Now thd->mem_root is a pointer to thd->main_mem_root and THR_MALLOC is a pointer to thd->mem_root.
This gives us the following benefits:
- Allow us to easily detect if arena has already been swapped before (this fixes a bug in setup_conds() where arena was swaped twice in some cases)
- Faster swaps of arenas (as we don't have to copy the whole MEM_ROOT)
- We don't anymore have to call my_pthread_setspecific_ptr(THR_MALLOC,...) to change where memory is alloced. Now it's enough to set thd->mem_root
client/mysqltest.c:
Remove some not needed defines
(Things like this should be done in config-win.h)
include/config-win.h:
Added popen() and pclose() compatibility macros
mysql-test/t/rpl_failed_optimize-master.opt:
Portability fix
sql/ha_berkeley.cc:
New thd->memroot handling
sql/item_cmpfunc.cc:
Simpler arena swapping code
sql/item_func.cc:
Simpler arena swapping code
sql/item_subselect.cc:
Simpler arena swapping code
New thd->mem_root handling
sql/item_sum.cc:
New thd->mem_root handling
sql/item_timefunc.cc:
Fixed not-initalized usage errors found by valgrind
sql/log_event.cc:
New thd->mem_root handling
sql/mysql_priv.h:
New thd->mem_root handling
sql/mysqld.cc:
New thd->mem_root handling
sql/opt_range.cc:
New thd->mem_root handling
sql/repl_failsafe.cc:
New thd->mem_root handling
sql/set_var.cc:
New thd->mem_root handling
sql/sql_acl.cc:
New thd->mem_root handling
sql/sql_base.cc:
Simpler arena swapping code
New thd->mem_root handling
sql/sql_class.cc:
New thd->mem_root handling
sql/sql_class.h:
Simpler arena swapping code
New thd->mem_root handling
sql/sql_db.cc:
New thd->mem_root handling
sql/sql_error.cc:
New thd->mem_root handling
sql/sql_help.cc:
New thd->mem_root handling
sql/sql_insert.cc:
New thd->mem_root handling
sql/sql_parse.cc:
New thd->mem_root handling
Added some extra checking of return value of new
sql/sql_prepare.cc:
New thd->mem_root handling
sql/sql_select.cc:
New thd->mem_root handling
sql/sql_select.h:
New thd->mem_root handling
sql/sql_union.cc:
Simpler arena swapping code
sql/sql_yacc.yy:
New thd->mem_root handling
sql/table.cc:
New thd->mem_root handling
sql/thr_malloc.cc:
New thd->mem_root handling
tests/client_test.c:
Added drop table to some tests
Changed some table names to 't1'
of client equals to character set of connection, possibly required
conversion to character set of column is not performed
(prepared statements, data is supplied using placeholders).
sql/item.cc:
Fix for the bug reported by Reggie: if character_set_connection
is equal to character_set_client, placeholder's value is not converted
furhter to character set of column when it's different.
This is because the original implementation left placeholder's character
set intact (binary) if there were no need for client->connection conversion.
tests/client_test.c:
A test case for the conversion bug.
(prepared staements)": adding the test case to close the bug
(the bug was fixed along with other conversion incompatibilities
in 4.1.7)
tests/client_test.c:
A test case for Bug#4172: "Floating point conversion loses precision
(prepared staements)"
Fix memleak.
Added protocol::flush() for easier embedded-server code
Increase block allocation variables a bit as they where a bit too small for MySQL 4.1
Added option --silent to client_test
client/mysqltest.c:
Removed compiler warning
Fixed identation & comments from earlier push
Renamed variable 'disable_abort_on_error' to 'abort_on_error'
Ensure that '$mysql_errno' also with --ps-protocol
include/mysql_com.h:
Removed special handling of net_flush for embedded server
mysql-test/r/mysqltest.result:
Remove usage of !$ in tests
mysql-test/t/client_test.test:
Use --silent
mysql-test/t/comments.test:
Remove usage of !$ in tests
mysql-test/t/join_outer.test:
Remove usage of !$ in tests
mysql-test/t/key.test:
Remove usage of !$ in tests
mysql-test/t/mysqltest.test:
Remove usage of !$ in tests
mysql-test/t/show_check.test:
Remove usage of !$ in tests
mysql-test/t/temp_table.test:
Remove usage of !$ in tests
mysql-test/t/type_ranges.test:
Remove usage of !$ in tests
sql/mysqld.cc:
Increase block allocation variables a bit as they where a bit too small for MySQL 4.1
sql/net_serv.cc:
Remove special usage of net_flush in embedded server
sql/protocol.cc:
Added protocol::flush() for easier embedded-server code
sql/protocol.h:
Added protocol::flush() for easier embedded-server code
sql/sql_prepare.cc:
Added protocol::flush() for easier embedded-server code
Remove one extra flush() for prepared statements
sql/sql_show.cc:
Added protocol::flush() for easier embedded-server code
tests/client_test.c:
Added option --silent
numeric columns (stmt_resultset_metadata)"
libmysql/libmysql.c:
A fix for Bug#6096 "field.max_length is always zero for numeric columns
(stmt_resultset_metadata)": set field->max_lenght for numeric
columns when we set up skip_result functions. A minor drawback for this
approach is that field->max_length can be not zero even if
STMT_ATTR_UPDATE_MAX_LENGTH is not set.
sql-common/my_time.c:
Fix valgrind warning.
tests/client_test.c:
A test case for Bug#6096 "field.max_length is always zero for numeric
columns (stmt_resultset_metadata)
positive numbers when no resultset is available": when sending
result set metadata we need to use virtual select_result::send_fields,
and not address protocol directly, because select_result descendents may
intercept result set (it's the case for example for SELECT INTO OUTFILE).
sql/sql_class.h:
A fix for bug#6059 "mysql_stmt_field_count returns positive numbers
when no resultset is available": introducing select_result::field_count()
method to report actual number of fields in a result set, if any result
set is sent to client.
sql/sql_prepare.cc:
A fix for bug#6059 "mysql_stmt_field_count returns
positive numbers when no resultset is available": when sending
result set metadata we need to use virtual select_result::send_fields,
and not address protocol directly, because select_result descendents may
intercept result set (it's the case for example for SELECT INTO OUTFILE).
Now we need to always have lex->result set if we're in prepared statements.
tests/client_test.c:
A test case for Bug#6059 "mysql_stmt_field_count returns positive
numbers when no resultset is available"
into mysql.com:/home/kostja/work/mysql-4.1-6049
libmysql/libmysql.c:
Auto merged
sql/field.cc:
Auto merged
sql/item.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
tests/client_test.c:
Auto merged
(date) as empty string": preserve time type (date, time, or datetime) for
zero dates, times, and datetimes.
libmysql/libmysql.c:
A fix for bug#6058 "Prepared statements return '0000-00-00' (date) as empty
string": preserve time type (date, time, or datetime) for zero
dates, times, and datetimes.
tests/client_test.c:
A test case for Bug#6058, the existing tests required some adjustments too.
Simple optimzations and cleanups
Removed compiler warnings and fixed portability issues
Added client functions 'mysql_embedded()' to allow client to check if we are using embedded server
Fixes for purify
client/mysqlimport.c:
Remove not used variable
client/mysqltest.c:
Remove usage of MAXPATHLEN (all MySQL code uses FN_REFLEN)
Simplified code
Remove usage of sprintf("%llu") as this is not portable
include/mysql.h:
Added mysql_embedded() to be able to easily check if we are using the embedded server
innobase/srv/srv0start.c:
Don't use memcmp() when using purify (to avoid false warnings)
libmysql/libmysql.c:
Added mysql_embedded() to be able to easily check if we are using the embedded server
libmysql/libmysql.def:
Added mysql_embedded() to be able to easily check if we are using the embedded server
myisam/myisam_ftdump.c:
Remove compiler warning
myisam/myisamchk.c:
Remove compiler warning
myisam/rt_test.c:
#ifdef not used code
mysys/hash.c:
Remove compiler warning (from last push)
mysys/my_gethwaddr.c:
Remove compiler warning
ndb/src/ndbapi/ndberror.c:
#ifdef not used code
regex/regcomp.c:
Remove not used code
regex/regcomp.ih:
Remove not used code (to remove compiler warnings)
sql-common/client.c:
Remove compiler warnings
sql/field.cc:
Simple optimization
sql/ha_innodb.cc:
Rename mysql_embedded -> mysqld_embedded
sql/item.cc:
Fix comments
Move variables first on block
Remove else after return
Simple optimizations
(no logic changes)
sql/item_cmpfunc.cc:
Added comment
sql/mysql_priv.h:
Rename mysql_embedded -> mysqld_embedded
sql/mysqld.cc:
Rename mysql_embedded -> mysqld_embedded
sql/sql_acl.cc:
Added comments
simple optimization
Fixed 'very unlikely' bug when doing REVOKE ALL PRIVILEGES
sql/sql_select.cc:
More comments
Simple optimization
sql/sql_show.cc:
Simple changes to make similar code similar
More comments
sql/sql_string.cc:
Trivial optimization and better code layout
strings/Makefile.am:
Change xml.c to use bcmp to avoid warnings from purify
strings/xml.c:
Change xml.c to use bcmp to avoid warnings from purify
tests/client_test.c:
Remove usage of MAXPATHLEN (all MySQL code uses FN_REFLEN)
server".
