mariadb/sql/sql_select.h

855 lines
30 KiB
C
Raw Normal View History

#ifndef SQL_SELECT_INCLUDED
#define SQL_SELECT_INCLUDED
2011-06-30 17:46:53 +02:00
/* Copyright (c) 2000, 2011, Oracle and/or its affiliates. All rights reserved.
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
2011-06-30 17:46:53 +02:00
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */
2000-07-31 21:29:14 +02:00
2007-10-16 22:11:50 +02:00
/**
@file
@brief
classes to use when handling where clause
*/
2000-07-31 21:29:14 +02:00
#ifdef USE_PRAGMA_INTERFACE
2000-07-31 21:29:14 +02:00
#pragma interface /* gcc class implementation */
#endif
#include "procedure.h"
#include <myisam.h>
#include "sql_array.h" /* Array */
#include "records.h" /* READ_RECORD */
#include "opt_range.h" /* SQL_SELECT, QUICK_SELECT_I */
2000-07-31 21:29:14 +02:00
typedef struct keyuse_t {
TABLE *table;
2007-10-16 22:11:50 +02:00
Item *val; /**< or value if no field */
2000-07-31 21:29:14 +02:00
table_map used_tables;
uint key, keypart, optimize;
2003-10-11 13:06:55 +02:00
key_part_map keypart_map;
ha_rows ref_table_rows;
2007-10-16 22:11:50 +02:00
/**
If true, the comparison this value was created from will not be
satisfied if val has NULL 'value'.
*/
bool null_rejecting;
/*
!NULL - This KEYUSE was created from an equality that was wrapped into
an Item_func_trig_cond. This means the equality (and validity of
this KEYUSE element) can be turned on and off. The on/off state
is indicted by the pointed value:
*cond_guard == TRUE <=> equality condition is on
*cond_guard == FALSE <=> equality condition is off
NULL - Otherwise (the source equality can't be turned off)
*/
bool *cond_guard;
2000-07-31 21:29:14 +02:00
} KEYUSE;
class store_key;
typedef struct st_table_ref
{
bool key_err;
/** True if something was read into buffer in join_read_key. */
bool has_record;
2007-10-16 22:11:50 +02:00
uint key_parts; ///< num of ...
uint key_length; ///< length of key_buff
int key; ///< key no
uchar *key_buff; ///< value to look for with key
uchar *key_buff2; ///< key_buff+key_length
2000-07-31 21:29:14 +02:00
store_key **key_copy; //
2007-10-16 22:11:50 +02:00
Item **items; ///< val()'s for each keypart
/*
Array of pointers to trigger variables. Some/all of the pointers may be
NULL. The ref access can be used iff
for each used key part i, (!cond_guards[i] || *cond_guards[i])
This array is used by subquery code. The subquery code may inject
triggered conditions, i.e. conditions that can be 'switched off'. A ref
access created from such condition is not valid when at least one of the
underlying conditions is switched off (see subquery code for more details)
*/
bool **cond_guards;
2007-10-16 22:11:50 +02:00
/**
(null_rejecting & (1<<i)) means the condition is '=' and no matching
rows will be produced if items[i] IS NULL (see add_not_null_conds())
*/
key_part_map null_rejecting;
2007-10-16 22:11:50 +02:00
table_map depend_map; ///< Table depends on these tables.
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
/* null byte position in the key_buf. Used for REF_OR_NULL optimization */
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 *null_ref_key;
/*
The number of times the record associated with this key was used
in the join.
