Question

我的ES索引产品包含这样的数据（简化版，实际上大约有20个字段）：

{_id: 1, _score: 1, color: red, size: S}
{_id: 2, _score: 1, color: red, size: M}
{_id: 3, _score: 1, color: red, size: L}
{_id: 4, _score: 1, color: blue, size: S}
{_id: 5, _score: 1, color: blue, size: M}
{_id: 6, _score: 1, color: blue, size: L}

我想按属性（颜色和大小）过滤产品，但我需要在结果中显示所有产品，过滤条件应仅影响分数。例子：

Query: color == red
Result:
{_id: 1, _score: 1}
{_id: 2, _score: 1}
{_id: 3, _score: 1}
{_id: 4, _score: 0}
{_id: 5, _score: 0}
{_id: 6, _score: 0}

Query: size == M
Result:
{_id: 1, _score: 0}
{_id: 2, _score: 1}
{_id: 3, _score: 0}
{_id: 4, _score: 0}
{_id: 5, _score: 1}
{_id: 6, _score: 0}

Query: color == red && size == M
Result:
{_id: 1, _score: 0}
{_id: 2, _score: 1}
{_id: 3, _score: 0}
{_id: 4, _score: 0}
{_id: 5, _score: 0}
{_id: 6, _score: 0}

任何想法我怎样才能做到这一点？它看起来像弹性搜索的工作？也许我应该考虑切换一些其他商店/数据库。 ES版本是1.7.5

Answer 1

我不确定您的用例是什么，但我不建议您尝试修复相关性得分。以下是其中一个案例的示例，有多种方法可以执行此操作

POST /_search
{
   "query":{
      "bool":{
         "should":[
            {
               "constant_score":{
                  "filter":{
                     "term":{
                        "color":"red"
                     }
                  },
                  "boost":1
               }
            },
            {
               "bool":{
                  "must_not":{
                     "term":{
                        "color":"red"
                     }
                  },
                  "boost":0
               }
            }
         ]
      }
   }
}

，结果是

{
    "took": 3,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": 6,
        "max_score": 1,
        "hits": [
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "2",
                "_score": 1,
                "_source": {
                    "color": "red",
                    "size": "M"
                }
            },
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "1",
                "_score": 1,
                "_source": {
                    "color": "red",
                    "size": "S"
                }
            },
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "3",
                "_score": 1,
                "_source": {
                    "color": "red",
                    "size": "L"
                }
            },
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "5",
                "_score": 0,
                "_source": {
                    "color": "blue",
                    "size": "M"
                }
            },
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "4",
                "_score": 0,
                "_source": {
                    "color": "blue",
                    "size": "S"
                }
            },
            {
                "_index": "test",
                "_type": "_doc",
                "_id": "6",
                "_score": 0,
                "_source": {
                    "color": "blue",
                    "size": "L"
                }
            }
        ]
    }
}

Elasticsearch：基于过滤器构建二进制掩码

1 个答案: