ES术语汇总，包括总计数，每个存储桶计数和过滤结果
[英]ES term aggregation with total count, count per bucket and filtered results
问题描述
我对Elasticsearch还是很陌生，想使用术语聚合来完成以下工作：
- 获取与类别和选定的规格（功能）匹配的产品列表
- 规格和可能的值的列表，以及如果选择它会得到的产品数量
- 规范列表应始终包含整个数据集的所有规范，即使其计数为0，但仅在所选类别内
- 产品必须至少包含选定规范的选定值之一
这是我的映射：
{
"mapping": {
"product": {
"properties": {
"category": {
"type": "keyword"
},
"features": {
"properties": {
"breedte": {
"properties": {
"id": {
"type": "long"
},
"measure": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"name": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"value": {
"type": "float"
}
}
},
"diepte": {
"properties": {
"id": {
"type": "long"
},
"measure": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"name": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"value": {
"type": "float"
}
}
},
"hoogte": {
"properties": {
"id": {
"type": "long"
},
"measure": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"name": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"value": {
"type": "float"
}
}
}
}
},
"name": {
"type": "keyword"
},
"url": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
}
}
}
这是我目前拥有的（用2个值选择“ breedte”，用1个选择“ hoogte”）
GET _search
{
"from": 0,
"size": 15,
"sort":{
"name":"desc"
},
"stored_fields":[
"_id",
"name"
],
"query":{
"constant_score":{
"filter":{
"bool":{
"must":[
{
"term":{
"category":"Gaskookplaat"
}
},
{
"term":{
"features.hoogte.value":"55"
}
}
],
"should":[
{
"term":{
"features.breedte.value":"750"
}
},
{
"term":{
"features.breedte.value":"590"
}
}
]
}
}
}
},
"aggs":{
"categories":{
"filter":{
"bool":{
"should":[
]
}
},
"aggs":{
"breedte":{
"terms":{
"field":"features.breedte.value",
"min_doc_count":0
},
"meta":{
"id":"id",
"measure":"mm",
"title":"Breedte"
}
},
"hoogte":{
"terms":{
"field":"features.hoogte.value",
"min_doc_count":0
},
"meta":{
"value":"features.hoogte.value"
}
}
}
}
}
}
结果
{
"took": 10,
"timed_out": false,
"_shards": {
"total": 6,
"successful": 5,
"skipped": 0,
"failed": 1,
"failures": [
{
"shard": 0,
"index": ".kibana",
"node": "2fjW10M0RSa_imRTIMutww",
"reason": {
"type": "query_shard_exception",
"reason": "No mapping found for [name] in order to sort on",
"index_uuid": "x4687TCDRTalALulM9xLSg",
"index": ".kibana"
}
}
]
},
"hits": {
"total": 9,
"max_score": null,
"hits": [
{
"_index": "products",
"_type": "product",
"_id": "2014195",
"_score": null,
"sort": [
"5ZTAKGF87"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014049",
"_score": null,
"sort": [
"5RVSAKGF87"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2023697",
"_score": null,
"sort": [
"9ZTAKG95"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2009530",
"_score": null,
"sort": [
"RVSAKG959"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2016478",
"_score": null,
"sort": [
"RVSAKG675"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2015225",
"_score": null,
"sort": [
"5IXBBGW7"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2010352",
"_score": null,
"sort": [
"75IXBGW"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2018012",
"_score": null,
"sort": [
"0IXBBGW6"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2009793",
"_score": null,
"sort": [
"60IXBGW"
]
}
]
},
"aggregations": {
"categories": {
"meta": {},
"doc_count": 9,
"hoogte": {
"meta": {
"value": "features.hoogte.value"
},
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 55,
"doc_count": 9
},
{
"key": 5,
"doc_count": 0
},
{
"key": 6,
"doc_count": 0
},
{
"key": 8,
"doc_count": 0
},
{
"key": 10,
"doc_count": 0
},
{
"key": 13,
"doc_count": 0
},
{
"key": 16,
"doc_count": 0
},
{
"key": 18,
"doc_count": 0
},
{
"key": 30,
"doc_count": 0
},
{
"key": 32,
"doc_count": 0
}
]
},
"breedte": {
"meta": {
"measure": "mm",
"id": "id",
"title": "Breedte"
},
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 750,
"doc_count": 5
},
{
"key": 590,
"doc_count": 4
},
{
"key": 70,
"doc_count": 0
},
{
"key": 83,
"doc_count": 0
},
{
"key": 85,
"doc_count": 0
},
{
"key": 94,
"doc_count": 0
},
{
"key": 98,
"doc_count": 0
},
{
"key": 110,
"doc_count": 0
},
{
"key": 120,
"doc_count": 0
},
{
"key": 165,
"doc_count": 0
}
]
}
}
}
}
