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
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
|
/**
* 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.
*/
#pragma once
#include "mongo/base/static_assert.h"
#include "mongo/bson/bson_validate.h"
#include "mongo/client/constants.h"
#include "mongo/db/jsobj.h"
#include "mongo/db/server_options.h"
#include "mongo/rpc/message.h"
namespace mongo {
class OperationContext;
/* db response format
Query or GetMore: // see struct QueryResult
int resultFlags;
int64 cursorID;
int startingFrom;
int nReturned;
list of marshalled JSObjects;
*/
/* db request message format
unsigned opid; // arbitary; will be echoed back
byte operation;
int options;
then for:
dbInsert:
std::string collection;
a series of JSObjects
dbDelete:
std::string collection;
int flags=0; // 1=DeleteSingle
JSObject query;
dbUpdate:
std::string collection;
int flags; // 1=upsert
JSObject query;
JSObject objectToUpdate;
objectToUpdate may include { $inc: <field> } or { $set: ... }, see struct Mod.
dbQuery:
std::string collection;
int nToSkip;
int nToReturn; // how many you want back as the beginning of the cursor data (0=no limit)
// greater than zero is simply a hint on how many objects to send back per
// "cursor batch".
// a negative number indicates a hard limit.
JSObject query;
[JSObject fieldsToReturn]
dbGetMore:
std::string collection; // redundant, might use for security.
int nToReturn;
int64 cursorID;
dbKillCursors=2007:
int n;
int64 cursorIDs[n];
Note that on Update, there is only one object, which is different
from insert where you can pass a list of objects to insert in the db.
Note that the update field layout is very similar layout to Query.
*/
namespace QueryResult {
#pragma pack(1)
/**
* See http://dochub.mongodb.org/core/mongowireprotocol.
*/
struct Layout {
MsgData::Layout msgdata;
int64_t cursorId;
int32_t startingFrom;
int32_t nReturned;
};
#pragma pack()
class ConstView {
public:
ConstView(const char* storage) : _storage(storage) {}
const char* view2ptr() const {
return storage().view();
}
MsgData::ConstView msgdata() const {
return storage().view(offsetof(Layout, msgdata));
}
int64_t getCursorId() const {
return storage().read<LittleEndian<int64_t>>(offsetof(Layout, cursorId));
}
int32_t getStartingFrom() const {
return storage().read<LittleEndian<int32_t>>(offsetof(Layout, startingFrom));
}
int32_t getNReturned() const {
return storage().read<LittleEndian<int32_t>>(offsetof(Layout, nReturned));
}
const char* data() const {
return storage().view(sizeof(Layout));
}
int32_t dataLen() const {
return msgdata().getLen() - sizeof(Layout);
}
protected:
const ConstDataView& storage() const {
return _storage;
}
private:
ConstDataView _storage;
};
class View : public ConstView {
public:
View(char* data) : ConstView(data) {}
using ConstView::view2ptr;
char* view2ptr() {
return storage().view();
}
using ConstView::msgdata;
MsgData::View msgdata() {
return storage().view(offsetof(Layout, msgdata));
}
void setCursorId(int64_t value) {
storage().write(tagLittleEndian(value), offsetof(Layout, cursorId));
}
void setStartingFrom(int32_t value) {
storage().write(tagLittleEndian(value), offsetof(Layout, startingFrom));
}
void setNReturned(int32_t value) {
storage().write(tagLittleEndian(value), offsetof(Layout, nReturned));
}
int32_t getResultFlags() {
return DataView(msgdata().data()).read<LittleEndian<int32_t>>();
}
void setResultFlags(int32_t value) {
DataView(msgdata().data()).write(tagLittleEndian(value));
}
void setResultFlagsToOk() {
setResultFlags(ResultFlag_AwaitCapable);
}
void initializeResultFlags() {
setResultFlags(0);
}
private:
DataView storage() const {
return const_cast<char*>(ConstView::view2ptr());
}
};
class Value : public EncodedValueStorage<Layout, ConstView, View> {
public:
Value() {
MONGO_STATIC_ASSERT(sizeof(Value) == sizeof(Layout));
}
Value(ZeroInitTag_t zit) : EncodedValueStorage<Layout, ConstView, View>(zit) {}
};
} // namespace QueryResult
/**
* For the database/server protocol, these objects and functions encapsulate the various messages
* transmitted over the connection.
*
* See http://dochub.mongodb.org/core/mongowireprotocol.
*/
class DbMessage {
// Assume sizeof(int) == 4 bytes
MONGO_STATIC_ASSERT(sizeof(int) == 4);
public:
/**
* Note: DbMessage constructor reads the first 4 bytes and stores it in reserved
*/
DbMessage(const Message& msg);
/**
* Indicates whether this message is expected to have a ns.
*/
bool messageShouldHaveNs() const {
return (_msg.operation() >= dbUpdate) & (_msg.operation() <= dbDelete);
}
/**
* Returns the 32 bit field before the ns.
