, @arg00 ;
execute stmt1 using @id, @arg01 ;
execute stmt1 using @id, @arg02 ;
execute stmt1 using @id, @arg03 ;
execute stmt1 using @id, @arg04 ;
drop table t2;
################ CREATE/DROP/ALTER/RENAME TESTS ################
--disable_query_log
select '------ create/drop/alter/rename tests ------' as test_sequence ;
--enable_query_log
--disable_warnings
drop table if exists t2, t3;
--enable_warnings
## DROP TABLE
prepare stmt_drop from ' drop table if exists t2 ' ;
--disable_warnings
execute stmt_drop;
--enable_warnings
## CREATE TABLE
prepare stmt_create from ' create table t2 (
a int primary key, b char(10)) ';
execute stmt_create;
prepare stmt3 from ' create table t3 like t2 ';
execute stmt3;
drop table t3;
## CREATE TABLE .. SELECT
set @arg00=1;
prepare stmt3 from ' create table t3 (m int) select ? as m ' ;
# Bug#4280 server hangs, prepared "create table .. as select ? .."
execute stmt3 using @arg00;
select m from t3;
drop table t3;
prepare stmt3 from ' create index t2_idx on t2(b) ';
prepare stmt3 from ' drop index t2_idx on t2 ' ;
prepare stmt3 from ' alter table t2 drop primary key ';
## RENAME TABLE
--disable_warnings
drop table if exists new_t2;
--enable_warnings
prepare stmt3 from ' rename table t2 to new_t2 ';
execute stmt3;
--error ER_NO_SUCH_TABLE
execute stmt3;
create table t2 like new_t2;
--error ER_TABLE_EXISTS_ERROR
execute stmt3;
drop table new_t2,t2;
## RENAME more than on TABLE within one statement
# cases derived from client_test.c: test_rename()
prepare stmt1 from ' rename table t5 to t6, t7 to t8 ' ;
create table t5 (a int) ;
--error ER_NO_SUCH_TABLE
execute stmt1 ;
create table t7 (a int) ;
# rename, t5 -> t6 and t7 -> t8
execute stmt1 ;
# rename must fail, t5 and t7 does not exist t6 and t8 already exist
--error ER_NO_SUCH_TABLE
execute stmt1 ;
rename table t6 to t5, t8 to t7 ;
# rename, t5 -> t6 and t7 -> t8
execute stmt1 ;
drop table t6, t8 ;
################ BIG STATEMENT TESTS ################
--disable_query_log
select '------ big statement tests ------' as test_sequence ;
--enable_query_log
# The following tests use huge numbers of lines, characters or parameters
# per prepared statement.
# I assume the server and also the client (mysqltest) are stressed.
#
# Attention: The limits used are NOT derived from the manual
# or other sources.
## many lines ( 50 )
let $my_stmt= select 'ABC' as my_const_col from t1 where
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 AND
1 = 1 ;
eval ($my_stmt) ;
eval prepare stmt1 from "$my_stmt" ;
execute stmt1 ;
execute stmt1 ;
## many characters ( about 1400 )
let $my_stmt= select 'ABC' as my_const_col FROM t1 WHERE
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' AND
'1234567890123456789012345678901234567890123456789012345678901234567890'
= '1234567890123456789012345678901234567890123456789012345678901234567890' ;
eval ($my_stmt) ;
eval prepare stmt1 from "$my_stmt" ;
execute stmt1 ;
execute stmt1 ;
## many parameters ( 50 )
--disable_query_log
set @arg00= 1;
set @arg01= 1;
set @arg02= 1;
set @arg03= 1;
set @arg04= 1;
set @arg05= 1;
set @arg06= 1;
set @arg07= 1;
set @arg10= 1;
set @arg11= 1;
set @arg12= 1;
set @arg13= 1;
set @arg14= 1;
set @arg15= 1;
set @arg16= 1;
set @arg17= 1;
set @arg20= 1;
set @arg21= 1;
set @arg22= 1;
set @arg23= 1;
set @arg24= 1;
set @arg25= 1;
set @arg26= 1;
set @arg27= 1;
set @arg30= 1;
set @arg31= 1;
set @arg32= 1;
set @arg33= 1;
set @arg34= 1;
set @arg35= 1;
set @arg36= 1;
set @arg37= 1;
set @arg40= 1;
set @arg41= 1;
set @arg42= 1;
set @arg43= 1;
set @arg44= 1;
set @arg45= 1;
set @arg46= 1;
set @arg47= 1;
set @arg50= 1;
set @arg51= 1;
set @arg52= 1;
set @arg53= 1;
set @arg54= 1;
set @arg55= 1;
set @arg56= 1;
set @arg57= 1;
set @arg60= 1;
set @arg61= 1;
--enable_query_log
select 'ABC' as my_const_col FROM t1 WHERE
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and @arg00=@arg00 and
@arg00=@arg00 ;
prepare stmt1 from ' select ''ABC'' as my_const_col FROM t1 WHERE
? = ? and ? = ? and ? = ? and ? = ? and
? = ? and ? = ? and ? = ? and ? = ? and
? = ? and ? = ? and ? = ? and ? = ? and
? = ? and ? = ? and ? = ? and ? = ? and
? = ? and ? = ? and ? = ? and ? = ? and
? = ? and ? = ? and ? = ? and ? = ? and
? = ? ' ;
execute stmt1 using
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00, @arg00,
@arg00, @arg00;
execute stmt1 using
@arg00, @arg01, @arg02, @arg03, @arg04, @arg05, @arg06, @arg07,
@arg10, @arg11, @arg12, @arg13, @arg14, @arg15, @arg16, @arg17,
@arg20, @arg21, @arg22, @arg23, @arg24, @arg25, @arg26, @arg27,
@arg30, @arg31, @arg32, @arg33, @arg34, @arg35, @arg36, @arg37,
@arg40, @arg41, @arg42, @arg43, @arg44, @arg45, @arg46, @arg47,
@arg50, @arg51, @arg52, @arg53, @arg54, @arg55, @arg56, @arg57,
@arg60, @arg61 ;
# cases derived from client_test.c: test_mem_overun()
--disable_warnings
drop table if exists t5 ;
