Skip to content

Commit aaba83e

Browse files
committed
Add missing value and sampler aggregation
All aggregations derived from ValuesSourceAggregationBuilder in Java have a missing field that will be used when a document has a missing value [1]. Also added the missing Sampler aggregation [2]. [1] https://www.elastic.co/guide/en/elasticsearch/reference/current/search-aggregations-bucket-terms-aggregation.html#_missing_value_12 [2] https://www.elastic.co/guide/en/elasticsearch/reference/current/search-aggregations-bucket-sampler-aggregation.html
1 parent b51c3b3 commit aaba83e

13 files changed

+388
-1
lines changed

README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -243,7 +243,7 @@ See the [wiki](https://github.com/olivere/elastic/wiki) for more details.
243243
- [x] Nested
244244
- [x] Range
245245
- [x] Reverse Nested
246-
- [ ] Sampler
246+
- [x] Sampler
247247
- [x] Significant Terms
248248
- [x] Terms
249249
- Pipeline Aggregations

search_aggs.go

+15
Original file line numberDiff line numberDiff line change
@@ -323,6 +323,21 @@ func (a Aggregations) SignificantTerms(name string) (*AggregationBucketSignifica
323323
return nil, false
324324
}
325325

326+
// Sampler returns sampler aggregation results.
327+
// See: https://www.elastic.co/guide/en/elasticsearch/reference/current/search-aggregations-bucket-sampler-aggregation.html
328+
func (a Aggregations) Sampler(name string) (*AggregationSingleBucket, bool) {
329+
if raw, found := a[name]; found {
330+
agg := new(AggregationSingleBucket)
331+
if raw == nil {
332+
return agg, true
333+
}
334+
if err := json.Unmarshal(*raw, agg); err == nil {
335+
return agg, true
336+
}
337+
}
338+
return nil, false
339+
}
340+
326341
// Range returns range aggregation results.
327342
// See: http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/search-aggregations-bucket-range-aggregation.html
328343
func (a Aggregations) Range(name string) (*AggregationBucketRangeItems, bool) {

search_aggs_bucket_date_histogram.go

+10
Original file line numberDiff line numberDiff line change
@@ -10,6 +10,7 @@ package elastic
1010
type DateHistogramAggregation struct {
1111
field string
1212
script *Script
13+
missing interface{}
1314
subAggregations map[string]Aggregation
1415
meta map[string]interface{}
1516

@@ -42,6 +43,12 @@ func (a *DateHistogramAggregation) Script(script *Script) *DateHistogramAggregat
4243
return a
4344
}
4445

46+
// Missing configures the value to use when documents miss a value.
47+
func (a *DateHistogramAggregation) Missing(missing interface{}) *DateHistogramAggregation {
48+
a.missing = missing
49+
return a
50+
}
51+
4552
func (a *DateHistogramAggregation) SubAggregation(name string, subAggregation Aggregation) *DateHistogramAggregation {
4653
a.subAggregations[name] = subAggregation
4754
return a
@@ -219,6 +226,9 @@ func (a *DateHistogramAggregation) Source() (interface{}, error) {
219226
}
220227
opts["script"] = src
221228
}
229+
if a.missing != nil {
230+
opts["missing"] = a.missing
231+
}
222232

223233
opts["interval"] = a.interval
224234
if a.minDocCount != nil {

search_aggs_bucket_date_histogram_test.go

+17
Original file line numberDiff line numberDiff line change
@@ -30,3 +30,20 @@ func TestDateHistogramAggregation(t *testing.T) {
3030
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
3131
}
3232
}
33+
34+
func TestDateHistogramAggregationWithMissing(t *testing.T) {
35+
agg := NewDateHistogramAggregation().Field("date").Interval("year").Missing("1900")
36+
src, err := agg.Source()
37+
if err != nil {
38+
t.Fatal(err)
39+
}
40+
data, err := json.Marshal(src)
41+
if err != nil {
42+
t.Fatalf("marshaling to JSON failed: %v", err)
43+
}
44+
got := string(data)
45+
expected := `{"date_histogram":{"field":"date","interval":"year","missing":"1900"}}`
46+
if got != expected {
47+
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
48+
}
49+
}

search_aggs_bucket_histogram.go

+10
Original file line numberDiff line numberDiff line change
@@ -12,6 +12,7 @@ package elastic
1212
type HistogramAggregation struct {
1313
field string
1414
script *Script
15+
missing interface{}
1516
subAggregations map[string]Aggregation
1617
meta map[string]interface{}
1718

@@ -40,6 +41,12 @@ func (a *HistogramAggregation) Script(script *Script) *HistogramAggregation {
4041
return a
4142
}
4243

44+
// Missing configures the value to use when documents miss a value.
45+
func (a *HistogramAggregation) Missing(missing interface{}) *HistogramAggregation {
46+
a.missing = missing
47+
return a
48+
}
49+
4350
func (a *HistogramAggregation) SubAggregation(name string, subAggregation Aggregation) *HistogramAggregation {
4451
a.subAggregations[name] = subAggregation
4552
return a
@@ -193,6 +200,9 @@ func (a *HistogramAggregation) Source() (interface{}, error) {
193200
}
194201
opts["script"] = src
195202
}
203+
if a.missing != nil {
204+
opts["missing"] = a.missing
205+
}
196206

