Question

我正在django中使用elasticsearch-dsl。而且我定义了一个DocType文档和一个包含值列表的关键字。

这是我的相同代码。

from elasticsearch_dsl import DocType, Text, Keyword

class ProductIndex(DocType):
    """
    Index for products
    """
    id = Keyword()
    slug = Keyword()
    name = Text()
    filter_list = Keyword()

filter_list是此处包含多个值的数组。现在我有了一些值，例如sample_filter_list，它们是与之不同的值，并且其中一些元素可以出现在某些产品的filter_list数组中。因此，鉴于此sample_filter_list，我希望filter_list与sample_filter_list相交的所有产品的filter_list的所有唯一元素都不为空。

for example I have 5 products whose filter_list is like :
1) ['a', 'b', 'c']
2) ['d', 'e', 'f']
3) ['g', 'h', 'i']
4) ['j', 'k', 'l']
5) ['m', 'n', 'o']
and if my sample filter_list is ['a', 'd', 'g', 'j', 'm']
then elasticsearch should return an array containg distinct element 
i.e. ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o']

Answer 1

            Writing Answer not specific to django but general,
            Suppose you have some ES index some_index2 with mapping

            PUT some_index2
            {
              "mappings": {
                "some_type": {
                  "dynamic_templates": [
                    {
                      "strings": {
                        "mapping": {
                          "type": "string"
                        },
                        "match_mapping_type": "string"
                      }
                    }
                  ],
                  "properties": {
                    "field1": {
                      "type": "string"
                    },
                    "field2": {
                      "type": "string"
                    }
                  }
                }
              }
            }

        Also you have inserted the documents 
        {
            "field1":"id1",
            "field2":["a","b","c","d]
        }
        {
            "field1":"id2",
            "field2":["e","f","g"]
        }
        {
            "field1":"id3",
            "field2":["e","l","k"]
        }

    Now as you stated you want all the distinct values of field2(filter_list) in your case, You can easily get that by using ElasticSearch term aggregation

    GET some_index2/_search
    {
    "aggs": {
      "some_name": {
        "terms": {
          "field": "field2",
          "size": 10000
        }
      }
    },
    "size": 0
    }

    Which will give you result as:

    {
      "took": 2,
      "timed_out": false,
      "_shards": {
        "total": 5,
        "successful": 5,
        "failed": 0
      },
      "hits": {
        "total": 3,
        "max_score": 0,
        "hits": []
      },
      "aggregations": {
        "some_name": {
          "doc_count_error_upper_bound": 0,
          "sum_other_doc_count": 0,
          "buckets": [
            {
              "key": "e",
              "doc_count": 2
            },
            {
              "key": "a",
              "doc_count": 1
            },
            {
              "key": "b",
              "doc_count": 1
            },
            {
              "key": "c",
              "doc_count": 1
            },
            {
              "key": "d",
              "doc_count": 1
            },
            {
              "key": "f",
              "doc_count": 1
            },
            {
              "key": "g",
              "doc_count": 1
            },
            {
              "key": "k",
              "doc_count": 1
            },
            {
              "key": "l",
              "doc_count": 1
            }
          ]
        }
      }
    }

    where buckets contains the list of all the distinct values.
    you can easily iterate through bucket and find the value under KEY.

Hope this is what is required to you.

Answer 2

对于您想要的东西，我有点困惑，只查询filter_list与sample_filter_list相交的产品，只需运行terms查询即可。

ProductIndex.search().filter('terms', filter_list=sample_filter_list)

希望这会有所帮助！

如何使用elasticsearch-dsl在数组中的所有索引中找到不同的值？

2 个答案: