mirror of https://github.com/mongodb/mongo
817 lines
32 KiB
JavaScript
817 lines
32 KiB
JavaScript
// Tests that an aggregation pipeline can be optimized away and the query can be answered using
|
|
// just the query layer if the pipeline has only one $cursor source, or if the pipeline can be
|
|
// collapsed into a single $cursor source pipeline. The resulting cursor in this case will look
|
|
// like what the client would have gotten from find command.
|
|
//
|
|
// Relies on the pipeline stages to be collapsed into a single $cursor stage, so pipelines cannot be
|
|
// wrapped into a facet stage to not prevent this optimization. Also, this test is not prepared to
|
|
// handle explain output for sharded collections.
|
|
// This test makes assumptions about how the explain output will be formatted, so cannot be run when
|
|
// pipeline optimization is disabled.
|
|
// @tags: [
|
|
// assumes_unsharded_collection,
|
|
// do_not_wrap_aggregations_in_facets,
|
|
// requires_pipeline_optimization,
|
|
// requires_profiling,
|
|
// ]
|
|
(function() {
|
|
"use strict";
|
|
|
|
load("jstests/concurrency/fsm_workload_helpers/server_types.js"); // For isWiredTiger.
|
|
load("jstests/libs/analyze_plan.js"); // For 'aggPlanHasStage' and other explain helpers.
|
|
load("jstests/libs/fixture_helpers.js"); // For 'isMongos' and 'isSharded'.
|
|
load("jstests/libs/sbe_util.js"); // For checkSBEEnabled.
|
|
|
|
const coll = db.optimize_away_pipeline;
|
|
coll.drop();
|
|
assert.commandWorked(coll.insert({_id: 1, x: 10}));
|
|
assert.commandWorked(coll.insert({_id: 2, x: 20}));
|
|
assert.commandWorked(coll.insert({_id: 3, x: 30}));
|
|
|
|
// Asserts that the give pipeline has *not* been optimized away and the request is answered
|
|
// using the aggregation module. There should be pipeline stages present in the explain output.
|
|
// The functions also asserts that a query stage passed in the 'stage' argument is present in
|
|
// the explain output. If 'expectedResult' is provided, the pipeline is executed and the
|
|
// returned result as validated agains the expected result without respecting the order of the
|
|
// documents. If 'preserveResultOrder' is 'true' - the order is respected.
|
|
//
|
|
// If 'optimizedAwayStages' is non-null, then it should contain a list of agg plan stages that
|
|
// should *not* be present in the pipeline, since their execution was pushed down into the query
|
|
// layer. The test will verify that this pushdown is reflected in explain output.
|
|
function assertPipelineUsesAggregation({
|
|
pipeline = [],
|
|
pipelineOptions = {},
|
|
expectedStages = null,
|
|
expectedResult = null,
|
|
preserveResultOrder = false,
|
|
optimizedAwayStages = null
|
|
} = {}) {
|
|
const explainOutput = coll.explain().aggregate(pipeline, pipelineOptions);
|
|
|
|
assert(isAggregationPlan(explainOutput), explainOutput);
|
|
assert(!isQueryPlan(explainOutput), explainOutput);
|
|
|
|
if (optimizedAwayStages) {
|
|
for (let stage of optimizedAwayStages) {
|
|
assert(!aggPlanHasStage(explainOutput, stage), explainOutput);
|
|
}
|
|
}
|
|
|
|
let cursor = getAggPlanStage(explainOutput, "$cursor");
|
|
if (cursor) {
|
|
cursor = cursor.$cursor;
|
|
} else {
|
|
cursor = getAggPlanStage(explainOutput, "$geoNearCursor").$geoNearCursor;
|
|
}
|
|
|
|
assert(cursor, explainOutput);
|
|
assert(cursor.queryPlanner.optimizedPipeline === undefined, explainOutput);
|
|
|
|
if (expectedStages) {
|
|
for (let expectedStage of expectedStages) {
|
|
assert(aggPlanHasStage(explainOutput, expectedStage), explainOutput);
|
|
}
|
|
}
|
|
|
|
if (expectedResult) {
|
|
const actualResult = coll.aggregate(pipeline, pipelineOptions).toArray();
|
|
if (preserveResultOrder) {
|
|
assert.docEq(actualResult, expectedResult);
|
|
} else {
|
|
assert.sameMembers(actualResult, expectedResult);
|
|
}
|
|
}
|
|
|
|
return explainOutput;
|
|
}
|
|
|
|
// Asserts that the give pipeline has been optimized away and the request is answered using
|
|
// just the query module. There should be no pipeline stages present in the explain output.
