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
|
/**
* 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.
*/
#define MONGO_LOG_DEFAULT_COMPONENT ::mongo::logger::LogComponent::kQuery
#include "mongo/platform/basic.h"
#include <memory>
#include "mongo/base/disallow_copying.h"
#include "mongo/db/auth/authorization_session.h"
#include "mongo/db/catalog/collection.h"
#include "mongo/db/catalog/database.h"
#include "mongo/db/client.h"
#include "mongo/db/clientcursor.h"
#include "mongo/db/commands.h"
#include "mongo/db/commands/run_aggregate.h"
#include "mongo/db/db_raii.h"
#include "mongo/db/exec/working_set_common.h"
#include "mongo/db/matcher/extensions_callback_real.h"
#include "mongo/db/query/cursor_response.h"
#include "mongo/db/query/explain.h"
#include "mongo/db/query/find.h"
#include "mongo/db/query/find_common.h"
#include "mongo/db/query/get_executor.h"
#include "mongo/db/repl/replication_coordinator.h"
#include "mongo/db/s/collection_sharding_state.h"
#include "mongo/db/server_parameters.h"
#include "mongo/db/service_context.h"
#include "mongo/db/stats/counters.h"
#include "mongo/rpc/get_status_from_command_result.h"
#include "mongo/util/log.h"
namespace mongo {
namespace {
const char kTermField[] = "term";
} // namespace
/**
* A command for running .find() queries.
*/
class FindCmd : public BasicCommand {
MONGO_DISALLOW_COPYING(FindCmd);
public:
FindCmd() : BasicCommand("find") {}
virtual bool supportsWriteConcern(const BSONObj& cmd) const override {
return false;
}
bool slaveOk() const override {
return false;
}
bool slaveOverrideOk() const override {
return true;
}
bool maintenanceOk() const override {
return false;
}
bool adminOnly() const override {
return false;
}
bool supportsNonLocalReadConcern(const std::string& dbName, const BSONObj& cmdObj) const final {
return true;
}
void help(std::stringstream& help) const override {
help << "query for documents";
}
LogicalOp getLogicalOp() const override {
return LogicalOp::opQuery;
}
ReadWriteType getReadWriteType() const {
return ReadWriteType::kRead;
}
std::size_t reserveBytesForReply() const override {
return FindCommon::kInitReplyBufferSize;
}
/**
* A find command does not increment the command counter, but rather increments the
* query counter.
*/
bool shouldAffectCommandCounter() const override {
return false;
}
Status checkAuthForCommand(Client* client,
const std::string& dbname,
const BSONObj& cmdObj) override {
const NamespaceString nss(parseNs(dbname, cmdObj));
auto hasTerm = cmdObj.hasField(kTermField);
return AuthorizationSession::get(client)->checkAuthForFind(nss, hasTerm);
}
Status explain(OperationContext* opCtx,
const std::string& dbname,
const BSONObj& cmdObj,
ExplainOptions::Verbosity verbosity,
BSONObjBuilder* out) const override {
const NamespaceString nss(parseNs(dbname, cmdObj));
if (!nss.isValid()) {
return {ErrorCodes::InvalidNamespace,
str::stream() << "Invalid collection name: " << nss.ns()};
}
// Parse the command BSON to a QueryRequest.
const bool isExplain = true;
auto qrStatus = QueryRequest::makeFromFindCommand(nss, cmdObj, isExplain);
if (!qrStatus.isOK()) {
return qrStatus.getStatus();
}
// Finish the parsing step by using the QueryRequest to create a CanonicalQuery.
ExtensionsCallbackReal extensionsCallback(opCtx, &nss);
auto statusWithCQ =
CanonicalQuery::canonicalize(opCtx, std::move(qrStatus.getValue()), extensionsCallback);
if (!statusWithCQ.isOK()) {
return statusWithCQ.getStatus();
}
std::unique_ptr<CanonicalQuery> cq = std::move(statusWithCQ.getValue());
// Acquire locks. If the namespace is a view, we release our locks and convert the query
// request into an aggregation command.
AutoGetCollectionOrViewForReadCommand ctx(opCtx, nss);
if (ctx.getView()) {
// Relinquish locks. The aggregation command will re-acquire them.
ctx.releaseLocksForView();
// Convert the find command into an aggregation using $match (and other stages, as
// necessary), if possible.
const auto& qr = cq->getQueryRequest();
auto viewAggregationCommand = qr.asAggregationCommand();
if (!viewAggregationCommand.isOK())
return viewAggregationCommand.getStatus();
// Create the agg request equivalent of the find operation, with the explain verbosity
// included.
auto aggRequest = AggregationRequest::parseFromBSON(
nss, viewAggregationCommand.getValue(), verbosity);
if (!aggRequest.isOK()) {
return aggRequest.getStatus();
}
try {
return runAggregate(
opCtx, nss, aggRequest.getValue(), viewAggregationCommand.getValue(), *out);
} catch (DBException& error) {
if (error.code() == ErrorCodes::InvalidPipelineOperator) {
return {ErrorCodes::InvalidPipelineOperator,
str::stream() << "Unsupported in view pipeline: " << error.what()};
}
return error.toStatus();
}
}
// The collection may be NULL. If so, getExecutor() should handle it by returning an
// execution tree with an EOFStage.
Collection* collection = ctx.getCollection();
// We have a parsed query. Time to get the execution plan for it.
auto statusWithPlanExecutor =
getExecutorFind(opCtx, collection, nss, std::move(cq), PlanExecutor::YIELD_AUTO);
if (!statusWithPlanExecutor.isOK()) {
return statusWithPlanExecutor.getStatus();
}
auto exec = std::move(statusWithPlanExecutor.getValue());
// Got the execution tree. Explain it.
Explain::explainStages(exec.get(), collection, verbosity, out);
return Status::OK();
}
/**
* Runs a query using the following steps:
* --Parsing.
* --Acquire locks.
* --Plan query, obtaining an executor that can run it.
* --Generate the first batch.
* --Save state for getMore, transferring ownership of the executor to a ClientCursor.
* --Generate response to send to the client.
*/
bool run(OperationContext* opCtx,
const std::string& dbname,
const BSONObj& cmdObj,
BSONObjBuilder& result) override {
const NamespaceString nss(parseNsOrUUID(opCtx, dbname, cmdObj));
// Although it is a command, a find command gets counted as a query.
globalOpCounters.gotQuery();
// Parse the command BSON to a QueryRequest.
const bool isExplain = false;
auto qrStatus = QueryRequest::makeFromFindCommand(nss, cmdObj, isExplain);
if (!qrStatus.isOK()) {
return appendCommandStatus(result, qrStatus.getStatus());
}
auto& qr = qrStatus.getValue();
// Validate term before acquiring locks, if provided.
if (auto term = qr->getReplicationTerm()) {
auto replCoord = repl::ReplicationCoordinator::get(opCtx);
Status status = replCoord->updateTerm(opCtx, *term);
// Note: updateTerm returns ok if term stayed the same.
if (!status.isOK()) {
return appendCommandStatus(result, status);
}
}
// Fill out curop information.
//
// We pass negative values for 'ntoreturn' and 'ntoskip' to indicate that these values
// should be omitted from the log line. Limit and skip information is already present in the
// find command parameters, so these fields are redundant.
const int ntoreturn = -1;
const int ntoskip = -1;
beginQueryOp(opCtx, nss, cmdObj, ntoreturn, ntoskip);
// Finish the parsing step by using the QueryRequest to create a CanonicalQuery.
ExtensionsCallbackReal extensionsCallback(opCtx, &nss);
auto statusWithCQ = CanonicalQuery::canonicalize(opCtx, std::move(qr), extensionsCallback);
if (!statusWithCQ.isOK()) {
return appendCommandStatus(result, statusWithCQ.getStatus());
}
std::unique_ptr<CanonicalQuery> cq = std::move(statusWithCQ.getValue());
// Acquire locks. If the query is on a view, we release our locks and convert the query
// request into an aggregation command.
AutoGetCollectionOrViewForReadCommand ctx(opCtx, nss);
Collection* collection = ctx.getCollection();
if (ctx.getView()) {
// Relinquish locks. The aggregation command will re-acquire them.
ctx.releaseLocksForView();
// Convert the find command into an aggregation using $match (and other stages, as
// necessary), if possible.
