1
0
mirror of https://github.com/MariaDB/server.git synced 2025-07-11 15:22:09 +03:00

A joint patch for MDEV-19284 and MDEV-19285 (INSTANT ALTER)

This patch fixes:

- MDEV-19284 INSTANT ALTER with ucs2-to-utf16 conversion produces bad data
- MDEV-19285 INSTANT ALTER from ascii_general_ci to latin1_general_ci produces corrupt data

These regressions were introduced in 10.4.3 by:
- MDEV-15564 Avoid table rebuild in ALTER TABLE on collation or charset changes

Changes:

1. Cleanup: Adding a helper method
   Field_longstr::csinfo_change_allows_instant_alter(),
   to remove some duplicate code in field.cc.

2. Cleanup: removing Type_handler::Charsets_are_compatible() and static
   function charsets_are_compatible() and
   introducing new methods in the recently added class Charset instead:
   - encoding_allows_reinterpret_as()
   - encoding_and_order_allow_reinterpret_as()

3. Bug fix: Removing the code that allowed instant conversion for
   ascii-to->8bit and ucs2-to->utf16.
   This actually fixes MDEV-19284 and MDEV-19285.

4. Bug fix: Adding a helper method Charset::collation_specific_name().
   The old corresponding code in Type_handler::Charsets_are_compatible()
   was not safe against (badly named) user-defined collations whose
   character set name can be longer than collation name.
This commit is contained in:
Alexander Barkov
2019-04-19 15:18:38 +04:00
parent 9aa80fcf46
commit c59d6395a6
8 changed files with 925 additions and 847 deletions

View File

@ -1,10 +1,10 @@
--- instant_alter_charset.result --- instant_alter_charset.result 2019-04-23 17:42:23.324326518 +0400
+++ instant_alter_charset,redundant.result +++ instant_alter_charset,redundant.result 2019-04-23 17:42:46.047531591 +0400
@@ -254,7 +254,6 @@ @@ -279,7 +279,6 @@
alter table boundary_255 alter table boundary_255
modify b varchar(200) charset utf8mb3, modify a varchar(70) charset utf8mb4,
algorithm=instant; algorithm=instant;
-ERROR 0A000: ALGORITHM=INSTANT is not supported. Reason: Cannot change column type. Try ALGORITHM=COPY -ERROR 0A000: ALGORITHM=INSTANT is not supported. Reason: Cannot change column type. Try ALGORITHM=COPY
alter table boundary_255 drop table boundary_255;
modify c varchar(300) charset utf8mb3, create table fully_compatible (
algorithm=instant; id int auto_increment unique key,

