1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
|
/* Copyright (c) 2000, 2025, Oracle and/or its affiliates.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License, version 2.0,
as published by the Free Software Foundation.
This program is designed to work with certain software (including
but not limited to OpenSSL) that is licensed under separate terms,
as designated in a particular file or component or in included license
documentation. The authors of MySQL hereby grant you an additional
permission to link the program and your derivative works with the
separately licensed software that they have either included with
the program or referenced in the documentation.
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, version 2.0, for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */
/* Functions to check if a row is unique */
#include <sys/types.h>
#include <algorithm>
#include "m_ctype.h"
#include "my_byteorder.h"
#include "my_dbug.h"
#include "my_inttypes.h"
#include "my_macros.h"
#include "storage/myisam/myisamdef.h"
bool mi_check_unique(MI_INFO *info, MI_UNIQUEDEF *def, uchar *record,
ha_checksum unique_hash, my_off_t disk_pos) {
my_off_t lastpos = info->lastpos;
MI_KEYDEF *key = &info->s->keyinfo[def->key];
uchar *key_buff = info->lastkey2;
DBUG_TRACE;
mi_unique_store(record + key->seg->start, unique_hash);
_mi_make_key(info, def->key, key_buff, record, 0);
if (_mi_search(info, info->s->keyinfo + def->key, key_buff,
MI_UNIQUE_HASH_LENGTH, SEARCH_FIND,
info->s->state.key_root[def->key])) {
info->page_changed = true; /* Can't optimize read next */
info->lastpos = lastpos;
return false; /* No matching rows */
}
for (;;) {
if (info->lastpos != disk_pos &&
!(*info->s->compare_unique)(info, def, record, info->lastpos)) {
set_my_errno(HA_ERR_FOUND_DUPP_UNIQUE);
info->errkey = (int)def->key;
info->dupp_key_pos = info->lastpos;
info->page_changed = true; /* Can't optimize read next */
info->lastpos = lastpos;
DBUG_PRINT("info", ("Found duplicate"));
return true; /* Found identical */
}
if (_mi_search_next(info, info->s->keyinfo + def->key, info->lastkey,
MI_UNIQUE_HASH_LENGTH, SEARCH_BIGGER,
info->s->state.key_root[def->key]) ||
memcmp(info->lastkey, key_buff, MI_UNIQUE_HASH_LENGTH)) {
info->page_changed = true; /* Can't optimize read next */
info->lastpos = lastpos;
return false; /* end of tree */
}
}
}
/*
Calculate a hash for a row
TODO
Add support for bit fields
*/
ha_checksum mi_unique_hash(MI_UNIQUEDEF *def, const uchar *record) {
const uchar *pos;
const uchar *end;
ha_checksum crc = 0;
HA_KEYSEG *keyseg;
for (keyseg = def->seg; keyseg < def->end; keyseg++) {
enum ha_base_keytype type = (enum ha_base_keytype)keyseg->type;
uint length = keyseg->length;
if (keyseg->null_bit) {
if (record[keyseg->null_pos] & keyseg->null_bit) {
/*
Change crc in a way different from an empty string or 0.
(This is an optimisation; The code will work even if this isn't
done)
*/
crc = ((crc << 8) + 511 + (crc >> (8 * sizeof(ha_checksum) - 8)));
continue;
}
}
pos = record + keyseg->start;
if (keyseg->flag & HA_VAR_LENGTH_PART) {
uint pack_length = keyseg->bit_start;
uint tmp_length = (pack_length == 1 ? (uint)*pos : uint2korr(pos));
pos += pack_length; /* Skip VARCHAR length */
length = std::min(length, tmp_length);
} else if (keyseg->flag & HA_BLOB_PART) {
uint tmp_length = _mi_calc_blob_length(keyseg->bit_start, pos);
memcpy(&pos, pos + keyseg->bit_start, sizeof(char *));
if (!length || length > tmp_length)
length = tmp_length; /* The whole blob */
}
end = pos + length;
if (type == HA_KEYTYPE_TEXT || type == HA_KEYTYPE_VARTEXT1 ||
type == HA_KEYTYPE_VARTEXT2) {
uint64 seed1 = 0, seed2 = 4;
keyseg->charset->coll->hash_sort(keyseg->charset, (const uchar *)pos,
length, &seed1, &seed2);
crc ^= static_cast<ulong>(seed1);
} else
while (pos != end)
crc = ((crc << 8) + (((uchar)*pos++))) +
(crc >> (8 * sizeof(ha_checksum) - 8));
}
return crc;
}
/*
compare unique key for two rows
TODO
Add support for bit fields
RETURN
0 if both rows have equal unique value
# Rows are different
*/
int mi_unique_comp(MI_UNIQUEDEF *def, const uchar *a, const uchar *b,
bool null_are_equal) {
const uchar *pos_a, *pos_b;
const uchar *end;
HA_KEYSEG *keyseg;
for (keyseg = def->seg; keyseg < def->end; keyseg++) {
enum ha_base_keytype type = (enum ha_base_keytype)keyseg->type;
uint a_length, b_length;
a_length = b_length = keyseg->length;
/* If part is NULL it's regarded as different */
if (keyseg->null_bit) {
uint tmp;
if ((tmp = (a[keyseg->null_pos] & keyseg->null_bit)) !=
(uint)(b[keyseg->null_pos] & keyseg->null_bit))
return 1;
if (tmp) {
if (!null_are_equal) return 1;
continue;
}
}
pos_a = a + keyseg->start;
pos_b = b + keyseg->start;
if (keyseg->flag & HA_VAR_LENGTH_PART) {
uint pack_length = keyseg->bit_start;
if (pack_length == 1) {
a_length = (uint)*pos_a++;
b_length = (uint)*pos_b++;
} else {
a_length = uint2korr(pos_a);
b_length = uint2korr(pos_b);
pos_a += 2; /* Skip VARCHAR length */
pos_b += 2;
}
a_length = std::min(a_length, uint(keyseg->length)); /* Safety */
b_length = std::min(b_length, uint(keyseg->length)); /* safety */
} else if (keyseg->flag & HA_BLOB_PART) {
/* Only compare 'length' characters if length != 0 */
a_length = _mi_calc_blob_length(keyseg->bit_start, pos_a);
b_length = _mi_calc_blob_length(keyseg->bit_start, pos_b);
/* Check that a and b are of equal length */
if (keyseg->length) {
/*
This is used in some cases when we are not interested in comparing
the whole length of the blob.
*/
a_length = std::min(a_length, uint(keyseg->length));
b_length = std::min(b_length, uint(keyseg->length));
}
memcpy(&pos_a, pos_a + keyseg->bit_start, sizeof(char *));
memcpy(&pos_b, pos_b + keyseg->bit_start, sizeof(char *));
}
if (type == HA_KEYTYPE_TEXT || type == HA_KEYTYPE_VARTEXT1 ||
type == HA_KEYTYPE_VARTEXT2) {
if (ha_compare_text(keyseg->charset, pos_a, a_length, pos_b, b_length,
false))
return 1;
} else {
if (a_length != b_length) return 1;
end = pos_a + a_length;
while (pos_a != end) {
if (*pos_a++ != *pos_b++) return 1;
}
}
}
return 0;
}
|
