mariadb/sql/sql_lex.h

2369 lines
70 KiB
C
Raw Normal View History

/* Copyright 2000-2008 MySQL AB, 2008 Sun Microsystems, Inc.
2000-07-31 21:29:14 +02:00
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
2000-07-31 21:29:14 +02:00
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
2000-07-31 21:29:14 +02:00
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
/**
@defgroup Semantic_Analysis Semantic Analysis
*/
2000-07-31 21:29:14 +02:00
#ifndef SQL_LEX_INCLUDED
#define SQL_LEX_INCLUDED
#include "violite.h" /* SSL_type */
#include "sql_trigger.h"
#include "item.h" /* From item_subselect.h: subselect_union_engine */
#include "thr_lock.h" /* thr_lock_type, TL_UNLOCK */
2000-07-31 21:29:14 +02:00
/* YACC and LEX Definitions */
/* These may not be declared yet */
class Table_ident;
class sql_exchange;
class LEX_COLUMN;
class sp_head;
class sp_name;
class sp_instr;
class sp_pcontext;
2006-01-11 11:35:25 +01:00
class st_alter_tablespace;
2005-07-18 13:31:02 +02:00
class partition_info;
class Event_parse_data;
WL#4738 streamline/simplify @@variable creation process Bug#16565 mysqld --help --verbose does not order variablesBug#20413 sql_slave_skip_counter is not shown in show variables Bug#20415 Output of mysqld --help --verbose is incomplete Bug#25430 variable not found in SELECT @@global.ft_max_word_len; Bug#32902 plugin variables don't know their names Bug#34599 MySQLD Option and Variable Reference need to be consistent in formatting! Bug#34829 No default value for variable and setting default does not raise error Bug#34834 ? Is accepted as a valid sql mode Bug#34878 Few variables have default value according to documentation but error occurs Bug#34883 ft_boolean_syntax cant be assigned from user variable to global var. Bug#37187 `INFORMATION_SCHEMA`.`GLOBAL_VARIABLES`: inconsistent status Bug#40988 log_output_basic.test succeeded though syntactically false. Bug#41010 enum-style command-line options are not honoured (maria.maria-recover fails) Bug#42103 Setting key_buffer_size to a negative value may lead to very large allocations Bug#44691 Some plugins configured as MYSQL_PLUGIN_MANDATORY in can be disabled Bug#44797 plugins w/o command-line options have no disabling option in --help Bug#46314 string system variables don't support expressions Bug#46470 sys_vars.max_binlog_cache_size_basic_32 is broken Bug#46586 When using the plugin interface the type "set" for options caused a crash. Bug#47212 Crash in DBUG_PRINT in mysqltest.cc when trying to print octal number Bug#48758 mysqltest crashes on sys_vars.collation_server_basic in gcov builds Bug#49417 some complaints about mysqld --help --verbose output Bug#49540 DEFAULT value of binlog_format isn't the default value Bug#49640 ambiguous option '--skip-skip-myisam' (double skip prefix) Bug#49644 init_connect and \0 Bug#49645 init_slave and multi-byte characters Bug#49646 mysql --show-warnings crashes when server dies
2009-12-22 10:35:56 +01:00
class set_var_base;
class sys_var;
class Item_func_match;
class Alter_drop;
class Alter_column;
class Key;
class File_parser;
class Key_part_spec;
WL#4738 streamline/simplify @@variable creation process Bug#16565 mysqld --help --verbose does not order variablesBug#20413 sql_slave_skip_counter is not shown in show variables Bug#20415 Output of mysqld --help --verbose is incomplete Bug#25430 variable not found in SELECT @@global.ft_max_word_len; Bug#32902 plugin variables don't know their names Bug#34599 MySQLD Option and Variable Reference need to be consistent in formatting! Bug#34829 No default value for variable and setting default does not raise error Bug#34834 ? Is accepted as a valid sql mode Bug#34878 Few variables have default value according to documentation but error occurs Bug#34883 ft_boolean_syntax cant be assigned from user variable to global var. Bug#37187 `INFORMATION_SCHEMA`.`GLOBAL_VARIABLES`: inconsistent status Bug#40988 log_output_basic.test succeeded though syntactically false. Bug#41010 enum-style command-line options are not honoured (maria.maria-recover fails) Bug#42103 Setting key_buffer_size to a negative value may lead to very large allocations Bug#44691 Some plugins configured as MYSQL_PLUGIN_MANDATORY in can be disabled Bug#44797 plugins w/o command-line options have no disabling option in --help Bug#46314 string system variables don't support expressions Bug#46470 sys_vars.max_binlog_cache_size_basic_32 is broken Bug#46586 When using the plugin interface the type "set" for options caused a crash. Bug#47212 Crash in DBUG_PRINT in mysqltest.cc when trying to print octal number Bug#48758 mysqltest crashes on sys_vars.collation_server_basic in gcov builds Bug#49417 some complaints about mysqld --help --verbose output Bug#49540 DEFAULT value of binlog_format isn't the default value Bug#49640 ambiguous option '--skip-skip-myisam' (double skip prefix) Bug#49644 init_connect and \0 Bug#49645 init_slave and multi-byte characters Bug#49646 mysql --show-warnings crashes when server dies
2009-12-22 10:35:56 +01:00
/**
used by the parser to store internal variable name
*/
struct sys_var_with_base
{
sys_var *var;
LEX_STRING base_name;
};
2000-07-31 21:29:14 +02:00
#ifdef MYSQL_SERVER
/*
The following hack is needed because mysql_yacc.cc does not define
YYSTYPE before including this file
*/
2000-07-31 21:29:14 +02:00
#ifdef MYSQL_YACC
#define LEX_YYSTYPE void *
#else
#include "lex_symbol.h"
#if MYSQL_LEX
#include "item_func.h" /* Cast_target used in sql_yacc.h */
2000-07-31 21:29:14 +02:00
#include "sql_yacc.h"
#define LEX_YYSTYPE YYSTYPE *
#else
#define LEX_YYSTYPE void *
#endif
2000-07-31 21:29:14 +02:00
#endif
#endif
2000-07-31 21:29:14 +02:00
/*
When a command is added here, be sure it's also added in mysqld.cc
in "struct show_var_st status_vars[]= {" ...
If the command returns a result set or is not allowed in stored
functions or triggers, please also make sure that
sp_get_flags_for_command (sp_head.cc) returns proper flags for the
added SQLCOM_.
*/
2000-07-31 21:29:14 +02:00
enum enum_sql_command {
SQLCOM_SELECT, SQLCOM_CREATE_TABLE, SQLCOM_CREATE_INDEX, SQLCOM_ALTER_TABLE,
SQLCOM_UPDATE, SQLCOM_INSERT, SQLCOM_INSERT_SELECT,
SQLCOM_DELETE, SQLCOM_TRUNCATE, SQLCOM_DROP_TABLE, SQLCOM_DROP_INDEX,
SQLCOM_SHOW_DATABASES, SQLCOM_SHOW_TABLES, SQLCOM_SHOW_FIELDS,
SQLCOM_SHOW_KEYS, SQLCOM_SHOW_VARIABLES, SQLCOM_SHOW_STATUS,
SQLCOM_SHOW_ENGINE_LOGS, SQLCOM_SHOW_ENGINE_STATUS, SQLCOM_SHOW_ENGINE_MUTEX,
SQLCOM_SHOW_PROCESSLIST, SQLCOM_SHOW_MASTER_STAT, SQLCOM_SHOW_SLAVE_STAT,
SQLCOM_SHOW_GRANTS, SQLCOM_SHOW_CREATE, SQLCOM_SHOW_CHARSETS,
SQLCOM_SHOW_COLLATIONS, SQLCOM_SHOW_CREATE_DB, SQLCOM_SHOW_TABLE_STATUS,
SQLCOM_SHOW_TRIGGERS,
2000-07-31 21:29:14 +02:00
SQLCOM_LOAD,SQLCOM_SET_OPTION,SQLCOM_LOCK_TABLES,SQLCOM_UNLOCK_TABLES,
SQLCOM_GRANT,
SQLCOM_CHANGE_DB, SQLCOM_CREATE_DB, SQLCOM_DROP_DB, SQLCOM_ALTER_DB,
SQLCOM_REPAIR, SQLCOM_REPLACE, SQLCOM_REPLACE_SELECT,
SQLCOM_CREATE_FUNCTION, SQLCOM_DROP_FUNCTION,
SQLCOM_REVOKE,SQLCOM_OPTIMIZE, SQLCOM_CHECK,
SQLCOM_ASSIGN_TO_KEYCACHE, SQLCOM_PRELOAD_KEYS,
SQLCOM_FLUSH, SQLCOM_KILL, SQLCOM_ANALYZE,
SQLCOM_ROLLBACK, SQLCOM_ROLLBACK_TO_SAVEPOINT,
SQLCOM_COMMIT, SQLCOM_SAVEPOINT, SQLCOM_RELEASE_SAVEPOINT,
SQLCOM_SLAVE_START, SQLCOM_SLAVE_STOP,
SQLCOM_BEGIN, SQLCOM_CHANGE_MASTER,
SQLCOM_RENAME_TABLE,
SQLCOM_RESET, SQLCOM_PURGE, SQLCOM_PURGE_BEFORE, SQLCOM_SHOW_BINLOGS,
SQLCOM_SHOW_OPEN_TABLES,
SQLCOM_HA_OPEN, SQLCOM_HA_CLOSE, SQLCOM_HA_READ,
SQLCOM_SHOW_SLAVE_HOSTS, SQLCOM_DELETE_MULTI, SQLCOM_UPDATE_MULTI,
2001-12-17 22:24:51 +01:00
SQLCOM_SHOW_BINLOG_EVENTS, SQLCOM_SHOW_NEW_MASTER, SQLCOM_DO,
SQLCOM_SHOW_WARNS, SQLCOM_EMPTY_QUERY, SQLCOM_SHOW_ERRORS,
SQLCOM_SHOW_STORAGE_ENGINES, SQLCOM_SHOW_PRIVILEGES,
SQLCOM_HELP, SQLCOM_CREATE_USER, SQLCOM_DROP_USER, SQLCOM_RENAME_USER,
SQLCOM_REVOKE_ALL, SQLCOM_CHECKSUM,
SQLCOM_CREATE_PROCEDURE, SQLCOM_CREATE_SPFUNCTION, SQLCOM_CALL,
SQLCOM_DROP_PROCEDURE, SQLCOM_ALTER_PROCEDURE,SQLCOM_ALTER_FUNCTION,
SQLCOM_SHOW_CREATE_PROC, SQLCOM_SHOW_CREATE_FUNC,
SQLCOM_SHOW_STATUS_PROC, SQLCOM_SHOW_STATUS_FUNC,
SQLCOM_PREPARE, SQLCOM_EXECUTE, SQLCOM_DEALLOCATE_PREPARE,
2004-07-16 00:15:55 +02:00
SQLCOM_CREATE_VIEW, SQLCOM_DROP_VIEW,
SQLCOM_CREATE_TRIGGER, SQLCOM_DROP_TRIGGER,
2005-01-16 13:16:23 +01:00
SQLCOM_XA_START, SQLCOM_XA_END, SQLCOM_XA_PREPARE,
SQLCOM_XA_COMMIT, SQLCOM_XA_ROLLBACK, SQLCOM_XA_RECOVER,
SQLCOM_SHOW_PROC_CODE, SQLCOM_SHOW_FUNC_CODE,
2006-01-11 11:35:25 +01:00
SQLCOM_ALTER_TABLESPACE,
SQLCOM_INSTALL_PLUGIN, SQLCOM_UNINSTALL_PLUGIN,
SQLCOM_SHOW_AUTHORS, SQLCOM_BINLOG_BASE64_EVENT,
SQLCOM_SHOW_PLUGINS,
2006-05-03 02:53:26 +02:00
SQLCOM_SHOW_CONTRIBUTORS,
SQLCOM_CREATE_SERVER, SQLCOM_DROP_SERVER, SQLCOM_ALTER_SERVER,
2005-12-02 13:07:02 +01:00
SQLCOM_CREATE_EVENT, SQLCOM_ALTER_EVENT, SQLCOM_DROP_EVENT,
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
SQLCOM_SHOW_CREATE_EVENT, SQLCOM_SHOW_EVENTS,
SQLCOM_SHOW_CREATE_TRIGGER,
SQLCOM_ALTER_DB_UPGRADE,
SQLCOM_SHOW_PROFILE, SQLCOM_SHOW_PROFILES,
SQLCOM_SIGNAL, SQLCOM_RESIGNAL,
SQLCOM_SHOW_RELAYLOG_EVENTS,
Prevent bugs by making DBUG_* expressions syntactically equivalent to a single statement. --- Bug#24795: SHOW PROFILE Profiling is only partially functional on some architectures. Where there is no getrusage() system call, presently Null values are returned where it would be required. Notably, Windows needs some love applied to make it as useful. Syntax this adds: SHOW PROFILES SHOW PROFILE [types] [FOR QUERY n] [OFFSET n] [LIMIT n] where "n" is an integer and "types" is zero or many (comma-separated) of "CPU" "MEMORY" (not presently supported) "BLOCK IO" "CONTEXT SWITCHES" "PAGE FAULTS" "IPC" "SWAPS" "SOURCE" "ALL" It also adds a session variable (boolean) "profiling", set to "no" by default, and (integer) profiling_history_size, set to 15 by default. This patch abstracts setting THDs' "proc_info" behind a macro that can be used as a hook into the profiling code when profiling support is compiled in. All future code in this line should use that mechanism for setting thd->proc_info. --- Tests are now set to omit the statistics. --- Adds an Information_schema table, "profiling" for access to "show profile" data. --- Merge zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community-3--bug24795 into zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community --- Fix merge problems. --- Fixed one bug in the query_source being NULL. Updated test results. --- Include more thorough profiling tests. Improve support for prepared statements. Use session-specific query IDs, starting at zero. --- Selecting from I_S.profiling is no longer quashed in profiling, as requested by Giuseppe. Limit the size of captured query text. No longer log queries that are zero length.
2007-02-22 16:03:08 +01:00
/*
When a command is added here, be sure it's also added in mysqld.cc
in "struct show_var_st status_vars[]= {" ...
*/
/* This should be the last !!! */
SQLCOM_END
2000-07-31 21:29:14 +02:00
};
// describe/explain types
#define DESCRIBE_NORMAL 1
#define DESCRIBE_EXTENDED 2
2005-12-22 10:29:00 +01:00
/*
This is not within #ifdef because we want "EXPLAIN PARTITIONS ..." to produce
2005-12-22 10:29:00 +01:00
additional "partitions" column even if partitioning is not compiled in.
*/
#define DESCRIBE_PARTITIONS 4
#ifdef MYSQL_SERVER
enum enum_sp_suid_behaviour
{
SP_IS_DEFAULT_SUID= 0,
SP_IS_NOT_SUID,
SP_IS_SUID
};
enum enum_sp_data_access
{
SP_DEFAULT_ACCESS= 0,
SP_CONTAINS_SQL,
SP_NO_SQL,
SP_READS_SQL_DATA,
SP_MODIFIES_SQL_DATA
};
const LEX_STRING sp_data_access_name[]=
{
{ C_STRING_WITH_LEN("") },
{ C_STRING_WITH_LEN("CONTAINS SQL") },
{ C_STRING_WITH_LEN("NO SQL") },
{ C_STRING_WITH_LEN("READS SQL DATA") },
{ C_STRING_WITH_LEN("MODIFIES SQL DATA") }
};
2004-07-16 00:15:55 +02:00
#define DERIVED_SUBQUERY 1
#define DERIVED_VIEW 2
enum enum_view_create_mode
{
VIEW_CREATE_NEW, // check that there are not such VIEW/table
VIEW_ALTER, // check that VIEW .frm with such name exists
VIEW_CREATE_OR_REPLACE // check only that there are not such table
};
enum enum_drop_mode
{
DROP_DEFAULT, // mode is not specified
DROP_CASCADE, // CASCADE option
DROP_RESTRICT // RESTRICT option
};
/* Options to add_table_to_list() */
#define TL_OPTION_UPDATING 1
#define TL_OPTION_FORCE_INDEX 2
#define TL_OPTION_IGNORE_LEAVES 4
#define TL_OPTION_ALIAS 8
2000-07-31 21:29:14 +02:00
typedef List<Item> List_item;
/* SERVERS CACHE CHANGES */
typedef struct st_lex_server_options
{
long port;
uint server_name_length;
char *server_name, *host, *db, *username, *password, *scheme, *socket, *owner;
} LEX_SERVER_OPTIONS;
/**
Structure to hold parameters for CHANGE MASTER or START/STOP SLAVE
or SHOW NEW MASTER.
Remark: this should not be confused with Master_info (and perhaps
would better be renamed to st_lex_replication_info). Some fields,
e.g., delay, are saved in Relay_log_info, not in Master_info.
