admin管理员组文章数量:1278978
I am querying product links from an Elasticsearch database, where I previously imported results from various online shops. The query returns 123 records, but many products are sets, different colors, or similar variations. I want to limit the results to a maximum of 3 products per shop. However, all attempts still return the full list. How can I modify my query to achieve this per-shop limit? Here is my current query:
{
"size": 100,
"query": {
"bool": {
"must": [
{
"match": {
"key": {
"query": "Some Keyword",
"fuzziness": "1",
"prefix_length": "0",
"max_expansions": 3,
"operator": "and",
"analyzer": "product_analyzer"
}
}
}
]
}
},
"aggs": {
"group_by_shop": {
"terms": {
"field": "shop_id.keyword",
"size": 100
},
"aggs": {
"top_hits_per_shop": {
"top_hits": {
"size": 100,
"sort": [
{ "_score": { "order": "desc" } }
]
}
}
}
}
}
}
I am querying product links from an Elasticsearch database, where I previously imported results from various online shops. The query returns 123 records, but many products are sets, different colors, or similar variations. I want to limit the results to a maximum of 3 products per shop. However, all attempts still return the full list. How can I modify my query to achieve this per-shop limit? Here is my current query:
{
"size": 100,
"query": {
"bool": {
"must": [
{
"match": {
"key": {
"query": "Some Keyword",
"fuzziness": "1",
"prefix_length": "0",
"max_expansions": 3,
"operator": "and",
"analyzer": "product_analyzer"
}
}
}
]
}
},
"aggs": {
"group_by_shop": {
"terms": {
"field": "shop_id.keyword",
"size": 100
},
"aggs": {
"top_hits_per_shop": {
"top_hits": {
"size": 100,
"sort": [
{ "_score": { "order": "desc" } }
]
}
}
}
}
}
}
1 Answer
Reset to default 0Tdlr;
I think you are super close to what you want to achieve. You might want to tweak the size parameters for both the query and the top-hit aggregation.
Demo
Set up
Below is a small dataset for running the demo
DELETE 79462081
POST 79462081/_bulk
{"index":{}}
{"shop_id": 1, "product": "product_1", "priority": 1}
{"index":{}}
{"shop_id": 1, "product": "product_2", "priority": 2}
{"index":{}}
{"shop_id": 1, "product": "product_3", "priority": 3}
{"index":{}}
{"shop_id": 1, "product": "product_4", "priority": 4}
{"index":{}}
{"shop_id": 2, "product": "product_1", "priority": 1}
{"index":{}}
{"shop_id": 2, "product": "product_2", "priority": 2}
{"index":{}}
{"shop_id": 2, "product": "product_3", "priority": 3}
{"index":{}}
{"shop_id": 2, "product": "product_4", "priority": 4}
Query
Here is a variation of your query
GET 79462081/_search
{
"size": 0, # The result for the search query do not really interest me.
"query": {
"match_all": {}. # For demo purposes I am matching the whole dataset
},
"aggs": {
"top_shops": {
"terms": {
"field": "shop_id"
},
"aggs": {
"top_products": {
"top_hits": {
"size": 2, # I only want to top best product with regard to priority, but you might also use _score
"sort": { "priority": { "order": "desc" } }
}
}
}
}
}
}
Results
{
"took": 2,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 8,
"relation": "eq"
},
"max_score": null,
"hits": []. # Notice how this section is empty, while "value" is 8, this is because of the size: 0. Documents are found, used in the aggregation, but not returned.
},
"aggregations": {
"top_shops": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 1,
"doc_count": 4,
"top_products": {
"hits": {
"total": {
"value": 4,
"relation": "eq"
},
"max_score": null,
"hits": [ # I only get the top 2 documents with regards to the priority, out of the 4 that were matched.
{
"_index": "79462081",
"_id": "CnUaN5UBJxYdI_ZD5ifg",
"_score": null,
"_source": {
"shop_id": 1,
"product": "product_4",
"priority": 4
},
"sort": [
4
]
},
{
"_index": "79462081",
"_id": "CXUaN5UBJxYdI_ZD5ifg",
"_score": null,
"_source": {
"shop_id": 1,
"product": "product_3",
"priority": 3
},
"sort": [
3
]
}
]
}
}
},
{
"key": 2,
"doc_count": 4,
"top_products": {
"hits": {
"total": {
"value": 4,
"relation": "eq"
},
"max_score": null,
"hits": [
{
"_index": "79462081",
"_id": "DnUaN5UBJxYdI_ZD5ifg",
"_score": null,
"_source": {
"shop_id": 2,
"product": "product_4",
"priority": 4
},
"sort": [
4
]
},
{
"_index": "79462081",
"_id": "DXUaN5UBJxYdI_ZD5ifg",
"_score": null,
"_source": {
"shop_id": 2,
"product": "product_3",
"priority": 3
},
"sort": [
3
]
}
]
}
}
}
]
}
}
}
本文标签： elasticsearchElastic Search 8 queryStack Overflow
版权声明：本文标题：elasticsearch - Elastic Search 8 query - Stack Overflow 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.betaflare.com/web/1741300510a2371064.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。
发表评论