197207
opts["interval"] = a.interval
198208
if a.order != "" {

search_aggs_bucket_histogram_test.go

+17
Original file line numberDiff line numberDiff line change
@@ -42,3 +42,20 @@ func TestHistogramAggregationWithMetaData(t *testing.T) {
4242
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
4343
}
4444
}
45+
46+
func TestHistogramAggregationWithMissing(t *testing.T) {
47+
agg := NewHistogramAggregation().Field("price").Interval(50).Missing("n/a")
48+
src, err := agg.Source()
49+
if err != nil {
50+
t.Fatal(err)
51+
}
52+
data, err := json.Marshal(src)
53+
if err != nil {
54+
t.Fatalf("marshaling to JSON failed: %v", err)
55+
}
56+
got := string(data)
57+
expected := `{"histogram":{"field":"price","interval":50,"missing":"n/a"}}`
58+
if got != expected {
59+
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
60+
}
61+
}

search_aggs_bucket_range.go

+10
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,7 @@ import (
1818
type RangeAggregation struct {
1919
field string
2020
script *Script
21+
missing interface{}
2122
subAggregations map[string]Aggregation
2223
meta map[string]interface{}
2324
keyed *bool
@@ -48,6 +49,12 @@ func (a *RangeAggregation) Script(script *Script) *RangeAggregation {
4849
return a
4950
}
5051

52+
// Missing configures the value to use when documents miss a value.
53+
func (a *RangeAggregation) Missing(missing interface{}) *RangeAggregation {
54+
a.missing = missing
55+
return a
56+
}
57+
5158
func (a *RangeAggregation) SubAggregation(name string, subAggregation Aggregation) *RangeAggregation {
5259
a.subAggregations[name] = subAggregation
5360
return a
@@ -163,6 +170,9 @@ func (a *RangeAggregation) Source() (interface{}, error) {
163170
}
164171
opts["script"] = src
165172
}
173+
if a.missing != nil {
174+
opts["missing"] = a.missing
175+
}
166176

167177
if a.keyed != nil {
168178
opts["keyed"] = *a.keyed

search_aggs_bucket_range_test.go

+20
Original file line numberDiff line numberDiff line change
@@ -134,3 +134,23 @@ func TestRangeAggregationWithMetaData(t *testing.T) {
134134
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
135135
}
136136
}
137+
138+
func TestRangeAggregationWithMissing(t *testing.T) {
139+
agg := NewRangeAggregation().Field("price").Missing(0)
140+
agg = agg.AddRange(nil, 50)
141+
agg = agg.AddRange(50, 100)
142+
agg = agg.AddRange(100, nil)
143+
src, err := agg.Source()
144+
if err != nil {
145+
t.Fatal(err)
146+
}
147+
data, err := json.Marshal(src)
148+
if err != nil {
149+
t.Fatalf("marshaling to JSON failed: %v", err)
150+
}
151+
got := string(data)
152+
expected := `{"range":{"field":"price","missing":0,"ranges":[{"to":50},{"from":50,"to":100},{"from":100}]}}`
153+
if got != expected {
154+
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
155+
}
156+
}