*/
2000-07-31 21:29:14 +02:00
typedef struct st_lex_master_info
{
char *host, *user, *password, *log_file_name;
2000-07-31 21:29:14 +02:00
uint port, connect_retry;
float heartbeat_period;
2000-07-31 21:29:14 +02:00
ulonglong pos;
2001-07-05 01:14:31 +02:00
ulong server_id;
/*
Enum is used for making it possible to detect if the user
changed variable or if it should be left at old value
*/
enum {LEX_MI_UNCHANGED, LEX_MI_DISABLE, LEX_MI_ENABLE}
2009-10-01 18:44:53 +02:00
ssl, ssl_verify_server_cert, heartbeat_opt, repl_ignore_server_ids_opt;
char *ssl_key, *ssl_cert, *ssl_ca, *ssl_capath, *ssl_cipher;
char *relay_log_name;
ulong relay_log_pos;
2009-10-01 18:44:53 +02:00
DYNAMIC_ARRAY repl_ignore_server_ids;
2000-07-31 21:29:14 +02:00
} LEX_MASTER_INFO;
enum sub_select_type
{
UNSPECIFIED_TYPE,UNION_TYPE, INTERSECT_TYPE,
EXCEPT_TYPE, GLOBAL_OPTIONS_TYPE, DERIVED_TABLE_TYPE, OLAP_TYPE
};
enum olap_type
{
UNSPECIFIED_OLAP_TYPE, CUBE_TYPE, ROLLUP_TYPE
};
enum tablespace_op_type
{
NO_TABLESPACE_OP, DISCARD_TABLESPACE, IMPORT_TABLESPACE
};
/*
String names used to print a statement with index hints.
Keep in sync with index_hint_type.
*/
extern const char * index_hint_type_name[];
WL#3817: Simplify string / memory area types and make things more consistent (first part) The following type conversions was done: - Changed byte to uchar - Changed gptr to uchar* - Change my_string to char * - Change my_size_t to size_t - Change size_s to size_t Removed declaration of byte, gptr, my_string, my_size_t and size_s. Following function parameter changes was done: - All string functions in mysys/strings was changed to use size_t instead of uint for string lengths. - All read()/write() functions changed to use size_t (including vio). - All protocoll functions changed to use size_t instead of uint - Functions that used a pointer to a string length was changed to use size_t* - Changed malloc(), free() and related functions from using gptr to use void * as this requires fewer casts in the code and is more in line with how the standard functions work. - Added extra length argument to dirname_part() to return the length of the created string. - Changed (at least) following functions to take uchar* as argument: - db_dump() - my_net_write() - net_write_command() - net_store_data() - DBUG_DUMP() - decimal2bin() & bin2decimal() - Changed my_compress() and my_uncompress() to use size_t. Changed one argument to my_uncompress() from a pointer to a value as we only return one value (makes function easier to use). - Changed type of 'pack_data' argument to packfrm() to avoid casts. - Changed in readfrm() and writefrom(), ha_discover and handler::discover() the type for argument 'frmdata' to uchar** to avoid casts. - Changed most Field functions to use uchar* instead of char* (reduced a lot of casts). - Changed field->val_xxx(xxx, new_ptr) to take const pointers. Other changes: - Removed a lot of not needed casts - Added a few new cast required by other changes - Added some cast to my_multi_malloc() arguments for safety (as string lengths needs to be uint, not size_t). - Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done explicitely as this conflict was often hided by casting the function to hash_get_key). - Changed some buffers to memory regions to uchar* to avoid casts. - Changed some string lengths from uint to size_t. - Changed field->ptr to be uchar* instead of char*. This allowed us to get rid of a lot of casts. - Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar - Include zlib.h in some files as we needed declaration of crc32() - Changed MY_FILE_ERROR to be (size_t) -1. - Changed many variables to hold the result of my_read() / my_write() to be size_t. This was needed to properly detect errors (which are returned as (size_t) -1). - Removed some very old VMS code - Changed packfrm()/unpackfrm() to not be depending on uint size (portability fix) - Removed windows specific code to restore cursor position as this causes slowdown on windows and we should not mix read() and pread() calls anyway as this is not thread safe. Updated function comment to reflect this. Changed function that depended on original behavior of my_pwrite() to itself restore the cursor position (one such case). - Added some missing checking of return value of malloc(). - Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow. - Changed type of table_def::m_size from my_size_t to ulong to reflect that m_size is the number of elements in the array, not a string/memory length. - Moved THD::max_row_length() to table.cc (as it's not depending on THD). Inlined max_row_length_blob() into this function. - More function comments - Fixed some compiler warnings when compiled without partitions. - Removed setting of LEX_STRING() arguments in declaration (portability fix). - Some trivial indentation/variable name changes. - Some trivial code simplifications: - Replaced some calls to alloc_root + memcpy to use strmake_root()/strdup_root(). - Changed some calls from memdup() to strmake() (Safety fix) - Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
typedef uchar index_clause_map;
/*
Bits in index_clause_map : one for each possible FOR clause in
USE/FORCE/IGNORE INDEX index hint specification
*/
#define INDEX_HINT_MASK_JOIN (1)
#define INDEX_HINT_MASK_GROUP (1 << 1)
#define INDEX_HINT_MASK_ORDER (1 << 2)
#define INDEX_HINT_MASK_ALL (INDEX_HINT_MASK_JOIN | INDEX_HINT_MASK_GROUP | \
INDEX_HINT_MASK_ORDER)
/* Single element of an USE/FORCE/IGNORE INDEX list specified as a SQL hint */
2007-07-23 18:09:48 +02:00
class Index_hint : public Sql_alloc
{
public:
/* The type of the hint : USE/FORCE/IGNORE */
enum index_hint_type type;
/* Where the hit applies to. A bitmask of INDEX_HINT_MASK_<place> values */
index_clause_map clause;
/*
The index name. Empty (str=NULL) name represents an empty list
USE INDEX () clause
*/
LEX_STRING key_name;
2007-07-23 18:09:48 +02:00
Index_hint (enum index_hint_type type_arg, index_clause_map clause_arg,
char *str, uint length) :
type(type_arg), clause(clause_arg)
{
key_name.str= str;
key_name.length= length;
}
void print(THD *thd, String *str);
};
2002-05-06 23:04:16 +02:00
/*
The state of the lex parsing for selects
master and slaves are pointers to select_lex.
master is pointer to upper level node.
slave is pointer to lower level node
select_lex is a SELECT without union
unit is container of either
- One SELECT
- UNION of selects
select_lex and unit are both inherited form select_lex_node
neighbors are two select_lex or units on the same level
2002-05-06 23:04:16 +02:00
All select describing structures linked with following pointers:
- list of neighbors (next/prev) (prev of first element point to slave
2002-05-06 23:04:16 +02:00
pointer of upper structure)
- For select this is a list of UNION's (or one element list)
- For units this is a list of sub queries for the upper level select
- pointer to master (master), which is
If this is a unit
- pointer to outer select_lex
If this is a select_lex
- pointer to outer unit structure for select
- pointer to slave (slave), which is either:
If this is a unit:
- first SELECT that belong to this unit
If this is a select_lex
- first unit that belong to this SELECT (subquries or derived tables)
- list of all select_lex (link_next/link_prev)
This is to be used for things like derived tables creation, where we
go through this list and create the derived tables.
If unit contain several selects (UNION now, INTERSECT etc later)
then it have special select_lex called fake_select_lex. It used for
storing global parameters (like ORDER BY, LIMIT) and executing union.
Subqueries used in global ORDER BY clause will be attached to this
fake_select_lex, which will allow them correctly resolve fields of
'upper' UNION and outer selects.
For example for following query:
2002-05-06 23:04:16 +02:00
select *
from table1
where table1.field IN (select * from table1_1_1 union
select * from table1_1_2)
union
select *
from table2
where table2.field=(select (select f1 from table2_1_1_1_1
where table2_1_1_1_1.f2=table2_1_1.f3)
from table2_1_1
where table2_1_1.f1=table2.f2)
union
select * from table3;
we will have following structure:
select1: (select * from table1 ...)
select2: (select * from table2 ...)
select3: (select * from table3)
select1.1.1: (select * from table1_1_1)
...
2002-05-06 23:04:16 +02:00
main unit
fake0
2002-05-06 23:04:16 +02:00
select1 select2 select3
|^^ |^
s||| ||master
l||| |+---------------------------------+
a||| +---------------------------------+|
v|||master slave ||
e||+-------------------------+ ||
V| neighbor | V|
unit1.1<+==================>unit1.2 unit2.1
2004-06-21 23:00:36 +02:00
fake1.1
select1.1.1 select 1.1.2 select1.2.1 select2.1.1
2002-05-06 23:04:16 +02:00
|^
||
V|
unit2.1.1.1
select2.1.1.1.1
relation in main unit will be following:
(bigger picture for:
main unit
fake0
select1 select2 select3
in the above picture)
2002-05-06 23:04:16 +02:00
main unit
|^^^^|fake_select_lex
|||||+--------------------------------------------+
||||+--------------------------------------------+|
|||+------------------------------+ ||
||+--------------+ | ||
slave||master | | ||
V| neighbor | neighbor | master|V
select1<========>select2<========>select3 fake0
2002-05-06 23:04:16 +02:00
list of all select_lex will be following (as it will be constructed by
parser):
select1->select2->select3->select2.1.1->select 2.1.2->select2.1.1.1.1-+
|
+---------------------------------------------------------------------+
|
+->select1.1.1->select1.1.2
*/
/*
Base class for st_select_lex (SELECT_LEX) &
st_select_lex_unit (SELECT_LEX_UNIT)
*/
struct LEX;
class st_select_lex;
class st_select_lex_unit;
class st_select_lex_node {
protected:
2002-05-06 23:04:16 +02:00
st_select_lex_node *next, **prev, /* neighbor list */
*master, *slave, /* vertical links */
*link_next, **link_prev; /* list of whole SELECT_LEX */
public:
2003-05-17 09:05:07 +02:00
ulonglong options;
/*
In sql_cache we store SQL_CACHE flag as specified by user to be
able to restore SELECT statement from internal structures.
*/
enum e_sql_cache { SQL_CACHE_UNSPECIFIED, SQL_NO_CACHE, SQL_CACHE };
e_sql_cache sql_cache;
/*
result of this query can't be cached, bit field, can be :
UNCACHEABLE_DEPENDENT
UNCACHEABLE_RAND
UNCACHEABLE_SIDEEFFECT
UNCACHEABLE_EXPLAIN
UNCACHEABLE_PREPARE
*/
uint8 uncacheable;
enum sub_select_type linkage;
2003-01-14 17:00:34 +01:00
bool no_table_names_allowed; /* used for global order by */
bool no_error; /* suppress error message (convert it to warnings) */
static void *operator new(size_t size) throw ()
{
WL#3817: Simplify string / memory area types and make things more consistent (first part) The following type conversions was done: - Changed byte to uchar - Changed gptr to uchar* - Change my_string to char * - Change my_size_t to size_t - Change size_s to size_t Removed declaration of byte, gptr, my_string, my_size_t and size_s. Following function parameter changes was done: - All string functions in mysys/strings was changed to use size_t instead of uint for string lengths. - All read()/write() functions changed to use size_t (including vio). - All protocoll functions changed to use size_t instead of uint - Functions that used a pointer to a string length was changed to use size_t* - Changed malloc(), free() and related functions from using gptr to use void * as this requires fewer casts in the code and is more in line with how the standard functions work. - Added extra length argument to dirname_part() to return the length of the created string. - Changed (at least) following functions to take uchar* as argument: - db_dump() - my_net_write() - net_write_command() - net_store_data() - DBUG_DUMP() - decimal2bin() & bin2decimal() - Changed my_compress() and my_uncompress() to use size_t. Changed one argument to my_uncompress() from a pointer to a value as we only return one value (makes function easier to use). - Changed type of 'pack_data' argument to packfrm() to avoid casts. - Changed in readfrm() and writefrom(), ha_discover and handler::discover() the type for argument 'frmdata' to uchar** to avoid casts. - Changed most Field functions to use uchar* instead of char* (reduced a lot of casts). - Changed field->val_xxx(xxx, new_ptr) to take const pointers. Other changes: - Removed a lot of not needed casts - Added a few new cast required by other changes - Added some cast to my_multi_malloc() arguments for safety (as string lengths needs to be uint, not size_t). - Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done explicitely as this conflict was often hided by casting the function to hash_get_key). - Changed some buffers to memory regions to uchar* to avoid casts. - Changed some string lengths from uint to size_t. - Changed field->ptr to be uchar* instead of char*. This allowed us to get rid of a lot of casts. - Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar - Include zlib.h in some files as we needed declaration of crc32() - Changed MY_FILE_ERROR to be (size_t) -1. - Changed many variables to hold the result of my_read() / my_write() to be size_t. This was needed to properly detect errors (which are returned as (size_t) -1). - Removed some very old VMS code - Changed packfrm()/unpackfrm() to not be depending on uint size (portability fix) - Removed windows specific code to restore cursor position as this causes slowdown on windows and we should not mix read() and pread() calls anyway as this is not thread safe. Updated function comment to reflect this. Changed function that depended on original behavior of my_pwrite() to itself restore the cursor position (one such case). - Added some missing checking of return value of malloc(). - Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow. - Changed type of table_def::m_size from my_size_t to ulong to reflect that m_size is the number of elements in the array, not a string/memory length. - Moved THD::max_row_length() to table.cc (as it's not depending on THD). Inlined max_row_length_blob() into this function. - More function comments - Fixed some compiler warnings when compiled without partitions. - Removed setting of LEX_STRING() arguments in declaration (portability fix). - Some trivial indentation/variable name changes. - Some trivial code simplifications: - Replaced some calls to alloc_root + memcpy to use strmake_root()/strdup_root(). - Changed some calls from memdup() to strmake() (Safety fix) - Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
return sql_alloc(size);
}
static void *operator new(size_t size, MEM_ROOT *mem_root) throw ()
{ return (void*) alloc_root(mem_root, (uint) size); }
static void operator delete(void *ptr,size_t size) { TRASH(ptr, size); }
static void operator delete(void *ptr, MEM_ROOT *mem_root) {}
// Ensures that at least all members used during cleanup() are initialized.
st_select_lex_node()
: next(NULL), prev(NULL),
master(NULL), slave(NULL),
link_next(NULL), link_prev(NULL),
linkage(UNSPECIFIED_TYPE)
{
}
virtual ~st_select_lex_node() {}
inline st_select_lex_node* get_master() { return master; }
virtual void init_query();
virtual void init_select();
2002-05-06 23:04:16 +02:00
void include_down(st_select_lex_node *upper);
void include_neighbour(st_select_lex_node *before);
void include_standalone(st_select_lex_node *sel, st_select_lex_node **ref);
2002-05-06 23:04:16 +02:00
void include_global(st_select_lex_node **plink);
void exclude();
virtual st_select_lex_unit* master_unit()= 0;
virtual st_select_lex* outer_select()= 0;
virtual st_select_lex* return_after_parsing()= 0;
virtual bool set_braces(bool value);
virtual bool inc_in_sum_expr();
virtual uint get_in_sum_expr();
virtual TABLE_LIST* get_table_list();
virtual List<Item>* get_item_list();
virtual ulong get_table_join_options();
virtual TABLE_LIST *add_table_to_list(THD *thd, Table_ident *table,
LEX_STRING *alias,
ulong table_options,
thr_lock_type flags= TL_UNLOCK,
2007-07-23 18:09:48 +02:00
List<Index_hint> *hints= 0,
LEX_STRING *option= 0);
2002-11-21 21:25:53 +01:00
virtual void set_lock_for_tables(thr_lock_type lock_type) {}
friend class st_select_lex_unit;
friend bool mysql_new_select(LEX *lex, bool move_down);
friend bool mysql_make_view(THD *thd, File_parser *parser,
TABLE_LIST *table, uint flags);
2002-05-06 23:04:16 +02:00
private:
void fast_exclude();
};
typedef class st_select_lex_node SELECT_LEX_NODE;
2002-05-06 23:04:16 +02:00
/*
SELECT_LEX_UNIT - unit of selects (UNION, INTERSECT, ...) group
SELECT_LEXs
*/
class THD;
class select_result;
class JOIN;
class select_union;
class Procedure;
class st_select_lex_unit: public st_select_lex_node {
protected:
TABLE_LIST result_table_list;
select_union *union_result;
TABLE *table; /* temporary table using for appending UNION results */
select_result *result;
ulonglong found_rows_for_union;
bool saved_error;
public:
// Ensures that at least all members used during cleanup() are initialized.