File diff suppressed because it is too large Load Diff

View File

@ -20,9 +20,13 @@ select c.prtype, c.len from information_schema.innodb_sys_columns as c inner joi
alter table no_rebuild alter table no_rebuild
change a a char(150) charset utf8mb3 collate utf8mb3_spanish_ci, change a a char(150) charset utf8mb3 collate utf8mb3_spanish_ci,
algorithm=inplace; algorithm=inplace;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table rebuild alter table rebuild
change a a varchar(150) charset latin1 not null default 'asdf', change a a varchar(150) charset latin1 not null default 'asdf',
algorithm=inplace; algorithm=inplace;
alter table rebuild
change a a varchar(150) charset latin1 not null default 'asdf',
algorithm=copy;
select name, prtype, len from information_schema.innodb_sys_columns select name, prtype, len from information_schema.innodb_sys_columns
where table_id = @id; where table_id = @id;
select c.prtype, c.len from information_schema.innodb_sys_columns as c inner join information_schema.innodb_sys_tables t on c.table_id = t.table_id select c.prtype, c.len from information_schema.innodb_sys_columns as c inner join information_schema.innodb_sys_tables t on c.table_id = t.table_id
@ -41,7 +45,7 @@ create table supported_types (
) engine=innodb; ) engine=innodb;
alter table supported_types alter table supported_types
convert to charset latin1, convert to charset ascii collate ascii_bin,
algorithm=instant; algorithm=instant;
drop table supported_types; drop table supported_types;
@ -53,7 +57,7 @@ create table various_cases (
) engine=innodb; ) engine=innodb;
alter table various_cases alter table various_cases
change a a char(150) charset latin1, change a a char(150) charset ascii collate ascii_bin,
algorithm=inplace; algorithm=inplace;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON --error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
@ -88,7 +92,7 @@ create table all_texts (
) engine=innodb; ) engine=innodb;
alter table all_texts alter table all_texts
convert to charset latin1 collate latin1_general_ci, convert to charset ascii collate ascii_bin,
algorithm=instant; algorithm=instant;
drop table all_texts; drop table all_texts;
@ -244,10 +248,15 @@ alter table latin1_swedish_special_case
modify copy1 varchar(150) charset latin1 collate latin1_swedish_ci, modify copy1 varchar(150) charset latin1 collate latin1_swedish_ci,
modify copy2 char(150) charset latin1 collate latin1_swedish_ci, modify copy2 char(150) charset latin1 collate latin1_swedish_ci,
algorithm=copy; algorithm=copy;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table latin1_swedish_special_case alter table latin1_swedish_special_case
modify instant1 varchar(150) charset latin1 collate latin1_swedish_ci, modify instant1 varchar(150) charset latin1 collate latin1_swedish_ci,
modify instant2 char(150) charset latin1 collate latin1_swedish_ci, modify instant2 char(150) charset latin1 collate latin1_swedish_ci,
algorithm=instant; algorithm=instant;
alter table latin1_swedish_special_case
modify instant1 varchar(150) charset latin1 collate latin1_swedish_ci,
modify instant2 char(150) charset latin1 collate latin1_swedish_ci,
algorithm=copy;
select c.name, c.prtype, c.mtype, c.len from information_schema.innodb_sys_columns as c inner join information_schema.innodb_sys_tables t on c.table_id = t.table_id select c.name, c.prtype, c.mtype, c.len from information_schema.innodb_sys_columns as c inner join information_schema.innodb_sys_tables t on c.table_id = t.table_id
where t.name = 'test/latin1_swedish_special_case'; where t.name = 'test/latin1_swedish_special_case';
alter table latin1_swedish_special_case alter table latin1_swedish_special_case
@ -275,26 +284,42 @@ create table boundary_255 (
c varchar(300) charset ascii c varchar(300) charset ascii