* track all bit usage here as its cross op
* 0: InsertOption_ContinueOnError
* 1: fromWriteback
*/
int reservedField() const {
return _reserved;
}
const char* getns() const;
int getQueryNToReturn() const;
int pullInt();
long long pullInt64();
const char* getArray(size_t count) const;
/**
* Used by insert and update msgs
*/
bool moreJSObjs() const {
return _nextjsobj != nullptr && _nextjsobj != _theEnd;
}
BSONObj nextJsObj();
const Message& msg() const {
return _msg;
}
const char* markGet() const {
return _nextjsobj;
}
void markSet() {
_mark = _nextjsobj;
}
void markReset(const char* toMark);
private:
// Check if we have enough data to read
template <typename T>
void checkRead(const char* start, size_t count = 0) const;
// Read some type without advancing our position
template <typename T>
T read() const;
// Read some type, and advance our position
template <typename T>
T readAndAdvance();
const Message& _msg;
int _reserved; // flags or zero depending on packet, starts the packet
const char* _nsStart; // start of namespace string, +4 from message start
const char* _nextjsobj; // current position reading packet
const char* _theEnd; // end of packet
const char* _mark;
unsigned int _nsLen;
};
/**
* The query field 'options' can have these bits set:
*/
enum QueryOptions {
/**
* Tailable means cursor is not closed when the last data is retrieved. Rather, the cursor
* marks the final object's position. You can resume using the cursor later, from where it was
* located, if more data were received. Set on dbQuery and dbGetMore.
*
* like any "latent cursor", the cursor may become invalid at some point -- for example if that
* final object it references were deleted.
*/
QueryOption_CursorTailable = 1 << 1,
/**
* Allow query of replica secondary. Normally these return an error except for namespace
* "local".
*/
QueryOption_SecondaryOk = 1 << 2,
/**
* In previous versions of the server, clients were required to set this option in order to
* enable an optimized oplog scan. As of 4.4, the server will apply the optimization for
* eligible queries regardless of whether this flag is set.
*
* This bit is reserved for compatibility with old clients, but it should not be set by modern
* clients.
*
* New server code should not use this flag.
*/
QueryOption_OplogReplay_DEPRECATED = 1 << 3,
/**
* The server normally times out idle cursors after an inactivity period to prevent excess
* memory uses. Set this option to prevent that.
*/
QueryOption_NoCursorTimeout = 1 << 4,
/**
* Use with QueryOption_CursorTailable. If we are at the end of the data, block for a while
* rather than returning no data. After a timeout period, we do return as normal.
*/
QueryOption_AwaitData = 1 << 5,
/**
* Stream the data down full blast in multiple "more" packages, on the assumption that the
* client will fully read all data queried. Faster when you are pulling a lot of data and know
* you want to pull it all down. Note: it is not allowed to not read all the data unless you
* close the connection.
*
* Use the query( std::function<void(const BSONObj&)> f, ... ) version of the connection's
* query() method, and it will take care of all the details for you.
*/
QueryOption_Exhaust = 1 << 6,
/**
* When sharded, this means its ok to return partial results. Usually we will fail a query if
* all required shards aren't up. If this is set, it'll be a partial result set.
*/
QueryOption_PartialResults = 1 << 7,
// DBClientCursor reserves flag 1 << 30 to force the use of OP_QUERY.
QueryOption_AllSupported = QueryOption_CursorTailable | QueryOption_SecondaryOk |
QueryOption_NoCursorTimeout | QueryOption_AwaitData | QueryOption_Exhaust |
QueryOption_PartialResults,
QueryOption_AllSupportedForSharding = QueryOption_CursorTailable | QueryOption_SecondaryOk |
QueryOption_NoCursorTimeout | QueryOption_AwaitData | QueryOption_PartialResults,
};
/**
* A request to run a query, received from the database.
*/
class QueryMessage {
public:
const char* ns;
int ntoskip;
int ntoreturn;
int queryOptions;
BSONObj query;
BSONObj fields;
/**
* Parses the message into the above fields.
* Warning: constructor mutates DbMessage.
*/
explicit QueryMessage(DbMessage& d) {
ns = d.getns();
ntoskip = d.pullInt();
ntoreturn = d.pullInt();
query = d.nextJsObj();
if (d.moreJSObjs()) {
fields = d.nextJsObj();
}
queryOptions = DataView(d.msg().header().data()).read<LittleEndian<int32_t>>();
}
/**
* A non-mutating constructor from the whole message.
*/
explicit QueryMessage(const Message& message) {
DbMessage dbm(message);
*this = QueryMessage(dbm);
}
};
template <typename Func>
Message makeMessage(NetworkOp op, Func&& bodyBuilder) {
BufBuilder b;
b.skip(sizeof(MSGHEADER::Layout));
bodyBuilder(b);
const int size = b.len();
auto out = Message(b.release());
out.header().setOperation(op);
out.header().setLen(size);
return out;
}
/**
* Builds a legacy OP_QUERY message.
*/
Message makeDeprecatedQueryMessage(StringData ns,
BSONObj query,
int nToReturn,
int nToSkip,
const BSONObj* fieldsToReturn,
int queryOptions);
enum InsertOptions {
/**
* With muli-insert keep processing inserts if one fails.
*/
InsertOption_ContinueOnError = 1 << 0
};
/**
* Builds a legacy OP_INSERT message.
*/
Message makeDeprecatedInsertMessage(StringData ns,
const BSONObj* objs,
size_t count,
int flags = 0);
/**
* A response to a DbMessage.
*
* Order of fields makes DbResponse{funcReturningMessage()} valid.
*/
struct DbResponse {
// If empty, nothing will be returned to the client.
Message response;
// For exhaust commands, indicates whether the command should be run again.
bool shouldRunAgainForExhaust = false;
// The next invocation for an exhaust command. If this is boost::none, the previous invocation
// should be reused for the next invocation.
boost::optional<BSONObj> nextInvocation;
};
/**
* Helper to build an error DbResponse for OP_QUERY and OP_GET_MORE.
*/
DbResponse makeErrorResponseToDeprecatedOpQuery(StringData errorMsg);
} // namespace mongo
|