*/
ha_rows use_count;
2000-07-31 21:29:14 +02:00
} TABLE_REF;
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
#define CACHE_BLOB 1 /* blob field */
#define CACHE_STRIPPED 2 /* field stripped of trailing spaces */
2007-10-16 22:11:50 +02:00
/**
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
CACHE_FIELD and JOIN_CACHE is used on full join to cache records in outer
table
2000-07-31 21:29:14 +02:00
*/
typedef struct st_cache_field {
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 *str;
uint length, blob_length;
Field *field;
uint type; /**< category of the of the copied field (CACHE_BLOB et al.) */
2000-07-31 21:29:14 +02:00
} CACHE_FIELD;
typedef struct st_join_cache {
uchar *buff,*pos,*end;
uint records,record_nr,ptr_record,fields,length,blobs;
CACHE_FIELD *field,**blob_ptr;
SQL_SELECT *select;
} JOIN_CACHE;
/*
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
The structs which holds the join connections and join states
2000-07-31 21:29:14 +02:00
*/
enum join_type { JT_UNKNOWN,JT_SYSTEM,JT_CONST,JT_EQ_REF,JT_REF,JT_MAYBE_REF,
JT_ALL, JT_RANGE, JT_NEXT, JT_FT, JT_REF_OR_NULL,
JT_UNIQUE_SUBQUERY, JT_INDEX_SUBQUERY, JT_INDEX_MERGE};
2000-07-31 21:29:14 +02:00
class JOIN;
enum enum_nested_loop_state
{
NESTED_LOOP_KILLED= -2, NESTED_LOOP_ERROR= -1,
NESTED_LOOP_OK= 0, NESTED_LOOP_NO_MORE_ROWS= 1,
NESTED_LOOP_QUERY_LIMIT= 3, NESTED_LOOP_CURSOR_LIMIT= 4
};
/* Values for JOIN_TAB::packed_info */
#define TAB_INFO_HAVE_VALUE 1
#define TAB_INFO_USING_INDEX 2
#define TAB_INFO_USING_WHERE 4
#define TAB_INFO_FULL_SCAN_ON_NULL 8
typedef enum_nested_loop_state
(*Next_select_func)(JOIN *, struct st_join_table *, bool);
Next_select_func setup_end_select_func(JOIN *join);
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
2000-07-31 21:29:14 +02:00
typedef struct st_join_table {
st_join_table() {} /* Remove gcc warning */
2000-07-31 21:29:14 +02:00
TABLE *table;
2007-10-16 22:11:50 +02:00
KEYUSE *keyuse; /**< pointer to first used key */
SQL_SELECT *select;
COND *select_cond;
2003-11-13 15:52:02 +01:00
QUICK_SELECT_I *quick;
2007-10-16 22:11:50 +02:00
Item **on_expr_ref; /**< pointer to the associated on expression */
COND_EQUAL *cond_equal; /**< multiple equalities for the on expression */
st_join_table *first_inner; /**< first inner table for including outerjoin */
bool found; /**< true after all matches or null complement */
bool not_null_compl;/**< true before null complement is added */
st_join_table *last_inner; /**< last table table for embedding outer join */
st_join_table *first_upper; /**< first inner table for embedding outer join */
st_join_table *first_unmatched; /**< used for optimization purposes only */
/* Special content for EXPLAIN 'Extra' column or NULL if none */
const char *info;
/*
Bitmap of TAB_INFO_* bits that encodes special line for EXPLAIN 'Extra'
column, or 0 if there is no info.
*/
uint packed_info;
READ_RECORD::Setup_func read_first_record;
Next_select_func next_select;
2000-07-31 21:29:14 +02:00
READ_RECORD read_record;
/*
Currently the following two fields are used only for a [NOT] IN subquery
if it is executed by an alternative full table scan when the left operand of
the subquery predicate is evaluated to NULL.
*/
READ_RECORD::Setup_func save_read_first_record;/* to save read_first_record */
READ_RECORD::Read_func save_read_record;/* to save read_record.read_record */
double worst_seeks;
2007-10-16 22:11:50 +02:00
key_map const_keys; /**< Keys with constant part */
key_map checked_keys; /**< Keys checked in find_best */
2000-07-31 21:29:14 +02:00
key_map needed_reg;
2007-10-16 22:11:50 +02:00
key_map keys; /**< all keys with can be used */
/* Either #rows in the table or 1 for const table. */
ha_rows records;
/*
Number of records that will be scanned (yes scanned, not returned) by the
best 'independent' access method, i.e. table scan or QUICK_*_SELECT)
*/
ha_rows found_records;
/*
Cost of accessing the table using "ALL" or range/index_merge access
method (but not 'index' for some reason), i.e. this matches method which
E(#records) is in found_records.
*/
ha_rows read_time;
2000-07-31 21:29:14 +02:00
table_map dependent,key_dependent;
uint use_quick,index;
2007-10-16 22:11:50 +02:00
uint status; ///< Save status for cache
2000-07-31 21:29:14 +02:00
uint used_fields,used_fieldlength,used_blobs;
enum join_type type;
bool cached_eq_ref_table,eq_ref_table,not_used_in_distinct;
2005-07-18 13:31:02 +02:00
bool sorted;
/*
If it's not 0 the number stored this field indicates that the index
scan has been chosen to access the table data and we expect to scan
this number of rows for the table.
*/
ha_rows limit;
2000-07-31 21:29:14 +02:00
TABLE_REF ref;
JOIN_CACHE cache;
JOIN *join;
2007-10-16 22:11:50 +02:00
/** Bitmap of nested joins this table is part of */
nested_join_map embedding_map;
void cleanup();
inline bool is_using_loose_index_scan()
{
return (select && select->quick &&
(select->quick->get_type() ==
QUICK_SELECT_I::QS_TYPE_GROUP_MIN_MAX));
}
2009-09-28 09:21:25 +02:00
bool is_using_agg_loose_index_scan ()
{
return (is_using_loose_index_scan() &&
((QUICK_GROUP_MIN_MAX_SELECT *)select->quick)->is_agg_distinct());
}
2000-07-31 21:29:14 +02:00
} JOIN_TAB;
enum_nested_loop_state sub_select_cache(JOIN *join, JOIN_TAB *join_tab, bool
end_of_records);
enum_nested_loop_state sub_select(JOIN *join,JOIN_TAB *join_tab, bool
end_of_records);
2007-10-16 22:11:50 +02:00
/**
Information about a position of table within a join order. Used in join
optimization.
*/
typedef struct st_position
{
/*
The "fanout": number of output rows that will be produced (after
pushed down selection condition is applied) per each row combination of
previous tables.
*/
double records_read;
/*
Cost accessing the table in course of the entire complete join execution,
i.e. cost of one access method use (e.g. 'range' or 'ref' scan ) times
number the access method will be invoked.
*/
double read_time;
2000-07-31 21:29:14 +02:00
JOIN_TAB *table;
/*
NULL - 'index' or 'range' or 'index_merge' or 'ALL' access is used.
Other - [eq_]ref[_or_null] access is used. Pointer to {t.keypart1 = expr}
*/
2000-07-31 21:29:14 +02:00
KEYUSE *key;
/* If ref-based access is used: bitmap of tables this table depends on */
table_map ref_depend_map;
2000-07-31 21:29:14 +02:00
} POSITION;
typedef struct st_rollup
{
enum State { STATE_NONE, STATE_INITED, STATE_READY };
State state;
Item_null_result **null_items;
Item ***ref_pointer_arrays;
List<Item> *fields;
} ROLLUP;
2000-07-31 21:29:14 +02:00
class JOIN :public Sql_alloc
{
2007-10-16 22:11:50 +02:00
JOIN(const JOIN &rhs); /**< not implemented */
JOIN& operator=(const JOIN &rhs); /**< not implemented */
2006-02-23 14:21:43 +01:00
public:
JOIN_TAB *join_tab,**best_ref;
2007-10-16 22:11:50 +02:00
JOIN_TAB **map2table; ///< mapping between table indexes and JOIN_TABs
JOIN_TAB *join_tab_save; ///< saved join_tab for subquery reexecution
TABLE **all_tables,*sort_by_table;
2000-07-31 21:29:14 +02:00
uint tables,const_tables;
uint send_group_parts;
/**
Indicates that grouping will be performed on the result set during
query execution. This field belongs to query execution.
@see make_group_fields, alloc_group_fields, JOIN::exec
*/
bool sort_and_group;
bool first_record,full_join,group, no_field_update;
bool do_send_rows;
table_map const_table_map,found_const_table_map;
/*
Bitmap of all inner tables from outer joins
*/
table_map outer_join;
ha_rows send_records,found_records,examined_rows,row_limit, select_limit;
2007-10-16 22:11:50 +02:00
/**
Used to fetch no more than given amount of rows per one
fetch operation of server side cursor.
The value is checked in end_send and end_send_group in fashion, similar
to offset_limit_cnt:
- fetch_limit= HA_POS_ERROR if there is no cursor.
- when we open a cursor, we set fetch_limit to 0,
- on each fetch iteration we add num_rows to fetch to fetch_limit
*/
ha_rows fetch_limit;
2000-07-31 21:29:14 +02:00
POSITION positions[MAX_TABLES+1],best_positions[MAX_TABLES+1];
2007-10-16 22:11:50 +02:00
/* *
Bitmap of nested joins embedding the position at the end of the current
partial join (valid only during join optimizer run).