--enable_warnings
set @col_num= 1000 ;
--disable_query_log
set @string= 'create table t5( ' ;
let $1=`select @col_num - 1` ;
while ($1)
{
eval set @string= concat(@string, 'c$1 int,') ;
dec $1 ;
}
set @string= concat(@string, 'c0 int)' );
--enable_query_log
select @string as "" ;
prepare stmt1 from @string ;
execute stmt1 ;
--disable_query_log
set @string= 'insert into t5 values(' ;
let $1=`select @col_num - 1` ;
while ($1)
{
eval set @string= concat(@string, '1 ,') ;
dec $1 ;
}
eval set @string= concat(@string, '1 )') ;
--enable_query_log
select @string as "" ;
prepare stmt1 from @string ;
execute stmt1 ;
prepare stmt1 from ' select * from t5 ' ;
--enable_metadata
# prevent too long lines
--vertical_results
--disable_result_log
execute stmt1 ;
--enable_result_log
--disable_metadata
--horizontal_results
drop table t1, t5, t9;
--echo #
--echo # testcase for bug#11765413 - Crash with dependent subquery and
--echo # prepared statement
create table t1 (c1 int);
insert into t1 values (1);
prepare stmt1 from "select 1 from t1 where 1=(select 1 from t1 having c1)";
execute stmt1;
drop prepare stmt1;
drop table t1;
##### RULES OF THUMB TO PRESERVE THE SYSTEMATICS OF THE PS TEST CASES #####
#
# 0. You don't have the time to
# - read and pay attention to these rules of thumb
# - accept that QA may move your test case to a different place
# (I will not change your code!!) .
# Please append your test case to
# t/ps.test
#
# 1. You have more time and want to get as much value from you test case as
# possible. Please try to make the following decisions:
#
# Will the execution or result of the sub test case depend on the
# properties of a storage engine ?
#
# NO --> alter t/ps_1general.test (Example: Command with syntax error)
# If you need a table, please try to use
# t1 - very simple table
# t9 - table with nearly all available column types
# whenever possible.
#
# The structure and the content of these tables can be found in
# include/ps_create.inc CREATE TABLE ...
# include/ps_renew.inc DELETE all rows and INSERT some rows
#
# Both tables are managed by the same storage engine.
# The type of the storage engine is stored in the variable
# '$type' . In ps_1general.test $type is set to 'MYISAM'.
#
# Please feel free to source ps_create.inc or ps_renew.inc
# whenever you think it helps. But please restore the original
# state of these tables after your tests, because the following
# statements may depend on it.
#
# YES
# |
# |
# Is it possible to apply the sub test case to all table types ?
# YES --> alter include/ps_query.inc (for SELECTs)
# include/ps_modify.inc (for INSERT/UPDATE/DELETE)
# include/ps_modify1.inc (also for INSERT/UPDATE/DELETE,
# but t/ps_5merge.test will not source that file)
# Please try to find an appropriate place within the file.
# It would be nice if we have some systematics in the
# order of the sub test cases (if possible).
#
# Please be aware, that
# include: ps_query.inc, ps_modify.inc, ps_modify1.inc
# will be sourced by several test case files stored within the
# subdirectory 't'. So every change here will affect several test
# cases.
#
# NO
# |
# |
# Append the sub test case to the appropriate
# ps_<number><table type>.test .
#
# 2. The current structure of the PS tests
#
# t/ps_1general.test Check of basic PS features, SHOW commands and DDL
# The tests should not depend on the table type.
#
# t/ps_2myisam Check of PS on tables of type MYISAM .
# t/ps_3innodb Check of PS on tables of type InnoDB .
# ...
# t/ps_6bdb Check of PS on tables of type BDB .
# All storage engine related tests use the variable $type to hold the
# name of the storage engine.
#
# include/ps_query.inc test cases with SELECT/...
# These test cases should not modify the content or
# the structure (DROP/ALTER..) of the tables
# 't1' and 't9'.
# include/ps_modify.inc test cases with INSERT/UPDATE/...
# These test cases should not modify the structure
# (DROP/ALTER..) of the tables
# 't1' and 't9'.
# These two test sequences will be applied to all table types .
#
# include/ps_modify1.inc test cases with INSERT/UPDATE/...
# This test sequences will be applied to all table types
# except MERGE tables.
#
# include/ps_create.inc DROP and CREATE of the tables
# 't1' and 't9' .
# include/ps_renew.inc DELETE all rows and INSERT some rows, that means
# recreate the original content of these tables.
# Please do not alter the commands concerning these two tables.
#
# Please feel free and encouraged to exploit the current code sharing
# mechanism of the 'ps_<number><table type>' test cases. It is an convenient
# way to check all storage engines.
#
# Thank you for reading these rules of thumb.
#
# Matthias
# End of 4.1 tests
#
# MDEV-10318 unset params in --ps --embedded
#
--error ER_PARSE_ERROR,2031
select ?+1;
Reference in a new issue View git blame Copy permalink