) engine=innodb; ) engine=innodb;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table boundary_255 alter table boundary_255
modify a varchar(50) charset utf8mb3, modify a varchar(50) charset utf8mb3,
algorithm=instant; algorithm=instant;
alter table boundary_255
modify a varchar(50) charset utf8mb3,
algorithm=copy;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table boundary_255
modify b varchar(200) charset utf8mb3,
algorithm=instant;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table boundary_255
modify c varchar(300) charset utf8mb3,
algorithm=instant;
drop table boundary_255;
create table boundary_255 (
a varchar(70) charset utf8mb3
) engine=innodb;
if ($row_format == 'redundant') { if ($row_format == 'redundant') {
alter table boundary_255 alter table boundary_255
modify b varchar(200) charset utf8mb3, modify a varchar(70) charset utf8mb4,
algorithm=instant; algorithm=instant;
} }
if ($row_format != 'redundant') { if ($row_format != 'redundant') {
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON --error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
alter table boundary_255 alter table boundary_255
modify b varchar(200) charset utf8mb3, modify a varchar(70) charset utf8mb4,
algorithm=instant; algorithm=instant;
} }
alter table boundary_255
modify c varchar(300) charset utf8mb3,
algorithm=instant;
drop table boundary_255; drop table boundary_255;
create table fully_compatible ( create table fully_compatible (
@ -338,30 +363,7 @@ insert into fully_compatible (from_charset, from_collate, to_charset, to_collate
('utf8mb3', 'utf8mb3_general_nopad_ci', 'utf8mb4', 'utf8mb4_general_nopad_ci'), ('utf8mb3', 'utf8mb3_general_nopad_ci', 'utf8mb4', 'utf8mb4_general_nopad_ci'),
('utf8mb3', 'utf8mb3_nopad_bin', 'utf8mb4', 'utf8mb4_nopad_bin'), ('utf8mb3', 'utf8mb3_nopad_bin', 'utf8mb4', 'utf8mb4_nopad_bin'),
('utf8mb3', 'utf8mb3_unicode_nopad_ci', 'utf8mb4', 'utf8mb4_unicode_nopad_ci'), ('utf8mb3', 'utf8mb3_unicode_nopad_ci', 'utf8mb4', 'utf8mb4_unicode_nopad_ci'),
('utf8mb3', 'utf8mb3_unicode_520_nopad_ci', 'utf8mb4', 'utf8mb4_unicode_520_nopad_ci'), ('utf8mb3', 'utf8mb3_unicode_520_nopad_ci', 'utf8mb4', 'utf8mb4_unicode_520_nopad_ci')
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_general_ci'),
('ucs2', 'ucs2_unicode_ci', 'utf16', 'utf16_unicode_ci'),
('ucs2', 'ucs2_icelandic_ci', 'utf16', 'utf16_icelandic_ci'),
('ucs2', 'ucs2_latvian_ci', 'utf16', 'utf16_latvian_ci'),
('ucs2', 'ucs2_romanian_ci', 'utf16', 'utf16_romanian_ci'),
('ucs2', 'ucs2_slovenian_ci', 'utf16', 'utf16_slovenian_ci'),
('ucs2', 'ucs2_polish_ci', 'utf16', 'utf16_polish_ci'),
('ucs2', 'ucs2_estonian_ci', 'utf16', 'utf16_estonian_ci'),
('ucs2', 'ucs2_spanish_ci', 'utf16', 'utf16_spanish_ci'),
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_general_ci'),
('ascii', 'ascii_general_ci', 'utf8mb3', 'utf8mb3_general_ci'),
('ascii', 'ascii_general_ci', 'utf8mb4', 'utf8mb4_general_ci'),
('ascii', 'ascii_general_ci', 'latin1', 'latin1_general_ci'),
('ascii', 'ascii_bin', 'latin1', 'latin1_bin'),
('ascii', 'ascii_nopad_bin', 'latin1', 'latin1_nopad_bin'),
('ascii', 'ascii_general_ci', 'latin2', 'latin2_general_ci'),
('ascii', 'ascii_general_ci', 'latin7', 'latin7_general_ci'),
('ascii', 'ascii_bin', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_bin', 'ujis', 'ujis_bin'),
('ascii', 'ascii_bin', 'big5', 'big5_bin'),
('ascii', 'ascii_bin', 'gbk', 'gbk_bin')
; ;
let $data_size = `select count(*) from fully_compatible`; let $data_size = `select count(*) from fully_compatible`;
@ -404,47 +406,19 @@ create table compatible_without_index (
); );
insert into compatible_without_index (from_charset, from_collate, to_charset, to_collate) values insert into compatible_without_index (from_charset, from_collate, to_charset, to_collate) values
