summaryrefslogtreecommitdiff
path: root/src/mongo/util/represent_as_test.cpp
blob: 99b02748983c77a7d553901769b3ff72f478f65c (plain)
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
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295

/**
 *    Copyright (C) 2018-present MongoDB, Inc.
 *
 *    This program is free software: you can redistribute it and/or modify
 *    it under the terms of the Server Side Public License, version 1,
 *    as published by MongoDB, Inc.
 *
 *    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
 *    Server Side Public License for more details.
 *
 *    You should have received a copy of the Server Side Public License
 *    along with this program. If not, see
 *    <http://www.mongodb.com/licensing/server-side-public-license>.
 *
 *    As a special exception, the copyright holders give permission to link the
 *    code of portions of this program with the OpenSSL library under certain
 *    conditions as described in each individual source file and distribute
 *    linked combinations including the program with the OpenSSL library. You
 *    must comply with the Server Side Public License in all respects for
 *    all of the code used other than as permitted herein. If you modify file(s)
 *    with this exception, you may extend this exception to your version of the
 *    file(s), but you are not obligated to do so. If you do not wish to do so,
 *    delete this exception statement from your version. If you delete this
 *    exception statement from all source files in the program, then also delete
 *    it in the license file.
 */

#include "mongo/platform/basic.h"

#include <cmath>
#include <limits>

#include <boost/optional.hpp>

#include "mongo/unittest/unittest.h"

#include "mongo/util/represent_as.h"

namespace mongo {

namespace {

// Char values
const signed char kCharMax = std::numeric_limits<signed char>::max();
const int kCharMaxAsInt = kCharMax;

// Unsigned char values
const unsigned char kUCharMax = std::numeric_limits<unsigned char>::max();
const unsigned char kUCharMin = std::numeric_limits<unsigned char>::lowest();
const int kUCharMaxAsInt = kUCharMax;

// Int values
const int kIntMax = std::numeric_limits<int>::max();
const int kIntMin = std::numeric_limits<int>::lowest();
const long long kIntMaxAsLongLong = kIntMax;
const long long kIntMinAsLongLong = kIntMin;
const unsigned long long kIntMaxAsULongLong = kIntMax;
const unsigned long long kIntMinAsULongLong = kIntMin;

// 32-bit integer values
const int32_t kInt32Zero = 0;
const int32_t kInt32Max = std::numeric_limits<int32_t>::max();
const int32_t kInt32Min = std::numeric_limits<int32_t>::lowest();
const uint32_t kInt32MaxAsUInt32 = kInt32Max;
const uint64_t kInt32MaxAsUInt64 = kInt32Max;
const double kInt32MaxAsDouble = kInt32Max;
const double kInt32MinAsDouble = kInt32Min;

// Unsigned 32-bit integer values
const uint32_t kUInt32Zero = 0;
const uint32_t kUInt32Max = std::numeric_limits<uint32_t>::max();
const int64_t kUInt32MaxAsInt64 = kUInt32Max;
const float kUInt32MaxAsFloat = static_cast<float>(kUInt32Max);
const double kUInt32MaxAsDouble = kUInt32Max;

// 64-bit integer values
const int64_t kInt64Zero = 0;
const int64_t kInt64Max = std::numeric_limits<int64_t>::max();
const int64_t kInt64Min = std::numeric_limits<int64_t>::lowest();
const uint64_t kInt64MaxAsUInt64 = kInt64Max;
const double kInt64MaxAsDouble = static_cast<double>(kInt64Max);
const double kInt64MinAsDouble = kInt64Min;

// Unsigned 64-bit integer values
const uint64_t kUInt64Zero = 0;
const uint64_t kUInt64Max = std::numeric_limits<uint64_t>::max();
const float kUInt64MaxAsFloat = static_cast<float>(kUInt64Max);
const double kUInt64MaxAsDouble = static_cast<double>(kUInt64Max);


// Long long values
const long long kLongLongMax = std::numeric_limits<long long>::max();

// Unsigned long long values
const unsigned long long kULongLongMax = std::numeric_limits<unsigned long long>::max();

// Float values
const float kFloatZero = 0;
const float kFloatMax = std::numeric_limits<float>::max();
const float kFloatMin = std::numeric_limits<float>::lowest();
const double kFloatMaxAsDouble = kFloatMax;
const double kFloatMinAsDouble = kFloatMin;

// Double values
const double kDoubleZero = 0;
const double kDoubleMax = std::numeric_limits<double>::max();
const double kDoubleMin = std::numeric_limits<double>::lowest();

