Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hanwenbo/fast-search
elasticsearch客户端封装
https://github.com/hanwenbo/fast-search
Last synced: 6 days ago
JSON representation
elasticsearch客户端封装
- Host: GitHub
- URL: https://github.com/hanwenbo/fast-search
- Owner: hanwenbo
- Created: 2018-08-27T14:57:15.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2018-08-27T09:46:06.000Z (over 6 years ago)
- Last Synced: 2024-11-07T16:12:28.400Z (about 2 months ago)
- Language: Java
- Size: 11.7 MB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Search
Search项目是基于ElasticSearch搜索服务提供的简易API调用，支持以下功能
---------------
## 数据索引
* [添加索引数据](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/BaseInsert.java)
* [更新索引数据(UpdateById/UpdateByQuery)](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/BaseUpdate.java)
* [数据批量操作(批量添加/删除/修改)](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/BulkOperation.java)
* [路由(routing)](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/RoutingOperation.java)
* [建议查询(Suggest)](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/SuggestSearch.java)
* [Mustache模板查询](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/TemplateSearch.java)
* [索引数据删除(DeleteById/DeleteByQuery/DeleteByType)](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/index/BaseDelete.java)
---------------
## 检索功能
### 检索功能,支持基本逻辑查询：
1. “& | ! ” 查询
2. “ 嵌套 & | ! ”查询
3. EQL查询(自定义SQL)
4. 聚合查询方式
### 每个简单数据类型查询单元支持
1. {term:中国西域} --- 精确查询
2. {match:中国西域} --- 匹配(支持分词)
3. {prefix:中国西域} --- 前缀
4. {wildcard:W?F*HW} --- 通配符
5. {regexp:W[0-9].+} --- 正则
---------------
## 类SQL查询规则
### 概念：
> 搜索单元
> * 概念:用于搜索的“一个”逻辑单元,小括号括起,并且每个括号是一个field的比较
> * 书写形式:(field:value)
> * Note:(a:1)、(1<a<10) 、(a>5&a<10) 都是单个field的比较
---------------
> 值函数
> * 概念:函数解析使用{}括起(未使用函数解析的均认为是term解析查询)
> * 书写形式:{term:中国西域}
> * 函数穷举:
> * {term:中国西域} ——词条查询
> * {match:中国西域} ——匹配(支持分词)
> * {prefix:中国西域} ——前缀
> * {wildcard:W?F*HW} ——通配符
> * {regexp:W[0-9].+} ——正则
> * Node:(supplierName:{wildcard:北京商贸*}) 查询已北京商贸开头的相关信息
---------------
> 范围查询
> * 概念:范围查询针对集合区间在后台处理做了抽象,支持任意的开闭原则及范围区间查询 支持基本数据类型范围查询及日期类型范围查询
> * 书写形式:(value1<field<=value2)、(field<v2)、(field<v2)
> * Node:
> * 闭区间查询保证字段在值的中间如: v1<field<v2
> * 开区间查询保证字段在值的左边如: field>v1 或者 field<v2
> * 日期范围查询日期格式为 “yyyy-MM-dd HH:mm:ss”
---------------
> 取反查询运算
> * 概念:满足条件的反向,只能用于查询语句的前面 用!表示
> * 书写形式: !(a<=10) 、!(&(a:1|2|3)&(b:zhangsan))
> * Node: !(a<10) = (a>=10)
---------------
> 逻辑运算符
> * 概念: 用于单元中与 或 的逻辑运算,用& | 表示
> * 书写形式: !(a<=10) 、!(&(a:1|2|3)&(b:zhangsan))
> * Node: !(a<10) = (a>=10)
---------------
### 搜索案例：
* 精确查找：(supplierId:2241527253818753)
* 单字段或查询：
* (cityId:1|18|241)
* !(cityId:1|18|241)
* 简单范围查询:
* (goodsStorage>100000000)
* (goodsStorage<=10)
* (goodsStorage<=10|goodsStorage>100000000)
* !(goodsStorage<=10|goodsStorage>100000000)同(10<goodsStorage<=100000000)等价
* 日期范围查询：yyyy-MM-dd HH:mm:ss
* (createTime<2018-05-15 00:00:00)
* (createTime*gt;=2018-05-15 00:00:00)
* !(createTime<=2018-05-15 00:00:00)
* (createTime<=2018-05-15 00:00:00|createTime>2018-06-15 00:00:00)
* !(createTime<=2018-05-15 00:00:00|createTime>2018-06-15 00:00:00)与(2018-05-15 00:00:00<createTime<=2018-06-15 00:00:00)等价
* 函数查询：
* (supplierName:{prefix:测试}) 前缀
* (supplierName:{wildcard:*测试*}) 通配符
* (supplierName:{regexp:[^Baker]+测试.*}) 正则
* 组合查询：
* (&(2018-05-15 00:00:00<=createTime<=2018-06-15 00:00:00)&(goodsStorage<1000))
* !(&(2018-05-15 00:00:00<=createTime<=2018-06-15 00:00:00)&(goodsStorage<1000))
* (&(2018-05-15 00:00:00<=createTime<=2018-06-15 00:00:00)&(goodsStorage<1000)&(supplierId:2241527253818753))
* !(supplierName:{regexp:[^Baker]+测试.*})
* (&(supplierName:{regexp:[^Baker]+测试.*})&!(5<goodsNum<25))
* &!(&(supplierName:{regexp:[^Baker]+测试.*})&!(5<goodsNum<25))
---------------
### EQL数据组织
```java
String eql = "|(attrIds:2232012366099328|189)|(attrIds:2292774003989889|2300097498406272)";
BoolPager boolPager = EqlToPagerConverterUtils.convertToPager(eql);
boolPager.setPageNo(pageNo);
boolPager.setPageSize(pageSize);
boolPager.set_index(dto.getIndexName().trim());
boolPager.set_type(dto.getIndexType().trim());
if(StringUtils.isNotBlank(dto.getSortName())) {
if("desc".equals(dto.getSortOrder())){
sortField.put(dto.getSortName().trim(), SearchFactor.DESC);
}else {
sortField.put(dto.getSortName().trim(), SearchFactor.ASC);
}
boolPager.setSortFields(sortField);
}
BoolPager resultPager = BaseSearch.boolQuery(transportClient, boolPager);
System.out.println(resultPager.getResult());