('ascii', 'ascii_general_ci', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_bin', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_general_nopad_ci', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_nopad_bin', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_general_ci', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_general_nopad_ci', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_nopad_bin', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_general_ci', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_bin', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_general_nopad_ci', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_nopad_bin', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_general_ci', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_bin', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_general_nopad_ci', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_nopad_bin', 'utf8mb4', 'utf8mb4_danish_ci'),
('utf8mb3', 'utf8mb3_general_ci', 'utf8mb4', 'utf8mb4_vietnamese_ci'), ('utf8mb3', 'utf8mb3_general_ci', 'utf8mb4', 'utf8mb4_vietnamese_ci'),
('utf8mb3', 'utf8mb3_bin', 'utf8mb4', 'utf8mb4_vietnamese_ci'), ('utf8mb3', 'utf8mb3_bin', 'utf8mb4', 'utf8mb4_vietnamese_ci'),
('utf8mb3', 'utf8mb3_general_nopad_ci', 'utf8mb4', 'utf8mb4_vietnamese_ci'), ('utf8mb3', 'utf8mb3_general_nopad_ci', 'utf8mb4', 'utf8mb4_vietnamese_ci'),
('utf8mb3', 'utf8mb3_nopad_bin', 'utf8mb4', 'utf8mb4_vietnamese_ci'), ('utf8mb3', 'utf8mb3_nopad_bin', 'utf8mb4', 'utf8mb4_vietnamese_ci'),
('ascii', 'ascii_general_ci', 'gbk', 'gbk_chinese_ci'),
('ascii', 'ascii_general_ci', 'gbk', 'gbk_chinese_nopad_ci'),
('ucs2', 'ucs2_myanmar_ci', 'utf16', 'utf16_thai_520_w2'),
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_unicode_nopad_ci'),
('ucs2', 'ucs2_general_mysql500_ci', 'utf16', 'utf16_spanish2_ci'),
('ascii', 'ascii_general_ci', 'ascii', 'ascii_bin'), ('ascii', 'ascii_general_ci', 'ascii', 'ascii_bin'),
('utf8mb3', 'utf8mb3_roman_ci', 'utf8mb3', 'utf8mb3_lithuanian_ci'), ('utf8mb3', 'utf8mb3_roman_ci', 'utf8mb3', 'utf8mb3_lithuanian_ci'),
('utf8mb4', 'utf8mb4_thai_520_w2', 'utf8mb4', 'utf8mb4_persian_ci'), ('utf8mb4', 'utf8mb4_thai_520_w2', 'utf8mb4', 'utf8mb4_persian_ci'),
('utf8mb3', 'utf8mb3_myanmar_ci', 'utf8mb4', 'utf8mb4_german2_ci'), ('utf8mb3', 'utf8mb3_myanmar_ci', 'utf8mb4', 'utf8mb4_german2_ci'),
('utf8mb3', 'utf8mb3_general_ci', 'utf8mb3', 'utf8mb3_unicode_ci'), ('utf8mb3', 'utf8mb3_general_ci', 'utf8mb3', 'utf8mb3_unicode_ci'),
('latin1', 'latin1_general_cs', 'latin1', 'latin1_general_ci'), ('latin1', 'latin1_general_cs', 'latin1', 'latin1_general_ci'),
('ascii', 'ascii_general_ci', 'ujis', 'ujis_japanese_ci'),
('ascii', 'ascii_general_ci', 'big5', 'big5_chinese_ci'),
('ascii', 'ascii_general_ci', 'latin2', 'latin2_croatian_ci'),
('ascii', 'ascii_general_ci', 'latin7', 'latin7_estonian_cs'),
('utf16', 'utf16_general_ci', 'utf16', 'utf16_german2_ci') ('utf16', 'utf16_general_ci', 'utf16', 'utf16_german2_ci')
; ;
@ -494,6 +468,59 @@ create table fully_incompatible (
); );
insert into fully_incompatible (from_charset, from_collate, to_charset, to_collate) values insert into fully_incompatible (from_charset, from_collate, to_charset, to_collate) values
('ascii', 'ascii_general_ci', 'utf8mb3', 'utf8mb3_general_ci'),
('ascii', 'ascii_general_ci', 'utf8mb4', 'utf8mb4_general_ci'),
('ascii', 'ascii_general_ci', 'latin1', 'latin1_general_ci'),
('ascii', 'ascii_bin', 'latin1', 'latin1_bin'),
('ascii', 'ascii_nopad_bin', 'latin1', 'latin1_nopad_bin'),
('ascii', 'ascii_general_ci', 'latin2', 'latin2_general_ci'),
('ascii', 'ascii_general_ci', 'latin7', 'latin7_general_ci'),
('ascii', 'ascii_bin', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_bin', 'ujis', 'ujis_bin'),
('ascii', 'ascii_bin', 'big5', 'big5_bin'),
('ascii', 'ascii_bin', 'gbk', 'gbk_bin'),