Altough mysql_create_db()/mysql_drop_db() API calls are deprecated
since 4.0, they should not crash server and should not stall connection
in case of errors.
sql/sql_parse.cc:
Handling of COM_CREATE_DB, COM_DROP_DB:
mysql_create_db() requires from its second parameter to be non-zero.
We also should call send_error() if mysql_create_db or mysql_drop_db
return error (like we do it for SQL versions of these commands).
tests/client_test.c:
Added test for bug #6081 "Execution of deprecated mysql_create_db()
crashes server".
statements and negative time/date values".
The bug was in wrong sprintf format used in the client library.
The fix moves TIME -> string conversion functions to sql-common and
utilized them in the client library.
include/my_time.h:
Declarations for new functions shared between the client and server.
libmysql/libmysql.c:
Fix for Bug#6049 "Loss of sign when using prepared statements and negative
time/date values": use the same function as the server to convert
date/time/datetime values to strings.
sql-common/my_time.c:
Implementation of my_{time,datetime,date,TIME}_to_str: it's
needed by the client library, so it should be shared.
sql/field.cc:
Don't create String object if it's not needed.
sql/item.cc:
Don't create String object if it's not needed: TIME_to_string was
moved to my_TIME_to_str, with different arguments.
sql/item_timefunc.cc:
Don't create String object if it's not needed.
sql/mysql_priv.h:
TIME_to_string and MAX_DATE_REP_LENGTH moved to the client library.
MAX_DATE_REP_LENGTH was renamed to MAX_DATE_STRING_REP_LENGTH to not
conflict with the same name in libmysql.c
sql/protocol.cc:
Don't create String object if it's not needed.
sql/time.cc:
Implementation of my_{time,date,datetime,TIME}_to_str moved to my_time.c
shared between the client and the server.
tests/client_test.c:
A test case for Bug#6049.
Make checks (asserts) performed in client_test.c work in non-debug
builds.
tests/client_test.c:
Make checks (asserts) performed in client_test.c work in non-debug
builds: all asserts/DBUG_ASSERTs replaced with if (!(expr)) abort();
crashes mysqld": implementation for a generic item tree modifications
registry. Every item tree modification which should be rolled back for
subsequent execution of a prepared statement or stored procedure should
be saved in the registry. All such modifications are rolled back at once
during cleanup stage of PS.
Actual fix for the bug just adds a call to register modifications to
convert_constant_item.