*/
nested_join_map cur_embedding_map;
2000-07-31 21:29:14 +02:00
double best_read;
List<Item> *fields;
List<Cached_item> group_fields, group_fields_cache;
2000-07-31 21:29:14 +02:00
TABLE *tmp_table;
2007-10-16 22:11:50 +02:00
/// used to store 2 possible tmp table of SELECT
TABLE *exec_tmp_table1, *exec_tmp_table2;
2000-07-31 21:29:14 +02:00
THD *thd;
Item_sum **sum_funcs, ***sum_funcs_end;
2007-10-16 22:11:50 +02:00
/** second copy of sumfuncs (for queries with 2 temporary tables */
Item_sum **sum_funcs2, ***sum_funcs_end2;
2000-07-31 21:29:14 +02:00
Procedure *procedure;
Item *having;
2007-10-16 22:11:50 +02:00
Item *tmp_having; ///< To store having when processed temporary table
Item *having_history; ///< Store having for explain
ulonglong select_options;
2000-07-31 21:29:14 +02:00
select_result *result;
TMP_TABLE_PARAM tmp_table_param;
MYSQL_LOCK *lock;
2007-10-16 22:11:50 +02:00
/// unit structure (with global parameters) for this select
SELECT_LEX_UNIT *unit;
2007-10-16 22:11:50 +02:00
/// select that processed
SELECT_LEX *select_lex;
2007-10-16 22:11:50 +02:00
/**
TRUE <=> optimizer must not mark any table as a constant table.
This is needed for subqueries in form "a IN (SELECT .. UNION SELECT ..):
when we optimize the select that reads the results of the union from a
temporary table, we must not mark the temp. table as constant because
the number of rows in it may vary from one subquery execution to another.
*/
bool no_const_tables;
/**
Copy of this JOIN to be used with temporary tables.
tmp_join is used when the JOIN needs to be "reusable" (e.g. in a subquery
that gets re-executed several times) and we know will use temporary tables
for materialization. The materialization to a temporary table overwrites the
JOIN structure to point to the temporary table after the materialization is
done. This is where tmp_join is used : it's a copy of the JOIN before the
materialization and is used in restoring before re-execution by overwriting
the current JOIN structure with the saved copy.
Because of this we should pay extra care of not freeing up helper structures
that are referenced by the original contents of the JOIN. We can check for
this by making sure the "current" join is not the temporary copy, e.g.
!tmp_join || tmp_join != join
We should free these sub-structures at JOIN::destroy() if the "current" join
has a copy is not that copy.
*/
JOIN *tmp_join;
2007-10-16 22:11:50 +02:00
ROLLUP rollup; ///< Used with rollup
2007-10-16 22:11:50 +02:00
bool select_distinct; ///< Set if SELECT DISTINCT
/**
If we have the GROUP BY statement in the query,
but the group_list was emptied by optimizer, this
flag is TRUE.
It happens when fields in the GROUP BY are from
constant table
*/
bool group_optimized_away;
/*
simple_xxxxx is set if ORDER/GROUP BY doesn't include any references
to other tables than the first non-constant table in the JOIN.
It's also set if ORDER/GROUP BY is empty.
Used for deciding for or against using a temporary table to compute
GROUP/ORDER BY.
*/
bool simple_order, simple_group;
2007-10-16 22:11:50 +02:00
/**
Is set only in case if we have a GROUP BY clause
and no ORDER BY after constant elimination of 'order'.
*/
bool no_order;
2007-10-16 22:11:50 +02:00
/** Is set if we have a GROUP BY and we have ORDER BY on a constant. */
bool skip_sort_order;
bool need_tmp, hidden_group_fields;
DYNAMIC_ARRAY keyuse;
Item::cond_result cond_value, having_value;
2007-10-16 22:11:50 +02:00
List<Item> all_fields; ///< to store all fields that used in query
///Above list changed to use temporary table
List<Item> tmp_all_fields1, tmp_all_fields2, tmp_all_fields3;
2007-10-16 22:11:50 +02:00
///Part, shared with list above, emulate following list
List<Item> tmp_fields_list1, tmp_fields_list2, tmp_fields_list3;
2007-10-16 22:11:50 +02:00
List<Item> &fields_list; ///< hold field list passed to mysql_select
List<Item> procedure_fields_list;
int error;
ORDER *order, *group_list, *proc_param; //hold parameters of mysql_select
COND *conds; // ---"---
Item *conds_history; // store WHERE for explain
2007-10-16 22:11:50 +02:00
TABLE_LIST *tables_list; ///<hold 'tables' parameter of mysql_select
List<TABLE_LIST> *join_list; ///< list of joined tables in reverse order
COND_EQUAL *cond_equal;
2007-10-16 22:11:50 +02:00
SQL_SELECT *select; ///<created in optimisation phase
JOIN_TAB *return_tab; ///<used only for outer joins
Item **ref_pointer_array; ///<used pointer reference for this select
// Copy of above to be used with different lists
Item **items0, **items1, **items2, **items3, **current_ref_pointer_array;
2007-10-16 22:11:50 +02:00
uint ref_pointer_array_size; ///< size of above in bytes
const char *zero_result_cause; ///< not 0 if exec must return zero result
2002-05-26 21:50:32 +02:00
2007-10-16 22:11:50 +02:00
bool union_part; ///< this subselect is part of union
bool optimized; ///< flag to avoid double optimization in EXPLAIN
/*
storage for caching buffers allocated during query execution.