|
|
// The functions also asserts that a query stage passed in the 'stage' argument is present in
|
|
// the explain output. If 'expectedResult' is provided, the pipeline is executed and the
|
|
// returned result as validated agains the expected result without respecting the order of the
|
|
// documents. If 'preserveResultOrder' is 'true' - the order is respected.
|
|
function assertPipelineDoesNotUseAggregation({
|
|
pipeline = [],
|
|
pipelineOptions = {},
|
|
expectedStages = null,
|
|
expectedResult = null,
|
|
preserveResultOrder = false
|
|
} = {}) {
|
|
const explainOutput = coll.explain().aggregate(pipeline, pipelineOptions);
|
|
|
|
assert(!isAggregationPlan(explainOutput), explainOutput);
|
|
assert(isQueryPlan(explainOutput), explainOutput);
|
|
if (explainOutput.hasOwnProperty("shards")) {
|
|
Object.keys(explainOutput.shards)
|
|
.forEach((shard) =>
|
|
assert(explainOutput.shards[shard].queryPlanner.optimizedPipeline === true,
|
|
explainOutput));
|
|
} else {
|
|
assert(explainOutput.queryPlanner.optimizedPipeline === true, explainOutput);
|
|
}
|
|
|
|
if (expectedStages) {
|
|
for (let expectedStage of expectedStages) {
|
|
assert(planHasStage(db, explainOutput, expectedStage), explainOutput);
|
|
}
|
|
}
|
|
|
|
if (expectedResult) {
|
|
const actualResult = coll.aggregate(pipeline, pipelineOptions).toArray();
|
|
if (preserveResultOrder) {
|
|
assert.docEq(actualResult, expectedResult);
|
|
} else {
|
|
assert.sameMembers(actualResult, expectedResult);
|
|
}
|
|
}
|
|
|
|
return explainOutput;
|
|
}
|
|
|
|
// Test that getMore works with the optimized query.
|
|
function testGetMore({command = null, expectedResult = null} = {}) {
|
|
const documents =
|
|
new DBCommandCursor(db, assert.commandWorked(db.runCommand(command)), 1 /* batchsize */)
|
|
.toArray();
|
|
assert.sameMembers(documents, expectedResult);
|
|
}
|
|
|
|
const groupPushdownEnabled = checkSBEEnabled(db, ["featureFlagSBEGroupPushdown"]);
|
|
|
|
// Calls 'assertPushdownEnabled' if groupPushdownEnabled is 'true'. Otherwise, it calls
|
|
// 'assertPushdownDisabled'.
|
|
function assertPipelineIfGroupPushdown(assertPushdownEnabled, assertPushdownDisabled) {
|
|
return groupPushdownEnabled ? assertPushdownEnabled() : assertPushdownDisabled();
|
|
}
|
|
|
|
let explainOutput;
|
|
|
|
// Basic pipelines.
|
|
|
|
// Test basic scenarios when a pipeline has a single $cursor stage or can be collapsed into a
|
|
// single cursor stage.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [],
|
|
expectedStages: ["COLLSCAN"],
|
|
expectedResult: [{_id: 1, x: 10}, {_id: 2, x: 20}, {_id: 3, x: 30}]
|
|
});
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: [{$match: {x: 20}}], expectedStage: "COLLSCAN", expectedResult: [{_id: 2, x: 20}]});
|
|
|
|
// Pipelines with a collation.
|
|
|
|
// Test a simple pipeline with a case-insensitive collation.
|
|
assert.commandWorked(coll.insert({_id: 4, x: 40, b: "abc"}));
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {b: "ABC"}}],
|
|
pipelineOptions: {collation: {locale: "en_US", strength: 2}},
|
|
expectedStages: ["COLLSCAN"],
|
|
expectedResult: [{_id: 4, x: 40, b: "abc"}]
|
|
});
|
|
assert.commandWorked(coll.deleteOne({_id: 4}));
|
|
|
|
// Pipelines with covered queries.
|
|
|
|
// We can collapse a covered query into a single $cursor when $project and $sort are present and
|
|
// the latter is near the front of the pipeline. Skip this test in sharded modes as we cannot
|
|
// correctly handle explain output in plan analyzer helper functions.
