Elasticsearch query_string嵌套查询

问题描述

我正在尝试使用query_string编写查询以检索嵌套对象的数据查询.

I am trying to write a query using query_string to retrieve data querying by nested objects.

我要执行的查询示例如下:

An example of query I would like to do is this one:

{
  "query": {
    "query_string": {
      "query": "a.id:2"
    }
  }
}

其中"a"是嵌套对象，"id"是"a"的字段.

Where "a" is a nested object, and "id" is a field of "a".

我知道我可以使用嵌套查询成功地执行此任务，编写如下查询:

I know I can successfully perform this task using using a nested query, writing a query like:

{
  "nested": {
    "path": "a"
    "query_string": {
      "query": "a.id:2"
    }
  }
}

但是，我想避免这种情况.我不想自己弄清楚用户正在搜索嵌套字段并修改查询.我尝试使用"fields"参数，但它似乎不适用于嵌套对象.

However, I would like to avoid it. I don't want to figure out by myself that the user is searching for a nested field and modify the query.I tried to use the "fields" parameter, but it looks like it doesn't work with nested objects.

是否可以使用"query_string"查询直接编写此查询?有可能获得什么语义?(例如，如果我写"a.id:2 AND a.b:10"，我是在匹配同一对象还是不同对象中的两个字段?)

Is it possible to write this query directly using "query_string" queries?What semantic is it possible to obtain? (for instance, if I write "a.id:2 AND a.b:10" I am matching the two fields in the same object or in different objects?)

推荐答案

我进行了更多研究，发现可以通过在映射中将include_in_parent设置为true来实现.

I was doing more research and found this can be achieved by setting the include_in_parent setting to true in the mapping.

现在您应该可以进行

{
  "query": {
    "query_string": {
      "query": "fields.fieldvalue:sometext"
    }
  }
}

例如:-

"mappings": {
         "documentmetadatavalue": {
            "properties": {
              "id": {
                  "type": "string"
               },
               "fields": {
                 "type": "nested",
                 "include_in_parent": true,
                 "properties": {
                   "fieldId": {"type": "string"},
                   "fieldvalue": {"type": "string"}
                 }
               }
           }
        }
     }

让我知道是否有帮助.

这篇关于Elasticsearch query_string嵌套查询的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！