These buffers allocations need to be cached as the thread memory pool is
cleared only at the end of the execution of the whole query and not caching
allocations that occur in repetition at execution time will result in
excessive memory usage.
Note: make_simple_join always creates an execution plan that accesses
a single table, thus it is sufficient to have a one-element array for
table_reexec.
*/
SORT_FIELD *sortorder; // make_unireg_sortorder()
TABLE *table_reexec[1]; // make_simple_join()
JOIN_TAB *join_tab_reexec; // make_simple_join()
/* end of allocation caching storage */
JOIN(THD *thd_arg, List<Item> &fields_arg, ulonglong select_options_arg,
select_result *result_arg)
:fields_list(fields_arg)
{
init(thd_arg, fields_arg, select_options_arg, result_arg);
}
void init(THD *thd_arg, List<Item> &fields_arg, ulonglong select_options_arg,
select_result *result_arg)
{
join_tab= join_tab_save= 0;
all_tables= 0;
tables= 0;
const_tables= 0;
join_list= 0;
implicit_grouping= FALSE;
sort_and_group= 0;
first_record= 0;
do_send_rows= 1;
send_records= 0;
found_records= 0;
fetch_limit= HA_POS_ERROR;
examined_rows= 0;
exec_tmp_table1= 0;
exec_tmp_table2= 0;
sortorder= 0;
table_reexec[0]= 0;
join_tab_reexec= 0;
thd= thd_arg;
2003-07-07 17:49:54 +02:00
sum_funcs= sum_funcs2= 0;
procedure= 0;
having= tmp_having= having_history= 0;
select_options= select_options_arg;
result= result_arg;
lock= thd_arg->lock;
select_lex= 0; //for safety
tmp_join= 0;
select_distinct= test(select_options & SELECT_DISTINCT);
no_order= 0;
simple_order= 0;
simple_group= 0;
skip_sort_order= 0;
need_tmp= 0;
hidden_group_fields= 0; /*safety*/
error= 0;
select= 0;
return_tab= 0;
ref_pointer_array= items0= items1= items2= items3= 0;
ref_pointer_array_size= 0;
zero_result_cause= 0;
optimized= 0;
cond_equal= 0;
group_optimized_away= 0;
2004-11-24 10:49:00 +01:00
all_fields= fields_arg;
if (&fields_list != &fields_arg) /* Avoid valgrind-warning */
fields_list= fields_arg;
bzero((char*) &keyuse,sizeof(keyuse));
tmp_table_param.init();
tmp_table_param.end_write_records= HA_POS_ERROR;
rollup.state= ROLLUP::STATE_NONE;
no_const_tables= FALSE;
}
int prepare(Item ***rref_pointer_array, TABLE_LIST *tables, uint wind_num,
COND *conds, uint og_num, ORDER *order, ORDER *group,
Item *having, ORDER *proc_param, SELECT_LEX *select,
SELECT_LEX_UNIT *unit);
int optimize();
2002-05-26 21:50:32 +02:00
int reinit();
void exec();
int destroy();
void restore_tmp();
bool alloc_func_list();
bool make_sum_func_list(List<Item> &all_fields, List<Item> &send_fields,
bool before_group_by, bool recompute= FALSE);
inline void set_items_ref_array(Item **ptr)
{
memcpy((char*) ref_pointer_array, (char*) ptr, ref_pointer_array_size);
current_ref_pointer_array= ptr;
}
inline void init_items_ref_array()
{
items0= ref_pointer_array + all_fields.elements;
memcpy(items0, ref_pointer_array, ref_pointer_array_size);
current_ref_pointer_array= items0;
}
bool rollup_init();
bool rollup_process_const_fields();
bool rollup_make_fields(List<Item> &all_fields, List<Item> &fields,
Item_sum ***func);
int rollup_send_data(uint idx);
int rollup_write_data(uint idx, TABLE *table);
void remove_subq_pushed_predicates(Item **where);
2007-10-16 22:11:50 +02:00
/**
Release memory and, if possible, the open tables held by this execution
plan (and nested plans). It's used to release some tables before
the end of execution in order to increase concurrency and reduce
memory consumption.
