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
|
/* Copyright (c) 2017, 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 */
/**
@file sql/histograms/value_map.cc
Value map (implementation).
*/
#include "sql/histograms/value_map.h"
#include <assert.h>
#include <algorithm>
#include <new>
#include <string> // std::string
#include "my_inttypes.h"
#include "my_sys.h"
#include "my_time.h"
#include "mysql_time.h" // MYSQL_TIME
#include "sql/histograms/histogram.h"
#include "sql/my_decimal.h" // my_decimal_cmp
#include "sql/psi_memory_key.h" // key_memory_histograms
#include "sql_string.h" // String
#include "template_utils.h" // down_cast
namespace histograms {
// Overloading the Histogram_comparator for various data types.
template <>
bool Histogram_comparator::operator()(const String &lhs,
const String &rhs) const {
// The collation MUST be the same
assert(lhs.charset()->number == rhs.charset()->number);
// The number of characters should already be limited.
assert(lhs.numchars() <= HISTOGRAM_MAX_COMPARE_LENGTH);
assert(rhs.numchars() <= HISTOGRAM_MAX_COMPARE_LENGTH);
return sortcmp(&lhs, &rhs, lhs.charset()) < 0;
}
template <>
bool Histogram_comparator::operator()(const MYSQL_TIME &lhs,
const MYSQL_TIME &rhs) const {
longlong lhs_packed = TIME_to_longlong_packed(lhs);
longlong rhs_packed = TIME_to_longlong_packed(rhs);
return lhs_packed < rhs_packed;
}
template <>
bool Histogram_comparator::operator()(const my_decimal &lhs,
const my_decimal &rhs) const {
return my_decimal_cmp(&lhs, &rhs) < 0;
}
Value_map_base::Value_map_base(const CHARSET_INFO *charset,
Value_map_type data_type)
: m_sampling_rate(1.0),
m_charset(charset),
m_num_null_values(0),
m_data_type(data_type),
m_mem_root(key_memory_histograms, 256) {}
template <class T>
bool Value_map_base::add_values(const T &value, const ha_rows count) {
Value_map<T> *value_map = down_cast<Value_map<T> *>(this);
return value_map->add_values(value, count);
}
template <class T>
bool Value_map<T>::add_values(const T &value, const ha_rows count) {
try {
auto res = m_value_map.emplace(value, count);
if (!res.second) res.first->second += count;
} catch (const std::bad_alloc &) {
// Out of memory.
return true;
}
return false;
}
template <>
bool Value_map<String>::add_values(const String &value, const ha_rows count) {
/*
We only consider the substring. That is, if the strings differs after
character number HISTOGRAM_MAX_COMPARE_LENGTH, they will be considered
equal.
When inserting a new string, we make a duplicate of it so that it survives
when the original string goes out of scope. The duplicate string is
allocated in the MEM_ROOT of the Value_map, so that is is automatically
reclaimed when the Value_map is destroyed.
*/
String substring = value.substr(0, HISTOGRAM_MAX_COMPARE_LENGTH);
auto found = m_value_map.find(substring);
if (found == m_value_map.end()) {
// Not found, insert a new value.
try {
char *string_data = substring.dup(&m_mem_root);
if (string_data == nullptr) return true; /* purecov: deadcode */
String string_dup(string_data, substring.length(), substring.charset());
m_value_map.emplace(string_dup, count);
} catch (const std::bad_alloc &) {
// Out of memory.
return true;
}
} else {
found->second += count;
}
return false;
}
template <class T>
bool Value_map<T>::insert(typename value_map_type::const_iterator begin,
typename value_map_type::const_iterator end) {
try {
assert(m_value_map.empty());
m_value_map.insert(begin, end);
} catch (const std::bad_alloc &) {
// Out of memory.
return true;
}
return false;
}
template <class T>
Histogram *Value_map<T>::build_histogram(MEM_ROOT *mem_root, size_t num_buckets,
const std::string &db_name,
const std::string &tbl_name,
const std::string &col_name) const {
return histograms::build_histogram(mem_root, *this, num_buckets, db_name,
tbl_name, col_name);
}
// Explicit template instantiations.
template class Value_map<double>;
template class Value_map<String>;
template class Value_map<ulonglong>;
template class Value_map<longlong>;
template class Value_map<MYSQL_TIME>;
template class Value_map<my_decimal>;
template bool Value_map_base::add_values(const double &, const ha_rows);
template bool Value_map_base::add_values(const String &, const ha_rows);
template bool Value_map_base::add_values(const ulonglong &, const ha_rows);
template bool Value_map_base::add_values(const longlong &, const ha_rows);
template bool Value_map_base::add_values(const MYSQL_TIME &, const ha_rows);
template bool Value_map_base::add_values(const my_decimal &, const ha_rows);
} // namespace histograms
|
