summaryrefslogtreecommitdiff
path: root/src/mongo/db/pipeline/document_source_bucket.cpp
blob: 4de8daee4918780db90ebbd44d213e8248ae97dc (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
/**
 *    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/db/pipeline/document_source_bucket.h"

#include "mongo/db/pipeline/document_source_group.h"
#include "mongo/db/pipeline/document_source_sort.h"
#include "mongo/db/pipeline/expression.h"
#include "mongo/db/pipeline/lite_parsed_document_source.h"

namespace mongo {

using boost::intrusive_ptr;
using std::list;
using std::vector;

REGISTER_DOCUMENT_SOURCE(bucket,
                         LiteParsedDocumentSourceDefault::parse,
                         DocumentSourceBucket::createFromBson,
                         AllowedWithApiStrict::kAlways);

namespace {
intrusive_ptr<ExpressionConstant> getExpressionConstant(ExpressionContext* const expCtx,
                                                        BSONElement expressionElem,
                                                        const VariablesParseState& vps) {
    auto expr = Expression::parseOperand(expCtx, expressionElem, vps)->optimize();
    return dynamic_cast<ExpressionConstant*>(expr.get());
}
}  // namespace

list<intrusive_ptr<DocumentSource>> DocumentSourceBucket::createFromBson(
    BSONElement elem, const intrusive_ptr<ExpressionContext>& pExpCtx) {
    uassert(40201,
            str::stream() << "Argument to $bucket stage must be an object, but found type: "
                          << typeName(elem.type()) << ".",
            elem.type() == BSONType::Object);

    const BSONObj bucketObj = elem.embeddedObject();
    BSONObjBuilder groupObjBuilder;
    BSONObjBuilder switchObjBuilder;

    VariablesParseState vps = pExpCtx->variablesParseState;

    vector<Value> boundaryValues;
    BSONElement groupByField;
    Value defaultValue;

    bool outputFieldSpecified = false;
    for (auto&& argument : bucketObj) {
        const auto argName = argument.fieldNameStringData();
        if ("groupBy" == argName) {
            groupByField = argument;

            const bool groupByIsExpressionInObject = groupByField.type() == BSONType::Object &&
                groupByField.embeddedObject().firstElementFieldName()[0] == '$';

            const bool groupByIsPrefixedPath =
                groupByField.type() == BSONType::String && groupByField.valueStringData()[0] == '$';
            uassert(40202,
                    str::stream() << "The $bucket 'groupBy' field must be defined as a $-prefixed "
                                     "path or an expression, but found: "
                                  << groupByField.toString(false, false) << ".",
                    groupByIsExpressionInObject || groupByIsPrefixedPath);
        } else if ("boundaries" == argName) {
            uassert(
                40200,
                str::stream() << "The $bucket 'boundaries' field must be an array, but found type: "
                              << typeName(argument.type()) << ".",
                argument.type() == BSONType::Array);

            for (auto&& boundaryElem : argument.embeddedObject()) {
                auto exprConst = getExpressionConstant(pExpCtx.get(), boundaryElem, vps);
                uassert(40191,
                        str::stream() << "The $bucket 'boundaries' field must be an array of "
                                         "constant values, but found value: "
                                      << boundaryElem.toString(false, false) << ".",
                        exprConst);
                boundaryValues.push_back(exprConst->getValue());
            }

            uassert(40192,
                    str::stream()
                        << "The $bucket 'boundaries' field must have at least 2 values, but found "
                        << boundaryValues.size() << " value(s).",
                    boundaryValues.size() >= 2);

            // Make sure that the boundaries are unique, sorted in ascending order, and have the
            // same canonical type.
            for (size_t i = 1; i < boundaryValues.size(); ++i) {
                Value lower = boundaryValues[i - 1];
                Value upper = boundaryValues[i];
                int lowerCanonicalType = canonicalizeBSONType(lower.getType());
                int upperCanonicalType = canonicalizeBSONType(upper.getType());

                uassert(40193,
                        str::stream() << "All values in the the 'boundaries' option to $bucket "
                                         "must have the same type. Found conflicting types "
                                      << typeName(lower.getType()) << " and "
                                      << typeName(upper.getType()) << ".",
                        lowerCanonicalType == upperCanonicalType);
                uassert(40194,
                        str::stream()
                            << "The 'boundaries' option to $bucket must be sorted, but elements "
                            << i - 1 << " and " << i << " are not in ascending order ("
                            << lower.toString() << " is not less than " << upper.toString() << ").",
                        pExpCtx->getValueComparator().evaluate(lower < upper));
            }
        } else if ("default" == argName) {
            // If there is a default, make sure that it parses to a constant expression then add
            // default to switch.
            auto exprConst = getExpressionConstant(pExpCtx.get(), argument, vps);
            uassert(40195,
                    str::stream()
                        << "The $bucket 'default' field must be a constant expression, but found: "
                        << argument.toString(false, false) << ".",
                    exprConst);

            defaultValue = exprConst->getValue();
            defaultValue.addToBsonObj(&switchObjBuilder, "default");
        } else if ("output" == argName) {
            outputFieldSpecified = true;
            uassert(
                40196,
                str::stream() << "The $bucket 'output' field must be an object, but found type: "
                              << typeName(argument.type()) << ".",
                argument.type() == BSONType::Object);

            for (auto&& outputElem : argument.embeddedObject()) {
                groupObjBuilder.append(outputElem);
            }
        } else {
            uasserted(40197, str::stream() << "Unrecognized option to $bucket: " << argName << ".");
        }
    }

    const bool isMissingRequiredField = groupByField.eoo() || boundaryValues.empty();
    uassert(40198,
            "$bucket requires 'groupBy' and 'boundaries' to be specified.",
            !isMissingRequiredField);

    Value lowerValue = boundaryValues.front();
    Value upperValue = boundaryValues.back();
    if (canonicalizeBSONType(defaultValue.getType()) ==
        canonicalizeBSONType(lowerValue.getType())) {
        // If the default has the same canonical type as the bucket's boundaries, then make sure the
        // default is less than the lowest boundary or greater than or equal to the highest
        // boundary.
        const auto& valueCmp = pExpCtx->getValueComparator();
        const bool hasValidDefault = valueCmp.evaluate(defaultValue < lowerValue) ||
            valueCmp.evaluate(defaultValue >= upperValue);
        uassert(40199,
                "The $bucket 'default' field must be less than the lowest boundary or greater than "
                "or equal to the highest boundary.",
                hasValidDefault);
    }

    // Make the branches for the $switch expression.
    BSONArrayBuilder branchesBuilder;
    for (size_t i = 1; i < boundaryValues.size(); ++i) {
        Value lower = boundaryValues[i - 1];
        Value upper = boundaryValues[i];
        BSONObj caseExpr =
            BSON("$and" << BSON_ARRAY(BSON("$gte" << BSON_ARRAY(groupByField << lower))
                                      << BSON("$lt" << BSON_ARRAY(groupByField << upper))));
        branchesBuilder.append(BSON("case" << caseExpr << "then" << lower));
    }

    // Add the $switch expression to the group BSON object.
    switchObjBuilder.append("branches", branchesBuilder.arr());
    groupObjBuilder.append("_id", BSON("$switch" << switchObjBuilder.obj()));

    // If no output is specified, add a count field by default.
    if (!outputFieldSpecified) {
        groupObjBuilder.append("count", BSON("$sum" << 1));
    }

    BSONObj groupObj = BSON("$group" << groupObjBuilder.obj());
    BSONObj sortObj = BSON("$sort" << BSON("_id" << 1));

    auto groupSource = DocumentSourceGroup::createFromBson(groupObj.firstElement(), pExpCtx);
    auto sortSource = DocumentSourceSort::createFromBson(sortObj.firstElement(), pExpCtx);

    return {groupSource, sortSource};
}
}  // namespace mongo