// Precision values
const int kFloatMantissa = std::numeric_limits<float>::digits;
const int kDoubleMantissa = std::numeric_limits<double>::digits;
const int32_t kInt32TooPreciseForFloat =
    static_cast<int32_t>(std::ldexp(1, kFloatMantissa + 1)) + 1;
const uint32_t kUInt32TooPreciseForFloat = kInt32TooPreciseForFloat;
const int64_t kInt64TooPreciseForFloat = kInt32TooPreciseForFloat;
const int64_t kInt64TooPreciseForDouble =
    static_cast<int64_t>(std::ldexp(1, kDoubleMantissa + 1)) + 1;
const uint64_t kUInt64TooPreciseForFloat = kInt32TooPreciseForFloat;
const uint64_t kUInt64TooPreciseForDouble = kInt64TooPreciseForDouble;

}  // namespace

TEST(RepresentAs, Int32ToDouble) {
    ASSERT(*(representAs<double>(kInt32Zero)) == 0);
    ASSERT(*(representAs<double>(5)) == 5);
}

TEST(RepresentAs, Int64ToDouble) {
    ASSERT(*(representAs<double>(kInt64Zero)) == 0);
    ASSERT(*(representAs<double>(5)) == 5);

    // kInt64Max is too precise for double
    ASSERT(!(representAs<double>(kInt64Max)));
    ASSERT(*(representAs<double>(kInt64Min)) == kInt64MinAsDouble);
}

TEST(RepresentAs, DoubleToInt32) {
    ASSERT(*(representAs<int32_t>(kDoubleZero)) == 0);
    ASSERT(*(representAs<int32_t>(-12345)) == -12345);
    ASSERT(!(representAs<int32_t>(10.3)));

    // Int32 edge cases
    ASSERT(*(representAs<int32_t>(kInt32Max)) == kInt32Max);
    ASSERT(!(representAs<int32_t>(kInt32MaxAsDouble + 1)));
    ASSERT(*(representAs<int32_t>(kInt32Min)) == kInt32Min);
    ASSERT(!(representAs<int32_t>(kInt32MinAsDouble - 1)));

    // Very large and small values
    ASSERT(!(representAs<int32_t>(kDoubleMax)));
    ASSERT(!(representAs<int32_t>(kDoubleMin)));
}

TEST(RepresentAs, DoubleToInt64) {
    ASSERT(*(representAs<int64_t>(kDoubleZero)) == 0);
    ASSERT(*(representAs<int64_t>(-12345)) == -12345);
    ASSERT(!(representAs<int64_t>(10.3)));

    // Int64 edge cases, max can't be represented as doubles, min can
    ASSERT(!(representAs<int64_t>(kInt64MaxAsDouble)));
    ASSERT(*(representAs<int64_t>(kInt64MinAsDouble)) == kInt64Min);

    // Very large and small values
    ASSERT(!(representAs<int64_t>(kDoubleMax)));
    ASSERT(!(representAs<int64_t>(kDoubleMin)));
}

TEST(RepresentAs, DoubleToFloat) {
    ASSERT(*(representAs<float>(kDoubleZero)) == 0);
    ASSERT(*(representAs<float>(-12345)) == -12345);

    // Float edge casees
    ASSERT(*(representAs<float>(kFloatMax)) == (representAs<float>(kFloatMaxAsDouble + 1)));
    ASSERT(*(representAs<float>(kFloatMin)) == (representAs<float>(kFloatMinAsDouble - 1)));

    // Very large and small values
    ASSERT(!(representAs<float>(kDoubleMax)));
    ASSERT(!(representAs<float>(kDoubleMin)));
}

TEST(RepresentAs, DoubleToUnsignedInt) {
    ASSERT(!(representAs<uint32_t>(-1.23)));
    ASSERT(*(representAs<uint64_t>(kDoubleZero)) == kUInt64Zero);
    ASSERT(!(representAs<uint32_t>(kDoubleMax)));
    ASSERT(!(representAs<uint64_t>(kDoubleMax)));
}

TEST(RepresentAs, FloatToDouble) {
    ASSERT(*(representAs<double>(kFloatZero)) == 0);
    ASSERT(*(representAs<double>(-12345)) == -12345);

    ASSERT(*(representAs<double>(kFloatMax)) == kFloatMax);
    ASSERT(*(representAs<double>(kFloatMin)) == kFloatMin);
}

TEST(RepresentAs, FloatToUnsignedInt) {
    ASSERT(!(representAs<uint32_t>(-1.23)));
    ASSERT(!(representAs<uint32_t>(-1)));
    ASSERT(*(representAs<uint64_t>(kUInt64Zero)) == kUInt64Zero);
    ASSERT(*(representAs<uint64_t>(10)) == static_cast<uint64_t>(10));
    ASSERT(!(representAs<uint32_t>(kFloatMax)));
    ASSERT(!(representAs<uint64_t>(kFloatMax)));
}

TEST(RepresentAs, SignedAndUnsigned32BitIntegers) {
    ASSERT(!(representAs<uint32_t>(kInt32Min)));
    ASSERT(*(representAs<uint32_t>(kInt32Max)) == kInt32MaxAsUInt32);

    ASSERT(!(representAs<int32_t>(kUInt32Max)));
    ASSERT(!(representAs<int32_t>(kInt32MaxAsUInt32 + 1)));
}

TEST(RepresentAs, SignedAndUnsigned64BitIntegers) {
    ASSERT(!(representAs<uint64_t>(kInt64Min)));
    ASSERT(*(representAs<uint64_t>(kInt64Max)) == kInt64MaxAsUInt64);

    ASSERT(!(representAs<int64_t>(kUInt64Max)));
    ASSERT(!(representAs<int64_t>(kInt64MaxAsUInt64 + 1)));
}

TEST(RepresentAs, SignedAndUnsignedMixedSizeIntegers) {
    ASSERT(!(representAs<uint32_t>(kInt64Min)));
    ASSERT(!(representAs<uint32_t>(kInt64Max)));
    ASSERT(*(representAs<int64_t>(kUInt32Max)) == kUInt32MaxAsInt64);

    ASSERT(!(representAs<uint64_t>(kInt32Min)));
    ASSERT(*(representAs<uint64_t>(kInt32Max)) == kInt32MaxAsUInt64);
    ASSERT(!(representAs<int32_t>(kUInt64Max)));
}

TEST(RepresentAs, UnsignedIntToFloat) {
    // kUInt32Max and kUInt64Max are too precise for float.
    ASSERT(!(representAs<float>(kUInt32Max)));
    ASSERT(!(representAs<float>(kUInt64Max)));
}

TEST(RepresentAs, UnsignedIntToDouble) {
    // kUInt64Max is too precise for double.
    ASSERT(*(representAs<double>(kUInt32Max)) == kUInt32MaxAsDouble);
    ASSERT(!(representAs<double>(kUInt64Max)));
}

TEST(RepresentAs, PlatformDependent) {
    // signed char
    ASSERT(*(representAs<int>(kCharMax)) == kCharMaxAsInt);
    ASSERT(!(representAs<signed char>(kIntMax)));

    // unsigned char
    ASSERT(*(representAs<int>(kUCharMax)) == kUCharMaxAsInt);
    ASSERT(!(representAs<unsigned char>(kIntMin)));

    // long long
    ASSERT(!(representAs<int>(kLongLongMax)));
    ASSERT(*(representAs<long long>(kIntMin)) == kIntMinAsLongLong);

    // unsigned long long
    ASSERT(!(representAs<int>(kULongLongMax)));
    ASSERT(*(representAs<unsigned long long>(kIntMax)) == kIntMaxAsULongLong);
}

TEST(RepresentAs, NaN) {
    ASSERT(!(representAs<int>(std::nanf("1"))));
    ASSERT(!(representAs<unsigned int>(std::nanf("1"))));

    // NaN Identities
    ASSERT(std::isnan(*representAs<float, float>(std::nanf("1"))));
    ASSERT(std::isnan(*representAs<double>(std::nanf("1"))));
    ASSERT(std::isnan(*representAs<float>(std::nan("1"))));
    ASSERT(std::isnan(*representAs<double>(std::nan("1"))));
}

TEST(RepresentAs, LostPrecision) {
    // A loss of precision should result in a disengaged optional
    ASSERT(!(representAs<float>(kInt32TooPreciseForFloat)));
    ASSERT(!(representAs<float>(kUInt32TooPreciseForFloat)));
    ASSERT(!(representAs<float>(kInt64TooPreciseForFloat)));
    ASSERT(!(representAs<float>(kUInt64TooPreciseForFloat)));

    ASSERT(!(representAs<double>(kInt64TooPreciseForDouble)));
    ASSERT(!(representAs<double>(kUInt64TooPreciseForDouble)));
}

TEST(RepresentAs, Identity) {
    ASSERT(*(representAs<int32_t>(kInt32Max)) == kInt32Max);
    ASSERT(*(representAs<int64_t>(kInt64Max)) == kInt64Max);
    ASSERT(*(representAs<long long>(50)) == 50);
    ASSERT(*(representAs<float>(kFloatMin)) == kFloatMin);
    ASSERT(*(representAs<double>(kDoubleMax)) == kDoubleMax);
    ASSERT(*(representAs<uint32_t>(kUInt32Max)) == kUInt32Max);
    ASSERT(*(representAs<uint64_t>(kUInt64Max)) == kUInt64Max);
}

}  // namespace mongo