st_select_lex_unit()
: union_result(NULL), table(NULL), result(NULL),
cleaned(false),
fake_select_lex(NULL)
{
}
bool prepared, // prepare phase already performed for UNION (unit)
2002-10-27 20:29:40 +01:00
optimized, // optimize phase already performed for UNION (unit)
executed, // already executed
cleaned;
// list of fields which points to temporary table for union
List<Item> item_list;
2003-11-23 20:26:43 +01:00
/*
list of types of items inside union (used for union & derived tables)
Item_type_holders from which this list consist may have pointers to Field,
pointers is valid only after preparing SELECTS of this unit and before
any SELECT of this unit execution
*/
List<Item> types;
2002-05-06 23:04:16 +02:00
/*
Pointer to 'last' select or pointer to unit where stored
global parameters for union
*/
st_select_lex *global_parameters;
//node on wich we should return current_select pointer after parsing subquery
st_select_lex *return_to;
2002-05-06 23:04:16 +02:00
/* LIMIT clause runtime counters */
ha_rows select_limit_cnt, offset_limit_cnt;
/* not NULL if unit used in subselect, point to subselect item */
Item_subselect *item;
/* thread handler */
THD *thd;
2004-05-06 23:32:44 +02:00
/*
SELECT_LEX for hidden SELECT in onion which process global
ORDER BY and LIMIT
*/
st_select_lex *fake_select_lex;
st_select_lex *union_distinct; /* pointer to the last UNION DISTINCT */
bool describe; /* union exec() called for EXPLAIN */
Procedure *last_procedure; /* Pointer to procedure, if such exists */
2002-05-06 23:04:16 +02:00
void init_query();
st_select_lex_unit* master_unit();
st_select_lex* outer_select();
st_select_lex* first_select()
{
return my_reinterpret_cast(st_select_lex*)(slave);
}
st_select_lex_unit* next_unit()
{
return my_reinterpret_cast(st_select_lex_unit*)(next);
2003-03-06 16:02:10 +01:00
}
st_select_lex* return_after_parsing() { return return_to; }
void exclude_level();
void exclude_tree();
/* UNION methods */
bool prepare(THD *thd, select_result *result, ulong additional_options);
bool exec();
bool cleanup();
inline void unclean() { cleaned= 0; }
void reinit_exec_mechanism();
void print(String *str, enum_query_type query_type);
bool add_fake_select_lex(THD *thd);
void init_prepare_fake_select_lex(THD *thd);
inline bool is_prepared() { return prepared; }
bool change_result(select_subselect *result, select_subselect *old_result);
void set_limit(st_select_lex *values);
void set_thd(THD *thd_arg) { thd= thd_arg; }
inline bool is_union ();
friend void lex_start(THD *thd);
friend int subselect_union_engine::exec();
List<Item> *get_unit_column_types();
2002-05-06 23:04:16 +02:00
};
2002-07-05 07:15:57 +02:00
typedef class st_select_lex_unit SELECT_LEX_UNIT;
2002-05-06 23:04:16 +02:00
/*
SELECT_LEX - store information of parsed SELECT statment
2002-05-06 23:04:16 +02:00
*/
class st_select_lex: public st_select_lex_node
{
public:
Name_resolution_context context;
char *db;
2002-05-06 23:04:16 +02:00
Item *where, *having; /* WHERE & HAVING clauses */
Item *prep_where; /* saved WHERE clause for prepared statement processing */
Item *prep_having;/* saved HAVING clause for prepared statement processing */
/* Saved values of the WHERE and HAVING clauses*/
Item::cond_result cond_value, having_value;
2004-07-16 00:15:55 +02:00
/* point on lex in which it was created, used in view subquery detection */
LEX *parent_lex;
enum olap_type olap;
/* FROM clause - points to the beginning of the TABLE_LIST::next_local list. */
SQL_LIST table_list;
SQL_LIST group_list; /* GROUP BY clause. */
List<Item> item_list; /* list of fields & expressions */
List<String> interval_list;
2005-10-12 01:32:14 +02:00
bool is_item_list_lookup;
/*
Usualy it is pointer to ftfunc_list_alloc, but in union used to create fake
select_lex for calling mysql_select under results of union
*/
List<Item_func_match> *ftfunc_list;
List<Item_func_match> ftfunc_list_alloc;
2002-07-01 18:06:24 +02:00
JOIN *join; /* after JOIN::prepare it is pointer to corresponding JOIN */
List<TABLE_LIST> top_join_list; /* join list of the top level */
List<TABLE_LIST> *join_list; /* list for the currently parsed join */
TABLE_LIST *embedding; /* table embedding to the above list */
/*
Beginning of the list of leaves in a FROM clause, where the leaves
inlcude all base tables including view tables. The tables are connected
by TABLE_LIST::next_leaf, so leaf_tables points to the left-most leaf.
*/
TABLE_LIST *leaf_tables;
const char *type; /* type of select for EXPLAIN */
SQL_LIST order_list; /* ORDER clause */
SQL_LIST *gorder_list;
Item *select_limit, *offset_limit; /* LIMIT clause parameters */
// Arrays of pointers to top elements of all_fields list
Item **ref_pointer_array;
2003-08-05 08:56:21 +02:00
/*
number of items in select_list and HAVING clause used to get number
bigger then can be number of entries that will be added to all item
list during split_sum_func
*/
uint select_n_having_items;
uint cond_count; /* number of arguments of and/or/xor in where/having/on */
uint between_count; /* number of between predicates in where/having/on */
uint max_equal_elems; /* maximal number of elements in multiple equalities */
/*
Number of fields used in select list or where clause of current select
and all inner subselects.
*/
uint select_n_where_fields;
enum_parsing_place parsing_place; /* where we are parsing expression */
bool with_sum_func; /* sum function indicator */
ulong table_join_options;
uint in_sum_expr;
uint select_number; /* number of select (used for EXPLAIN) */
int nest_level; /* nesting level of select */
Item_sum *inner_sum_func_list; /* list of sum func in nested selects */
uint with_wild; /* item list contain '*' */
bool braces; /* SELECT ... UNION (SELECT ... ) <- this braces */
/* TRUE when having fix field called in processing of this SELECT */
bool having_fix_field;
Bug #45640: optimizer bug produces wrong results Grouping by a subquery in a query with a distinct aggregate function lead to a wrong result (wrong and unordered grouping values). There are two related problems: 1) The query like this: SELECT (SELECT t1.a) aa, COUNT(DISTINCT b) c FROM t1 GROUP BY aa returned wrong result, because the outer reference "t1.a" in the subquery was substituted with the Item_ref item. The Item_ref item obtains data from the result_field object that refreshes once after the end of each group. This data is not applicable to filesort since filesort() doesn't care about groups (and doesn't update result_field objects with copy_fields() and so on). Also that data is not applicable to group separation algorithm: end_send_group() checks every record with test_if_group_changed() that evaluates Item_ref items, but it refreshes those Item_ref-s only after the end of group, that is a vicious circle and the grouped column values in the output are shifted. Fix: if a) we grouping by a subquery and b) that subquery has outer references to FROM list of the grouping query, then we substitute these outer references with Item_direct_ref like references under aggregate functions: Item_direct_ref obtains data directly from the current record. 2) The query with a non-trivial grouping expression like: SELECT (SELECT t1.a) aa, COUNT(DISTINCT b) c FROM t1 GROUP BY aa+0 also returned wrong result, since JOIN::exec() substitutes references to top-level aliases in SELECT list with Item_copy caching items. Item_copy items have same refreshing policy as Item_ref items, so the whole groping expression with Item_copy inside returns wrong result in filesort() and end_send_group(). Fix: include aliased items into GROUP BY item tree instead of Item_ref references to them.
2010-02-06 20:54:30 +01:00
/* TRUE when GROUP BY fix field called in processing of this SELECT */
bool group_fix_field;
/* List of references to fields referenced from inner selects */
List<Item_outer_ref> inner_refs_list;
/* Number of Item_sum-derived objects in this SELECT */
uint n_sum_items;
/* Number of Item_sum-derived objects in children and descendant SELECTs */
uint n_child_sum_items;
/* explicit LIMIT clause was used */
bool explicit_limit;
/*
there are subquery in HAVING clause => we can't close tables before
query processing end even if we use temporary table
*/
bool subquery_in_having;
/* TRUE <=> this SELECT is correlated w.r.t. some ancestor select */
bool is_correlated;
/*
This variable is required to ensure proper work of subqueries and
stored procedures. Generally, one should use the states of
Query_arena to determine if it's a statement prepare or first
execution of a stored procedure. However, in case when there was an
error during the first execution of a stored procedure, the SP body
is not expelled from the SP cache. Therefore, a deeply nested
subquery might be left unoptimized. So we need this per-subquery
variable to inidicate the optimization/execution state of every
subquery. Prepared statements work OK in that regard, as in
case of an error during prepare the PS is not created.
*/
bool first_execution;
bool first_natural_join_processing;
bool first_cond_optimization;
2004-07-23 08:20:58 +02:00
/* do not wrap view fields with Item_ref */
bool no_wrap_view_item;
/* exclude this select from check of unique_table() */
bool exclude_from_table_unique_test;
/* List of fields that aren't under an aggregate function */
List<Item_field> non_agg_fields;
/* index in the select list of the expression currently being fixed */
int cur_pos_in_select_list;
List<udf_func> udf_list; /* udf function calls stack */
/*
This is a copy of the original JOIN USING list that comes from
the parser. The parser :
1. Sets the natural_join of the second TABLE_LIST in the join
and the st_select_lex::prev_join_using.
2. Makes a parent TABLE_LIST and sets its is_natural_join/
join_using_fields members.
3. Uses the wrapper TABLE_LIST as a table in the upper level.
We cannot assign directly to join_using_fields in the parser because
at stage (1.) the parent TABLE_LIST is not constructed yet and
the assignment will override the JOIN USING fields of the lower level
joins on the right.
*/
List<String> *prev_join_using;
/*
Bitmap used in the ONLY_FULL_GROUP_BY_MODE to prevent mixture of aggregate
functions and non aggregated fields when GROUP BY list is absent.
Bits:
0 - non aggregated fields are used in this select,
defined as NON_AGG_FIELD_USED.
1 - aggregate functions are used in this select,
defined as SUM_FUNC_USED.
*/
uint8 full_group_by_flag;
2002-05-06 23:04:16 +02:00
void init_query();
void init_select();
st_select_lex_unit* master_unit();
st_select_lex_unit* first_inner_unit()
{
return (st_select_lex_unit*) slave;
}
st_select_lex* outer_select();
2002-05-26 21:50:32 +02:00
st_select_lex* next_select() { return (st_select_lex*) next; }
st_select_lex* next_select_in_list()
{
return (st_select_lex*) link_next;
}
st_select_lex_node** next_select_in_list_addr()
{
return &link_next;
}
st_select_lex* return_after_parsing()
{
return master_unit()->return_after_parsing();
}
void mark_as_dependent(st_select_lex *last);
bool set_braces(bool value);
bool inc_in_sum_expr();
uint get_in_sum_expr();
bool add_item_to_list(THD *thd, Item *item);
bool add_group_to_list(THD *thd, Item *item, bool asc);
bool add_ftfunc_to_list(Item_func_match *func);
bool add_order_to_list(THD *thd, Item *item, bool asc);
TABLE_LIST* add_table_to_list(THD *thd, Table_ident *table,
LEX_STRING *alias,
ulong table_options,
thr_lock_type flags= TL_UNLOCK,
2007-07-23 18:09:48 +02:00
List<Index_hint> *hints= 0,
LEX_STRING *option= 0);
TABLE_LIST* get_table_list();
bool init_nested_join(THD *thd);
TABLE_LIST *end_nested_join(THD *thd);
TABLE_LIST *nest_last_join(THD *thd);
void add_joined_table(TABLE_LIST *table);
TABLE_LIST *convert_right_join();
List<Item>* get_item_list();
ulong get_table_join_options();
2002-11-21 21:25:53 +01:00
void set_lock_for_tables(thr_lock_type lock_type);
2002-11-05 00:10:05 +01:00
inline void init_order()
{
order_list.elements= 0;
order_list.first= 0;
WL#3817: Simplify string / memory area types and make things more consistent (first part) The following type conversions was done: - Changed byte to uchar - Changed gptr to uchar* - Change my_string to char * - Change my_size_t to size_t - Change size_s to size_t Removed declaration of byte, gptr, my_string, my_size_t and size_s. Following function parameter changes was done: - All string functions in mysys/strings was changed to use size_t instead of uint for string lengths. - All read()/write() functions changed to use size_t (including vio). - All protocoll functions changed to use size_t instead of uint - Functions that used a pointer to a string length was changed to use size_t* - Changed malloc(), free() and related functions from using gptr to use void * as this requires fewer casts in the code and is more in line with how the standard functions work. - Added extra length argument to dirname_part() to return the length of the created string. - Changed (at least) following functions to take uchar* as argument: - db_dump() - my_net_write() - net_write_command() - net_store_data() - DBUG_DUMP() - decimal2bin() & bin2decimal() - Changed my_compress() and my_uncompress() to use size_t. Changed one argument to my_uncompress() from a pointer to a value as we only return one value (makes function easier to use). - Changed type of 'pack_data' argument to packfrm() to avoid casts. - Changed in readfrm() and writefrom(), ha_discover and handler::discover() the type for argument 'frmdata' to uchar** to avoid casts. - Changed most Field functions to use uchar* instead of char* (reduced a lot of casts). - Changed field->val_xxx(xxx, new_ptr) to take const pointers. Other changes: - Removed a lot of not needed casts - Added a few new cast required by other changes - Added some cast to my_multi_malloc() arguments for safety (as string lengths needs to be uint, not size_t). - Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done explicitely as this conflict was often hided by casting the function to hash_get_key). - Changed some buffers to memory regions to uchar* to avoid casts. - Changed some string lengths from uint to size_t. - Changed field->ptr to be uchar* instead of char*. This allowed us to get rid of a lot of casts. - Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar - Include zlib.h in some files as we needed declaration of crc32() - Changed MY_FILE_ERROR to be (size_t) -1. - Changed many variables to hold the result of my_read() / my_write() to be size_t. This was needed to properly detect errors (which are returned as (size_t) -1). - Removed some very old VMS code - Changed packfrm()/unpackfrm() to not be depending on uint size (portability fix) - Removed windows specific code to restore cursor position as this causes slowdown on windows and we should not mix read() and pread() calls anyway as this is not thread safe. Updated function comment to reflect this. Changed function that depended on original behavior of my_pwrite() to itself restore the cursor position (one such case). - Added some missing checking of return value of malloc(). - Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow. - Changed type of table_def::m_size from my_size_t to ulong to reflect that m_size is the number of elements in the array, not a string/memory length. - Moved THD::max_row_length() to table.cc (as it's not depending on THD). Inlined max_row_length_blob() into this function. - More function comments - Fixed some compiler warnings when compiled without partitions. - Removed setting of LEX_STRING() arguments in declaration (portability fix). - Some trivial indentation/variable name changes. - Some trivial code simplifications: - Replaced some calls to alloc_root + memcpy to use strmake_root()/strdup_root(). - Changed some calls from memdup() to strmake() (Safety fix) - Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
order_list.next= (uchar**) &order_list.first;
2002-11-05 00:10:05 +01:00
}
/*
This method created for reiniting LEX in mysql_admin_table() and can be
used only if you are going remove all SELECT_LEX & units except belonger
to LEX (LEX::unit & LEX::select, for other purposes there are
SELECT_LEX_UNIT::exclude_level & SELECT_LEX_UNIT::exclude_tree
*/
void cut_subtree() { slave= 0; }
bool test_limit();
friend void lex_start(THD *thd);
st_select_lex() : n_sum_items(0), n_child_sum_items(0) {}
void make_empty_select()
2003-03-06 16:02:10 +01:00
{
init_query();
init_select();
}
2003-08-16 12:26:48 +02:00
bool setup_ref_array(THD *thd, uint order_group_num);
void print(THD *thd, String *str, enum_query_type query_type);
static void print_order(String *str,
ORDER *order,
enum_query_type query_type);
void print_limit(THD *thd, String *str, enum_query_type query_type);
void fix_prepare_information(THD *thd, Item **conds, Item **having_conds);
/*
Destroy the used execution plan (JOIN) of this subtree (this
SELECT_LEX and all nested SELECT_LEXes and SELECT_LEX_UNITs).
*/
bool cleanup();
/*
Recursively cleanup the join of this select lex and of all nested
select lexes.
*/
void cleanup_all_joins(bool full);
void set_index_hint_type(enum index_hint_type type, index_clause_map clause);
/*
Add a index hint to the tagged list of hints. The type and clause of the
hint will be the current ones (set by set_index_hint())
*/
bool add_index_hint (THD *thd, char *str, uint length);
/* make a list to hold index hints */
void alloc_index_hints (THD *thd);
/* read and clear the index hints */
2007-07-23 18:09:48 +02:00
List<Index_hint>* pop_index_hints(void)
{
2007-07-23 18:09:48 +02:00
List<Index_hint> *hints= index_hints;
index_hints= NULL;
return hints;
}
void clear_index_hints(void) { index_hints= NULL; }
private:
/* current index hint kind. used in filling up index_hints */
enum index_hint_type current_index_hint_type;
index_clause_map current_index_hint_clause;
/* a list of USE/FORCE/IGNORE INDEX */
2007-07-23 18:09:48 +02:00
List<Index_hint> *index_hints;
2002-05-06 23:04:16 +02:00
};
2002-07-05 07:15:57 +02:00
typedef class st_select_lex SELECT_LEX;
inline bool st_select_lex_unit::is_union ()
{
return first_select()->next_select() &&
first_select()->next_select()->linkage == UNION_TYPE;
}
#define ALTER_ADD_COLUMN (1L << 0)
#define ALTER_DROP_COLUMN (1L << 1)
#define ALTER_CHANGE_COLUMN (1L << 2)
#define ALTER_ADD_INDEX (1L << 3)
#define ALTER_DROP_INDEX (1L << 4)
#define ALTER_RENAME (1L << 5)
#define ALTER_ORDER (1L << 6)
#define ALTER_OPTIONS (1L << 7)
#define ALTER_CHANGE_COLUMN_DEFAULT (1L << 8)
#define ALTER_KEYS_ONOFF (1L << 9)
#define ALTER_CONVERT (1L << 10)
#define ALTER_FORCE (1L << 11)
#define ALTER_RECREATE (1L << 12)
#define ALTER_ADD_PARTITION (1L << 13)
#define ALTER_DROP_PARTITION (1L << 14)
#define ALTER_COALESCE_PARTITION (1L << 15)
#define ALTER_REORGANIZE_PARTITION (1L << 16)
#define ALTER_PARTITION (1L << 17)
#define ALTER_ADMIN_PARTITION (1L << 18)
#define ALTER_TABLE_REORG (1L << 19)
#define ALTER_REBUILD_PARTITION (1L << 20)
#define ALTER_ALL_PARTITION (1L << 21)
#define ALTER_REMOVE_PARTITIONING (1L << 22)
#define ALTER_FOREIGN_KEY (1L << 23)
2007-06-04 12:03:15 +02:00
enum enum_alter_table_change_level
{
ALTER_TABLE_METADATA_ONLY= 0,
ALTER_TABLE_DATA_CHANGED= 1,
ALTER_TABLE_INDEX_CHANGED= 2
};
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
/**
@brief Parsing data for CREATE or ALTER TABLE.
