逗号分隔值的Elasticseach或查询

Question

我将id作为逗号保存在数据库中，并将其索引到ElasticSearch。现在我需要检索user_id是否与值匹配。

例如，它在列user_ids的索引中这样保存(在elasticsearch中，数据库类型是varchar(500)，它是文本)

8938,8936,8937

$userId = 8936; // For example expecting to return that row
$whereCondition = [];
$whereCondition[]  = [
                "query_string" => [
                    "query"=> $userId,
                    "default_field" => "user_ids",
                    "default_operator" => "OR"
                ]
            ];

$searchParams = [
    'query' => [
        'bool' => [
            'must' => [
                $whereCondition
            ],
            'must_not' => [
                ['exists' => ['field' => 'deleted_at']]
            ]
        ]
    ],
    "size" => 10000
];

User::search($searchParams);

Json查询

{
    "query": {
        "bool": {
            "must": [
                [{
                    "query_string": {
                        "query": 8936,
                        "default_field": "user_ids",
                        "default_operator": "OR"
                    }
                }]
            ],
            "must_not": [
                [{
                    "exists": {
                        "field": "deleted_at"
                    }
                }]
            ]
        }
    },
    "size": 10000
}

映射细节

{
    "user_details_index": {
        "aliases": {},
        "mappings": {
            "test_type": {
                "properties": {
                    "created_at": {
                        "type": "date",
                        "format": "yyyy-MM-dd HH:mm:ss"
                    },
                    "deleted_at": {
                        "type": "date",
                        "format": "yyyy-MM-dd HH:mm:ss"
                    },
                    "updated_at": {
                        "type": "date",
                        "format": "yyyy-MM-dd HH:mm:ss"
                    },
                    "user_ids": {
                        "type": "text",
                        "fields": {
                            "keyword": {
                                "type": "keyword",
                                "ignore_above": 256
                            }
                        }
                    }
                }
            }
        },
        "settings": {
            "index": {
                "creation_date": "1546404165500",
                "number_of_shards": "5",
                "number_of_replicas": "1",
                "uuid": "krpph26NTv2ykt6xE05klQ",
                "version": {
                    "created": "6020299"
                },
                "provided_name": "user_details_index"
            }
        }
    }
}

我正在尝试上述逻辑，但并不是无法检索。有人能帮上忙吗。

Answer 1

由于字段user_ids是text类型，因此默认情况下没有为其指定任何分析器，它将使用standard分析器，它不会将8938,8936,8937分解为术语8938、8936和8937，因此id不能匹配。

为了解决这个问题，我建议您将I数组存储到user_ids字段，而不是csv。因此，在索引您的json输入时，应该如下所示：

{
   ...

   "user_ids": [
      8938,
      8936,
      8937
   ]

   ...
}

由于用户in是整数值，所以在映射中应该进行更改：

{
   "user_ids": {
      "type": "integer"
   }
}

现在查询如下：

{
  "query": {
    "bool": {
      "filter": [
        [
          {
            "terms": {
              "userIds": [
                8936
              ]
            }
          }
        ]
      ],
      "must_not": [
        [
          {
            "exists": {
              "field": "deleted_at"
            }
          }
        ]
      ]
    }
  },
  "size": 10000
}