('ascii', 'ascii_general_ci', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_bin', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_general_nopad_ci', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_nopad_bin', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_general_ci', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_general_nopad_ci', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_nopad_bin', 'koi8u', 'koi8u_bin'),
('ascii', 'ascii_general_ci', 'latin1', 'latin1_swedish_ci'),
('ascii', 'ascii_bin', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_general_nopad_ci', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_nopad_bin', 'utf8mb3', 'utf8mb3_swedish_ci'),
('ascii', 'ascii_general_ci', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_bin', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_general_nopad_ci', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_nopad_bin', 'utf8mb4', 'utf8mb4_danish_ci'),
('ascii', 'ascii_general_ci', 'gbk', 'gbk_chinese_ci'),
('ascii', 'ascii_general_ci', 'gbk', 'gbk_chinese_nopad_ci'),
('ascii', 'ascii_general_ci', 'ujis', 'ujis_japanese_ci'),
('ascii', 'ascii_general_ci', 'big5', 'big5_chinese_ci'),
('ascii', 'ascii_general_ci', 'latin2', 'latin2_croatian_ci'),
('ascii', 'ascii_general_ci', 'latin7', 'latin7_estonian_cs'),
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_general_ci'),
('ucs2', 'ucs2_unicode_ci', 'utf16', 'utf16_unicode_ci'),
('ucs2', 'ucs2_icelandic_ci', 'utf16', 'utf16_icelandic_ci'),
('ucs2', 'ucs2_latvian_ci', 'utf16', 'utf16_latvian_ci'),
('ucs2', 'ucs2_romanian_ci', 'utf16', 'utf16_romanian_ci'),
('ucs2', 'ucs2_slovenian_ci', 'utf16', 'utf16_slovenian_ci'),
('ucs2', 'ucs2_polish_ci', 'utf16', 'utf16_polish_ci'),
('ucs2', 'ucs2_estonian_ci', 'utf16', 'utf16_estonian_ci'),
('ucs2', 'ucs2_spanish_ci', 'utf16', 'utf16_spanish_ci'),
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_general_ci'),
('ucs2', 'ucs2_myanmar_ci', 'utf16', 'utf16_thai_520_w2'),
('ucs2', 'ucs2_general_ci', 'utf16', 'utf16_unicode_nopad_ci'),
('ucs2', 'ucs2_general_mysql500_ci', 'utf16', 'utf16_spanish2_ci'),
('utf8mb4', 'utf8mb4_general_ci', 'utf8mb3', 'utf8mb3_general_ci'), ('utf8mb4', 'utf8mb4_general_ci', 'utf8mb3', 'utf8mb3_general_ci'),
('utf8mb4', 'utf8mb4_general_ci', 'ascii', 'ascii_general_ci'), ('utf8mb4', 'utf8mb4_general_ci', 'ascii', 'ascii_general_ci'),
('utf8mb3', 'utf8mb3_general_ci', 'ascii', 'ascii_general_ci'), ('utf8mb3', 'utf8mb3_general_ci', 'ascii', 'ascii_general_ci'),
@ -536,3 +563,33 @@ while ($counter <= $data_size) {
} }
drop table fully_incompatible; drop table fully_incompatible;
--echo #
--echo # MDEV-19284 INSTANT ALTER with ucs2-to-utf16 conversion produces bad data
--echo #
CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET ucs2, PRIMARY KEY(a)) ENGINE=InnoDB;
INSERT INTO t1 VALUES ('a'),(0xD800);
--error ER_TRUNCATED_WRONG_VALUE_FOR_FIELD
ALTER TABLE t1 ALGORITHM=COPY, MODIFY a VARCHAR(10) CHARACTER SET utf16;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
ALTER TABLE t1 ALGORITHM=INSTANT, MODIFY a VARCHAR(10) CHARACTER SET utf16;
--enable_info ONCE
ALTER IGNORE TABLE t1 MODIFY a VARCHAR(10) CHARACTER SET utf16;
SELECT HEX(a) FROM t1;
DROP TABLE t1;
--echo #
--echo # MDEV-19285 INSTANT ALTER from ascii_general_ci to latin1_general_ci produces currupt data
--echo #
CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET ascii COLLATE ascii_general_ci, PRIMARY KEY(a)) ENGINE=InnoDB;
INSERT INTO t1 VALUES ('a'),(0xC0),('b');
--error ER_TRUNCATED_WRONG_VALUE_FOR_FIELD
ALTER TABLE t1 ALGORITHM=COPY, MODIFY a VARCHAR(10) CHARACTER SET latin1 COLLATE latin1_general_ci;
--error ER_ALTER_OPERATION_NOT_SUPPORTED_REASON
ALTER TABLE t1 ALGORITHM=INSTANT, MODIFY a VARCHAR(10) CHARACTER SET latin1 COLLATE latin1_general_ci;
--enable_info ONCE
ALTER IGNORE TABLE t1 MODIFY a VARCHAR(10) CHARACTER SET latin1 COLLATE latin1_general_ci;
SELECT HEX(a) FROM t1;
DROP TABLE t1;