Post review fixes implemented.
mysql-test/r/ps.result:
A fix for bug#5748, test results fixed.
mysql-test/t/ps.test:
A test case for Bug#5748 "Prepared statement with BETWEEN and bigint
values crashes mysqld"
sql/item.cc:
Fix for Bug#5748 "Prepared statement with BETWEEN and bigint values
crashes mysqld":
First step in removing up item-specific cleanups: now all such
tree modifications should be done using the genericm mechanism implemented
in this changeset.
sql/item.h:
Fix for Bug#5748 "Prepared statement with BETWEEN and bigint values
crashes mysqld": no need for an item-specific change record any more.
sql/item_cmpfunc.cc:
A fix for Bug#5748 "Prepared statement with BETWEEN and bigint
values crashes mysqld": register item tree transformation performed by
convert_constant_item.
sql/sql_class.cc:
Implementation for item tree transformations registry.
sql/sql_class.h:
Declarations, necessary for the tree transformations registry.
sql/sql_parse.cc:
Assert that the item tree transformations registry is not used for
conventional execution.
sql/sql_prepare.cc:
Use of the item tree modifications registry in prepared statements:
rollback all modifications in the end of statement prepare and execute.
Also we now always set thd->current_arena to be able to determine that
this is an execution of prepared statement inside the registry code.
tests/client_test.c:
A typo fixed.
Run client_test as a testcase
new file
mysql-test-run.sh:
More clearly report failure if --force, also exit 1
client_test.c:
Use MAXPATHLEN in test_frm_bug()
Renable test cases disabled for running from mysql-test-run
mysqltest.c:
Don't check errno from popen, may not be set
client/mysqltest.c:
Don't check errno from popen, may not be set
tests/client_test.c:
Use MAXPATHLEN in test_frm_bug()
Renable test cases disabled for running from mysql-test-run
mysql-test/mysql-test-run.sh:
More clearly report failure if --force, also exit 1
mysql-test/t/client_test.test:
Run client_test as a testcase
Return exit(1) instead of exit(0) on failure.
Allow longer path names using MAXPATHLEN.
Added option --testcase that skips some failing tests.
Replaced 'return' with exit(1) in test_frm_bug().
mysqltest.c:
Let --exec fail if command fails
client/mysqltest.c:
Let --exec fail if command fails
tests/client_test.c:
Return exit(1) instead of exit(0) on failure.
Allow longer path names using MAXPATHLEN.
Added option --testcase that skips some failing tests.
Replaced 'return' with exit(1) in test_frm_bug().
prepared statements."
include/hash.h:
New declaration for hash_reset() function. The old version was not used.
libmysql/client_settings.h:
Declaration for mysql_detach_stmt_list().
libmysql/libmysql.c:
Fix for bug#5315 "mysql_change_user() doesn't free prepared statements":
add call to mysql_detach_stmt_list(prepared statements) to
mysql_change_user(): all statements are freed by server, so client
counterparts need to be marked as not usable.
mysys/hash.c:
Fix for bug#5315 "mysql_change_user() doesn't free prepared statements":
implementation of hash_reset(), which frees all hash elements
and prepares the hash for reuse.
sql-common/client.c:
Fix for bug#5315 "mysql_change_user() doesn't free prepared statements":
implementation of mysql_detach_stmt_list(): zero connection pointer
in given statement list, thus marking given statements as not usable.
sql/sql_class.cc:
Fix for bug#5315 "mysql_change_user() doesn't free prepared statements":
reset prepared statements map in THD::change_user().
sql/sql_class.h:
Fix for bug#5315 "mysql_change_user() doesn't free prepared statements":
implementation of Statement_map::reset().
A little cleanup of ~Statement_map(): first empty names_hash, as st_hash
has a free function, which will delete statements.
tests/client_test.c:
A test case for bug #5315 "mysql_change_user() doesn't free prepared
statements".
Statements":
- fix a couple of net->buff overruns in libmysql,
- check in the server that statement parameter count is less than
65535 (maximum value supported by prepared statements protocol).
libmysql/libmysql.c:
Bug#5194 "Bulk Insert Failures with Prepared Statements":
- clean up my_realloc_str()
- ensure that net buffer has space when storing null bits and
parameter typecodes.
sql/net_serv.cc:
- set net->last_errno if packet is too big, even on client
(Why was it ifdefed before?)
sql/sql_prepare.cc:
Bug#5194 "Bulk Insert Failures with Prepared Statements":
- if placeholder count is bigger than 65535, give error.
We have only 2 bytes reserved for transferring placeholder count
in 4.1 protocol.