|
|
assert.commandWorked(coll.createIndex({x: 1}));
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$sort: {x: 1}}, {$project: {x: 1, _id: 0}}],
|
|
expectedStages: ["IXSCAN"],
|
|
expectedResult: [{x: 10}, {x: 20}, {x: 30}],
|
|
preserveResultOrder: true
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$sort: {x: 1}}, {$project: {x: 1, _id: 0}}],
|
|
expectedStages: ["IXSCAN"],
|
|
expectedResult: [{x: 20}, {x: 30}],
|
|
preserveResultOrder: true
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$sort: {x: 1}}, {$limit: 1}, {$project: {x: 1, _id: 0}}],
|
|
expectedStages: ["IXSCAN"],
|
|
expectedResult: [{x: 20}]
|
|
});
|
|
// However, when the $project is computed, pushing it down into the find() layer would sometimes
|
|
// have the effect of reordering it before the $sort and $limit. This can cause a valid query to
|
|
// throw an error, as in SERVER-54128.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [
|
|
{$match: {x: {$gte: 20}}},
|
|
{$sort: {x: 1}},
|
|
{$limit: 1},
|
|
{$project: {x: {$substr: ["$y", 0, 1]}, _id: 0}}
|
|
],
|
|
expectedStages: ["IXSCAN"],
|
|
expectedResult: [{x: ""}]
|
|
});
|
|
assert.commandWorked(coll.dropIndexes());
|
|
|
|
assert.commandWorked(coll.insert({_id: 4, x: 40, a: {b: "ab1"}}));
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$project: {x: 1, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
expectedResult: [{x: 10}, {x: 20}, {x: 30}, {x: 40}]
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: 20}}, {$project: {x: 1, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
expectedResult: [{x: 20}]
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$project: {x: 1, "a.b": 1, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT"],
|
|
expectedResult: [{x: 10}, {x: 20}, {x: 30}, {x: 40, a: {b: "ab1"}}]
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: 40}}, {$project: {"a.b": 1, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT"],
|
|
expectedResult: [{a: {b: "ab1"}}]
|
|
});
|
|
// We can collapse a $project stage if it has a complex pipeline expression.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$project: {x: {$substr: ["$y", 0, 1]}, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT"]
|
|
});
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: 20}}, {$project: {x: {$substr: ["$y", 0, 1]}, _id: 0}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT"]
|
|
});
|
|
assert.commandWorked(coll.deleteOne({_id: 4}));
|
|
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$skip: 1}],
|
|
expectedStages: ["COLLSCAN", "SKIP"],
|
|
expectedResult: [{_id: 3, x: 30}]
|
|
});
|
|
|
|
// Pipelines which cannot be optimized away.
|
|
|
|
// We cannot optimize away a pipeline if there are stages which have no equivalent in the
|
|
// find command.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$count: "count"}],
|
|
expectedStages: ["COLLSCAN"],
|
|
expectedResult: [{count: 2}]
|
|
});
|
|
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
return assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$group: {_id: "null", s: {$sum: "$x"}}}],
|
|
expectedStages: ["COLLSCAN", "GROUP"],
|
|
expectedResult: [{_id: "null", s: 50}],
|
|
});
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$group: {_id: "null", s: {$sum: "$x"}}}],
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
expectedResult: [{_id: "null", s: 50}],
|
|
});
|
|
});
|
|
|
|
// Test that we can optimize away a pipeline with a $text search predicate.
|
|
assert.commandWorked(coll.createIndex({y: "text"}));
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: [{$match: {$text: {$search: "abc"}}}], expectedStages: ["IXSCAN"]});
|
|
// Test that $match, $sort, and $project all get answered by the PlanStage layer for a $text query.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline:
|
|
[{$match: {$text: {$search: "abc"}}}, {$sort: {sortField: 1}}, {$project: {a: 1, b: 1}}],
|
|
expectedStages: ["TEXT_MATCH", "SORT", "PROJECTION_SIMPLE"],
|
|
optimizedAwayStages: ["$match", "$sort", "$project"]
|
|
});
|
|
assert.commandWorked(coll.dropIndexes());
|
|
|
|
// We cannot optimize away geo near queries.
|
|
assert.commandWorked(coll.createIndex({"y": "2d"}));
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$geoNear: {near: [0, 0], distanceField: "y", spherical: true}}],
|
|
expectedStages: ["GEO_NEAR_2D"],
|
|
});
|
|
assert.commandWorked(coll.dropIndexes());
|
|
|
|
// Test cases around pushdown of $limit.