*/
void join_free();
2007-10-16 22:11:50 +02:00
/** Cleanup this JOIN, possibly for reuse */
void cleanup(bool full);
void clear();
bool save_join_tab();
bool init_save_join_tab();
bool send_row_on_empty_set()
{
return (do_send_rows && tmp_table_param.sum_func_count != 0 &&
!group_list && having_value != Item::COND_FALSE);
}
bool change_result(select_result *result);
bool is_top_level_join() const
{
return (unit == &thd->lex->unit && (unit->fake_select_lex == 0 ||
select_lex == unit->fake_select_lex));
}
void cache_const_exprs();
private:
/**
TRUE if the query contains an aggregate function but has no GROUP
BY clause.
*/
bool implicit_grouping;
bool make_simple_join(JOIN *join, TABLE *tmp_table);
void cleanup_item_list(List<Item> &items) const;
2000-07-31 21:29:14 +02:00
};
typedef struct st_select_check {
uint const_ref,reg_ref;
} SELECT_CHECK;
extern const char *join_type_str[];
/* Extern functions in sql_select.cc */
bool store_val_in_field(Field *field, Item *val, enum_check_fields check_flag);
2000-07-31 21:29:14 +02:00
TABLE *create_tmp_table(THD *thd,TMP_TABLE_PARAM *param,List<Item> &fields,
ORDER *group, bool distinct, bool save_sum_fields,
ulonglong select_options, ha_rows rows_limit,
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
const char* alias);
2000-07-31 21:29:14 +02:00
void free_tmp_table(THD *thd, TABLE *entry);
void count_field_types(SELECT_LEX *select_lex, TMP_TABLE_PARAM *param,
List<Item> &fields, bool reset_with_sum_func);
bool setup_copy_fields(THD *thd, TMP_TABLE_PARAM *param,
Item **ref_pointer_array,
List<Item> &new_list1, List<Item> &new_list2,
uint elements, List<Item> &fields);
2000-07-31 21:29:14 +02:00
void copy_fields(TMP_TABLE_PARAM *param);
bool copy_funcs(Item **func_ptr, const THD *thd);
bool create_myisam_from_heap(THD *thd, TABLE *table, TMP_TABLE_PARAM *param,
int error, bool ignore_last_dupp_error);
2003-11-21 22:20:48 +01:00
uint find_shortest_key(TABLE *table, const key_map *usable_keys);
2005-03-01 21:19:19 +01:00
Field* create_tmp_field_from_field(THD *thd, Field* org_field,
2005-03-30 15:57:56 +02:00
const char *name, TABLE *table,
Item_field *item, uint convert_blob_length);
2009-09-28 09:21:25 +02:00
bool is_indexed_agg_distinct(JOIN *join, List<Item_field> *out_args);
2000-07-31 21:29:14 +02:00
/* functions from opt_sum.cc */
bool simple_pred(Item_func *func_item, Item **args, bool *inv_order);
int opt_sum_query(THD* thd,
TABLE_LIST *tables, List<Item> &all_fields, COND *conds);
2000-07-31 21:29:14 +02:00
/* from sql_delete.cc, used by opt_range.cc */
extern "C" int refpos_order_cmp(void* arg, const void *a,const void *b);
2007-10-16 22:11:50 +02:00
/** class to copying an field/item to a key struct */
2000-07-31 21:29:14 +02:00
class store_key :public Sql_alloc
{
public:
bool null_key; /* TRUE <=> the value of the key has a null part */
enum store_key_result { STORE_KEY_OK, STORE_KEY_FATAL, STORE_KEY_CONV };
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
store_key(THD *thd, Field *field_arg, uchar *ptr, uchar *null, uint length)
:null_key(0), null_ptr(null), err(0)
2000-07-31 21:29:14 +02:00
{
if (field_arg->type() == MYSQL_TYPE_BLOB
|| field_arg->type() == MYSQL_TYPE_GEOMETRY)
{
/*
Key segments are always packed with a 2 byte length prefix.
See mi_rkey for details.
*/
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
to_field= new Field_varstring(ptr, length, 2, null, 1,
Field::NONE, field_arg->field_name,
field_arg->table->s, field_arg->charset());
to_field->init(field_arg->table);
}
2000-07-31 21:29:14 +02:00
else
to_field=field_arg->new_key_field(thd->mem_root, field_arg->table,
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
ptr, null, 1);
2000-07-31 21:29:14 +02:00
}
2007-10-16 22:11:50 +02:00
virtual ~store_key() {} /** Not actually needed */
2000-07-31 21:29:14 +02:00
virtual const char *name() const=0;
/**
@brief sets ignore truncation warnings mode and calls the real copy method
@details this function makes sure truncation warnings when preparing the
key buffers don't end up as errors (because of an enclosing INSERT/UPDATE).