- can't add a proper error code and message in 4.1 because of
possible merge difficulties."
tests/client_test.c:
A test case for Bug#5194 "Bulk Insert Failures with Prepared
Statements".
server" (use my_charset_bin for stmt id hash).
sql/sql_class.cc:
Fix for bug#5399 "Wrong statement executed by MySQL server":
we hash stmt ids bytewise, need to use binary collation for it.
tests/client_test.c:
Test case for bug#5399 "Wrong statement executed by MySQL server":
create number of statements returning different results, execute them
one by one.
parameters": when unpacking binary time recieved from client, handle
the case when length is 0: it means all MYSQL_TIME members are zero.
include/my_time.h:
Declaration for set_zero_time: a tiny piece of code, which I
see no reason to not reuse.
libmysql/libmysql.c:
set_zero_time implementation is now shared between client and
server.
sql-common/my_time.c:
set_zero_time implementation added.
sql/sql_prepare.cc:
A fix for Bug#4231 "Wrong result with MYSQL_TIME parameters":
when unpacking binary time recieved from client, handle the
case when length is 0: it means all MYSQL_TIME members are zero.
tests/client_test.c:
Test case for bug#4231 "Wrong result with MYSQL_TIME parameters"
"Mediumint and PS problem": just treat mediumint as long.
libmysql/libmysql.c:
A short fix for bug #5126 "Mediumint and PS problem": just
treat mediumint as long, as it's sent just like long.
tests/client_test.c:
A test case for bug#5126
libmysql/libmysql.c:
Cleanup of conversion part of libmysql (prepared statements protocol):
- now we have basic support for any conversion sequence:
when we don't implement direct conversion of given value to requested
buffer type (i.e. time -> double, or the other way around) we
first convert i.e. time -> string and then call string -> double
conversion.
param->offset is now handled only in one place.
- conversion functions renamed from send_data_{string, long, double}
to fetch_{string,long,double}_with_conversion. Don't be confused
with strange diff for send_data_long: I had to move send_data_string
before all other sends as it's used inside thesm. (Shall we have
a forward declaration instead?-)
- a little cleanup in read_binary_{date,time,datetime} - now type of
date value is set inside these functions, so we can be sure
that we always return fully filled MYSQL_TIME structure to the user
- float -> string conversion is fixed to honor param->precision.
This is a step forward in fixing bug#4172
tests/client_test.c:
test fix: now libmysql always sets MYSQL_TIME::time_type field.
We need to set these fields in the test as later _in values are compared
with canonical by plain memcmp.
work (prepared statements)" and after-review fixes:
- str_to_TIME renamed to str_to_datetime to pair with str_to_time
- functions str_to_time and str_to_TIME moved to sql-common
- send_data_str now supports MYSQL_TYPE_TIME, MYSQL_TIME_DATE,
MYSQL_TIME_DATETIME types of user input buffers.
- few more comments in the client library
- a test case added.
VC++Files/libmysql/libmysql.dsp:
new file: my_time.c
VC++Files/libmysqld/libmysqld.dsp:
new file: my_time.c
VC++Files/sql/mysqld.dsp:
new file: my_time.c
include/Makefile.am:
- mysql_time.h added to the list of installed client library headers
include/mysql.h:
- declarations for MYSQL_TIME and enum_mysql_timestamp_type moved to
mysql_time.h, which is in shared use of client library and mysys.
libmysql/Makefile.shared:
- my_time.lo added to the list of libmysql objects
libmysql/libmysql.c:
Fix for bug#4030 "Client side conversion string -> date type doesn't work
(prepared statements)" and cleanup.
- added case labels for TIME/DATE/DATETIME types to send_data_str
- comments for read_binary_{date,time,datetime}, fetch_result_*, fetch_results.
libmysqld/Makefile.am:
- my_time.c added
sql-common/Makefile.am:
- my_time.c added to the list of files included into source distribution.
sql/Makefile.am:
my_time.c added to the list of mysqld sources.
