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
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
|
/* Copyright (C) 2006 MySQL AB & MySQL Finland AB & TCX DataKonsult AB
Copyright (c) 2020, MariaDB Corporation.
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.
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.
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 Street, Fifth Floor, Boston, MA 02110-1335 USA */
/* Functions to check if a row is unique */
#include "maria_def.h"
#include <m_ctype.h>
/**
Check if there exist a row with the same hash
@notes
This function is not versioning safe. For the moment this is not a problem
as it's only used for internal temporary tables in MySQL for which there
isn't any versioning information.
*/
my_bool _ma_check_unique(MARIA_HA *info, MARIA_UNIQUEDEF *def,
const uchar *record,
ha_checksum unique_hash, my_off_t disk_pos)
{
my_off_t lastpos=info->cur_row.lastpos;
MARIA_KEYDEF *keyinfo= &info->s->keyinfo[def->key];
uchar *key_buff= info->lastkey_buff2;
MARIA_KEY key;
int error= 0;
DBUG_ENTER("_ma_check_unique");
DBUG_PRINT("enter",("unique_hash: %lu", (ulong) unique_hash));
/* We need to store the hash value as a key in the record, breaking const */
maria_unique_store(record+keyinfo->seg->start, unique_hash);
/* Can't be spatial so it's ok to call _ma_make_key directly here */
_ma_make_key(info, &key, def->key, key_buff, record, 0, 0);
/* The above changed info->lastkey_buff2. Inform maria_rnext_same(). */
info->update&= ~HA_STATE_RNEXT_SAME;
/* Setup that unique key is active key */
info->last_key.keyinfo= keyinfo;
/* any key pointer in data is destroyed */
info->lastinx= ~0;
DBUG_ASSERT(key.data_length == MARIA_UNIQUE_HASH_LENGTH);
if (_ma_search(info, &key, SEARCH_FIND | SEARCH_SAVE_BUFF,
info->s->state.key_root[def->key]))
{
info->page_changed=1; /* Can't optimize read next */
info->cur_row.lastpos= lastpos;
goto end;
}
for (;;)
{
if (info->cur_row.lastpos != disk_pos &&
!(*info->s->compare_unique)(info,def,record,info->cur_row.lastpos))
{
my_errno=HA_ERR_FOUND_DUPP_UNIQUE;
info->errkey= (int) def->key;
info->dup_key_pos= info->cur_row.lastpos;
info->page_changed= 1; /* Can't optimize read next */
info->cur_row.lastpos= lastpos;
DBUG_PRINT("info",("Found duplicate"));
error= 1; /* Found identical */
goto end;
}
DBUG_ASSERT(info->last_key.data_length == MARIA_UNIQUE_HASH_LENGTH);
if (_ma_search_next(info, &info->last_key, SEARCH_BIGGER,
info->s->state.key_root[def->key]) ||
bcmp(info->last_key.data, key_buff, MARIA_UNIQUE_HASH_LENGTH))
{
info->page_changed= 1; /* Can't optimize read next */
info->cur_row.lastpos= lastpos;
break; /* end of tree */
}
}
end:
DBUG_RETURN(error);
}
/*
Calculate a hash for a row
TODO
Add support for bit fields
*/
ha_checksum _ma_unique_hash(MARIA_UNIQUEDEF *def, const uchar *record)
{
const uchar *pos, *end;
ha_checksum crc= 0;
ulong seed1=0, seed2= 4;
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 */
set_if_smaller(length,tmp_length);
}
else if (keyseg->flag & HA_BLOB_PART)
{
uint tmp_length= _ma_calc_blob_length(keyseg->bit_start,pos);
memcpy((void*) &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)
{
my_ci_hash_sort(keyseg->charset,
(const uchar*) pos, length,
&seed1, &seed2);
crc+= seed1;
}
else
{
my_hash_sort_bin((CHARSET_INFO*) 0, pos, (size_t) (end-pos),
&seed1, &seed2);
crc+= seed1;
}
}
return crc;
}
/*
compare unique key for two rows
TODO
Add support for bit fields
RETURN
0 if both rows have equal unique value
1 Rows are different
*/
my_bool _ma_unique_comp(MARIA_UNIQUEDEF *def, const uchar *a, const uchar *b,
my_bool null_are_equal)
{
const uchar *pos_a, *pos_b, *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;
}
set_if_smaller(a_length, keyseg->length); /* Safety */
set_if_smaller(b_length, keyseg->length); /* safety */
}
else if (keyseg->flag & HA_BLOB_PART)
{
/* Only compare 'length' characters if length != 0 */
a_length= _ma_calc_blob_length(keyseg->bit_start,pos_a);
b_length= _ma_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.
*/
set_if_smaller(a_length, keyseg->length);
set_if_smaller(b_length, keyseg->length);
}
memcpy((void*) &pos_a, pos_a+keyseg->bit_start, sizeof(char*));
memcpy((void*) &pos_b, pos_b+keyseg->bit_start, sizeof(char*));
}
if (type == HA_KEYTYPE_TEXT/* the CHAR data type*/)
{
if (ha_compare_char_fixed(keyseg->charset,
pos_a, a_length,
pos_b, b_length,
keyseg->length / keyseg->charset->mbmaxlen,
FALSE/*b_is_prefix*/))
return 1;
}
else if (type == HA_KEYTYPE_VARTEXT1 ||
type == HA_KEYTYPE_VARTEXT2)
{
if (ha_compare_char_varying(keyseg->charset,
pos_a, a_length,
pos_b, b_length,
FALSE/*b_is_prefix*/))
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;
}
|