*/
enum store_key_result copy()
{
enum store_key_result result;
THD *thd= to_field->table->in_use;
enum_check_fields saved_count_cuted_fields= thd->count_cuted_fields;
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
ulonglong sql_mode= thd->variables.sql_mode;
thd->variables.sql_mode&= ~(MODE_NO_ZERO_IN_DATE | MODE_NO_ZERO_DATE);
thd->count_cuted_fields= CHECK_FIELD_IGNORE;
result= copy_inner();
thd->count_cuted_fields= saved_count_cuted_fields;
thd->variables.sql_mode= sql_mode;
return result;
}
protected:
Field *to_field; // Store data here
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 *null_ptr;
uchar err;
virtual enum store_key_result copy_inner()=0;
2000-07-31 21:29:14 +02:00
};
class store_key_field: public store_key
{
Copy_field copy_field;
const char *field_name;
public:
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
store_key_field(THD *thd, Field *to_field_arg, uchar *ptr,
uchar *null_ptr_arg,
2000-07-31 21:29:14 +02:00
uint length, Field *from_field, const char *name_arg)
:store_key(thd, to_field_arg,ptr,
2000-07-31 21:29:14 +02:00
null_ptr_arg ? null_ptr_arg : from_field->maybe_null() ? &err
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*) 0, length), field_name(name_arg)
2000-07-31 21:29:14 +02:00
{
if (to_field)
{
copy_field.set(to_field,from_field,0);
}
}
const char *name() const { return field_name; }
protected:
enum store_key_result copy_inner()
{
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
TABLE *table= copy_field.to_field->table;
my_bitmap_map *old_map= dbug_tmp_use_all_columns(table,
table->write_set);
copy_field.do_copy(&copy_field);
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
dbug_tmp_restore_column_map(table->write_set, old_map);
null_key= to_field->is_null();
return err != 0 ? STORE_KEY_FATAL : STORE_KEY_OK;
}
2000-07-31 21:29:14 +02:00
};
class store_key_item :public store_key
{
protected:
Item *item;
public:
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
store_key_item(THD *thd, Field *to_field_arg, uchar *ptr,
uchar *null_ptr_arg, uint length, Item *item_arg)
:store_key(thd, to_field_arg, ptr,
2000-07-31 21:29:14 +02:00
null_ptr_arg ? null_ptr_arg : item_arg->maybe_null ?
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
&err : (uchar*) 0, length), item(item_arg)
2000-07-31 21:29:14 +02:00
{}
const char *name() const { return "func"; }
protected:
enum store_key_result copy_inner()
2000-07-31 21:29:14 +02:00
{
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
TABLE *table= to_field->table;
my_bitmap_map *old_map= dbug_tmp_use_all_columns(table,
table->write_set);
int res= item->save_in_field(to_field, 1);
/*
Item::save_in_field() may call Item::val_xxx(). And if this is a subquery
we need to check for errors executing it and react accordingly
*/
if (!res && table->in_use->is_error())
res= 1; /* STORE_KEY_FATAL */
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
dbug_tmp_restore_column_map(table->write_set, old_map);
null_key= to_field->is_null() || item->null_value;
return ((err != 0 || res < 0 || res > 2) ? STORE_KEY_FATAL :
(store_key_result) res);
2000-07-31 21:29:14 +02:00
}
};
class store_key_const_item :public store_key_item
{
bool inited;
public:
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
store_key_const_item(THD *thd, Field *to_field_arg, uchar *ptr,
uchar *null_ptr_arg, uint length,
2000-07-31 21:29:14 +02:00
Item *item_arg)
:store_key_item(thd, to_field_arg,ptr,
2000-07-31 21:29:14 +02:00
null_ptr_arg ? null_ptr_arg : item_arg->maybe_null ?