|
|
assert.commandWorked(coll.createIndex({x: 1}));
|
|
|
|
// A lone $limit pipeline can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$limit: 1}],
|
|
expectedStages: ["COLLSCAN", "LIMIT"],
|
|
});
|
|
|
|
// $match followed by $limit can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: 20}}, {$limit: 1}],
|
|
expectedStages: ["IXSCAN", "LIMIT"],
|
|
expectedResult: [{_id: 2, x: 20}],
|
|
});
|
|
|
|
// $limit followed by $match cannot be fully optimized away. The $limit is pushed down, but the
|
|
// $match is executed in the agg layer.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$limit: 1}, {$match: {x: 20}}],
|
|
expectedStages: ["COLLSCAN", "LIMIT"],
|
|
optimizedAwayStages: ["$limit"],
|
|
});
|
|
|
|
// $match, $project, $limit can be optimized away when the projection is covered.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$project: {_id: 0, x: 1}}, {$limit: 1}],
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{x: 20}],
|
|
});
|
|
|
|
// $match, $project, and $limit can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$project: {_id: 0, x: 1, y: 1}}, {$limit: 1}],
|
|
expectedStages: ["IXSCAN", "FETCH", "LIMIT", "PROJECTION_SIMPLE"],
|
|
expectedResult: [{x: 20}],
|
|
optimizedAwayStages: ["$limit", "$project"],
|
|
});
|
|
|
|
// $match, $project, $limit, $sort cannot be optimized away because the $limit comes before the
|
|
// $sort.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$project: {_id: 0, x: 1}}, {$limit: 1}, {$sort: {x: 1}}],
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{x: 20}],
|
|
optimizedAwayStages: ["$project", "$limit"],
|
|
});
|
|
|
|
// $match, $sort, $limit can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$sort: {x: -1}}, {$limit: 2}],
|
|
expectedStages: ["IXSCAN", "LIMIT"],
|
|
expectedResult: [{_id: 3, x: 30}, {_id: 2, x: 20}],
|
|
});
|
|
|
|
// $match, $sort, $limit, $project can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline:
|
|
[{$match: {x: {$gte: 20}}}, {$sort: {x: -1}}, {$limit: 2}, {$project: {_id: 0, x: 1}}],
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{x: 30}, {x: 20}],
|
|
});
|
|
|
|
// $match, $sort, $project, $limit can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline:
|
|
[{$match: {x: {$gte: 20}}}, {$sort: {x: -1}}, {$project: {_id: 0, x: 1}}, {$limit: 2}],
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{x: 30}, {x: 20}],
|
|
});
|
|
|
|
// $match, $sort, $limit, $project can be optimized away, where limits must swap and combine to
|
|
// enable pushdown.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [
|
|
{$match: {x: {$gte: 20}}},
|
|
{$sort: {x: -1}},
|
|
{$limit: 3},
|
|
{$project: {_id: 0, x: 1}},
|
|
{$limit: 2}
|
|
],
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{x: 30}, {x: 20}],
|
|
});
|
|
|
|
let pipeline = [{$sort: {x: 1}}, {$limit: 2}, {$group: {_id: null, s: {$sum: "$x"}}}];
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
return assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT", "GROUP"],
|
|
expectedResult: [{_id: null, s: 30}],
|
|
});
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
expectedResult: [{_id: null, s: 30}],
|
|
optimizedAwayStages: ["$sort", "$limit"],
|
|
});
|
|
});
|
|
|
|
// Test that $limit can be pushed down before a group, but it prohibits the DISTINCT_SCAN
|
|
// optimization.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$group: {_id: "$x"}}],
|
|
expectedStages: ["DISTINCT_SCAN", "PROJECTION_COVERED"],
|
|
expectedResult: [{_id: 10}, {_id: 20}, {_id: 30}],
|
|
});
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$sort: {x: 1}}, {$group: {_id: "$x"}}],
|
|
expectedStages: ["DISTINCT_SCAN", "PROJECTION_COVERED"],
|
|
expectedResult: [{_id: 10}, {_id: 20}, {_id: 30}],
|
|
optimizedAwayStages: ["$sort"],
|
|
});
|
|
|
|
pipeline = [{$limit: 2}, {$group: {_id: "$x"}}];
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
return assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "LIMIT", "GROUP"],
|
|
});
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "LIMIT"],
|
|
optimizedAwayStages: ["$limit"],
|
|
});
|
|
});
|
|
|
|
pipeline = [{$sort: {x: 1}}, {$limit: 2}, {$group: {_id: "$x"}}];
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
return assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT", "GROUP"],
|
|
});
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
optimizedAwayStages: ["$sort", "$limit"],
|
|
});
|
|
});
|
|
|
|
// $limit after a group has no effect on our ability to produce a DISTINCT_SCAN plan.
