database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-08-01 03:47:19 +03:00
Code Activity

MDEV-32113: utf8mb3_key_col=utf8mb4_value cannot be used for ref

Browse Source 
(Variant#3: Allow cross-charset comparisons, use a special
CHARSET_INFO to create lookup keys. Review input addressed.)

Equalities that compare utf8mb{3,4}_general_ci strings, like:

  WHERE ... utf8mb3_key_col=utf8mb4_value    (MB3-4-CMP)

can now be used to construct ref[const] access and also participate
in multiple-equalities.
This means that utf8mb3_key_col can be used for key-lookups when
compared with an utf8mb4 constant, field or expression using '=' or
'<=>' comparison operators.

This is controlled by optimizer_switch='cset_narrowing=on', which is
OFF by default.

IMPLEMENTATION
Item value comparison in (MB3-4-CMP) is done using utf8mb4_general_ci.
This is valid as any utf8mb3 value is also an utf8mb4 value.

When making index lookup value for utf8mb3_key_col, we do "Charset
Narrowing": characters that are in the Basic Multilingual Plane (=BMP) are
copied as-is, as they can be represented in utf8mb3. Characters that are
outside the BMP cannot be represented in utf8mb3 and are replaced
with U+FFFD, the "Replacement Character".

In utf8mb4_general_ci, the Replacement Character compares as equal to any
character that's not in BMP. Because of this, the constructed lookup value
will find all index records that would be considered equal by the original
condition (MB3-4-CMP).

Approved-by: Monty <monty@mariadb.org>
This commit is contained in:
Sergei Petrunia
2023-09-19 18:22:49 +03:00
parent 6a674c3142
commit 4941ac9192
23 changed files with 1001 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
sql/sql_select.h
View File

@ -34,6 +34,8 @@
#include "opt_range.h" /* SQL_SELECT, QUICK_SELECT_I */
#include "filesort.h"
#include "cset_narrowing.h"
typedef struct st_join_table JOIN_TAB;
/* Values in optimize */
#define KEY_OPTIMIZE_EXISTS 1U
@ -1921,7 +1923,14 @@ public:
{
enum_check_fields org_count_cuted_fields= thd->count_cuted_fields;
Use_relaxed_field_copy urfc(to_field->table->in_use);
/* If needed, perform CharsetNarrowing for making ref access lookup keys. */
Utf8_narrow do_narrow(to_field, do_cset_narrowing);
store_key_result result= copy_inner();
do_narrow.stop();
thd->count_cuted_fields= org_count_cuted_fields;
return result;
}
@ -1931,6 +1940,12 @@ public:
uchar *null_ptr;
uchar err;
/*
This is set to true if we need to do Charset Narrowing when making a lookup
key.
*/
bool do_cset_narrowing= false;
virtual enum store_key_result copy_inner()=0;
};
@ -1950,6 +1965,7 @@ class store_key_field: public store_key
if (to_field)
{
copy_field.set(to_field,from_field,0);
setup_charset_narrowing();
}
}
@ -1960,6 +1976,15 @@ class store_key_field: public store_key
{
copy_field.set(to_field, fld_item->field, 0);
field_name= fld_item->full_name();
setup_charset_narrowing();
}
/* Setup CharsetNarrowing if necessary */
void setup_charset_narrowing()
{
do_cset_narrowing=
Utf8_narrow::should_do_narrowing(copy_field.to_field,
copy_field.from_field->charset());
}
protected:
@ -2000,7 +2025,12 @@ public:
:store_key(thd, to_field_arg, ptr,
null_ptr_arg ? null_ptr_arg : item_arg->maybe_null() ?
&err : (uchar*) 0, length), item(item_arg), use_value(val)
{}
{
/* Setup CharsetNarrowing to be done if necessary */
do_cset_narrowing=
Utf8_narrow::should_do_narrowing(to_field,
item->collation.collation);
}
store_key_item(store_key &arg, Item *new_item, bool val)
:store_key(arg), item(new_item), use_value(val)
{}
@ -2388,7 +2418,7 @@ Item_equal *find_item_equal(COND_EQUAL *cond_equal, Field *field,
extern bool test_if_ref(Item *,
Item_field *left_item,Item *right_item);
inline bool optimizer_flag(THD *thd, ulonglong flag)
inline bool optimizer_flag(const THD *thd, ulonglong flag)
{
return (thd->variables.optimizer_switch & flag);
}