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
&err : (uchar*) 0, length, item_arg), inited(0)
2000-07-31 21:29:14 +02:00
{
}
const char *name() const { return "const"; }
protected:
enum store_key_result copy_inner()
2000-07-31 21:29:14 +02:00
{
int res;
2000-07-31 21:29:14 +02:00
if (!inited)
{
inited=1;
if ((res= item->save_in_field(to_field, 1)))
{
if (!err)
err= res < 0 ? 1 : res; /* 1=STORE_KEY_FATAL */
}
/*
Item::save_in_field() may call Item::val_xxx(). And if this is a subquery
we need to check for errors executing it and react accordingly
*/
if (!err && to_field->table->in_use->is_error())
err= 1; /* STORE_KEY_FATAL */
2000-07-31 21:29:14 +02:00
}
null_key= to_field->is_null() || item->null_value;
return (err > 2 ? STORE_KEY_FATAL : (store_key_result) err);
2000-07-31 21:29:14 +02:00
}
};
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
bool cp_buffer_from_ref(THD *thd, TABLE *table, TABLE_REF *ref);
bool error_if_full_join(JOIN *join);
int report_error(TABLE *table, int error);
int safe_index_read(JOIN_TAB *tab);
COND *remove_eq_conds(THD *thd, COND *cond, Item::cond_result *cond_value);
int get_quick_record(SQL_SELECT *select);
SORT_FIELD * make_unireg_sortorder(ORDER *order, uint *length,
SORT_FIELD *sortorder);
int setup_order(THD *thd, Item **ref_pointer_array, TABLE_LIST *tables,
List<Item> &fields, List <Item> &all_fields, ORDER *order);
int setup_group(THD *thd, Item **ref_pointer_array, TABLE_LIST *tables,
List<Item> &fields, List<Item> &all_fields, ORDER *order,
bool *hidden_group_fields);
bool fix_inner_refs(THD *thd, List<Item> &all_fields, SELECT_LEX *select,
Item **ref_pointer_array, ORDER *group_list= NULL);
bool handle_select(THD *thd, LEX *lex, select_result *result,
ulong setup_tables_done_option);
bool mysql_select(THD *thd, Item ***rref_pointer_array,
TABLE_LIST *tables, uint wild_num, List<Item> &list,
COND *conds, uint og_num, ORDER *order, ORDER *group,
Item *having, ORDER *proc_param, ulonglong select_type,
select_result *result, SELECT_LEX_UNIT *unit,
SELECT_LEX *select_lex);
void free_underlaid_joins(THD *thd, SELECT_LEX *select);
bool mysql_explain_union(THD *thd, SELECT_LEX_UNIT *unit,
select_result *result);
Field *create_tmp_field(THD *thd, TABLE *table,Item *item, Item::Type type,
Item ***copy_func, Field **from_field,
Field **def_field,
bool group, bool modify_item,
bool table_cant_handle_bit_fields,
bool make_copy_field,
uint convert_blob_length);
/*
General routine to change field->ptr of a NULL-terminated array of Field
objects. Useful when needed to call val_int, val_str or similar and the
field data is not in table->record[0] but in some other structure.
set_key_field_ptr changes all fields of an index using a key_info object.
All methods presume that there is at least one field to change.
*/
TABLE *create_virtual_tmp_table(THD *thd, List<Create_field> &field_list);
inline bool optimizer_flag(THD *thd, uint flag)
{
return (thd->variables.optimizer_switch & flag);
}
Bug #30584: delete with order by and limit clauses does not use limit efficiently Bug #36569: UPDATE ... WHERE ... ORDER BY... always does a filesort even if not required Also two bugs reported after QA review (before the commit of bugs above to public trees, no documentation needed): Bug #53737: Performance regressions after applying patch for bug 36569 Bug #53742: UPDATEs have no effect after applying patch for bug 36569 Execution of single-table UPDATE and DELETE statements did not use the same optimizer as was used in the compilation of SELECT statements. Instead, it had an optimizer of its own that did not take into account that you can omit sorting by retrieving rows using an index. Extra optimization has been added: when applicable, single-table UPDATE/DELETE statements use an existing index instead of filesort. A corresponding SELECT query would do the former. Also handling of the DESC ordering expression has been added when reverse index scan is applicable. From now on most single table UPDATE and DELETE statements show the same disk access patterns as the corresponding SELECT query. We verify this by comparing the result of SHOW STATUS LIKE 'Sort% Currently the get_index_for_order function a) checks quick select index (if any) for compatibility with the ORDER expression list or b) chooses the cheapest available compatible index, but only if the index scan is cheaper than filesort. Second way is implemented by the new test_if_cheaper_ordering function (extracted part the test_if_skip_sort_order()).
2010-06-22 22:32:29 +02:00
uint get_index_for_order(ORDER *order, TABLE *table, SQL_SELECT *select,
ha_rows limit, bool *need_sort, bool *reverse);
ORDER *simple_remove_const(ORDER *order, COND *where);
bool const_expression_in_where(COND *cond, Item *comp_item,
Field *comp_field= NULL,
Item **const_item= NULL);
#endif /* SQL_SELECT_INCLUDED */