This structure contains a list of columns or indexes to be created,
altered or dropped.
*/
class Alter_info
{
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
public:
2007-06-04 12:03:15 +02:00
List<Alter_drop> drop_list;
List<Alter_column> alter_list;
List<Key> key_list;
2007-06-12 02:08:51 +02:00
List<Create_field> create_list;
2007-06-04 12:03:15 +02:00
uint flags;
enum enum_enable_or_disable keys_onoff;
enum tablespace_op_type tablespace_op;
List<char> partition_names;
uint num_parts;
2007-06-04 12:03:15 +02:00
enum_alter_table_change_level change_level;
2007-06-12 02:08:51 +02:00
Create_field *datetime_field;
2007-08-15 15:43:08 +02:00
bool error_if_not_empty;
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
Alter_info() :
flags(0),
keys_onoff(LEAVE_AS_IS),
tablespace_op(NO_TABLESPACE_OP),
num_parts(0),
2007-06-04 12:03:15 +02:00
change_level(ALTER_TABLE_METADATA_ONLY),
datetime_field(NULL),
error_if_not_empty(FALSE)
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
{}
void reset()
{
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
drop_list.empty();
alter_list.empty();
key_list.empty();
create_list.empty();
flags= 0;
keys_onoff= LEAVE_AS_IS;
tablespace_op= NO_TABLESPACE_OP;
num_parts= 0;
partition_names.empty();
2007-06-04 12:03:15 +02:00
change_level= ALTER_TABLE_METADATA_ONLY;
datetime_field= 0;
error_if_not_empty= FALSE;
}
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
Alter_info(const Alter_info &rhs, MEM_ROOT *mem_root);
private:
Alter_info &operator=(const Alter_info &rhs); // not implemented
Alter_info(const Alter_info &rhs); // not implemented
};
struct st_sp_chistics
{
LEX_STRING comment;
enum enum_sp_suid_behaviour suid;
bool detistic;
enum enum_sp_data_access daccess;
};
struct st_trg_chistics
{
enum trg_action_time_type action_time;
enum trg_event_type event;
};
extern sys_var *trg_new_row_fake_var;
2005-01-16 13:16:23 +01:00
enum xa_option_words {XA_NONE, XA_JOIN, XA_RESUME, XA_ONE_PHASE,
XA_SUSPEND, XA_FOR_MIGRATE};
extern const LEX_STRING null_lex_str;
Backport of revno 2630.28.10, 2630.28.31, 2630.28.26, 2630.33.1, 2630.39.1, 2630.28.29, 2630.34.3, 2630.34.2, 2630.34.1, 2630.29.29, 2630.29.28, 2630.31.1, 2630.28.13, 2630.28.10, 2617.23.14 and some other minor revisions. This patch implements: WL#4264 "Backup: Stabilize Service Interface" -- all the server prerequisites except si_objects.{h,cc} themselves (they can be just copied over, when needed). WL#4435: Support OUT-parameters in prepared statements. (and all issues in the initial patches for these two tasks, that were discovered in pushbuild and during testing). Bug#39519: mysql_stmt_close() should flush all data associated with the statement. After execution of a prepared statement, send OUT parameters of the invoked stored procedure, if any, to the client. When using the binary protocol, send the parameters in an additional result set over the wire. When using the text protocol, assign out parameters to the user variables from the CALL(@var1, @var2, ...) specification. The following refactoring has been made: - Protocol::send_fields() was renamed to Protocol::send_result_set_metadata(); - A new Protocol::send_result_set_row() was introduced to incapsulate common functionality for sending row data. - Signature of Protocol::prepare_for_send() was changed: this operation does not need a list of items, the number of items is fully sufficient. The following backward incompatible changes have been made: - CLIENT_MULTI_RESULTS is now enabled by default in the client; - CLIENT_PS_MULTI_RESUTLS is now enabled by default in the client.
2009-10-21 22:02:06 +02:00
extern const LEX_STRING empty_lex_str;
/*
Committing on behalf or Dmitry Lenev: Fix for bug #46947 "Embedded SELECT without FOR UPDATE is causing a lock", with after-review fixes. SELECT statements with subqueries referencing InnoDB tables were acquiring shared locks on rows in these tables when they were executed in REPEATABLE-READ mode and with statement or mixed mode binary logging turned on. This was a regression which were introduced when fixing bug 39843. The problem was that for tables belonging to subqueries parser set TL_READ_DEFAULT as a lock type. In cases when statement/mixed binary logging at open_tables() time this type of lock was converted to TL_READ_NO_INSERT lock at open_tables() time and caused InnoDB engine to acquire shared locks on reads from these tables. Although in some cases such behavior was correct (e.g. for subqueries in DELETE) in case of SELECT it has caused unnecessary locking. This patch tries to solve this problem by rethinking our approach to how we handle locking for SELECT and subqueries. Now we always set TL_READ_DEFAULT lock type for all cases when we read data. When at open_tables() time this lock is interpreted as TL_READ_NO_INSERT or TL_READ depending on whether this statement as a whole or call to function which uses particular table should be written to the binary log or not (if yes then statement should be properly serialized with concurrent statements and stronger lock should be acquired). Test coverage is added for both InnoDB and MyISAM. This patch introduces an "incompatible" change in locking scheme for subqueries used in SELECT ... FOR UPDATE and SELECT .. IN SHARE MODE. In 4.1 the server would use a snapshot InnoDB read for subqueries in SELECT FOR UPDATE and SELECT .. IN SHARE MODE statements, regardless of whether the binary log is on or off. If the user required a different type of read (i.e. locking read), he/she could request so explicitly by providing FOR UPDATE/IN SHARE MODE clause for each individual subquery. On of the patches for 5.0 broke this behaviour (which was not documented or tested), and started to use locking reads fora all subqueries in SELECT ... FOR UPDATE/IN SHARE MODE. This patch restored 4.1 behaviour.
2010-04-28 12:04:11 +02:00
Class representing list of all tables used by statement and other
information which is necessary for opening and locking its tables,
like SQL command for this statement.
Also contains information about stored functions used by statement
since during its execution we may have to add all tables used by its
stored functions/triggers to this list in order to pre-open and lock
them.
Also used by LEX::reset_n_backup/restore_backup_query_tables_list()
methods to save and restore this information.
*/
class Query_tables_list
{
public:
Committing on behalf or Dmitry Lenev: Fix for bug #46947 "Embedded SELECT without FOR UPDATE is causing a lock", with after-review fixes. SELECT statements with subqueries referencing InnoDB tables were acquiring shared locks on rows in these tables when they were executed in REPEATABLE-READ mode and with statement or mixed mode binary logging turned on. This was a regression which were introduced when fixing bug 39843. The problem was that for tables belonging to subqueries parser set TL_READ_DEFAULT as a lock type. In cases when statement/mixed binary logging at open_tables() time this type of lock was converted to TL_READ_NO_INSERT lock at open_tables() time and caused InnoDB engine to acquire shared locks on reads from these tables. Although in some cases such behavior was correct (e.g. for subqueries in DELETE) in case of SELECT it has caused unnecessary locking. This patch tries to solve this problem by rethinking our approach to how we handle locking for SELECT and subqueries. Now we always set TL_READ_DEFAULT lock type for all cases when we read data. When at open_tables() time this lock is interpreted as TL_READ_NO_INSERT or TL_READ depending on whether this statement as a whole or call to function which uses particular table should be written to the binary log or not (if yes then statement should be properly serialized with concurrent statements and stronger lock should be acquired). Test coverage is added for both InnoDB and MyISAM. This patch introduces an "incompatible" change in locking scheme for subqueries used in SELECT ... FOR UPDATE and SELECT .. IN SHARE MODE. In 4.1 the server would use a snapshot InnoDB read for subqueries in SELECT FOR UPDATE and SELECT .. IN SHARE MODE statements, regardless of whether the binary log is on or off. If the user required a different type of read (i.e. locking read), he/she could request so explicitly by providing FOR UPDATE/IN SHARE MODE clause for each individual subquery. On of the patches for 5.0 broke this behaviour (which was not documented or tested), and started to use locking reads fora all subqueries in SELECT ... FOR UPDATE/IN SHARE MODE. This patch restored 4.1 behaviour.
2010-04-28 12:04:11 +02:00
/**
SQL command for this statement. Part of this class since the
process of opening and locking tables for the statement needs
this information to determine correct type of lock for some of
the tables.
*/
enum_sql_command sql_command;
/* Global list of all tables used by this statement */
TABLE_LIST *query_tables;
/* Pointer to next_global member of last element in the previous list. */
TABLE_LIST **query_tables_last;
/*
If non-0 then indicates that query requires prelocking and points to
next_global member of last own element in query table list (i.e. last
table which was not added to it as part of preparation to prelocking).
0 - indicates that this query does not need prelocking.
*/
TABLE_LIST **query_tables_own_last;
/*
Set of stored routines called by statement.
(Note that we use lazy-initialization for this hash).
*/
enum { START_SROUTINES_HASH_SIZE= 16 };
HASH sroutines;
/*
List linking elements of 'sroutines' set. Allows you to add new elements
to this set as you iterate through the list of existing elements.
'sroutines_list_own_last' is pointer to ::next member of last element of
this list which represents routine which is explicitly used by query.
'sroutines_list_own_elements' number of explicitly used routines.
We use these two members for restoring of 'sroutines_list' to the state
in which it was right after query parsing.
*/
SQL_LIST sroutines_list;
WL#3817: Simplify string / memory area types and make things more consistent (first part) The following type conversions was done: - Changed byte to uchar - Changed gptr to uchar* - Change my_string to char * - Change my_size_t to size_t - Change size_s to size_t Removed declaration of byte, gptr, my_string, my_size_t and size_s. Following function parameter changes was done: - All string functions in mysys/strings was changed to use size_t instead of uint for string lengths. - All read()/write() functions changed to use size_t (including vio). - All protocoll functions changed to use size_t instead of uint - Functions that used a pointer to a string length was changed to use size_t* - Changed malloc(), free() and related functions from using gptr to use void * as this requires fewer casts in the code and is more in line with how the standard functions work. - Added extra length argument to dirname_part() to return the length of the created string. - Changed (at least) following functions to take uchar* as argument: - db_dump() - my_net_write() - net_write_command() - net_store_data() - DBUG_DUMP() - decimal2bin() & bin2decimal() - Changed my_compress() and my_uncompress() to use size_t. Changed one argument to my_uncompress() from a pointer to a value as we only return one value (makes function easier to use). - Changed type of 'pack_data' argument to packfrm() to avoid casts. - Changed in readfrm() and writefrom(), ha_discover and handler::discover() the type for argument 'frmdata' to uchar** to avoid casts. - Changed most Field functions to use uchar* instead of char* (reduced a lot of casts). - Changed field->val_xxx(xxx, new_ptr) to take const pointers. Other changes: - Removed a lot of not needed casts - Added a few new cast required by other changes - Added some cast to my_multi_malloc() arguments for safety (as string lengths needs to be uint, not size_t). - Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done explicitely as this conflict was often hided by casting the function to hash_get_key). - Changed some buffers to memory regions to uchar* to avoid casts. - Changed some string lengths from uint to size_t. - Changed field->ptr to be uchar* instead of char*. This allowed us to get rid of a lot of casts. - Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar - Include zlib.h in some files as we needed declaration of crc32() - Changed MY_FILE_ERROR to be (size_t) -1. - Changed many variables to hold the result of my_read() / my_write() to be size_t. This was needed to properly detect errors (which are returned as (size_t) -1). - Removed some very old VMS code - Changed packfrm()/unpackfrm() to not be depending on uint size (portability fix) - Removed windows specific code to restore cursor position as this causes slowdown on windows and we should not mix read() and pread() calls anyway as this is not thread safe. Updated function comment to reflect this. Changed function that depended on original behavior of my_pwrite() to itself restore the cursor position (one such case). - Added some missing checking of return value of malloc(). - Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow. - Changed type of table_def::m_size from my_size_t to ulong to reflect that m_size is the number of elements in the array, not a string/memory length. - Moved THD::max_row_length() to table.cc (as it's not depending on THD). Inlined max_row_length_blob() into this function. - More function comments - Fixed some compiler warnings when compiled without partitions. - Removed setting of LEX_STRING() arguments in declaration (portability fix). - Some trivial indentation/variable name changes. - Some trivial code simplifications: - Replaced some calls to alloc_root + memcpy to use strmake_root()/strdup_root(). - Changed some calls from memdup() to strmake() (Safety fix) - Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
uchar **sroutines_list_own_last;
uint sroutines_list_own_elements;
/*
These constructor and destructor serve for creation/destruction
of Query_tables_list instances which are used as backup storage.
*/
Query_tables_list() {}
~Query_tables_list() {}
/* Initializes (or resets) Query_tables_list object for "real" use. */
void reset_query_tables_list(bool init);
void destroy_query_tables_list();
void set_query_tables_list(Query_tables_list *state)
{
*this= *state;
}
A fix and a test case for Bug#19022 "Memory bug when switching db during trigger execution" Bug#17199 "Problem when view calls function from another database." Bug#18444 "Fully qualified stored function names don't work correctly in SELECT statements" Documentation note: this patch introduces a change in behaviour of prepared statements. This patch adds a few new invariants with regard to how THD::db should be used. These invariants should be preserved in future: - one should never refer to THD::db by pointer and always make a deep copy (strmake, strdup) - one should never compare two databases by pointer, but use strncmp or my_strncasecmp - TABLE_LIST object table->db should be always initialized in the parser or by creator of the object. For prepared statements it means that if the current database is changed after a statement is prepared, the database that was current at prepare remains active. This also means that you can not prepare a statement that implicitly refers to the current database if the latter is not set. This is not documented, and therefore needs documentation. This is NOT a change in behavior for almost all SQL statements except: - ALTER TABLE t1 RENAME t2 - OPTIMIZE TABLE t1 - ANALYZE TABLE t1 - TRUNCATE TABLE t1 -- until this patch t1 or t2 could be evaluated at the first execution of prepared statement. CURRENT_DATABASE() still works OK and is evaluated at every execution of prepared statement. Note, that in stored routines this is not an issue as the default database is the database of the stored procedure and "use" statement is prohibited in stored routines. This patch makes obsolete the use of check_db_used (it was never used in the old code too) and all other places that check for table->db and assign it from THD::db if it's NULL, except the parser. How this patch was created: THD::{db,db_length} were replaced with a LEX_STRING, THD::db. All the places that refer to THD::{db,db_length} were manually checked and: - if the place uses thd->db by pointer, it was fixed to make a deep copy - if a place compared two db pointers, it was fixed to compare them by value (via strcmp/my_strcasecmp, whatever was approproate) Then this intermediate patch was used to write a smaller patch that does the same thing but without a rename. TODO in 5.1: - remove check_db_used - deploy THD::set_db in mysql_change_db See also comments to individual files.
2006-06-26 22:47:52 +02:00
/*
Direct addition to the list of query tables.
If you are using this function, you must ensure that the table
object, in particular table->db member, is initialized.
*/
void add_to_query_tables(TABLE_LIST *table)
{
*(table->prev_global= query_tables_last)= table;
query_tables_last= &table->next_global;
}
bool requires_prelocking()
{
return test(query_tables_own_last);
}
void mark_as_requiring_prelocking(TABLE_LIST **tables_own_last)
{
query_tables_own_last= tables_own_last;
}
/* Return pointer to first not-own table in query-tables or 0 */
TABLE_LIST* first_not_own_table()
{
return ( query_tables_own_last ? *query_tables_own_last : 0);
}
void chop_off_not_own_tables()
{
if (query_tables_own_last)
{
*query_tables_own_last= 0;
query_tables_last= query_tables_own_last;
query_tables_own_last= 0;
}
}
/** Return a pointer to the last element in query table list. */
TABLE_LIST *last_table()
{
/* Don't use offsetof() macro in order to avoid warnings. */
return query_tables ?
(TABLE_LIST*) ((char*) query_tables_last -
((char*) &(query_tables->next_global) -
(char*) query_tables)) :
0;
}
/**
Enumeration listing of all types of unsafe statement.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
@note The order of elements of this enumeration type must
correspond to the order of the elements of the @c explanations
array defined in the body of @c THD::issue_unsafe_warnings.
*/
enum enum_binlog_stmt_unsafe {
/**
SELECT..LIMIT is unsafe because the set of rows returned cannot
be predicted.
*/
BINLOG_STMT_UNSAFE_LIMIT= 0,
/**
INSERT DELAYED is unsafe because the time when rows are inserted
cannot be predicted.
*/
BINLOG_STMT_UNSAFE_INSERT_DELAYED,
/**
Access to log tables is unsafe because slave and master probably
log different things.
*/
BINLOG_STMT_UNSAFE_SYSTEM_TABLE,
/**
Inserting into an autoincrement column in a stored routine is unsafe.