我认为我走在正确的轨道上，结果似乎还不错，但是每个规范的值是OR而不是AND，如果选择一个值，则规范的其他值将显示为0。 我想要的是将值添加到选择中时得到的计数。 我希望有人能指出我正确的方向。
1 个解决方案
解决方案1
0 2018-07-02 14:32:50
我想我离现在有点近了。 我将所选规格与每个存储桶一起发送。 我现在还有以下两个问题：
-
我无法过滤原始产品，因为选定规格的所有未选定值的存储桶计数将归零（请参见下文），我发现我可以使用post_filter解决此post_filter！ - 如果该值是从该特定规范中唯一选择的值，那么我将获得计数。 但是，如果我在规范中选择2个值，则需要将这些数字相加
这就是我现在所拥有的。 关于如何进行的任何想法？
GET _search
{
"from": 0,
"size": 15,
"sort":{
"name":"desc"
},
"stored_fields":[
"_id",
"name"
],
"query":{
"constant_score":{
"filter":{
"bool": {
"must": [
{
"terms": {
"features.breedte.value": [
"1200"
]
}
}
]
}
}
}
},
"aggs":{
"filtered_hoogte": {
"filter": {
"bool": {
"must": [
{
"terms": {
"features.breedte.value": [
"1200"
]
}
}
]
}
},
"aggs": {
"hoogte_aggs": {
"terms": {
"field": "features.hoogte.value",
"min_doc_count": 0
},
"meta":{
"value":"features.hoogte.value"
}
}
}
},
"filtered_breedte": {
"filter": {
"bool": {
"must": [
]
}
},
"aggs": {
"breedte_aggs": {
"terms": {
"field": "features.hoogte.value",
"min_doc_count": 0
},
"meta":{
"value":"features.hoogte.value"
}
}
}
}
}
}
结果：
{
"took": 19,
"timed_out": false,
"_shards": {
"total": 6,
"successful": 5,
"skipped": 0,
"failed": 1,
"failures": [
{
"shard": 0,
"index": ".kibana",
"node": "2fjW10M0RSa_imRTIMutww",
"reason": {
"type": "query_shard_exception",
"reason": "No mapping found for [name] in order to sort on",
"index_uuid": "x4687TCDRTalALulM9xLSg",
"index": ".kibana"
}
}
]
},
"hits": {
"total": 210,
"max_score": null,
"hits": [
{
"_index": "products",
"_type": "product",
"_id": "2011475",
"_score": null,
"sort": [
"0IXGBVFPS12"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014465",
"_score": null,
"sort": [
"20IXGVFPS1"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014455",
"_score": null,
"sort": [
"120IXBVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2013879",
"_score": null,
"sort": [
"120IXVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2011478",
"_score": null,
"sort": [
"20ANGBVFPS1"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014466",
"_score": null,
"sort": [
"20ANGVFPS1"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014457",
"_score": null,
"sort": [
"20ANBVFPS1"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2014444",
"_score": null,
"sort": [
"120ANVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2009698",
"_score": null,
"sort": [
"02IXGBVFPS12"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2011493",
"_score": null,
"sort": [
"202IXGVFPS1"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2011491",
"_score": null,
"sort": [
"02IXBVFPS12"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2012677",
"_score": null,
"sort": [
"1202IXVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2009701",
"_score": null,
"sort": [
"1202ANGBVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2009697",
"_score": null,
"sort": [
"1202ANGVFPS"
]
},
{
"_index": "products",
"_type": "product",
"_id": "2011492",
"_score": null,
"sort": [
"1202ANBVFPS"
]
}
]
},
"aggregations": {
"filtered_breedte": {
"meta": {},
"doc_count": 210,
"breedte_aggs": {
"meta": {
"value": "features.breedte.value"
},
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 1200,
"doc_count": 210
},
{
"key": 70,
"doc_count": 0
},
{
"key": 83,
"doc_count": 0
},
{
"key": 85,
"doc_count": 0
},
{
"key": 94,
"doc_count": 0
},
{
"key": 98,
"doc_count": 0
},
{
"key": 110,
"doc_count": 0
},
{
"key": 120,
"doc_count": 0
},
{
"key": 165,
"doc_count": 0
},
{
"key": 270,
"doc_count": 0
}
]
}
},
"filtered_hoogte": {
"doc_count": 210,
"hoogte_aggs": {
"meta": {
"value": "features.hoogte.value"
},
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 18,
"buckets": [
{
"key": 890,
"doc_count": 40
},
{
"key": 327,
"doc_count": 4
},
{
"key": 628,
"doc_count": 3
},
{
"key": 642,
"doc_count": 3
},
{
"key": 700,
"doc_count": 3
},
{
"key": 744,
"doc_count": 3
},
{
"key": 85,
"doc_count": 2
},
{
"key": 185,
"doc_count": 2
},
{
"key": 239,
"doc_count": 2
},
{
"key": 245,
"doc_count": 2
}
]
}
}
}
}
如何根据子桶聚合文档计数过滤桶聚合结果？
[英]How can I filter bucket aggregation results based on sub-bucket aggregation document count?
声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.