|
|
assertPipelineUsesAggregation({
|
|
pipeline: [{$group: {_id: "$x"}}, {$sort: {_id: 1}}, {$limit: 2}],
|
|
expectedStages: ["DISTINCT_SCAN", "PROJECTION_COVERED"],
|
|
expectedResult: [{_id: 10}, {_id: 20}],
|
|
});
|
|
|
|
// For $limit, $project, $limit, we can optimize away both $limit stages.
|
|
pipeline = [{$match: {x: {$gte: 0}}}, {$limit: 2}, {$project: {_id: 0, x: 1}}, {$limit: 1}];
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT"],
|
|
});
|
|
// Make sure that we end up using the smaller limit of 1.
|
|
let explain = coll.explain().aggregate(pipeline);
|
|
let limitStage = getAggPlanStage(explain, "LIMIT");
|
|
assert.neq(null, limitStage, explain);
|
|
assert.eq(1, limitStage.limitAmount, explain);
|
|
|
|
// We can optimize away interleaved $limit and $skip after a project.
|
|
pipeline = [
|
|
{$match: {x: {$gte: 0}}},
|
|
{$project: {_id: 0, x: 1}},
|
|
{$skip: 20},
|
|
{$limit: 15},
|
|
{$skip: 10},
|
|
{$limit: 7}
|
|
];
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["IXSCAN", "PROJECTION_COVERED", "LIMIT", "SKIP"],
|
|
optimizedAwayStages: ["$match", "$limit", "$skip"],
|
|
});
|
|
explain = coll.explain().aggregate(pipeline);
|
|
|
|
let skipStage = getAggPlanStage(explain, "SKIP");
|
|
assert.neq(null, skipStage, explain);
|
|
assert.eq(30, skipStage.skipAmount, explain);
|
|
|
|
limitStage = getAggPlanStage(explain, "LIMIT");
|
|
assert.neq(null, limitStage, explain);
|
|
assert.eq(5, limitStage.limitAmount, explain);
|
|
|
|
assert.commandWorked(coll.dropIndexes());
|
|
|
|
// $sort can be optimized away even if there is no index to provide the sort.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [
|
|
{$sort: {x: -1}},
|
|
],
|
|
expectedStages: ["COLLSCAN", "SORT"],
|
|
expectedResult: [{_id: 3, x: 30}, {_id: 2, x: 20}, {_id: 1, x: 10}],
|
|
});
|
|
|
|
// $match, $sort, $limit can be optimized away even if there is no index to provide the sort.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [{$match: {x: {$gte: 0}}}, {$sort: {x: -1}}, {$limit: 1}],
|
|
expectedStages: ["COLLSCAN", "SORT"],
|
|
expectedResult: [{_id: 3, x: 30}],
|
|
});
|
|
|
|
// $match, $sort, $project, $limit can be optimized away.
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline:
|
|
[{$match: {x: {$gte: 20}}}, {$sort: {x: -1}}, {$project: {_id: 0, x: 1}}, {$limit: 2}],
|
|
expectedStages: ["COLLSCAN", "SORT", "PROJECTION_SIMPLE"],
|
|
expectedResult: [{x: 30}, {x: 20}],
|
|
});
|
|
|
|
// Test a case where there is a projection that can be covered by an index, but a blocking sort is
|
|
// still required. In this case, the entire pipeline can be optimized away.
|
|
assert.commandWorked(coll.createIndex({y: 1, x: 1}));
|
|
assertPipelineDoesNotUseAggregation({
|
|
pipeline: [
|
|
{$match: {y: {$gt: 0}, x: {$gte: 20}}},
|
|
{$sort: {x: -1}},
|
|
{$project: {_id: 0, y: 1, x: 1}},
|
|
{$limit: 2}
|
|
],
|
|
expectedStages: ["IXSCAN", "SORT", "PROJECTION_COVERED"],
|
|
expectedResult: [],
|
|
});
|
|
assert.commandWorked(coll.dropIndexes());
|
|
|
|
// Test that even if we don't have a projection stage at the front of the pipeline but there is a
|
|
// finite dependency set, a projection representing this dependency set is pushed down.