Even with just one autoincrement column, if the routine is invoked more than
once slave is not guaranteed to execute the statement graph same way as
the master.
And since it's impossible to estimate how many times a routine can be invoked at
the query pre-execution phase (see lock_tables), the statement is marked
pessimistically unsafe.
*/
BINLOG_STMT_UNSAFE_AUTOINC_COLUMNS,
/**
Using a UDF (user-defined function) is unsafe.
*/
BINLOG_STMT_UNSAFE_UDF,
/**
Using most system variables is unsafe, because slave may run
with different options than master.
*/
BINLOG_STMT_UNSAFE_SYSTEM_VARIABLE,
/**
Using some functions is unsafe (e.g., UUID).
*/
BINLOG_STMT_UNSAFE_SYSTEM_FUNCTION,
/**
Mixing transactional and non-transactional statements are unsafe if
non-transactional reads or writes are occur after transactional
reads or writes inside a transaction.
*/
BINLOG_STMT_UNSAFE_NONTRANS_AFTER_TRANS,
/**
Mixing self-logging and non-self-logging engines in a statement
is unsafe.
*/
BINLOG_STMT_UNSAFE_MULTIPLE_ENGINES_AND_SELF_LOGGING_ENGINE,
BUG#51291 Unfortunate effect around variable binlog_direct_non_transactional_updates BUG#46364 introduced the flag binlog_direct_non_transactional_updates which would make N-changes to be written to the binary log upon committing the statement when "ON". On the other hand, when "OFF" the option was supposed to mimic the behavior in 5.1. However, the implementation was not mimicking the behavior correctly and the following bugs popped up: Case #1: N-changes executed within a transaction would go into the S-cache. When later in the same transaction a T-change occurs, N-changes following it were written to the T-cache instead of the S-cache. In some cases, this raises problems. For example, a Table_map_log_event being written initially into the S-cache, together with the initial N-changes, would be absent from the T-cache. This would log N-changes orphaned from a Table_map_log_event (thence discarded at the slave). (MIXED and ROW) Case #2: When rolling back a transaction, the N-changes that might be in the T-cache were disregarded and truncated along with the T-changes. (MIXED and ROW) Case #3: When a MIXED statement (TN) is ahead of any other T-changes in the transaction and it fails, it is kept in the T-cache until the transaction ends. This is not the case in 5.1 or Betony (5.5.2). In these, the failed TN statement would be written to the binlog at the same instant it had failed and not deferred until transaction end. (SBR) To fix these problems, we have decided to do what follows: For Case #1 and #2, we circumvent them: 1. by not letting binlog_direct_non_transactional_updates affect MIXED and RBR. These modes will keep the behavior provided by WL#2687. Although this will make Celosia to behave differently from 5.1, an execution will be always safe under such modes in the sense that slaves will never go out sync. In 5.1, using either MIXED or ROW while mixing N-statements and T-statements was not safe. For Case #3, we don't actually fix it. We: 1. keep it and make all MIXED statements whether they end up failing or not or whether they are up front in the transaction or after some transactional change to always be stored in the T-cache. This means that it is written to the binary log on transaction commit/rollback only. 2. We make the warning message even more specific about the MIXED statement and SBR.
2010-03-31 15:22:47 +02:00
/**
Statements that read from both transactional and non-transactional
tables and write to any of them are unsafe.
*/
BINLOG_STMT_UNSAFE_MIXED_STATEMENT,
/* The last element of this enumeration type. */
BINLOG_STMT_UNSAFE_COUNT
};
/**
This has all flags from 0 (inclusive) to BINLOG_STMT_FLAG_COUNT
(exclusive) set.
*/
static const int BINLOG_STMT_UNSAFE_ALL_FLAGS=
((1 << BINLOG_STMT_UNSAFE_COUNT) - 1);
/**
Maps elements of enum_binlog_stmt_unsafe to error codes.
*/
static const int binlog_stmt_unsafe_errcode[BINLOG_STMT_UNSAFE_COUNT];
/**
Determine if this statement is marked as unsafe.
@retval 0 if the statement is not marked as unsafe.
@retval nonzero if the statement is marked as unsafe.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
*/
inline bool is_stmt_unsafe() const {
return get_stmt_unsafe_flags() != 0;
}
/**
Flag the current (top-level) statement as unsafe.
The flag will be reset after the statement has finished.
@param unsafe_type The type of unsafety: one of the @c
BINLOG_STMT_FLAG_UNSAFE_* flags in @c enum_binlog_stmt_flag.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
*/
inline void set_stmt_unsafe(enum_binlog_stmt_unsafe unsafe_type) {
DBUG_ENTER("set_stmt_unsafe");
DBUG_ASSERT(unsafe_type >= 0 && unsafe_type < BINLOG_STMT_UNSAFE_COUNT);
binlog_stmt_flags|= (1U << unsafe_type);
DBUG_VOID_RETURN;
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
}
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
/**
Set the bits of binlog_stmt_flags determining the type of
unsafeness of the current statement. No existing bits will be
cleared, but new bits may be set.
@param flags A binary combination of zero or more bits, (1<<flag)
where flag is a member of enum_binlog_stmt_unsafe.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
*/
inline void set_stmt_unsafe_flags(uint32 flags) {
DBUG_ENTER("set_stmt_unsafe_flags");
DBUG_ASSERT((flags & ~BINLOG_STMT_UNSAFE_ALL_FLAGS) == 0);
binlog_stmt_flags|= flags;
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
DBUG_VOID_RETURN;
}
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
/**
Return a binary combination of all unsafe warnings for the
statement. If the statement has been marked as unsafe by the
'flag' member of enum_binlog_stmt_unsafe, then the return value
from this function has bit (1<<flag) set to 1.
*/
inline uint32 get_stmt_unsafe_flags() const {
DBUG_ENTER("get_stmt_unsafe_flags");
DBUG_RETURN(binlog_stmt_flags & BINLOG_STMT_UNSAFE_ALL_FLAGS);
}
/**
Mark the current statement as safe; i.e., clear all bits in
binlog_stmt_flags that correspond to elements of
enum_binlog_stmt_unsafe.
*/
inline void clear_stmt_unsafe() {
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
DBUG_ENTER("clear_stmt_unsafe");
binlog_stmt_flags&= ~BINLOG_STMT_UNSAFE_ALL_FLAGS;
DBUG_VOID_RETURN;
}
/**
Determine if this statement is a row injection.
@retval 0 if the statement is not a row injection
@retval nonzero if the statement is a row injection
*/
inline bool is_stmt_row_injection() const {
return binlog_stmt_flags &
(1U << (BINLOG_STMT_UNSAFE_COUNT + BINLOG_STMT_TYPE_ROW_INJECTION));
}
/**
Flag the statement as a row injection. A row injection is either
a BINLOG statement, or a row event in the relay log executed by
the slave SQL thread.
*/
inline void set_stmt_row_injection() {
DBUG_ENTER("set_stmt_row_injection");
binlog_stmt_flags|=
(1U << (BINLOG_STMT_UNSAFE_COUNT + BINLOG_STMT_TYPE_ROW_INJECTION));
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
DBUG_VOID_RETURN;
}
A fix and a test case for Bug#21483 "Server abort or deadlock on INSERT DELAYED with another implicit insert" Also fixes and adds test cases for bugs: 20497 "Trigger with INSERT DELAYED causes Error 1165" 21714 "Wrong NEW.value and server abort on INSERT DELAYED to a table with a trigger". Post-review fixes. Problem: In MySQL INSERT DELAYED is a way to pipe all inserts into a given table through a dedicated thread. This is necessary for simplistic storage engines like MyISAM, which do not have internal concurrency control or threading and thus can not achieve efficient INSERT throughput without support from SQL layer. DELAYED INSERT works as follows: For every distinct table, which can accept DELAYED inserts and has pending data to insert, a dedicated thread is created to write data to disk. All user connection threads that attempt to delayed-insert into this table interact with the dedicated thread in producer/consumer fashion: all records to-be inserted are pushed into a queue of the dedicated thread, which fetches the records and writes them. In this design, client connection threads never open or lock the delayed insert table. This functionality was introduced in version 3.23 and does not take into account existence of triggers, views, or pre-locking. E.g. if INSERT DELAYED is called from a stored function, which, in turn, is called from another stored function that uses the delayed table, a deadlock can occur, because delayed locking by-passes pre-locking. Besides: * the delayed thread works directly with the subject table through the storage engine API and does not invoke triggers * even if it was patched to invoke triggers, if triggers, in turn, used other tables, the delayed thread would have to open and lock involved tables (use pre-locking). * even if it was patched to use pre-locking, without deadlock detection the delayed thread could easily lock out user connection threads in case when the same table is used both in a trigger and on the right side of the insert query: the delayed thread would not release locks until all inserts are complete, and user connection can not complete inserts without having locks on the tables used on the right side of the query. Solution: These considerations suggest two general alternatives for the future of INSERT DELAYED: * it is considered a full-fledged alternative to normal INSERT * it is regarded as an optimisation that is only relevant for simplistic engines. Since we missed our chance to provide complete support of new features when 5.0 was in development, the first alternative currently renders infeasible. However, even the second alternative, which is to detect new features and convert DELAYED insert into a normal insert, is not easy to implement. The catch-22 is that we don't know if the subject table has triggers or is a view before we open it, and we only open it in the delayed thread. We don't know if the query involves pre-locking until we have opened all tables, and we always first create the delayed thread, and only then open the remaining tables. This patch detects the problematic scenarios and converts DELAYED INSERT to a normal INSERT using the following approach: * if the statement is executed under pre-locking (e.g. from within a stored function or trigger) or the right side may require pre-locking, we detect the situation before creating a delayed insert thread and convert the statement to a conventional INSERT. * if the subject table is a view or has triggers, we shutdown the delayed thread and convert the statement to a conventional INSERT.
2007-05-16 07:51:05 +02:00
/**
true if the parsed tree contains references to stored procedures
or functions, false otherwise
*/
bool uses_stored_routines() const
{ return sroutines_list.elements != 0; }
private:
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
/**
Enumeration listing special types of statements.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
Currently, the only possible type is ROW_INJECTION.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
*/
enum enum_binlog_stmt_type {
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
/**
The statement is a row injection (i.e., either a BINLOG
statement or a row event executed by the slave SQL thread).
*/
BINLOG_STMT_TYPE_ROW_INJECTION = 0,
/** The last element of this enumeration type. */
BINLOG_STMT_TYPE_COUNT
};
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
/**
Bit field indicating the type of statement.
There are two groups of bits:
- The low BINLOG_STMT_UNSAFE_COUNT bits indicate the types of
unsafeness that the current statement has.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
- The next BINLOG_STMT_TYPE_COUNT bits indicate if the statement
is of some special type.
BUG#39934: Slave stops for engine that only support row-based logging General overview: The logic for switching to row format when binlog_format=MIXED had numerous flaws. The underlying problem was the lack of a consistent architecture. General purpose of this changeset: This changeset introduces an architecture for switching to row format when binlog_format=MIXED. It enforces the architecture where it has to. It leaves some bugs to be fixed later. It adds extensive tests to verify that unsafe statements work as expected and that appropriate errors are produced by problems with the selection of binlog format. It was not practical to split this into smaller pieces of work. Problem 1: To determine the logging mode, the code has to take several parameters into account (namely: (1) the value of binlog_format; (2) the capabilities of the engines; (3) the type of the current statement: normal, unsafe, or row injection). These parameters may conflict in several ways, namely: - binlog_format=STATEMENT for a row injection - binlog_format=STATEMENT for an unsafe statement - binlog_format=STATEMENT for an engine only supporting row logging - binlog_format=ROW for an engine only supporting statement logging - statement is unsafe and engine does not support row logging - row injection in a table that does not support statement logging - statement modifies one table that does not support row logging and one that does not support statement logging Several of these conflicts were not detected, or were detected with an inappropriate error message. The problem of BUG#39934 was that no appropriate error message was written for the case when an engine only supporting row logging executed a row injection with binlog_format=ROW. However, all above cases must be handled. Fix 1: Introduce new error codes (sql/share/errmsg.txt). Ensure that all conditions are detected and handled in decide_logging_format() Problem 2: The binlog format shall be determined once per statement, in decide_logging_format(). It shall not be changed before or after that. Before decide_logging_format() is called, all information necessary to determine the logging format must be available. This principle ensures that all unsafe statements are handled in a consistent way. However, this principle is not followed: thd->set_current_stmt_binlog_row_based_if_mixed() is called in several places, including from code executing UPDATE..LIMIT, INSERT..SELECT..LIMIT, DELETE..LIMIT, INSERT DELAYED, and SET @@binlog_format. After Problem 1 was fixed, that caused inconsistencies where these unsafe statements would not print the appropriate warnings or errors for some of the conflicts. Fix 2: Remove calls to THD::set_current_stmt_binlog_row_based_if_mixed() from code executed after decide_logging_format(). Compensate by calling the set_current_stmt_unsafe() at parse time. This way, all unsafe statements are detected by decide_logging_format(). Problem 3: INSERT DELAYED is not unsafe: it is logged in statement format even if binlog_format=MIXED, and no warning is printed even if binlog_format=STATEMENT. This is BUG#45825. Fix 3: Made INSERT DELAYED set itself to unsafe at parse time. This allows decide_logging_format() to detect that a warning should be printed or the binlog_format changed. Problem 4: LIMIT clause were not marked as unsafe when executed inside stored functions/triggers/views/prepared statements. This is BUG#45785. Fix 4: Make statements containing the LIMIT clause marked as unsafe at parse time, instead of at execution time. This allows propagating unsafe-ness to the view.
2009-07-14 21:31:19 +02:00
This must be a member of LEX, not of THD: each stored procedure
needs to remember its unsafeness state between calls and each
stored procedure has its own LEX object (but no own THD object).
*/
uint32 binlog_stmt_flags;
};
/*
st_parsing_options contains the flags for constructions that are
allowed in the current statement.
*/
struct st_parsing_options
{
bool allows_variable;
bool allows_select_into;
bool allows_select_procedure;
bool allows_derived;
st_parsing_options() { reset(); }
void reset();
};
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
The state of the lexical parser, when parsing comments.
*/
enum enum_comment_state
{
/**
Not parsing comments.
*/
NO_COMMENT,
/**
Parsing comments that need to be preserved.
Typically, these are user comments '/' '*' ... '*' '/'.
*/
PRESERVE_COMMENT,
/**
Parsing comments that need to be discarded.
Typically, these are special comments '/' '*' '!' ... '*' '/',
or '/' '*' '!' 'M' 'M' 'm' 'm' 'm' ... '*' '/', where the comment
markers should not be expanded.
*/
DISCARD_COMMENT
};
/**
2007-08-15 15:43:08 +02:00
@brief This class represents the character input stream consumed during
lexical analysis.
2007-08-15 15:43:08 +02:00
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
In addition to consuming the input stream, this class performs some
comment pre processing, by filtering out out of bound special text
from the query input stream.
Two buffers, with pointers inside each buffers, are maintained in
parallel. The 'raw' buffer is the original query text, which may
contain out-of-bound comments. The 'cpp' (for comments pre processor)
is the pre-processed buffer that contains only the query text that
should be seen once out-of-bound data is removed.
*/
2007-08-15 15:43:08 +02:00
class Lex_input_stream
{
public:
Lex_input_stream(THD *thd, const char* buff, unsigned int length);
~Lex_input_stream();
void reset(const char *buff, unsigned int length);
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Set the echo mode.
2007-08-15 15:43:08 +02:00
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
When echo is true, characters parsed from the raw input stream are
preserved. When false, characters parsed are silently ignored.
@param echo the echo mode.
*/
void set_echo(bool echo)
{
m_echo= echo;
}
void save_in_comment_state()
{
m_echo_saved= m_echo;
in_comment_saved= in_comment;
}
void restore_in_comment_state()
{
m_echo= m_echo_saved;
in_comment= in_comment_saved;
}
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Skip binary from the input stream.
@param n number of bytes to accept.
*/
void skip_binary(int n)
{
if (m_echo)
{
memcpy(m_cpp_ptr, m_ptr, n);
m_cpp_ptr += n;
}
m_ptr += n;
}
/**
Get a character, and advance in the stream.
@return the next character to parse.
*/
2009-08-10 20:46:20 +02:00
unsigned char yyGet()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
char c= *m_ptr++;
if (m_echo)
*m_cpp_ptr++ = c;
return c;
}
/**
Get the last character accepted.
@return the last character accepted.
*/
2009-08-10 20:46:20 +02:00
unsigned char yyGetLast()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_ptr[-1];
}
/**
Look at the next character to parse, but do not accept it.
*/
2009-08-10 20:46:20 +02:00
unsigned char yyPeek()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_ptr[0];
}
/**
Look ahead at some character to parse.
@param n offset of the character to look up
*/
2009-08-10 20:46:20 +02:00
unsigned char yyPeekn(int n)
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_ptr[n];
}
/**
Cancel the effect of the last yyGet() or yySkip().
Note that the echo mode should not change between calls to yyGet / yySkip
and yyUnget. The caller is responsible for ensuring that.
*/
void yyUnget()
{
m_ptr--;
if (m_echo)
m_cpp_ptr--;
}
/**
Accept a character, by advancing the input stream.
*/
void yySkip()
{
if (m_echo)
*m_cpp_ptr++ = *m_ptr++;
else
m_ptr++;
}
/**
Accept multiple characters at once.
@param n the number of characters to accept.
