ElasticSearch 电商搜索实现

最近打算自己做一个类似手机电商app的搜索功能，于是自然想到了elasticsearch，首先参考到的美团的app，根据类目和喜好，加上关键词，根据距离远近筛选出满足用户要求的店铺，以及店铺下的满足要求的商品（3个），如下图类似的效果

由于刚刚入坑，elasticsearch，所以看了些文档，elastcisearch，有nested和parent/child这两种格式的文档结构满足功能需求，nested是将店铺和商品信息看做一个文档，当店铺里的商品很多或者，要更改商品信息时需要更新整个文档。根据子集进行过滤后无法支持只取子集的前3个所以暂不考虑，parent/child这种格式的现在6.X现在统一叫join，反正性能好像不太好。elasticsearch属于NoSql的分支，数据尽量扁平化。根据网络上的做法是做宽表冗余。

我采用的就是这种做法。将店铺的部分信息冗余到商品里去。（店铺名称，图标，月销量和坐标）。下图是我的商品的mapping

ElasticSearch 电商搜索实现

看下product的文档数据 ElasticSearch 电商搜索实现要求就是根据店铺做折叠，对折叠功能有兴趣的同学可以看下链接：https://elasticsearch.cn/article/132

现在要搜索关键字为数码，且goodTypeId为2的所有商品，按店铺显示

GET product/_search
{
   "from": 0,
   "size": 10,
   "query": {
       "bool": {
           "must": [{
               "match": {
                   "goodsLabels": {
                       "query": "数码",
                       "operator": "OR",
                       "prefix_length": 0,
                       "max_expansions": 50,
                       "fuzzy_transpositions": true,
                       "lenient": false,
                       "zero_terms_query": "NONE",
                       "auto_generate_synonyms_phrase_query": true,
                       "boost": 1.0
                   }
               }
           },
           {
               "term": {
                   "goodTypeId": {
                       "value": 2,
                       "boost": 1.0
                   }
               }
           },
           {
               "geo_distance": {
                   "coordinate": [106.395645,
                   59.929986],
                   "distance": 15000.0,
                   "distance_type": "arc",
                   "validation_method": "STRICT",
                   "ignore_unmapped": false,
                   "boost": 1.0
               }
           }],
           "adjust_pure_negative": true,
           "boost": 1.0
       }
   },
   "_source": {
       "includes": ["shopId",
       "merchantId",
       "monthSalesVolume",
       "shopName",
       "coordinate"],
       "excludes": []
   },
   "script_fields": {
       "distance": {
           "script": {
               "source": "doc['coordinate'].arcDistance(params.lat,params.lon)",
               "lang": "painless",
               "params": {
                   "lon": 106.395645,
                   "lat": 59.929986
               }
           },
           "ignore_failure": false
       }
   },
   "sort": [{
       "_geo_distance": {
           "coordinate": [{
               "lat": 59.929986,
               "lon": 106.395645
           }],
           "unit": "km",
           "distance_type": "arc",
           "order": "asc",
           "validation_method": "STRICT",
           "ignore_unmapped": false
       }
   }],
   "collapse": {
       "field": "shopId",
       "inner_hits": {
           "name": "top_price",
           "ignore_unmapped": true,
           "from": 0,
           "size": 1,
           "version": false,
           "explain": false,
           "track_scores": true,
           "_source": {
               "includes": ["goodsId",
               "goodName",
               "price",
               "discountedPrice",
               "goodUrlThumb"],
               "excludes": []
           },
           "sort": [{
               "price": {
                   "order": "desc"
               }
           }]
       }
   }
}

这个es请求的json，有兴趣的自己看下，应该不难懂。

请求结果为：

{
"took" : 10,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : 2,
"max_score" : null,
"hits" : [
{
"_index" : "product",
"_type" : "goods",
"_id" : "3",
"_score" : null,
"_source" : {
"coordinate" : "59.929986,106.395645",
"merchantId" : null,
"monthSalesVolume" : null,
"shopName" : "华为荣耀官方旗舰店",
"shopId" : "2"
},
"fields" : {
"distance" : [
0.0
],
"shopId" : [
"2"
]
},
"sort" : [
0.0
],
"inner_hits" : {
"top_price" : {
"hits" : {
"total" : 2,
"max_score" : 2.3884578,
"hits" : [
{
"_index" : "product",
"_type" : "goods",
"_id" : "3",
"_score" : 2.3296995,
"_source" : {
"goodName" : "荣耀V20",
"discountedPrice" : null,
"goodsId" : 3,
"price" : 1800,
"goodUrlThumb" : null
},
"sort" : [
1800
]
}
]
}
}
}
}
]
}
}

贴出java代码，过程中的难点是es的api很难和spring-data-elasticsearch的api对应起来：

ElasticSearch 电商搜索实现

看下swagger请求效果

ElasticSearch 电商搜索实现

用到的主要是折叠，其次还是需要在模型上多思考。上面用到了elasticsearch的script_filed.来计算出传入的经纬度和店铺经纬度之间的距离，也就是这里的distance.

ElasticSearch 电商搜索实现

相关推荐