```
## 搜索平台
#### 分词功能
#### 检索功能
#### 聚合........
---------------
## 插件功能
IK分词器进行二次开发，支持数据库扫描热词、停止词功能。
添加ElasticSearch TokenFilter 扩展,增加以数据库为数据源的联想词功能,源码下载
* [elasticsearch-analysis-ik]([email protected]:business/project/platform/elasticsearch-analysis-ik.git)
IK分词器配置使用方式不变,内部自动集成MySQL热词数据源
联想词`ik_synonym`功能扩展
```bash
curl -XPUT http://XXX.XXX.XXX.XXX:9200/g_i -d '
{
"settings":{
"refresh_interval":"1s",
"number_of_replicas":1,
"number_of_shards":1,
"analysis":{
"filter":{
"by_tfr":{
"type":"stop",
"stopwords":[" "]
},
"by_sfr":{
"type":"ik_synonym",
"ignore_case":true,
"expand":true
}
},
"analyzer":{
"by_smart":{
"type":"custom",
"char_filter": [
"html_strip"
],
"tokenizer":"ik_smart",
"filter":[
"by_sfr"
]
},
"by_max_word":{
"type":"custom",
"char_filter": [
"html_strip"
],
"tokenizer":"standard",
"filter":[
"by_sfr"
]
}
}
}
},
"mappings":{
"g_t": {
"properties": {
"goodsName": {
"type": "text",
"analyzer": "by_smart"
}
}
}
}
}'
```
---------------
## 聚合功能
[聚合查询](https://github.com/ZhuBaker/fast-search/tree/master/fast-common/src/main/java/com/github/search/search/index/aggr)
---------------
# 自定义Spring标签
```xml
```
---------------
# 查询数据组织(统一入口)
## 针对查询采用统一的查询入口进行数据组织
## 组织形式
```java
BoolPager boolPager = new BoolPager();
boolPager.setPageNo(pageNo);
boolPager.setPageSize(pageSize);
List vps = new ArrayList<>();
List vs = new ArrayList<>();
vs.add(new ValueEntity.Builder("categoryId",new Object[]{categoryId.toString()}).setNot().build());
vs.add(new ValueEntity.Builder("supplierName",new Object[]{"*"+supplierName+"*"}).rule(SearchType.WILDCARD_QUERY).build());
vs.add(new ValueEntity.Builder("orderStatus",new Object[]{1}).build());//接单状态 1正常接单
vs.add(new ValueEntity.Builder("isDel",new Object[]{0}).build());//供应商状态 1正常 2暂停接单
if(isSpecial != null && isSpecial == 1){//自营订单传 1 限制可接自营供应商才可接该单 进行限制
vs.add(new ValueEntity.Builder("isSpecial",new Object[]{isSpecial}).build());//供应商状态 1正常 2暂停接单
}
if(isTest != null){//自营订单传 1 限制可接自营供应商才可接该单 进行限制
vs.add(new ValueEntity.Builder("isTest",new Object[]{isTest}).build());
}
ValuePackage vp = new ValuePackage();
vp.setEntitys(vs);
vps.add(vp);
boolPager.setQuery(vps);
boolPager.set_index(IndexConf.supplierIndex);
boolPager.set_type(IndexConf.categoryType);
boolPager.setFields(new String[]{"supplierId","supplierName"});//获取供应商id和供应商名称
boolPager = BaseSearch.boolQuery(client,boolPager);
```
ValuePackage类似于一个查询实体包 ,是包装了多个查询逻辑单元组合的查询集合,如果对ElasticSearch Restful调用
---------------
``