*/
void yySkipn(int n)
{
if (m_echo)
{
memcpy(m_cpp_ptr, m_ptr, n);
m_cpp_ptr += n;
}
m_ptr += n;
}
/**
End of file indicator for the query text to parse.
@return true if there are no more characters to parse
*/
bool eof()
{
return (m_ptr >= m_end_of_query);
}
/**
End of file indicator for the query text to parse.
@param n number of characters expected
@return true if there are less than n characters to parse
*/
bool eof(int n)
{
return ((m_ptr + n) >= m_end_of_query);
}
/** Get the raw query buffer. */
2007-07-20 17:52:25 +02:00
const char *get_buf()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_buf;
}
/** Get the pre-processed query buffer. */
2007-07-20 17:52:25 +02:00
const char *get_cpp_buf()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_cpp_buf;
}
/** Get the end of the raw query buffer. */
2007-07-20 17:52:25 +02:00
const char *get_end_of_query()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_end_of_query;
}
/** Mark the stream position as the start of a new token. */
void start_token()
{
m_tok_start_prev= m_tok_start;
m_tok_start= m_ptr;
m_tok_end= m_ptr;
m_cpp_tok_start_prev= m_cpp_tok_start;
m_cpp_tok_start= m_cpp_ptr;
m_cpp_tok_end= m_cpp_ptr;
}
/**
Adjust the starting position of the current token.
This is used to compensate for starting whitespace.
*/
void restart_token()
{
m_tok_start= m_ptr;
m_cpp_tok_start= m_cpp_ptr;
}
/** Get the token start position, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *get_tok_start()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_tok_start;
}
/** Get the token start position, in the pre-processed buffer. */
2007-07-20 17:52:25 +02:00
const char *get_cpp_tok_start()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_cpp_tok_start;
}
/** Get the token end position, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *get_tok_end()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_tok_end;
}
/** Get the token end position, in the pre-processed buffer. */
2007-07-20 17:52:25 +02:00
const char *get_cpp_tok_end()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_cpp_tok_end;
}
/** Get the previous token start position, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *get_tok_start_prev()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_tok_start_prev;
}
/** Get the current stream pointer, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *get_ptr()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_ptr;
}
/** Get the current stream pointer, in the pre-processed buffer. */
2007-07-20 17:52:25 +02:00
const char *get_cpp_ptr()
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
{
return m_cpp_ptr;
}
/** Get the length of the current token, in the raw buffer. */
uint yyLength()
{
/*
The assumption is that the lexical analyser is always 1 character ahead,
which the -1 account for.
*/
DBUG_ASSERT(m_ptr > m_tok_start);
return (uint) ((m_ptr - m_tok_start) - 1);
}
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
/** Get the utf8-body string. */
const char *get_body_utf8_str()
{
return m_body_utf8;
}
/** Get the utf8-body length. */
uint get_body_utf8_length()
{
return (uint) (m_body_utf8_ptr - m_body_utf8);
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
}
void body_utf8_start(THD *thd, const char *begin_ptr);
void body_utf8_append(const char *ptr);
void body_utf8_append(const char *ptr, const char *end_ptr);
void body_utf8_append_literal(THD *thd,
const LEX_STRING *txt,
CHARSET_INFO *txt_cs,
const char *end_ptr);
/** Current thread. */
THD *m_thd;
/** Current line number. */
uint yylineno;
/** Length of the last token parsed. */
uint yytoklen;
/** Interface with bison, value of the last token parsed. */
LEX_YYSTYPE yylval;
/**
LALR(2) resolution, look ahead token.
Value of the next token to return, if any,
or -1, if no token was parsed in advance.
Note: 0 is a legal token, and represents YYEOF.
*/
int lookahead_token;
/** LALR(2) resolution, value of the look ahead token.*/
LEX_YYSTYPE lookahead_yylval;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
private:
/** Pointer to the current position in the raw input stream. */
2007-07-20 17:52:25 +02:00
const char *m_ptr;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Starting position of the last token parsed, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *m_tok_start;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Ending position of the previous token parsed, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *m_tok_end;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** End of the query text in the input stream, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *m_end_of_query;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Starting position of the previous token parsed, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *m_tok_start_prev;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Begining of the query text in the input stream, in the raw buffer. */
2007-07-20 17:52:25 +02:00
const char *m_buf;
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
/** Length of the raw buffer. */
uint m_buf_length;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Echo the parsed stream to the pre-processed buffer. */
bool m_echo;
bool m_echo_saved;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Pre-processed buffer. */
2007-07-20 17:52:25 +02:00
char *m_cpp_buf;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** Pointer to the current position in the pre-processed input stream. */
2007-07-20 17:52:25 +02:00
char *m_cpp_ptr;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Starting position of the last token parsed,
in the pre-processed buffer.
*/
2007-07-20 17:52:25 +02:00
const char *m_cpp_tok_start;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Starting position of the previous token parsed,
in the pre-procedded buffer.
*/
2007-07-20 17:52:25 +02:00
const char *m_cpp_tok_start_prev;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Ending position of the previous token parsed,
in the pre-processed buffer.
*/
2007-07-20 17:52:25 +02:00
const char *m_cpp_tok_end;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
/** UTF8-body buffer created during parsing. */
char *m_body_utf8;
/** Pointer to the current position in the UTF8-body buffer. */
char *m_body_utf8_ptr;
/**
Position in the pre-processed buffer. The query from m_cpp_buf to
m_cpp_utf_processed_ptr is converted to UTF8-body.
*/
const char *m_cpp_utf8_processed_ptr;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
public:
/** Current state of the lexical analyser. */
enum my_lex_states next_state;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
Position of ';' in the stream, to delimit multiple queries.
This delimiter is in the raw buffer.
*/
2007-07-20 17:52:25 +02:00
const char *found_semicolon;
/** Token character bitmaps, to detect 7bit strings. */
uchar tok_bitmap;
2007-04-25 19:38:11 +02:00
/** SQL_MODE = IGNORE_SPACE. */
bool ignore_space;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/**
TRUE if we're parsing a prepared statement: in this mode
Backport of revno 2630.28.10, 2630.28.31, 2630.28.26, 2630.33.1, 2630.39.1, 2630.28.29, 2630.34.3, 2630.34.2, 2630.34.1, 2630.29.29, 2630.29.28, 2630.31.1, 2630.28.13, 2630.28.10, 2617.23.14 and some other minor revisions. This patch implements: WL#4264 "Backup: Stabilize Service Interface" -- all the server prerequisites except si_objects.{h,cc} themselves (they can be just copied over, when needed). WL#4435: Support OUT-parameters in prepared statements. (and all issues in the initial patches for these two tasks, that were discovered in pushbuild and during testing). Bug#39519: mysql_stmt_close() should flush all data associated with the statement. After execution of a prepared statement, send OUT parameters of the invoked stored procedure, if any, to the client. When using the binary protocol, send the parameters in an additional result set over the wire. When using the text protocol, assign out parameters to the user variables from the CALL(@var1, @var2, ...) specification. The following refactoring has been made: - Protocol::send_fields() was renamed to Protocol::send_result_set_metadata(); - A new Protocol::send_result_set_row() was introduced to incapsulate common functionality for sending row data. - Signature of Protocol::prepare_for_send() was changed: this operation does not need a list of items, the number of items is fully sufficient. The following backward incompatible changes have been made: - CLIENT_MULTI_RESULTS is now enabled by default in the client; - CLIENT_PS_MULTI_RESUTLS is now enabled by default in the client.
2009-10-21 22:02:06 +02:00
we should allow placeholders.
*/
bool stmt_prepare_mode;
Backport of revno 2630.28.10, 2630.28.31, 2630.28.26, 2630.33.1, 2630.39.1, 2630.28.29, 2630.34.3, 2630.34.2, 2630.34.1, 2630.29.29, 2630.29.28, 2630.31.1, 2630.28.13, 2630.28.10, 2617.23.14 and some other minor revisions. This patch implements: WL#4264 "Backup: Stabilize Service Interface" -- all the server prerequisites except si_objects.{h,cc} themselves (they can be just copied over, when needed). WL#4435: Support OUT-parameters in prepared statements. (and all issues in the initial patches for these two tasks, that were discovered in pushbuild and during testing). Bug#39519: mysql_stmt_close() should flush all data associated with the statement. After execution of a prepared statement, send OUT parameters of the invoked stored procedure, if any, to the client. When using the binary protocol, send the parameters in an additional result set over the wire. When using the text protocol, assign out parameters to the user variables from the CALL(@var1, @var2, ...) specification. The following refactoring has been made: - Protocol::send_fields() was renamed to Protocol::send_result_set_metadata(); - A new Protocol::send_result_set_row() was introduced to incapsulate common functionality for sending row data. - Signature of Protocol::prepare_for_send() was changed: this operation does not need a list of items, the number of items is fully sufficient. The following backward incompatible changes have been made: - CLIENT_MULTI_RESULTS is now enabled by default in the client; - CLIENT_PS_MULTI_RESUTLS is now enabled by default in the client.
2009-10-21 22:02:06 +02:00
/**
TRUE if we should allow multi-statements.
*/
bool multi_statements;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** State of the lexical analyser for comments. */
enum_comment_state in_comment;
enum_comment_state in_comment_saved;
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
/**
Starting position of the TEXT_STRING or IDENT in the pre-processed
buffer.
NOTE: this member must be used within MYSQLlex() function only.
*/
const char *m_cpp_text_start;
/**
Ending position of the TEXT_STRING or IDENT in the pre-processed
buffer.
NOTE: this member must be used within MYSQLlex() function only.
*/
const char *m_cpp_text_end;
/**
Character set specified by the character-set-introducer.
NOTE: this member must be used within MYSQLlex() function only.
*/
CHARSET_INFO *m_underscore_cs;
};
/**
Abstract representation of a statement.
This class is an interface between the parser and the runtime.
The parser builds the appropriate sub classes of Sql_statement
to represent a SQL statement in the parsed tree.
The execute() method in the sub classes contain the runtime implementation.
Note that this interface is used for SQL statement recently implemented,
the code for older statements tend to load the LEX structure with more
attributes instead.
The recommended way to implement new statements is to sub-class
Sql_statement, as this improves code modularity (see the 'big switch' in
dispatch_command()), and decrease the total size of the LEX structure
(therefore saving memory in stored programs).
*/
class Sql_statement : public Sql_alloc
{
public:
/**
Execute this SQL statement.
@param thd the current thread.
@return 0 on success.
*/
virtual bool execute(THD *thd) = 0;
protected:
/**
Constructor.
@param lex the LEX structure that represents parts of this statement.
*/
Sql_statement(LEX *lex)
: m_lex(lex)
{}
/** Destructor. */
virtual ~Sql_statement()
{
/*
Sql_statement objects are allocated in thd->mem_root.
In MySQL, the C++ destructor is never called, the underlying MEM_ROOT is
simply destroyed instead.
Do not rely on the destructor for any cleanup.
*/
DBUG_ASSERT(FALSE);
}
protected:
/**
The legacy LEX structure for this statement.
The LEX structure contains the existing properties of the parsed tree.
TODO: with time, attributes from LEX should move to sub classes of
Sql_statement, so that the parser only builds Sql_statement objects
with the minimum set of attributes, instead of a LEX structure that
contains the collection of every possible attribute.
*/
LEX *m_lex;
};
2000-07-31 21:29:14 +02:00
/* The state of the lex parsing. This is saved in the THD struct */
struct LEX: public Query_tables_list
{
2002-05-06 23:04:16 +02:00
SELECT_LEX_UNIT unit; /* most upper unit */
SELECT_LEX select_lex; /* first SELECT_LEX */
/* current SELECT_LEX in parsing */
SELECT_LEX *current_select;
/* list of all SELECT_LEX */
SELECT_LEX *all_selects_list;
2006-10-16 18:57:33 +02:00
char *length,*dec,*change;
LEX_STRING name;
char *help_arg;
char *backup_dir; /* For RESTORE/BACKUP */
char* to_log; /* For PURGE MASTER LOGS TO */
char* x509_subject,*x509_issuer,*ssl_cipher;
2000-07-31 21:29:14 +02:00
String *wild;
sql_exchange *exchange;
2002-10-16 15:55:08 +02:00
select_result *result;
Item *default_value, *on_update_value;
2005-01-16 13:16:23 +01:00
LEX_STRING comment, ident;
LEX_USER *grant_user;
2005-04-04 00:50:05 +02:00
XID *xid;
THD *thd;
/* maintain a list of used plugins for this LEX */
DYNAMIC_ARRAY plugins;
plugin_ref plugins_static_buffer[INITIAL_LEX_PLUGIN_LIST_SIZE];
Patch for the following bugs: - BUG#11986: Stored routines and triggers can fail if the code has a non-ascii symbol - BUG#16291: mysqldump corrupts string-constants with non-ascii-chars - BUG#19443: INFORMATION_SCHEMA does not support charsets properly - BUG#21249: Character set of SP-var can be ignored - BUG#25212: Character set of string constant is ignored (stored routines) - BUG#25221: Character set of string constant is ignored (triggers) There were a few general problems that caused these bugs: 1. Character set information of the original (definition) query for views, triggers, stored routines and events was lost. 2. mysqldump output query in client character set, which can be inappropriate to encode definition-query. 3. INFORMATION_SCHEMA used strings with mixed encodings to display object definition; 1. No query-definition-character set. In order to compile query into execution code, some extra data (such as environment variables or the database character set) is used. The problem here was that this context was not preserved. So, on the next load it can differ from the original one, thus the result will be different. The context contains the following data: - client character set; - connection collation (character set and collation); - collation of the owner database; The fix is to store this context and use it each time we parse (compile) and execute the object (stored routine, trigger, ...). 2. Wrong mysqldump-output. The original query can contain several encodings (by means of character set introducers). The problem here was that we tried to convert original query to the mysqldump-client character set. Moreover, we stored queries in different character sets for different objects (views, for one, used UTF8, triggers used original character set). The solution is - to store definition queries in the original character set; - to change SHOW CREATE statement to output definition query in the binary character set (i.e. without any conversion); - introduce SHOW CREATE TRIGGER statement; - to dump special statements to switch the context to the original one before dumping and restore it afterwards. Note, in order to preserve the database collation at the creation time, additional ALTER DATABASE might be used (to temporary switch the database collation back to the original value). In this case, ALTER DATABASE privilege will be required. This is a backward-incompatible change. 3. INFORMATION_SCHEMA showed non-UTF8 strings The fix is to generate UTF8-query during the parsing, store it in the object and show it in the INFORMATION_SCHEMA. Basically, the idea is to create a copy of the original query convert it to UTF8. Character set introducers are removed and all text literals are converted to UTF8. This UTF8 query is intended to provide user-readable output. It must not be used to recreate the object. Specialized SHOW CREATE statements should be used for this. The reason for this limitation is the following: the original query can contain symbols from several character sets (by means of character set introducers). Example: - original query: CREATE VIEW v1 AS SELECT _cp1251 'Hello' AS c1; - UTF8 query (for INFORMATION_SCHEMA): CREATE VIEW v1 AS SELECT 'Hello' AS c1;
2007-06-28 19:34:54 +02:00
CHARSET_INFO *charset;
bool text_string_is_7bit;
/* store original leaf_tables for INSERT SELECT and PS/SP */
TABLE_LIST *leaf_tables_insert;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
/** SELECT of CREATE VIEW statement */
LEX_STRING create_view_select;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
2007-08-16 17:52:55 +02:00
/** Start of 'ON table', in trigger statements. */
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
const char* raw_trg_on_table_name_begin;
2007-08-16 17:52:55 +02:00
/** End of 'ON table', in trigger statements. */
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
const char* raw_trg_on_table_name_end;
2005-07-18 13:31:02 +02:00
/* Partition info structure filled in by PARTITION BY parse part */
partition_info *part_info;
2000-07-31 21:29:14 +02:00
/*
The definer of the object being created (view, trigger, stored routine).
I.e. the value of DEFINER clause.
*/
LEX_USER *definer;
List<Key_part_spec> col_list;
List<Key_part_spec> ref_list;
List<String> interval_list;
List<LEX_USER> users_list;
2000-07-31 21:29:14 +02:00
List<LEX_COLUMN> columns;
List<Item> *insert_list,field_list,value_list,update_list;
List<List_item> many_values;
List<set_var_base> var_list;
List<Item_func_set_user_var> set_var_list; // in-query assignment list
List<Item_param> param_list;
2004-07-16 00:15:55 +02:00
List<LEX_STRING> view_list; // view list (list of field names in view)
/*
A stack of name resolution contexts for the query. This stack is used
at parse time to set local name resolution contexts for various parts
of a query. For example, in a JOIN ... ON (some_condition) clause the
Items in 'some_condition' must be resolved only against the operands
of the the join, and not against the whole clause. Similarly, Items in
subqueries should be resolved against the subqueries (and outer queries).
The stack is used in the following way: when the parser detects that
all Items in some clause need a local context, it creates a new context
and pushes it on the stack. All newly created Items always store the
top-most context in the stack. Once the parser leaves the clause that
required a local context, the parser pops the top-most context.
