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
|
/**
* Copyright (C) 2014 MongoDB Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License, version 3,
* as published by the Free Software Foundation.
*
* 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 Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* 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 GNU Affero General 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/db/exec/count.h"
#include "mongo/db/index/index_cursor.h"
#include "mongo/db/index/index_descriptor.h"
namespace mongo {
// static
const char* Count::kStageType = "COUNT";
Count::Count(const CountParams& params, WorkingSet* workingSet)
: _workingSet(workingSet),
_descriptor(params.descriptor),
_iam(params.descriptor->getIndexCatalog()->getIndex(params.descriptor)),
_btreeCursor(NULL),
_params(params),
_hitEnd(false),
_shouldDedup(params.descriptor->isMultikey()),
_commonStats(kStageType) { }
void Count::initIndexCursor() {
CursorOptions cursorOptions;
cursorOptions.direction = CursorOptions::INCREASING;
IndexCursor *cursor;
Status s = _iam->newCursor(&cursor);
verify(s.isOK());
verify(cursor);
// Is this assumption always valid? See SERVER-12397
_btreeCursor.reset(static_cast<BtreeIndexCursor*>(cursor));
_btreeCursor->setOptions(cursorOptions);
// _btreeCursor points at our start position. We move it forward until it hits a cursor
// that points at the end.
_btreeCursor->seek(_params.startKey, !_params.startKeyInclusive);
// Create the cursor that points at our end position.
IndexCursor* endCursor;
verify(_iam->newCursor(&endCursor).isOK());
verify(endCursor);
// Is this assumption always valid? See SERVER-12397
_endCursor.reset(static_cast<BtreeIndexCursor*>(endCursor));
_endCursor->setOptions(cursorOptions);
// If the end key is inclusive we want to point *past* it since that's the end.
_endCursor->seek(_params.endKey, _params.endKeyInclusive);
// See if we've hit the end already.
checkEnd();
}
void Count::checkEnd() {
if (isEOF()) { return; }
if (_endCursor->isEOF()) {
// If the endCursor is EOF we're only done when our 'current count position' hits EOF.
_hitEnd = _btreeCursor->isEOF();
}
else {
// If not, we're only done when we hit the end cursor's (valid) position.
_hitEnd = _btreeCursor->pointsAt(*_endCursor.get());
}
}
PlanStage::StageState Count::work(WorkingSetID* out) {
if (NULL == _btreeCursor.get()) {
// First call to work(). Perform cursor init.
initIndexCursor();
checkEnd();
return PlanStage::NEED_TIME;
}
if (isEOF()) { return PlanStage::IS_EOF; }
DiskLoc loc = _btreeCursor->getValue();
_btreeCursor->next();
checkEnd();
if (_shouldDedup) {
if (_returned.end() != _returned.find(loc)) {
return PlanStage::NEED_TIME;
}
else {
_returned.insert(loc);
}
}
*out = WorkingSet::INVALID_ID;
return PlanStage::ADVANCED;
}
bool Count::isEOF() {
if (NULL == _btreeCursor.get()) {
// Have to call work() at least once.
return false;
}
return _hitEnd || _btreeCursor->isEOF();
}
void Count::prepareToYield() {
if (isEOF() || (NULL == _btreeCursor.get())) { return; }
verify(!_btreeCursor->isEOF());
_btreeCursor->savePosition();
if (!_endCursor->isEOF()) {
_endCursor->savePosition();
}
}
void Count::recoverFromYield() {
if (isEOF() || (NULL == _btreeCursor.get())) { return; }
if (!_btreeCursor->restorePosition().isOK()) {
_hitEnd = true;
}
if (_btreeCursor->isEOF()) {
_hitEnd = true;
return;
}
// See if we're somehow already past our end key (maybe the thing we were pointing at got
// deleted...)
int cmp = _btreeCursor->getKey().woCompare(_params.endKey, _descriptor->keyPattern(), false);
if (cmp > 0 || (cmp == 0 && !_params.endKeyInclusive)) {
_hitEnd = true;
return;
}
if (!_endCursor->isEOF()) {
if (!_endCursor->restorePosition().isOK()) {
_hitEnd = true;
return;
}
}
// If we were EOF when we yielded we don't always want to have _btreeCursor run until
// EOF. New documents may have been inserted after our endKey and our end marker
// may be before them.
//
// As an example, say we're counting from 5 to 10 and the index only has keys
// for 6, 7, 8, and 9. btreeCursor will point at a 6 key at the start and the
// endCursor will be EOF. If we insert documents with keys 11 during a yield we
// need to relocate the endCursor to point at them as the "end key" of our count.
//
// If we weren't EOF our end position might have moved around. Relocate it.
_endCursor->seek(_params.endKey, _params.endKeyInclusive);
// This can change during yielding.
_shouldDedup = _descriptor->isMultikey();
checkEnd();
}
void Count::invalidate(const DiskLoc& dl, InvalidationType type) {
// The only state we're responsible for holding is what DiskLocs to drop. If a document
// mutates the underlying index cursor will deal with it.
if (INVALIDATION_MUTATION == type) {
return;
}
// If we see this DiskLoc again, it may not be the same document it was before, so we want
// to return it if we see it again.
unordered_set<DiskLoc, DiskLoc::Hasher>::iterator it = _returned.find(dl);
if (it != _returned.end()) {
_returned.erase(it);
}
}
PlanStageStats* Count::getStats() {
// We don't collect stats since this stage is only used by the count command.
// If count ever collects stats we must implement this.
invariant(0);
return NULL;
}
} // namespace mongo
|
