Elasticsearch:reverse_nested 深度嵌套聚合下的聚合不起作用
Elasticsearch: reverse_nested aggregation under deep nested aggregation doesn't work
Elasticsearch 版本:2.3.3
标题基本上说明了一切。如果在第二个嵌套聚合下使用 reverse_nested
,尽管文档似乎通过 reverse_nested
来限定范围(请参阅结果中的最后一个 "doc_count"
字段),但它后面的聚合不会不知何故无法工作。
这里我准备了一个例子——一个文件是一个学生,有入学日期和考试历史。
映射:
{
"mappings": {
"students": {
"properties": {
"name": {
"type": "string"},
"enrollment": {
"type": "date"},
"exam_histories": {
"type": "nested",
"properties": {
"date": {
"type": "date"},
"subjects": {
"type": "nested",
"properties": {
"name": {
"type": "string"},
"score": {
"type": "short"}}}}}}}}}
测试文档:
{
"name": "John",
"enrollment": "2012-09-01T00:00:00+00:00",
"exam_histories": [
{
"date": "2016-05-05T00:00:00+00:00",
"subjects": [
{
"name": "math",
"score": 90}]}]}
聚合查询(无实际意义):
{
"aggs": {
"nested_exam_histories": {
"nested": {
"path": "exam_histories"},
"aggs": {
"date_buckets": {
"date_histogram": {
"field": "exam_histories.date",
"interval": "day"},
"aggs": {
"this_reverse_nested_does_work": {
"reverse_nested": {},
"aggs": {
"newest_enrollment": {
"max": {
"field": "enrollment"}}}},
"deep_nested_subjects": {
"nested": {
"path": "exam_histories.subjects"},
"aggs": {
"score_buckets": {
"terms": {
"field": "exam_histories.subjects.score"},
"aggs": {
"this_reverse_nested_doesnt_work": {
"reverse_nested": {},
"aggs": {
"newest_exam_date": {
"max": {
"field": "exam_histories.date"}}}}}}}}}}}}}}
结果:
...
"aggregations" : {
"nested_exam_histories" : {
"doc_count" : 1,
"date_buckets" : {
"buckets" : [ {
"key_as_string" : "2016-05-05T00:00:00.000Z",
"key" : 1462406400000,
"doc_count" : 1,
"this_reverse_nested_does_work" : {
"doc_count" : 1,
"newest_enrollment" : {
"value" : 1.3464576E12,
"value_as_string" : "2012-09-01T00:00:00.000Z"
}
},
"deep_nested_subjects" : {
"doc_count" : 1,
"score_buckets" : {
"doc_count_error_upper_bound" : 0,
"sum_other_doc_count" : 0,
"buckets" : [ {
"key" : 90,
"doc_count" : 1,
"this_reverse_nested_doesnt_work" : {
"doc_count" : 1,
"newest_exam_date" : {
"value" : null
}
...
...在这里您可以看到聚合 "newest_exam_date" 不起作用。是错误还是我做错了什么?
您需要使用 path
选项明确指定您想要 "reverse-aggregate" 的嵌套对象,否则它假定该字段位于根级别。
path - Which defines to what nested object field should be joined
back. The default is empty, which means that it joins back to the root
/ main document level. The path cannot contain a reference to a nested
object field that falls outside the nested aggregation’s nested
structure a reverse_nested is in.
Example:
{
"size":0,
"aggs": {
"nested_exam_histories": {
"nested": {
"path": "exam_histories"
},
"aggs": {
"date_buckets": {
"date_histogram": {
"field": "exam_histories.date",
"interval": "day"
},
"aggs": {
"this_reverse_nested_does_work": {
"reverse_nested": {},
"aggs": {
"newest_enrollment": {
"max": {
"field": "enrollment"
}
}
}
},
"deep_nested_subjects": {
"nested": {
"path": "exam_histories.subjects"
},
"aggs": {
"score_buckets": {
"terms": {
"field": "exam_histories.subjects.score"
},
"aggs": {
"this_reverse_nested_doesnt_work": {
"reverse_nested": {
"path": "exam_histories"
},
"aggs": {
"newest_exam_date": {
"max": {
"field": "exam_histories.date"
}
}
}
}
}
}
}
}
}
}
}
}
}
}
结果:
{
"took": 5,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 0,
"hits": []
},
"aggregations": {
"nested_exam_histories": {
"doc_count": 2,
"date_buckets": {
"buckets": [
{
"key_as_string": "2016-05-05T00:00:00.000Z",
"key": 1462406400000,
"doc_count": 2,
"this_reverse_nested_does_work": {
"doc_count": 2,
"newest_enrollment": {
"value": 1377993600000,
"value_as_string": "2013-09-01T00:00:00.000Z"
}
},
"deep_nested_subjects": {
"doc_count": 2,
"score_buckets": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 90,
"doc_count": 2,
"this_reverse_nested_doesnt_work": {
"doc_count": 2,
"newest_exam_date": {
"value": 1462406400000,
"value_as_string": "2016-05-05T00:00:00.000Z"
}
}
}
]
}
}
}
]
}
}
}
}
注意第二个"reverse-aggergation"中的path