|
|
pipeline = [{$group: {_id: "$a", b: {$sum: "$b"}}}];
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
return assertPipelineDoesNotUseAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "GROUP"],
|
|
});
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
});
|
|
});
|
|
pipeline = [{$group: {_id: "$a", b: {$sum: "$b"}}}, {$group: {_id: "$c", x: {$sum: "$b"}}}];
|
|
assertPipelineIfGroupPushdown(
|
|
function() {
|
|
const explain = coll.explain().aggregate(pipeline);
|
|
// Both $group must be pushed down.
|
|
assert.eq(getPlanStages(explain, "GROUP").length, 2);
|
|
// PROJECTION_SIMPLE must be optimized away.
|
|
assert.eq(getPlanStages(explain, "PROJECTION_SIMPLE").length, 0);
|
|
// At bottom, there must be a COLLSCAN.
|
|
assert.eq(getPlanStages(explain, "COLLSCAN").length, 1);
|
|
},
|
|
function() {
|
|
return assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
});
|
|
});
|
|
|
|
function assertTransformByShape(expected, actual, message) {
|
|
assert.eq(Object.keys(expected).sort(), Object.keys(actual).sort(), message);
|
|
for (let key in expected) {
|
|
assert.eq(expected[key], actual[key]);
|
|
}
|
|
}
|
|
|
|
assertPipelineIfGroupPushdown(
|
|
// When $group pushdown is enabled, $group will be lowered and the PROJECTION_SIMPLE will be
|
|
// erased.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
let projStage = getAggPlanStage(explain, "PROJECTION_SIMPLE");
|
|
assert.eq(null, projStage, explain);
|
|
},
|
|
// When $group pushdown is disabled, $group will not be lowered and the PROJECTION_SIMPLE will
|
|
// be preserved.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
let projStage = getAggPlanStage(explain, "PROJECTION_SIMPLE");
|
|
assert.neq(null, projStage, explain);
|
|
assertTransformByShape({a: 1, b: 1, _id: 0}, projStage.transformBy, explain);
|
|
});
|
|
|
|
// Similar as above, but with $addFields stage at the front of the pipeline.
|
|
pipeline = [{$addFields: {z: "abc"}}, {$group: {_id: "$a", b: {$sum: "$b"}}}];
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE"],
|
|
});
|
|
explain = coll.explain().aggregate(pipeline);
|
|
let projStage = getAggPlanStage(explain, "PROJECTION_SIMPLE");
|
|
assert.neq(null, projStage, explain);
|
|
assertTransformByShape({a: 1, b: 1, _id: 0}, projStage.transformBy, explain);
|
|
|
|
// Asserts that, if group pushdown is enabled, we can remove a redundant projection stage before a
|
|
// group.
|
|
function assertProjectionCanBeRemovedBeforeGroup(pipeline) {
|
|
assertPipelineIfGroupPushdown(
|
|
// The projection and group should both be pushed down, and we expect to optimize away the
|
|
// projection after realizing that it will not affect the output of the group.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: pipeline, expectedStages: ["COLLSCAN", "GROUP"]});
|
|
},
|
|
// If group pushdown is not enabled we still expect the projection to be pushed down.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "$group"],
|
|
});
|
|
});
|
|
}
|
|
|
|
// Asserts that a projection stage is not optimized out of a pipeline with a projection and a group
|
|
// stage.
|
|
function assertProjectionIsNotRemoved(pipeline) {
|
|
assertPipelineIfGroupPushdown(
|
|
// The projection and group should both be pushed down, and we expect NOT to optimize away
|
|
// the projection.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: pipeline, expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "GROUP"]});
|
|
},
|
|
// If group pushdown is not enabled we still expect the projection to be pushed down.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "$group"],
|
|
});
|
|
});
|
|
}
|
|
|
|
// Test that an inclusion projection is optimized away if it is redundant/unnecessary.
|
|
assertProjectionCanBeRemovedBeforeGroup(
|
|
[{$project: {a: 1, b: 1}}, {$group: {_id: "$a", s: {$sum: "$b"}}}]);
|
|
|
|
// TODO SERVER-66061 Test the same thing for an inclusion projection with a dotted path.