*/
List<Name_resolution_context> context_stack;
2006-07-04 00:07:41 +02:00
SQL_LIST proc_list, auxiliary_table_list, save_list;
Create_field *last_field;
Item_sum *in_sum_func;
2000-07-31 21:29:14 +02:00
udf_func udf;
2000-09-14 01:39:07 +02:00
HA_CHECK_OPT check_opt; // check/repair options
HA_CREATE_INFO create_info;
KEY_CREATE_INFO key_create_info;
LEX_MASTER_INFO mi; // used by CHANGE MASTER
LEX_SERVER_OPTIONS server_options;
USER_RESOURCES mqh;
ulong type;
/*
This variable is used in post-parse stage to declare that sum-functions,
or functions which have sense only if GROUP BY is present, are allowed.
For example in a query
SELECT ... FROM ...WHERE MIN(i) == 1 GROUP BY ... HAVING MIN(i) > 2
MIN(i) in the WHERE clause is not allowed in the opposite to MIN(i)
in the HAVING clause. Due to possible nesting of select construct
the variable can contain 0 or 1 for each nest level.
*/
nesting_map allow_sum_func;
Sql_statement *m_stmt;
/*
Usually `expr` rule of yacc is quite reused but some commands better
not support subqueries which comes standard with this rule, like
KILL, HA_READ, CREATE/ALTER EVENT etc. Set this to `false` to get
syntax error back.
*/
bool expr_allows_subselect;
enum SSL_type ssl_type; /* defined in violite.h */
2000-09-14 01:39:07 +02:00
enum enum_duplicates duplicates;
enum enum_tx_isolation tx_isolation;
enum enum_ha_read_modes ha_read_mode;
2005-01-16 13:16:23 +01:00
union {
2005-04-04 00:50:05 +02:00
enum ha_rkey_function ha_rkey_mode;
2005-01-16 13:16:23 +01:00
enum xa_option_words xa_opt;
};
enum enum_var_type option_type;
2004-07-16 00:15:55 +02:00
enum enum_view_create_mode create_view_mode;
enum enum_drop_mode drop_mode;
Prevent bugs by making DBUG_* expressions syntactically equivalent to a single statement. --- Bug#24795: SHOW PROFILE Profiling is only partially functional on some architectures. Where there is no getrusage() system call, presently Null values are returned where it would be required. Notably, Windows needs some love applied to make it as useful. Syntax this adds: SHOW PROFILES SHOW PROFILE [types] [FOR QUERY n] [OFFSET n] [LIMIT n] where "n" is an integer and "types" is zero or many (comma-separated) of "CPU" "MEMORY" (not presently supported) "BLOCK IO" "CONTEXT SWITCHES" "PAGE FAULTS" "IPC" "SWAPS" "SOURCE" "ALL" It also adds a session variable (boolean) "profiling", set to "no" by default, and (integer) profiling_history_size, set to 15 by default. This patch abstracts setting THDs' "proc_info" behind a macro that can be used as a hook into the profiling code when profiling support is compiled in. All future code in this line should use that mechanism for setting thd->proc_info. --- Tests are now set to omit the statistics. --- Adds an Information_schema table, "profiling" for access to "show profile" data. --- Merge zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community-3--bug24795 into zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community --- Fix merge problems. --- Fixed one bug in the query_source being NULL. Updated test results. --- Include more thorough profiling tests. Improve support for prepared statements. Use session-specific query IDs, starting at zero. --- Selecting from I_S.profiling is no longer quashed in profiling, as requested by Giuseppe. Limit the size of captured query text. No longer log queries that are zero length.
2007-02-22 16:03:08 +01:00
uint profile_query_id;
Prevent bugs by making DBUG_* expressions syntactically equivalent to a single statement. --- Bug#24795: SHOW PROFILE Profiling is only partially functional on some architectures. Where there is no getrusage() system call, presently Null values are returned where it would be required. Notably, Windows needs some love applied to make it as useful. Syntax this adds: SHOW PROFILES SHOW PROFILE [types] [FOR QUERY n] [OFFSET n] [LIMIT n] where "n" is an integer and "types" is zero or many (comma-separated) of "CPU" "MEMORY" (not presently supported) "BLOCK IO" "CONTEXT SWITCHES" "PAGE FAULTS" "IPC" "SWAPS" "SOURCE" "ALL" It also adds a session variable (boolean) "profiling", set to "no" by default, and (integer) profiling_history_size, set to 15 by default. This patch abstracts setting THDs' "proc_info" behind a macro that can be used as a hook into the profiling code when profiling support is compiled in. All future code in this line should use that mechanism for setting thd->proc_info. --- Tests are now set to omit the statistics. --- Adds an Information_schema table, "profiling" for access to "show profile" data. --- Merge zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community-3--bug24795 into zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-community --- Fix merge problems. --- Fixed one bug in the query_source being NULL. Updated test results. --- Include more thorough profiling tests. Improve support for prepared statements. Use session-specific query IDs, starting at zero. --- Selecting from I_S.profiling is no longer quashed in profiling, as requested by Giuseppe. Limit the size of captured query text. No longer log queries that are zero length.
2007-02-22 16:03:08 +01:00
uint profile_options;
uint uint_geom_type;
uint grant, grant_tot_col, which_columns;
enum Foreign_key::fk_match_opt fk_match_option;
enum Foreign_key::fk_option fk_update_opt;
enum Foreign_key::fk_option fk_delete_opt;
uint slave_thd_opt, start_transaction_opt;
int nest_level;
/*
In LEX representing update which were transformed to multi-update
stores total number of tables. For LEX representing multi-delete
holds number of tables from which we will delete records.
*/
uint table_count;
uint8 describe;
/*
A flag that indicates what kinds of derived tables are present in the
query (0 if no derived tables, otherwise a combination of flags
2005-10-29 11:11:34 +02:00
DERIVED_SUBQUERY and DERIVED_VIEW).
*/
2004-07-16 00:15:55 +02:00
uint8 derived_tables;
uint8 create_view_algorithm;
2004-09-03 14:18:40 +02:00
uint8 create_view_check;
bool drop_if_exists, drop_temporary, local_file, one_shot_set;
bool autocommit;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
bool verbose, no_write_to_binlog;
* Mixed replication mode * : 1) Fix for BUG#19630 "stored function inserting into two auto_increment breaks statement-based binlog": a stored function inserting into two such tables may fail to replicate (inserting wrong data in the slave's copy of the second table) if the slave's second table had an internal auto_increment counter different from master's. Because the auto_increment value autogenerated by master for the 2nd table does not go into binlog, only the first does, so the slave lacks information. To fix this, if running in mixed binlogging mode, if the stored function or trigger plans to update two different tables both having auto_increment columns, we switch to row-based for the whole function. We don't have a simple solution for statement-based binlogging mode, there the bug remains and will be documented as a known problem. Re-enabling rpl_switch_stm_row_mixed. 2) Fix for BUG#20630 "Mixed binlogging mode does not work with stored functions, triggers, views", which was a documented limitation (in mixed mode, we didn't detect that a stored function's execution needed row-based binlogging (due to some UUID() call for example); same for triggers, same for views (a view created from a SELECT UUID(), and doing INSERT INTO sometable SELECT theview; would not replicate row-based). This is implemented by, after parsing a routine's body, remembering in sp_head that this routine needs row-based binlogging. Then when this routine is used, the caller is marked to require row-based binlogging too. Same for views: when we parse a view and detect that its SELECT needs row-based binary logging, we mark the calling LEX as such. 3) Fix for BUG#20499 "mixed mode with temporary table breaks binlog": a temporary table containing e.g. UUID has its changes not binlogged, so any query updating a permanent table with data from the temporary table will run wrongly on slave. Solution: in mixed mode we don't switch back from row-based to statement-based when there exists temporary tables. 4) Attempt to test mysqlbinlog on a binlog generated by mysqlbinlog; impossible due to BUG#11312 and BUG#20329, but test is in place for when they are fixed.
2006-07-09 17:00:47 +02:00
bool tx_chain, tx_release;
2005-05-23 20:38:12 +02:00
/*
Special JOIN::prepare mode: changing of query is prohibited.
When creating a view, we need to just check its syntax omitting
any optimizations: afterwards definition of the view will be
reconstructed by means of ::print() methods and written to
to an .frm file. We need this definition to stay untouched.
*/
2004-07-16 00:15:55 +02:00
bool view_prepare_mode;
bool safe_to_cache_query;
bool subqueries, ignore;
st_parsing_options parsing_options;
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
Alter_info alter_info;
/*
For CREATE TABLE statement last element of table list which is not
part of SELECT or LIKE part (i.e. either element for table we are
creating or last of tables referenced by foreign keys).
*/
TABLE_LIST *create_last_non_select_table;
/* Prepared statements SQL syntax:*/
LEX_STRING prepared_stmt_name; /* Statement name (in all queries) */
2005-04-04 00:50:05 +02:00
/*
Prepared statement query text or name of variable that holds the
prepared statement (in PREPARE ... queries)
*/
2005-04-04 00:50:05 +02:00
LEX_STRING prepared_stmt_code;
/* If true, prepared_stmt_code is a name of variable that holds the query */
bool prepared_stmt_code_is_varref;
/* Names of user variables holding parameters (in EXECUTE) */
2005-04-04 00:50:05 +02:00
List<LEX_STRING> prepared_stmt_params;
sp_head *sphead;
sp_name *spname;
bool sp_lex_in_use; /* Keep track on lex usage in SPs for error handling */
bool all_privileges;
sp_pcontext *spcont;
st_sp_chistics sp_chistics;
2005-12-02 13:07:02 +01:00
Event_parse_data *event_parse_data;
2005-12-02 13:07:02 +01:00
2004-07-16 00:15:55 +02:00
bool only_view; /* used for SHOW CREATE TABLE/VIEW */
/*
field_list was created for view and should be removed before PS/SP
rexecuton
*/
bool empty_field_list_on_rset;
/*
view created to be run from definer (standard behaviour)
*/
uint8 create_view_suid;
/* Characterstics of trigger being created */
st_trg_chistics trg_chistics;
/*
List of all items (Item_trigger_field objects) representing fields in
old/new version of row in trigger. We use this list for checking whenever
all such fields are valid at trigger creation time and for binding these
fields to TABLE object at table open (altough for latter pointer to table
being opened is probably enough).
*/
SQL_LIST trg_table_fields;
/*
stmt_definition_begin is intended to point to the next word after
DEFINER-clause in the following statements:
- CREATE TRIGGER (points to "TRIGGER");
- CREATE PROCEDURE (points to "PROCEDURE");
- CREATE FUNCTION (points to "FUNCTION" or "AGGREGATE");
This pointer is required to add possibly omitted DEFINER-clause to the
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
DDL-statement before dumping it to the binlog.
*/
const char *stmt_definition_begin;
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
const char *stmt_definition_end;
/**
During name resolution search only in the table list given by
Name_resolution_context::first_name_resolution_table and
Name_resolution_context::last_name_resolution_table
(see Item_field::fix_fields()).
*/
bool use_only_table_context;
2006-01-11 11:35:25 +01:00
/*
Reference to a struct that contains information in various commands
to add/create/drop/change table spaces.
*/
st_alter_tablespace *alter_tablespace_info;
bool escape_used;
bool is_lex_started; /* If lex_start() did run. For debugging. */
/*
Special case for SELECT .. FOR UPDATE and LOCK TABLES .. WRITE.
Protect from a impending GRL as otherwise the thread might deadlock
if it starts waiting for the GRL in mysql_lock_tables.
The protection is needed because there is a race between setting
the global read lock and waiting for all open tables to be closed.
The problem is a circular wait where a thread holding "old" open
tables will wait for the global read lock to be released while the
thread holding the global read lock will wait for all "old" open
tables to be closed -- the flush part of flush tables with read
lock.
*/
bool protect_against_global_read_lock;
LEX();
virtual ~LEX()
{
destroy_query_tables_list();
plugin_unlock_list(NULL, (plugin_ref *)plugins.buffer, plugins.elements);
delete_dynamic(&plugins);
}
inline void uncacheable(uint8 cause)
{
safe_to_cache_query= 0;
/*
There are no sense to mark select_lex and union fields of LEX,
but we should merk all subselects as uncacheable from current till
most upper
*/
SELECT_LEX *sl;
2003-03-14 13:16:09 +01:00
SELECT_LEX_UNIT *un;
for (sl= current_select, un= sl->master_unit();
un != &unit;
sl= sl->outer_select(), un= sl->master_unit())
{
sl->uncacheable|= cause;
un->uncacheable|= cause;
}
}
void set_trg_event_type_for_tables();
2004-07-16 00:15:55 +02:00
TABLE_LIST *unlink_first_table(bool *link_to_local);
void link_first_table_back(TABLE_LIST *first, bool link_to_local);
void first_lists_tables_same();
bool can_be_merged();
bool can_use_merged();
bool can_not_use_merged();
2004-07-16 00:15:55 +02:00
bool only_view_structure();
bool need_correct_ident();
uint8 get_effective_with_check(TABLE_LIST *view);
/*
Is this update command where 'WHITH CHECK OPTION' clause is important
SYNOPSIS
LEX::which_check_option_applicable()
RETURN
TRUE have to take 'WHITH CHECK OPTION' clause into account
FALSE 'WHITH CHECK OPTION' clause do not need
*/
inline bool which_check_option_applicable()
{
switch (sql_command) {
case SQLCOM_UPDATE:
case SQLCOM_UPDATE_MULTI:
case SQLCOM_INSERT:
case SQLCOM_INSERT_SELECT:
case SQLCOM_REPLACE:
case SQLCOM_REPLACE_SELECT:
case SQLCOM_LOAD:
return TRUE;
default:
return FALSE;
}
}
void cleanup_after_one_table_open();
bool push_context(Name_resolution_context *context)
{
return context_stack.push_front(context);
}
void pop_context()
{
context_stack.pop();
}
bool copy_db_to(char **p_db, size_t *p_db_length) const;
Name_resolution_context *current_context()
{
return context_stack.head();
}
A fix for Bug#26750 "valgrind leak in sp_head" (and post-review fixes). The legend: on a replication slave, in case a trigger creation was filtered out because of application of replicate-do-table/ replicate-ignore-table rule, the parsed definition of a trigger was not cleaned up properly. LEX::sphead member was left around and leaked memory. Until the actual implementation of support of replicate-ignore-table rules for triggers by the patch for Bug 24478 it was never the case that "case SQLCOM_CREATE_TRIGGER" was not executed once a trigger was parsed, so the deletion of lex->sphead there worked and the memory did not leak. The fix: The real cause of the bug is that there is no 1 or 2 places where we can clean up the main LEX after parse. And the reason we can not have just one or two places where we clean up the LEX is asymmetric behaviour of MYSQLparse in case of success or error. One of the root causes of this behaviour is the code in Item::Item() constructor. There, a newly created item adds itself to THD::free_list - a single-linked list of Items used in a statement. Yuck. This code is unaware that we may have more than one statement active at a time, and always assumes that the free_list of the current statement is located in THD::free_list. One day we need to be able to explicitly allocate an item in a given Query_arena. Thus, when parsing a definition of a stored procedure, like CREATE PROCEDURE p1() BEGIN SELECT a FROM t1; SELECT b FROM t1; END; we actually need to reset THD::mem_root, THD::free_list and THD::lex to parse the nested procedure statement (SELECT *). The actual reset and restore is implemented in semantic actions attached to sp_proc_stmt grammar rule. The problem is that in case of a parsing error inside a nested statement Bison generated parser would abort immediately, without executing the restore part of the semantic action. This would leave THD in an in-the-middle-of-parsing state. This is why we couldn't have had a single place where we clean up the LEX after MYSQLparse - in case of an error we needed to do a clean up immediately, in case of success a clean up could have been delayed. This left the door open for a memory leak. One of the following possibilities were considered when working on a fix: - patch the replication logic to do the clean up. Rejected as breaks module borders, replication code should not need to know the gory details of clean up procedure after CREATE TRIGGER. - wrap MYSQLparse with a function that would do a clean up. Rejected as ideally we should fix the problem when it happens, not adjust for it outside of the problematic code. - make sure MYSQLparse cleans up after itself by invoking the clean up functionality in the appropriate places before return. Implemented in this patch. - use %destructor rule for sp_proc_stmt to restore THD - cleaner than the prevoius approach, but rejected because needs a careful analysis of the side effects, and this patch is for 5.0, and long term we need to use the next alternative anyway - make sure that sp_proc_stmt doesn't juggle with THD - this is a large work that will affect many modules. Cleanup: move main_lex and main_mem_root from Statement to its only two descendants Prepared_statement and THD. This ensures that when a Statement instance was created for purposes of statement backup, we do not involve LEX constructor/destructor, which is fairly expensive. In order to track that the transformation produces equivalent functionality please check the respective constructors and destructors of Statement, Prepared_statement and THD - these members were used only there. This cleanup is unrelated to the patch.
2007-03-07 10:24:46 +01:00
/*
Restore the LEX and THD in case of a parse error.