const auto& qr = cq->getQueryRequest();
auto viewAggregationCommand = qr.asAggregationCommand();
if (!viewAggregationCommand.isOK())
return appendCommandStatus(result, viewAggregationCommand.getStatus());
BSONObj aggResult = Command::runCommandDirectly(
opCtx,
OpMsgRequest::fromDBAndBody(dbname, std::move(viewAggregationCommand.getValue())));
auto status = getStatusFromCommandResult(aggResult);
if (status.code() == ErrorCodes::InvalidPipelineOperator) {
return appendCommandStatus(
result,
{ErrorCodes::InvalidPipelineOperator,
str::stream() << "Unsupported in view pipeline: " << status.reason()});
}
result.resetToEmpty();
result.appendElements(aggResult);
return status.isOK();
}
// Get the execution plan for the query.
auto statusWithPlanExecutor =
getExecutorFind(opCtx, collection, nss, std::move(cq), PlanExecutor::YIELD_AUTO);
if (!statusWithPlanExecutor.isOK()) {
return appendCommandStatus(result, statusWithPlanExecutor.getStatus());
}
auto exec = std::move(statusWithPlanExecutor.getValue());
{
stdx::lock_guard<Client> lk(*opCtx->getClient());
CurOp::get(opCtx)->setPlanSummary_inlock(Explain::getPlanSummary(exec.get()));
}
if (!collection) {
// No collection. Just fill out curop indicating that there were zero results and
// there is no ClientCursor id, and then return.
const long long numResults = 0;
const CursorId cursorId = 0;
endQueryOp(opCtx, collection, *exec, numResults, cursorId);
appendCursorResponseObject(cursorId, nss.ns(), BSONArray(), &result);
return true;
}
const QueryRequest& originalQR = exec->getCanonicalQuery()->getQueryRequest();
// Stream query results, adding them to a BSONArray as we go.
CursorResponseBuilder firstBatch(/*isInitialResponse*/ true, &result);
BSONObj obj;
PlanExecutor::ExecState state = PlanExecutor::ADVANCED;
long long numResults = 0;
while (!FindCommon::enoughForFirstBatch(originalQR, numResults) &&
PlanExecutor::ADVANCED == (state = exec->getNext(&obj, NULL))) {
// If we can't fit this result inside the current batch, then we stash it for later.
if (!FindCommon::haveSpaceForNext(obj, numResults, firstBatch.bytesUsed())) {
exec->enqueue(obj);
break;
}
// Add result to output buffer.
firstBatch.append(obj);
numResults++;
}
// Throw an assertion if query execution fails for any reason.
if (PlanExecutor::FAILURE == state || PlanExecutor::DEAD == state) {
firstBatch.abandon();
error() << "Plan executor error during find command: " << PlanExecutor::statestr(state)
<< ", stats: " << redact(Explain::getWinningPlanStats(exec.get()));
return appendCommandStatus(result,
Status(ErrorCodes::OperationFailed,
str::stream()
<< "Executor error during find command: "
<< WorkingSetCommon::toStatusString(obj)));
}
// Before saving the cursor, ensure that whatever plan we established happened with the
// expected collection version
auto css = CollectionShardingState::get(opCtx, nss);
css->checkShardVersionOrThrow(opCtx);
// Set up the cursor for getMore.
CursorId cursorId = 0;
if (shouldSaveCursor(opCtx, collection, state, exec.get())) {
// Create a ClientCursor containing this plan executor and register it with the cursor
// manager.
ClientCursorPin pinnedCursor = collection->getCursorManager()->registerCursor(
opCtx,
{std::move(exec),
nss,
AuthorizationSession::get(opCtx->getClient())->getAuthenticatedUserNames(),
opCtx->recoveryUnit()->isReadingFromMajorityCommittedSnapshot(),
cmdObj});
cursorId = pinnedCursor.getCursor()->cursorid();
invariant(!exec);
PlanExecutor* cursorExec = pinnedCursor.getCursor()->getExecutor();
// State will be restored on getMore.
cursorExec->saveState();
cursorExec->detachFromOperationContext();
// We assume that cursors created through a DBDirectClient are always used from their
// original OperationContext, so we do not need to move time to and from the cursor.
if (!opCtx->getClient()->isInDirectClient()) {
pinnedCursor.getCursor()->setLeftoverMaxTimeMicros(
opCtx->getRemainingMaxTimeMicros());
}
pinnedCursor.getCursor()->setPos(numResults);
// Fill out curop based on the results.
endQueryOp(opCtx, collection, *cursorExec, numResults, cursorId);
} else {
endQueryOp(opCtx, collection, *exec, numResults, cursorId);
}
// Generate the response object to send to the client.
firstBatch.done(cursorId, nss.ns());
return true;
}
} findCmd;
} // namespace mongo
|