View File

@ -7092,6 +7092,17 @@ int Field_str::store(double nr)
} }
bool Field_longstr::
csinfo_change_allows_instant_alter(const Create_field *to) const
{
Charset cs(field_charset);
const bool part_of_a_key= !to->field->part_of_key.is_clear_all();
return part_of_a_key ?
cs.encoding_and_order_allow_reinterpret_as(to->charset) :
cs.encoding_allows_reinterpret_as(to->charset);
}
uint Field_string::is_equal(Create_field *new_field) uint Field_string::is_equal(Create_field *new_field)
{ {
DBUG_ASSERT(!compression_method()); DBUG_ASSERT(!compression_method());
@ -7102,9 +7113,7 @@ uint Field_string::is_equal(Create_field *new_field)
if (new_field->char_length < char_length()) if (new_field->char_length < char_length())
return IS_EQUAL_NO; return IS_EQUAL_NO;
const bool part_of_a_key= !new_field->field->part_of_key.is_clear_all(); if (!csinfo_change_allows_instant_alter(new_field))
if (!Type_handler::Charsets_are_compatible(field_charset, new_field->charset,
part_of_a_key))
return IS_EQUAL_NO; return IS_EQUAL_NO;
if (new_field->length == max_display_length()) if (new_field->length == max_display_length())
@ -7954,9 +7963,7 @@ uint Field_varstring::is_equal(Create_field *new_field)
if (!new_field->compression_method() != !compression_method()) if (!new_field->compression_method() != !compression_method())
return IS_EQUAL_NO; return IS_EQUAL_NO;
bool part_of_a_key= !new_field->field->part_of_key.is_clear_all(); if (!csinfo_change_allows_instant_alter(new_field))
if (!Type_handler::Charsets_are_compatible(field_charset, new_field->charset,
part_of_a_key))
return IS_EQUAL_NO; return IS_EQUAL_NO;
const Type_handler *new_type_handler= new_field->type_handler(); const Type_handler *new_type_handler= new_field->type_handler();
@ -8751,12 +8758,8 @@ uint Field_blob::is_equal(Create_field *new_field)
return IS_EQUAL_NO; return IS_EQUAL_NO;
} }
bool part_of_a_key= !new_field->field->part_of_key.is_clear_all(); if (!csinfo_change_allows_instant_alter(new_field))
if (!Type_handler::Charsets_are_compatible(field_charset, new_field->charset,
part_of_a_key))
{
return IS_EQUAL_NO; return IS_EQUAL_NO;
}
if (field_charset != new_field->charset) if (field_charset != new_field->charset)
{ {

View File

@ -1926,6 +1926,7 @@ protected:
CHARSET_INFO *cs, size_t nchars); CHARSET_INFO *cs, size_t nchars);
String *uncompress(String *val_buffer, String *val_ptr, String *uncompress(String *val_buffer, String *val_ptr,
const uchar *from, uint from_length); const uchar *from, uint from_length);
bool csinfo_change_allows_instant_alter(const Create_field *to) const;
public: public:
Field_longstr(uchar *ptr_arg, uint32 len_arg, uchar *null_ptr_arg, Field_longstr(uchar *ptr_arg, uint32 len_arg, uchar *null_ptr_arg,
uchar null_bit_arg, utype unireg_check_arg, uchar null_bit_arg, utype unireg_check_arg,

View File

@ -159,6 +159,14 @@ public:
{ {
swap_variables(CHARSET_INFO*, m_charset, other.m_charset); swap_variables(CHARSET_INFO*, m_charset, other.m_charset);
} }
/*
Collation name without the character set name.
For example, in case of "latin1_swedish_ci",
this method returns "_swedish_ci".
*/
LEX_CSTRING collation_specific_name() const;
bool encoding_allows_reinterpret_as(CHARSET_INFO *cs) const;
bool encoding_and_order_allow_reinterpret_as(CHARSET_INFO *cs) const;
}; };

View File

@ -8219,48 +8219,51 @@ Type_handler_timestamp_common::Item_param_val_native(THD *thd,
TIME_to_native(thd, &ltime, to, item->datetime_precision(thd)); TIME_to_native(thd, &ltime, to, item->datetime_precision(thd));
} }
static bool charsets_are_compatible(const char *old_cs_name,
const CHARSET_INFO *new_ci) LEX_CSTRING Charset::collation_specific_name() const
{ {
const char *new_cs_name= new_ci->csname; /*
User defined collations can provide arbitrary names
for character sets and collations, so a collation
name not necessarily starts with the character set name.
*/
size_t csname_length= strlen(m_charset->csname);
if (strncmp(m_charset->name, m_charset->csname, csname_length))
return {NULL, 0};
const char *ptr= m_charset->name + csname_length;
return {ptr, strlen(ptr) };
}
if (!strcmp(old_cs_name, new_cs_name))
bool
Charset::encoding_allows_reinterpret_as(const CHARSET_INFO *cs) const
{
if (!strcmp(m_charset->csname, cs->csname))
return true; return true;
if (!strcmp(old_cs_name, MY_UTF8MB3) && !strcmp(new_cs_name, MY_UTF8MB4)) if (!strcmp(m_charset->csname, MY_UTF8MB3) &&
return true; !strcmp(cs->csname, MY_UTF8MB4))
if (!strcmp(old_cs_name, "ascii") && !(new_ci->state & MY_CS_NONASCII))
return true;
if (!strcmp(old_cs_name, "ucs2") && !strcmp(new_cs_name, "utf16"))
return true; return true;
/*
Originally we allowed here instat ALTER for ASCII-to-LATIN1
and UCS2-to-UTF16, but this was wrong:
- MariaDB's ascii is not a subset for 8-bit character sets
like latin1, because it allows storing bytes 0x80..0xFF as
"unassigned" characters (see MDEV-19285).
- MariaDB's ucs2 (as in Unicode-1.1) is not a subset for UTF16,
because they treat surrogate codes differently (MDEV-19284).
*/
return false; return false;
} }
bool Type_handler::Charsets_are_compatible(const CHARSET_INFO *old_ci,
const CHARSET_INFO *new_ci, bool
bool part_of_a_key) Charset::encoding_and_order_allow_reinterpret_as(CHARSET_INFO *cs) const
{ {
const char *old_cs_name= old_ci->csname; if (!encoding_allows_reinterpret_as(cs))
const char *new_cs_name= new_ci->csname;
if (!charsets_are_compatible(old_cs_name, new_ci))
{
return false; return false;
} LEX_CSTRING name0= collation_specific_name();
LEX_CSTRING name1= Charset(cs).collation_specific_name();
if (!part_of_a_key) return name0.length && !cmp(&name0, &name1);
{
return true;
}
if (strcmp(old_ci->name + strlen(old_cs_name),
new_ci->name + strlen(new_cs_name)))
{
return false;
}
return true;
} }

View File

@ -3705,10 +3705,6 @@ public:
virtual bool virtual bool
Vers_history_point_resolve_unit(THD *thd, Vers_history_point *point) const; Vers_history_point_resolve_unit(THD *thd, Vers_history_point *point) const;
static bool Charsets_are_compatible(const CHARSET_INFO *old_ci,
const CHARSET_INFO *new_ci,
bool part_of_a_key);
}; };