search_aggs_bucket_sampler.go

+145
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,145 @@
1+
// Copyright 2012-2016 Oliver Eilhard. All rights reserved.
2+
// Use of this source code is governed by a MIT-license.
3+
// See http://olivere.mit-license.org/license.txt for details.
4+
5+
package elastic
6+
7+
// SamplerAggregation is a filtering aggregation used to limit any
8+
// sub aggregations' processing to a sample of the top-scoring documents.
9+
// Optionally, diversity settings can be used to limit the number of matches
10+
// that share a common value such as an "author".
11+
// See: https://www.elastic.co/guide/en/elasticsearch/reference/2.x/search-aggregations-bucket-sampler-aggregation.html
12+
type SamplerAggregation struct {
13+
field string
14+
script *Script
15+
missing interface{}
16+
subAggregations map[string]Aggregation
17+
meta map[string]interface{}
18+
19+
shardSize int
20+
maxDocsPerValue int
21+
executionHint string
22+
}
23+
24+
func NewSamplerAggregation() *SamplerAggregation {
25+
return &SamplerAggregation{
26+
shardSize: -1,
27+
maxDocsPerValue: -1,
28+
subAggregations: make(map[string]Aggregation),
29+
}
30+
}
31+
32+
func (a *SamplerAggregation) Field(field string) *SamplerAggregation {
33+
a.field = field
34+
return a
35+
}
36+
37+
func (a *SamplerAggregation) Script(script *Script) *SamplerAggregation {
38+
a.script = script
39+
return a
40+
}
41+
42+
// Missing configures the value to use when documents miss a value.
43+
func (a *SamplerAggregation) Missing(missing interface{}) *SamplerAggregation {
44+
a.missing = missing
45+
return a
46+
}
47+
48+
func (a *SamplerAggregation) SubAggregation(name string, subAggregation Aggregation) *SamplerAggregation {
49+
a.subAggregations[name] = subAggregation
50+
return a
51+
}
52+
53+
// Meta sets the meta data to be included in the aggregation response.
54+
func (a *SamplerAggregation) Meta(metaData map[string]interface{}) *SamplerAggregation {
55+
a.meta = metaData
56+
return a
57+
}
58+
59+
// ShardSize sets the maximum number of docs returned from each shard.
60+
func (a *SamplerAggregation) ShardSize(shardSize int) *SamplerAggregation {
61+
a.shardSize = shardSize
62+
return a
63+
}
64+
65+
func (a *SamplerAggregation) MaxDocsPerValue(maxDocsPerValue int) *SamplerAggregation {
66+
a.maxDocsPerValue = maxDocsPerValue
67+
return a
68+
}
69+
70+
func (a *SamplerAggregation) ExecutionHint(hint string) *SamplerAggregation {
71+
a.executionHint = hint
72+
return a
73+
}
74+
75+
func (a *SamplerAggregation) Source() (interface{}, error) {
76+
// Example:
77+
// {
78+
// "aggs" : {
79+
// "sample" : {
80+
// "sampler" : {
81+
// "field" : "user.id",
82+
// "shard_size" : 200
83+
// },
84+
// "aggs": {
85+
// "keywords": {
86+
// "significant_terms": {
87+
// "field": "text"
88+
// }
89+
// }
90+
// }
91+
// }
92+
// }
93+
// }
94+
//
95+
// This method returns only the { "sampler" : { ... } } part.
96+
97+
source := make(map[string]interface{})
98+
opts := make(map[string]interface{})
99+
source["sampler"] = opts
100+
101+
// ValuesSourceAggregationBuilder
102+
if a.field != "" {
103+
opts["field"] = a.field
104+
}
105+
if a.script != nil {
106+
src, err := a.script.Source()
107+
if err != nil {
108+
return nil, err
109+
}
110+
opts["script"] = src
111+
}
112+
if a.missing != nil {
113+
opts["missing"] = a.missing
114+
}
115+
116+
if a.shardSize >= 0 {
117+
opts["shard_size"] = a.shardSize
118+
}
119+
if a.maxDocsPerValue >= 0 {
120+
opts["max_docs_per_value"] = a.maxDocsPerValue
121+
}
122+
if a.executionHint != "" {
123+
opts["execution_hint"] = a.executionHint
124+
}
125+
126+
// AggregationBuilder (SubAggregations)
127+
if len(a.subAggregations) > 0 {
128+
aggsMap := make(map[string]interface{})
129+
source["aggregations"] = aggsMap
130+
for name, aggregate := range a.subAggregations {
131+
src, err := aggregate.Source()
132+
if err != nil {
133+
return nil, err
134+
}
135+
aggsMap[name] = src
136+
}
137+
}
138+
139+
// Add Meta data if available
140+
if len(a.meta) > 0 {
141+
source["meta"] = a.meta
142+
}
143+
144+
return source, nil
145+
}

search_aggs_bucket_sampler_test.go

+52
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,52 @@
1+
// Copyright 2012-2016 Oliver Eilhard. All rights reserved.
2+
// Use of this source code is governed by a MIT-license.
3+
// See http://olivere.mit-license.org/license.txt for details.
4+
5+
package elastic
6+
7+
import (
8+
"encoding/json"
9+
"testing"
10+
)
11+
12+
func TestSamplerAggregation(t *testing.T) {
13+
keywordsAgg := NewSignificantTermsAggregation().Field("text")
14+
agg := NewSamplerAggregation().
15+
Field("user.id").
16+
ShardSize(200).
17+
SubAggregation("keywords", keywordsAgg)
18+
src, err := agg.Source()
19+
if err != nil {
20+
t.Fatal(err)
21+
}
22+
data, err := json.Marshal(src)
23+
if err != nil {
24+
t.Fatalf("marshaling to JSON failed: %v", err)
25+
}
26+
got := string(data)
27+
expected := `{"aggregations":{"keywords":{"significant_terms":{"field":"text"}}},"sampler":{"field":"user.id","shard_size":200}}`
28+
if got != expected {
29+
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
30+
}
31+
}
32+
33+
func TestSamplerAggregationWithMissing(t *testing.T) {
34+
keywordsAgg := NewSignificantTermsAggregation().Field("text")
35+
agg := NewSamplerAggregation().
36+
Field("user.id").
37+
Missing("n/a").
38+
SubAggregation("keywords", keywordsAgg)
39+
src, err := agg.Source()
40+
if err != nil {
41+
t.Fatal(err)
42+
}
43+
data, err := json.Marshal(src)
44+
if err != nil {
45+
t.Fatalf("marshaling to JSON failed: %v", err)
46+
}
47+
got := string(data)
48+
expected := `{"aggregations":{"keywords":{"significant_terms":{"field":"text"}}},"sampler":{"field":"user.id","missing":"n/a"}}`
49+
if got != expected {
50+
t.Errorf("expected\n%s\n,got:\n%s", expected, got)
51+
}
52+
}

0 commit comments

Comments
 (0)