选项:
reverse_nested": {
"path": "exam_histories"
}
Elasticsearch 版本:2.3.3
标题基本上说明了一切。如果在第二个嵌套聚合下使用 reverse_nested
,尽管文档似乎通过 reverse_nested
来限定范围(请参阅结果中的最后一个 "doc_count"
字段),但它后面的聚合不会不知何故无法工作。
这里我准备了一个例子——一个文件是一个学生,有入学日期和考试历史。
映射:
{
"mappings": {
"students": {
"properties": {
"name": {
"type": "string"},
"enrollment": {
"type": "date"},
"exam_histories": {
"type": "nested",
"properties": {
"date": {
"type": "date"},
"subjects": {
"type": "nested",
"properties": {
"name": {
"type": "string"},
"score": {
"type": "short"}}}}}}}}}
测试文档:
{
"name": "John",
"enrollment": "2012-09-01T00:00:00+00:00",
"exam_histories": [
{
"date": "2016-05-05T00:00:00+00:00",
"subjects": [
{
"name": "math",
"score": 90}]}]}
聚合查询(无实际意义):
{
"aggs": {
"nested_exam_histories": {
"nested": {
"path": "exam_histories"},
"aggs": {
"date_buckets": {
"date_histogram": {
"field": "exam_histories.date",
"interval": "day"},
"aggs": {
"this_reverse_nested_does_work": {
"reverse_nested": {},
"aggs": {
"newest_enrollment": {
"max": {
"field": "enrollment"}}}},
"deep_nested_subjects": {
"nested": {
"path": "exam_histories.subjects"},
"aggs": {
"score_buckets": {
"terms": {
"field": "exam_histories.subjects.score"},
"aggs": {
"this_reverse_nested_doesnt_work": {
"reverse_nested": {},
"aggs": {
"newest_exam_date": {
"max": {
"field": "exam_histories.date"}}}}}}}}}}}}}}
结果:
...
"aggregations" : {
"nested_exam_histories" : {
"doc_count" : 1,
"date_buckets" : {
"buckets" : [ {
"key_as_string" : "2016-05-05T00:00:00.000Z",
"key" : 1462406400000,
"doc_count" : 1,
"this_reverse_nested_does_work" : {
"doc_count" : 1,
"newest_enrollment" : {
"value" : 1.3464576E12,
"value_as_string" : "2012-09-01T00:00:00.000Z"
}
},
"deep_nested_subjects" : {
"doc_count" : 1,
"score_buckets" : {
"doc_count_error_upper_bound" : 0,
"sum_other_doc_count" : 0,
"buckets" : [ {
"key" : 90,
"doc_count" : 1,
"this_reverse_nested_doesnt_work" : {
"doc_count" : 1,
"newest_exam_date" : {
"value" : null
}
...
...在这里您可以看到聚合 "newest_exam_date" 不起作用。是错误还是我做错了什么?
您需要使用 path
选项明确指定您想要 "reverse-aggregate" 的嵌套对象,否则它假定该字段位于根级别。
path - Which defines to what nested object field should be joined back. The default is empty, which means that it joins back to the root / main document level. The path cannot contain a reference to a nested object field that falls outside the nested aggregation’s nested structure a reverse_nested is in. Example:
{
"size":0,
"aggs": {
"nested_exam_histories": {
"nested": {
"path": "exam_histories"
},
"aggs": {
"date_buckets": {
"date_histogram": {
"field": "exam_histories.date",
"interval": "day"
},
"aggs": {
"this_reverse_nested_does_work": {
"reverse_nested": {},
"aggs": {
"newest_enrollment": {
"max": {
"field": "enrollment"
}
}
}
},
"deep_nested_subjects": {
"nested": {
"path": "exam_histories.subjects"
},
"aggs": {
"score_buckets": {
"terms": {
"field": "exam_histories.subjects.score"
},
"aggs": {
"this_reverse_nested_doesnt_work": {
"reverse_nested": {
"path": "exam_histories"
},
"aggs": {
"newest_exam_date": {
"max": {
"field": "exam_histories.date"
}
}
}
}
}
}
}
}
}
}
}
}
}
}
结果:
{
"took": 5,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 0,
"hits": []
},
"aggregations": {
"nested_exam_histories": {
"doc_count": 2,
"date_buckets": {
"buckets": [
{
"key_as_string": "2016-05-05T00:00:00.000Z",
"key": 1462406400000,
"doc_count": 2,
"this_reverse_nested_does_work": {
"doc_count": 2,
"newest_enrollment": {
"value": 1377993600000,
"value_as_string": "2013-09-01T00:00:00.000Z"
}
},
"deep_nested_subjects": {
"doc_count": 2,
"score_buckets": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 90,
"doc_count": 2,
"this_reverse_nested_doesnt_work": {
"doc_count": 2,
"newest_exam_date": {
"value": 1462406400000,
"value_as_string": "2016-05-05T00:00:00.000Z"
}
}
}
]
}
}
}
]
}
}
}
}
注意第二个"reverse-aggergation"中的path
选项:
reverse_nested": {
"path": "exam_histories"
}