如果状态为真，则 Elasticsearch bool 查询按日期排序
Posted
技术标签:
【中文标题】如果状态为真，则 Elasticsearch bool 查询按日期排序【英文标题】：Elasticsearch bool query sort by date if status is true 【发布时间】：2020-11-30 09:44:47 【问题描述】：我在 Elasticsearch 索引中有一个如下所示的 JSON 文件。如果广告未过期且状态为真，我需要对数据进行排序，然后将它们排序为 desc。我怎样才能做到这一点？
我尝试使用 end_date 排序，但它不起作用。我还需要显示 end_date 已过期的所有过期数据。
advertisement = [
"id": 1,
"name": "test",
"status": True,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"id": 2,
"name": "test2",
"status": False,
"start_date": "2020-08-09",
"end_date": "2020-08-09",
]
这是我的弹性搜索方法。
def elastic_search(category=None):
client = Elasticsearch(host="localhost", port=9200)
query_all =
'size': 10000,
'query':
"bool":
"filter": [
"match":
"name": "test"
]
,
,
"sort": [
"end_date":
"type": "date",
"order": 'desc'
]
resp = client.search(
index="my-index",
body=query_all
)
return resp
这是我的回复
http://localhost:9200/my-index/_search
"took":96,
"timed_out":false,
"_shards":
"total":5,
"successful":5,
"skipped":0,
"failed":0
,
"hits":
"total":36,
"max_score":1.0,
"hits":[
"_index":"my-index",
"_type":"doc",
"_id":"52",
"_score":1.0,
"_source":
"id": 1,
"name": "test",
"status": True,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"_index":"my-index",
"_type":"doc",
"_id":"60",
"_score":1.0,
"_source":
"id": 1,
"name": "English test",
"status": True,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"_index":"my-index",
"_type":"doc",
"_id":"40",
"_score":1.0,
"_source":
"id": 1,
"name": "Designw test",
"status": false,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"_index":"my-index",
"_type":"doc",
"_id":"41",
"_score":1.0,
"_source":
"id": 1,
"name": "Designw New",
"status": false,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"_index":"my-index",
"_type":"doc",
"_id":"59",
"_score":1.0,
"_source":
"id": 1,
"name": "Designw New",
"status": false,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
,
"_index":"my-index",
"_type":"doc",
"_id":"62",
"_score":1.0,
"_source":
"id": 1,
"name": "Designw New",
"status": false,
"start_date": "2020-08-09",
"end_date": "2020-09-09",
]
这是我的映射 http://localhost:9200/my-index/_mapping 响应。
"my-index":
"mappings":
"_doc":
"properties":
"address":
"properties":
"name":
"type":"text",
"fields":
"keyword":
"type":"keyword",
"ignore_above":256
,
"start_date":
"type":"text",
"fields":
"keyword":
"type":"keyword",
"ignore_above":256
,
"end_date":
"type":"text",
"fields":
"keyword":
"type":"keyword",
"ignore_above":256
,
"id":
"type":"long"
,
"status":
"type":"text",
"fields":
"keyword":
"type":"keyword",
"ignore_above":256
【问题讨论】：
你的映射是什么？ 需要按过期日期对数据进行排序并排序 再来。你的索引映射是什么？GET my-index/_mapping
再说一次，我不需要您的映射的一般描述。我需要你复制粘贴http://localhost:9200/my-index/_mapping的回复
不是。这是 _search 响应，而不是 _mapping。
【参考方案1】：
关于映射的两件事：
-
有一个
address 字段在您的实际文档中找不到。删除它。
应使用date 数据类型正确映射您的日期。
正确的映射如下所示：
"properties":
"end_date":
"type":"date",
"format":"yyyy-MM-dd"
,
"start_date":
"type":"date",
"format":"yyyy-MM-dd"
,
//...other properties
一旦您获得正确的映射，此查询会查找所有未过期且具有真实状态的广告，并按运行时间最长的广告进行排序：
"query":
"bool":
"must": [
"range":
"end_date":
"gt": "now"
,
"term":
"status":
"value": true
]
,
"sort": [
"end_date":
"order": "desc"
]
或者，如果您要查找过期的，请将gt 更改为lt，它代表less-than。
【讨论】：
以上是关于如果状态为真，则 Elasticsearch bool 查询按日期排序的主要内容，如果未能解决你的问题，请参考以下文章