sql/field.cc:
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...}
sql/item.cc:
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...}
sql/item_timefunc.cc:
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...}
sql/mysql_priv.h:
- added typedefs for TIME and timestamp_type
- removed declarations for str_to_time and str_to_TIME (now this functions
reside in mysys)
sql/mysqld.cc:
- log_10_int moved to mysys (it's used by str_to_TIME and str_to_time)
- enum values TIMESTAMP_{TIME,DATE,DATETIME} were renamed to
MYSQL_TIMESTAMP_{TIME,DATE,DATETIME}
sql/set_var.cc:
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...}
sql/set_var.h:
- fixed timestamp_type usage to be compatible with typedef.
sql/sql_prepare.cc:
- TIMESTAMP_{TIME,DATE,DATETIME} were renamed to
MYSQL_TIMESTAMP_{TIME,DATE,DATETIME}
- embedded library implementation of set_param_{time,date,datetime} is
much simplier now, as MYSQL_TIME is the same as TIME.
sql/sql_yacc.yy:
- s/\<TIMESTAMP_/MYSQL_TIMESTAMP/gc
sql/structs.h:
- declarations for TIME and timestamp_type replaced with typedefs
- str_to_datetime arguments moved to mysys headers
sql/time.cc:
- str_to_time and str_to_TIME moved to mysys
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...} as these names are now
exported to client.
- str_to_TIME renamed to str_to_datetime to pair with str_to_time
- str_to_TIME_with_warn renamed accordingly
sql/tztime.cc:
- TIMESTAMP_{TIME,DATE,DATETIME,...} renamed to
MYSQL_TIMESTAMP_{TIME,DATETIME,DATE,...}
tests/client_test.c:
- a test case for Bug#4030 "Client side conversion string -> date type
doesn't work (prepared statements)"
test suite fixed.
libmysql/libmysql.c:
- fixed a bug in protocol -> host integer conversions in send_data_str
tests/client_test.c:
Fixes to make client_test run on 64 bit Sun:
- strlen() doesn't accept NULL input
- Sun is high byte first, so we can't store bind 32-bit variables to 16-bit
to MYSQL_TYPE_SHORT/TINY
is broken (prepared statements)": fixed date handling in many places
of prepared statements code.
libmysql/libmysql.c:
Fix for Bug#4026:
- now buffer_length is defined for any buffer type. Network buffer
preallocation cleaned up.
- added constants for maximum buffer sizes necessary for MYSQL_TYPE_DATE,
MYSQL_TYPE_TIME, MYSQL_TYPE_DATETIME types.
- TIME/DATETIME packing/unpacking functions fixed
- now result set metadata is always updated from fields sent to COM_EXECUTE.
This is necessary to make 'SELECT ?' queries work without conversions.
sql/item.cc:
- added implementatoin of Item_param::get_date
sql/item.h:
- added enum_field_types Item_param::param_type. First step for proper
handling of placeholders.
- added get_date() implementation to prevent date -> string -> date
conversions when MYSQL_TYPE_DATE/DATETIME parameter is used in temporal
context.
sql/protocol.cc:
Fix for Bug#4026:
- PACKET_BUFFET_EXTRA_ALLOC -> PACKET_BUFFER_EXTRA_ALLOC.
The define itself was moved to .cc as it's used only in protocol.cc
- fixed Protocol_prep::store_time() call.
sql/protocol.h:
- PACKET_BUFFER_EXTRA_ALLOC moved to protocol.cc
sql/sql_prepare.cc:
Fix for Bug#4026:
- MYSQL_TYPE_TIME/DATETIME handling fixed.
- added initialization for Item_param::param_type in
setup_one_conversion_function
tests/client_test.c:
Test case for Bug#4026
include/mysql.h:
mysql_stmt_field_count() declaration
libmysql/libmysql.c:
added mysql_stmt_field_count():
we need this function to ease use of mysql_stmt_result_metadata:
if mysql_stmt_field_count() != 0 mysql_stmt_result_metadata fails
only if OOM.
libmysql/libmysql.def:
declaration for mysql_stmt_field_count()
result": new Item_param member for use in val_str()
sql/item.cc:
Fix for Bug#3796:
- return str_value_ptr from val_str to prevent modifications of parameter
value.
sql/item.h:
Fix for bug#3796:
- Item::val_str() method semantics documented
- new member of Item_param
tests/client_test.c:
Test case for bug #3796 added. A few compile-time warnings removed.