|
|
// assertProjectionCanBeRemovedBeforeGroup(
|
|
// [{$project: {'a.b': 1, 'b.c': 1}}, {$group: {_id: "$a.b", s: {$sum: "$b.c"}}}]);
|
|
|
|
// Test that an inclusion projection is NOT optimized away if it is NOT redundant. This one fails to
|
|
// include a dependency of the $group and so will have an impact on the query results.
|
|
assertProjectionIsNotRemoved([{$project: {a: 1}}, {$group: {_id: "$a", s: {$sum: "$b"}}}]);
|
|
// TODO SERVER-66061 Analyze similar cases with dotted paths. For now, our ananlysis is limited to
|
|
// PROJECTION_SIMPLE which is not used for any dotted paths.
|
|
|
|
// TODO SERVER-66061 This one could be removed, but is left for future work.
|
|
assertProjectionIsNotRemoved(
|
|
[{$project: {a: 1, b: 1}}, {$group: {_id: "$a.b", s: {$sum: "$b.c"}}}]);
|
|
|
|
// Test that an inclusion projection is NOT optimized away if group depends on the entire document.
|
|
assertProjectionIsNotRemoved([{$project: {a: 1}}, {$group: {_id: "$$ROOT"}}]);
|
|
|
|
// Spinoff on the one above: Without supporting this kind of prefixing analysis, we can confuse
|
|
// ourselves with our dependency analysis. If the $group depends on both "path" and "path.subpath"
|
|
// then it will generate a $project on only "path" to express its dependency set. We then fail to
|
|
// optimize that out.
|
|
pipeline = [{$group: {_id: "$a.b", s: {$first: "$a"}}}];
|
|
assertPipelineIfGroupPushdown(
|
|
// The group should be pushed down, and we expect NOT to have any projection.
|
|
// TODO SERVER-66061 we should consider fixing this.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: pipeline, expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "GROUP"]});
|
|
},
|
|
// If group pushdown is not enabled we do expect a projection to be pushed down.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "$group"],
|
|
});
|
|
});
|
|
|
|
// We generate a projection stage from dependency analysis, even if the pipeline begins with an
|
|
// exclusion projection.
|
|
pipeline = [{$project: {c: 0}}, {$group: {_id: "$a", b: {$sum: "$b"}}}];
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "$project"],
|
|
});
|
|
explain = coll.explain().aggregate(pipeline);
|
|
projStage = getAggPlanStage(explain, "PROJECTION_SIMPLE");
|
|
assert.neq(null, projStage, explain);
|
|
assertTransformByShape({a: 1, b: 1, _id: 0}, projStage.transformBy, explain);
|
|
|
|
// Similar as above, but with a field 'a' presented both in the finite dependency set, and in the
|
|
// exclusion projection at the front of the pipeline.
|
|
pipeline = [{$project: {a: 0}}, {$group: {_id: "$a", b: {$sum: "$b"}}}];
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_SIMPLE", "$project"],
|
|
});
|
|
explain = coll.explain().aggregate(pipeline);
|
|
projStage = getAggPlanStage(explain, "PROJECTION_SIMPLE");
|
|
assert.neq(null, projStage, explain);
|
|
assertTransformByShape({a: 1, b: 1, _id: 0}, projStage.transformBy, explain);
|
|
|
|
// Test that an exclusion projection at the front of the pipeline is not pushed down, if there no
|
|
// finite dependency set.
|
|
pipeline = [{$project: {x: 0}}];
|
|
assertPipelineUsesAggregation({pipeline: pipeline, expectedStages: ["COLLSCAN"]});
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assert(!planHasStage(db, explain, "PROJECTION_SIMPLE"), explain);
|
|
assert(!planHasStage(db, explain, "PROJECTION_DEFAULT"), explain);
|
|
|
|
// Test that a computed projection at the front of the pipeline is pushed down, even if there's no
|
|
// finite dependency set.
|
|
pipeline = [{$project: {x: {$add: ["$a", 1]}}}];
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: pipeline, expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT"]});
|
|
|
|
pipeline = [{$project: {a: {$add: ["$a", 1]}}}, {$group: {_id: "$a", s: {$sum: "$b"}}}];
|
|
assertPipelineIfGroupPushdown(
|
|
// Test that a computed projection at the front of the pipeline is pushed down when there's a
|
|
// finite dependency set. Additionally, the group pushdown shouldn't erase the computed
|
|
// projection.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineDoesNotUseAggregation(
|
|
{pipeline: pipeline, expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT", "GROUP"]});
|
|
},
|
|
// Test that a computed projection at the front of the pipeline is pushed down when there's a
|
|
// finite dependency set.