*/
static void cleanup_lex_after_parse_error(THD *thd);
void reset_n_backup_query_tables_list(Query_tables_list *backup);
void restore_backup_query_tables_list(Query_tables_list *backup);
bool table_or_sp_used();
5.1 version of a fix and test cases for bugs: Bug#4968 ""Stored procedure crash if cursor opened on altered table" Bug#6895 "Prepared Statements: ALTER TABLE DROP COLUMN does nothing" Bug#19182 "CREATE TABLE bar (m INT) SELECT n FROM foo; doesn't work from stored procedure." Bug#19733 "Repeated alter, or repeated create/drop, fails" Bug#22060 "ALTER TABLE x AUTO_INCREMENT=y in SP crashes server" Bug#24879 "Prepared Statements: CREATE TABLE (UTF8 KEY) produces a growing key length" (this bug is not fixed in 5.0) Re-execution of CREATE DATABASE, CREATE TABLE and ALTER TABLE statements in stored routines or as prepared statements caused incorrect results (and crashes in versions prior to 5.0.25). In 5.1 the problem occured only for CREATE DATABASE, CREATE TABLE SELECT and CREATE TABLE with INDEX/DATA DIRECTOY options). The problem of bugs 4968, 19733, 19282 and 6895 was that functions mysql_prepare_table, mysql_create_table and mysql_alter_table are not re-execution friendly: during their operation they modify contents of LEX (members create_info, alter_info, key_list, create_list), thus making the LEX unusable for the next execution. In particular, these functions removed processed columns and keys from create_list, key_list and drop_list. Search the code in sql_table.cc for drop_it.remove() and similar patterns to find evidence. The fix is to supply to these functions a usable copy of each of the above structures at every re-execution of an SQL statement. To simplify memory management, LEX::key_list and LEX::create_list were added to LEX::alter_info, a fresh copy of which is created for every execution. The problem of crashing bug 22060 stemmed from the fact that the above metnioned functions were not only modifying HA_CREATE_INFO structure in LEX, but also were changing it to point to areas in volatile memory of the execution memory root. The patch solves this problem by creating and using an on-stack copy of HA_CREATE_INFO in mysql_execute_command. Additionally, this patch splits the part of mysql_alter_table that analizes and rewrites information from the parser into a separate function - mysql_prepare_alter_table, in analogy with mysql_prepare_table, which is renamed to mysql_prepare_create_table.
2007-05-28 13:30:01 +02:00
bool is_partition_management() const;
/**
@brief check if the statement is a single-level join
@return result of the check
@retval TRUE The statement doesn't contain subqueries, unions and
stored procedure calls.
@retval FALSE There are subqueries, UNIONs or stored procedure calls.
*/
bool is_single_level_stmt()
{
/*
This check exploits the fact that the last added to all_select_list is
on its top. So select_lex (as the first added) will be at the tail
of the list.
*/
if (&select_lex == all_selects_list && !sroutines.records)
{
DBUG_ASSERT(!all_selects_list->next_select_in_list());
return TRUE;
}
return FALSE;
}
};
2000-07-31 21:29:14 +02:00
/**
Set_signal_information is a container used in the parsed tree to represent
the collection of assignments to condition items in the SIGNAL and RESIGNAL
statements.
*/
class Set_signal_information
{
public:
/** Empty default constructor, use clear() */
Set_signal_information() {}
/** Copy constructor. */
Set_signal_information(const Set_signal_information& set);
/** Destructor. */
~Set_signal_information()
{}
/** Clear all items. */
void clear();
/**
For each condition item assignment, m_item[] contains the parsed tree
that represents the expression assigned, if any.
m_item[] is an array indexed by Diag_condition_item_name.
*/
Item *m_item[LAST_DIAG_SET_PROPERTY+1];
};
/**
The internal state of the syntax parser.
This object is only available during parsing,
and is private to the syntax parser implementation (sql_yacc.yy).
*/
class Yacc_state
{
public:
Yacc_state()
{
reset();
}
void reset()
{
yacc_yyss= NULL;
yacc_yyvs= NULL;
m_set_signal_info.clear();
m_lock_type= TL_READ_DEFAULT;
}
~Yacc_state();
Committing on behalf or Dmitry Lenev: Fix for bug #46947 "Embedded SELECT without FOR UPDATE is causing a lock", with after-review fixes. SELECT statements with subqueries referencing InnoDB tables were acquiring shared locks on rows in these tables when they were executed in REPEATABLE-READ mode and with statement or mixed mode binary logging turned on. This was a regression which were introduced when fixing bug 39843. The problem was that for tables belonging to subqueries parser set TL_READ_DEFAULT as a lock type. In cases when statement/mixed binary logging at open_tables() time this type of lock was converted to TL_READ_NO_INSERT lock at open_tables() time and caused InnoDB engine to acquire shared locks on reads from these tables. Although in some cases such behavior was correct (e.g. for subqueries in DELETE) in case of SELECT it has caused unnecessary locking. This patch tries to solve this problem by rethinking our approach to how we handle locking for SELECT and subqueries. Now we always set TL_READ_DEFAULT lock type for all cases when we read data. When at open_tables() time this lock is interpreted as TL_READ_NO_INSERT or TL_READ depending on whether this statement as a whole or call to function which uses particular table should be written to the binary log or not (if yes then statement should be properly serialized with concurrent statements and stronger lock should be acquired). Test coverage is added for both InnoDB and MyISAM. This patch introduces an "incompatible" change in locking scheme for subqueries used in SELECT ... FOR UPDATE and SELECT .. IN SHARE MODE. In 4.1 the server would use a snapshot InnoDB read for subqueries in SELECT FOR UPDATE and SELECT .. IN SHARE MODE statements, regardless of whether the binary log is on or off. If the user required a different type of read (i.e. locking read), he/she could request so explicitly by providing FOR UPDATE/IN SHARE MODE clause for each individual subquery. On of the patches for 5.0 broke this behaviour (which was not documented or tested), and started to use locking reads fora all subqueries in SELECT ... FOR UPDATE/IN SHARE MODE. This patch restored 4.1 behaviour.
2010-04-28 12:04:11 +02:00
/**
Reset part of the state which needs resetting before parsing
substatement.
*/
void reset_before_substatement()
{
m_lock_type= TL_READ_DEFAULT;
}
/**
Bison internal state stack, yyss, when dynamically allocated using
my_yyoverflow().
*/
uchar *yacc_yyss;
/**
Bison internal semantic value stack, yyvs, when dynamically allocated using
my_yyoverflow().
*/
uchar *yacc_yyvs;
/**
Fragments of parsed tree,
used during the parsing of SIGNAL and RESIGNAL.
*/
Set_signal_information m_set_signal_info;
Committing on behalf or Dmitry Lenev: Fix for bug #46947 "Embedded SELECT without FOR UPDATE is causing a lock", with after-review fixes. SELECT statements with subqueries referencing InnoDB tables were acquiring shared locks on rows in these tables when they were executed in REPEATABLE-READ mode and with statement or mixed mode binary logging turned on. This was a regression which were introduced when fixing bug 39843. The problem was that for tables belonging to subqueries parser set TL_READ_DEFAULT as a lock type. In cases when statement/mixed binary logging at open_tables() time this type of lock was converted to TL_READ_NO_INSERT lock at open_tables() time and caused InnoDB engine to acquire shared locks on reads from these tables. Although in some cases such behavior was correct (e.g. for subqueries in DELETE) in case of SELECT it has caused unnecessary locking. This patch tries to solve this problem by rethinking our approach to how we handle locking for SELECT and subqueries. Now we always set TL_READ_DEFAULT lock type for all cases when we read data. When at open_tables() time this lock is interpreted as TL_READ_NO_INSERT or TL_READ depending on whether this statement as a whole or call to function which uses particular table should be written to the binary log or not (if yes then statement should be properly serialized with concurrent statements and stronger lock should be acquired). Test coverage is added for both InnoDB and MyISAM. This patch introduces an "incompatible" change in locking scheme for subqueries used in SELECT ... FOR UPDATE and SELECT .. IN SHARE MODE. In 4.1 the server would use a snapshot InnoDB read for subqueries in SELECT FOR UPDATE and SELECT .. IN SHARE MODE statements, regardless of whether the binary log is on or off. If the user required a different type of read (i.e. locking read), he/she could request so explicitly by providing FOR UPDATE/IN SHARE MODE clause for each individual subquery. On of the patches for 5.0 broke this behaviour (which was not documented or tested), and started to use locking reads fora all subqueries in SELECT ... FOR UPDATE/IN SHARE MODE. This patch restored 4.1 behaviour.
2010-04-28 12:04:11 +02:00
/**
Type of lock to be used for tables being added to the statement's
table list in table_factor, table_alias_ref, single_multi and
table_wild_one rules.
Statements which use these rules but require lock type different
from one specified by this member have to override it by using
st_select_lex::set_lock_for_tables() method.
The default value of this member is TL_READ_DEFAULT. The only two
cases in which we change it are:
- When parsing SELECT HIGH_PRIORITY.
- Rule for DELETE. In which we use this member to pass information
about type of lock from delete to single_multi part of rule.
We should try to avoid introducing new use cases as we would like
to get rid of this member eventually.
*/
thr_lock_type m_lock_type;
/*
TODO: move more attributes from the LEX structure here.
*/
};
/**
Internal state of the parser.
The complete state consist of:
- state data used during lexical parsing,
- state data used during syntactic parsing.
*/
class Parser_state
{
public:
Parser_state(THD *thd, const char* buff, unsigned int length)
: m_lip(thd, buff, length), m_yacc()
{}
~Parser_state()
{}
Lex_input_stream m_lip;
Yacc_state m_yacc;
void reset(const char *found_semicolon, unsigned int length)
{
m_lip.reset(found_semicolon, length);
m_yacc.reset();
}
};
struct st_lex_local: public LEX
2004-07-16 14:20:51 +02:00
{
static void *operator new(size_t size) throw()
2004-07-16 14:20:51 +02:00
{
WL#3817: Simplify string / memory area types and make things more consistent (first part) The following type conversions was done: - Changed byte to uchar - Changed gptr to uchar* - Change my_string to char * - Change my_size_t to size_t - Change size_s to size_t Removed declaration of byte, gptr, my_string, my_size_t and size_s. Following function parameter changes was done: - All string functions in mysys/strings was changed to use size_t instead of uint for string lengths. - All read()/write() functions changed to use size_t (including vio). - All protocoll functions changed to use size_t instead of uint - Functions that used a pointer to a string length was changed to use size_t* - Changed malloc(), free() and related functions from using gptr to use void * as this requires fewer casts in the code and is more in line with how the standard functions work. - Added extra length argument to dirname_part() to return the length of the created string. - Changed (at least) following functions to take uchar* as argument: - db_dump() - my_net_write() - net_write_command() - net_store_data() - DBUG_DUMP() - decimal2bin() & bin2decimal() - Changed my_compress() and my_uncompress() to use size_t. Changed one argument to my_uncompress() from a pointer to a value as we only return one value (makes function easier to use). - Changed type of 'pack_data' argument to packfrm() to avoid casts. - Changed in readfrm() and writefrom(), ha_discover and handler::discover() the type for argument 'frmdata' to uchar** to avoid casts. - Changed most Field functions to use uchar* instead of char* (reduced a lot of casts). - Changed field->val_xxx(xxx, new_ptr) to take const pointers. Other changes: - Removed a lot of not needed casts - Added a few new cast required by other changes - Added some cast to my_multi_malloc() arguments for safety (as string lengths needs to be uint, not size_t). - Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done explicitely as this conflict was often hided by casting the function to hash_get_key). - Changed some buffers to memory regions to uchar* to avoid casts. - Changed some string lengths from uint to size_t. - Changed field->ptr to be uchar* instead of char*. This allowed us to get rid of a lot of casts. - Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar - Include zlib.h in some files as we needed declaration of crc32() - Changed MY_FILE_ERROR to be (size_t) -1. - Changed many variables to hold the result of my_read() / my_write() to be size_t. This was needed to properly detect errors (which are returned as (size_t) -1). - Removed some very old VMS code - Changed packfrm()/unpackfrm() to not be depending on uint size (portability fix) - Removed windows specific code to restore cursor position as this causes slowdown on windows and we should not mix read() and pread() calls anyway as this is not thread safe. Updated function comment to reflect this. Changed function that depended on original behavior of my_pwrite() to itself restore the cursor position (one such case). - Added some missing checking of return value of malloc(). - Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow. - Changed type of table_def::m_size from my_size_t to ulong to reflect that m_size is the number of elements in the array, not a string/memory length. - Moved THD::max_row_length() to table.cc (as it's not depending on THD). Inlined max_row_length_blob() into this function. - More function comments - Fixed some compiler warnings when compiled without partitions. - Removed setting of LEX_STRING() arguments in declaration (portability fix). - Some trivial indentation/variable name changes. - Some trivial code simplifications: - Replaced some calls to alloc_root + memcpy to use strmake_root()/strdup_root(). - Changed some calls from memdup() to strmake() (Safety fix) - Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
return sql_alloc(size);
2004-07-16 14:20:51 +02:00
}
static void *operator new(size_t size, MEM_ROOT *mem_root) throw()
2004-07-16 14:20:51 +02:00
{
return (void*) alloc_root(mem_root, (uint) size);
}
static void operator delete(void *ptr,size_t size)
{ TRASH(ptr, size); }
static void operator delete(void *ptr, MEM_ROOT *mem_root)
{ /* Never called */ }
2004-07-16 14:20:51 +02:00
};
2000-07-31 21:29:14 +02:00
extern void lex_init(void);
extern void lex_free(void);
extern void lex_start(THD *thd);
extern void lex_end(LEX *lex);
extern int MYSQLlex(void *arg, void *yythd);
Bug#25411 (trigger code truncated), PART II Bug 28127 (Some valid identifiers names are not parsed correctly) Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) This patch is the second part of a major cleanup, required to fix Bug 25411 (trigger code truncated). The root cause of the issue stems from the function skip_rear_comments, which was a work around to remove "extra" "*/" characters from the query text, when parsing a query and reusing the text fragments to represent a view, trigger, function or stored procedure. The reason for this work around is that "special comments", like /*!50002 XXX */, were not parsed properly, so that a query like: AAA /*!50002 BBB */ CCC would be seen by the parser as "AAA BBB */ CCC" when the current version is greater or equal to 5.0.2 The root cause of this stems from how special comments are parsed. Special comments are really out-of-bound text that appear inside a query, that affects how the parser behave. In nature, /*!50002 XXX */ in MySQL is similar to the C concept of preprocessing : #if VERSION >= 50002 XXX #endif Depending on the current VERSION of the server, either the special comment should be expanded or it should be ignored, but in all cases the "text" of the query should be re-written to strip the "/*!50002" and "*/" markers, which does not belong to the SQL language itself. Prior to this fix, these markers would leak into : - the storage format for VIEW, - the storage format for FUNCTION, - the storage format for FUNCTION parameters, in mysql.proc (param_list), - the storage format for PROCEDURE, - the storage format for PROCEDURE parameters, in mysql.proc (param_list), - the storage format for TRIGGER, - the binary log used for replication. In all cases, not only this cause format corruption, but also provide a vector for dormant security issues, by allowing to tunnel code that will be activated after an upgrade. The proper solution is to deal with special comments strictly during parsing, when accepting a query from the outside world. Once a query is parsed and an object is created with a persistant representation, this object should not arbitrarily mutate after an upgrade. In short, special comments are a useful but limited feature for MYSQLdump, when used at an *interface* level to facilitate import/export, but bloating the server *internal* storage format is *not* the proper way to deal with configuration management of the user logic. With this fix: - the Lex_input_stream class now acts as a comment pre-processor, and either expands or ignore special comments on the fly. - MYSQLlex and sql_yacc.yy have been cleaned up to strictly use the public interface of Lex_input_stream. In particular, how the input stream accepts or rejects a character is private to Lex_input_stream, and the internal buffer pointers of that class are strictly private, and should not be tempered with during parsing. This caused many changes mostly in sql_lex.cc. During the code cleanup in case MY_LEX_NUMBER_IDENT, Bug 28127 (Some valid identifiers names are not parsed correctly) was found and fixed. By parsing special comments properly, and removing the function 'skip_rear_comments' [sic], Bug 26302 (MySQL server cuts off trailing "*/" from comments in SP/func) has been fixed as well.
2007-06-12 23:23:58 +02:00
extern void trim_whitespace(CHARSET_INFO *cs, LEX_STRING *str);
2000-07-31 21:29:14 +02:00
Bug#18239 (Possible to overload internal functions with stored functions) Bug#21025 (misleading error message when creating functions named 'x', or 'y') Bug#22619 (Spaces considered harmful) This change contains a fix to report warnings or errors, and multiple tests cases. Before this fix, name collisions between: - Native functions - User Defined Functions - Stored Functions were not systematically reported, leading to confusing behavior. I) Native / User Defined Function Before this fix, is was possible to create a UDF named "foo", with the same name as a native function "foo", but it was impossible to invoke the UDF, since the syntax "foo()" always refer to the native function. After this fix, creating a UDF fails with an error if there is a name collision with a native function. II) Native / Stored Function Before this fix, is was possible to create a SF named "db.foo", with the same name as a native function "foo", but this was confusing since the syntax "foo()" would refer to the native function. To refer to the Stored Function, the user had to use the "db.foo()" syntax. After this fix, creating a Stored Function reports a warning if there is a name collision with a native function. III) User Defined Function / Stored Function Before this fix, creating a User Defined Function "foo" and a Stored Function "db.foo" are mutually exclusive operations. Whenever the second function is created, an error is reported. However, the test suite did not cover this behavior. After this fix, the behavior is unchanged, and is now covered by test cases. Note that the code change in this patch depends on the fix for Bug 21114.
2006-11-15 03:34:16 +01:00
extern bool is_lex_native_function(const LEX_STRING *name);
/**
@} (End of group Semantic_Analysis)
*/
void my_missing_function_error(const LEX_STRING &token, const char *name);
bool is_keyword(const char *name, uint len);
2009-05-27 17:19:44 +02:00
#endif /* MYSQL_SERVER */
#endif /* SQL_LEX_INCLUDED */