|
|
function() {
|
|
explain = coll.explain().aggregate(pipeline);
|
|
assertPipelineUsesAggregation({
|
|
pipeline: pipeline,
|
|
expectedStages: ["COLLSCAN", "PROJECTION_DEFAULT", "$group"],
|
|
});
|
|
});
|
|
|
|
// getMore cases.
|
|
|
|
// Test getMore on a collection with an optimized away pipeline.
|
|
testGetMore({
|
|
command: {aggregate: coll.getName(), pipeline: [], cursor: {batchSize: 1}},
|
|
expectedResult: [{_id: 1, x: 10}, {_id: 2, x: 20}, {_id: 3, x: 30}]
|
|
});
|
|
testGetMore({
|
|
command:
|
|
{aggregate: coll.getName(), pipeline: [{$match: {x: {$gte: 20}}}], cursor: {batchSize: 1}},
|
|
expectedResult: [{_id: 2, x: 20}, {_id: 3, x: 30}]
|
|
});
|
|
testGetMore({
|
|
command: {
|
|
aggregate: coll.getName(),
|
|
pipeline: [{$match: {x: {$gte: 20}}}, {$project: {x: 1, _id: 0}}],
|
|
cursor: {batchSize: 1}
|
|
},
|
|
expectedResult: [{x: 20}, {x: 30}]
|
|
});
|
|
// Test getMore on a view with an optimized away pipeline. Since views cannot be created when
|
|
// imlicit sharded collection mode is on, this test will be run only on a non-sharded
|
|
// collection.
|
|
let view;
|
|
if (!FixtureHelpers.isSharded(coll)) {
|
|
view = db.optimize_away_pipeline_view;
|
|
view.drop();
|
|
assert.commandWorked(db.createView(view.getName(), coll.getName(), []));
|
|
testGetMore({
|
|
command: {find: view.getName(), filter: {}, batchSize: 1},
|
|
expectedResult: [{_id: 1, x: 10}, {_id: 2, x: 20}, {_id: 3, x: 30}]
|
|
});
|
|
}
|
|
// Test getMore puts a correct namespace into profile data for a colletion with optimized away
|
|
// pipeline. Cannot be run on mongos as profiling can be enabled only on mongod. Also profiling
|
|
// is supported on WiredTiger only.
|
|
if (!FixtureHelpers.isMongos(db) && isWiredTiger(db)) {
|
|
db.system.profile.drop();
|
|
db.setProfilingLevel(2);
|
|
testGetMore({
|
|
command: {
|
|
aggregate: coll.getName(),
|
|
pipeline: [{$match: {x: 10}}],
|
|
cursor: {batchSize: 1},
|
|
comment: 'optimize_away_pipeline'
|
|
},
|
|
expectedResult: [{_id: 1, x: 10}]
|
|
});
|
|
db.setProfilingLevel(0);
|
|
let profile = db.system.profile.find({}, {op: 1, ns: 1}).sort({ts: 1}).toArray();
|
|
assert.sameMembers(
|
|
profile,
|
|
[{op: "command", ns: coll.getFullName()}, {op: "getmore", ns: coll.getFullName()}]);
|
|
// Test getMore puts a correct namespace into profile data for a view with an optimized away
|
|
// pipeline.
|
|
if (!FixtureHelpers.isSharded(coll)) {
|
|
db.system.profile.drop();
|
|
db.setProfilingLevel(2);
|
|
testGetMore({
|
|
command: {
|
|
find: view.getName(),
|
|
filter: {x: 10},
|
|
batchSize: 1,
|
|
comment: 'optimize_away_pipeline'
|
|
},
|
|
expectedResult: [{_id: 1, x: 10}]
|
|
});
|
|
db.setProfilingLevel(0);
|
|
profile = db.system.profile.find({}, {op: 1, ns: 1}).sort({ts: 1}).toArray();
|
|
assert.sameMembers(
|
|
profile,
|
|
[{op: "query", ns: view.getFullName()}, {op: "getmore", ns: view.getFullName()}]);
|
|
}
|
|
}
|
|
}());
|