elasticsearch ik配置动态同义词

使用环境

elasticsearch5.1.1
kibana5.1.1
同义词插件5.1.1

安装插件

下载对应的elasticsearch-analysis-dynamic-synonym-5.1.1.zip, 解压到本地的elasticsearch/plugins目录下, 重新启动es

第一种方式本地文件

说明：

对于本地文件：主要通过文件的修改时间戳(Modify time)来判断是否要重新加载
在elasticsearch/config目录下,建立analysis目录, 并在analysis目录下放入synonym.txt, 在文件首行加入下面一行同义词，来进行测试

西红柿, 番茄, 圣女

es设置索引和自定义解析器

PUT /megacorp
{"mappings": {"employee": {"properties": {"name":{"type": "text","analyzer": "ik-index",  //指定索引时候用的解析器"search_analyzer": "ik-smart" //指定搜索时候用的解析器}}}},"settings": {"analysis": {"filter": {"local_synonym" : {"type" : "dynamic_synonym","synonyms_path" : "analysis/synonym.txt"  }},"analyzer": {"ik-index": {"type": "custom","tokenizer": "ik_max_word","filter": ["local_synonym"   //对同义词进行了过滤]},"ik-smart": {"type": "custom","tokenizer": "ik_smart","filter": ["local_synonym"]}}}}
}

设置好后，可以用以下命令查看，同义词是否配置成功

GET /megacorp/_analyze
{"analyzer": "ik-index","text": "西红柿"
}

正确分词结果如下：

{"tokens": [{"token": "西红柿","start_offset": 0,"end_offset": 3,"type": "CN_WORD","position": 0},{"token": "番茄","start_offset": 0,"end_offset": 3,"type": "SYNONYM","position": 0},{"token": "圣女","start_offset": 0,"end_offset": 3,"type": "SYNONYM","position": 0}]
}

插入2条数据：

PUT /megacorp/employee/1
{"name" : "圣女果"
}PUT /megacorp/employee/2
{"name" : "番茄"
}

搜索西红柿, 会搜索出番茄和圣女果的记录：

GET /megacorp/employee/_search
{"query":{"match": {"name": "西红柿"} }
}

第二种方式远程接口

说明：

这个http请求需要返回两个头部，一个是 Last-Modified，一个是 ETag，只要有一个发生变化，该插件就会去获取新的同义词来更新相应的同义词。

本地写个接口

http://localhost/synonym/list

该接口返回的需要设置以下三个属性

$response->setLastModified($lastModified);
$response->setEtag($etag, true);
$response->headers->set('Content-Type', 'text/plain');

注：

nginx 在开启了 gzip 之后，如果有 ETAG 则会调用 ngx_http_clear_etag 将其清除,

解决的办法很简单：

只要 PHP 返回的 ETAG 是 weak ETAG，那么就一切都会正常起来了。而所谓的 weak ETAG，也就是弱 ETAG，它是相对于正常 ETAG 而言的，表现形式就是 ETAG 前面加上 W/

W/"db8b38e8a3257a2f195b727eceb2c5d3"

下面是设置远程, 本地同义词的配置

PUT /megacorp

{"settings": {"index": {"number_of_shards": "5","number_of_replicas": "1","analysis": {"filter": {"remote_synonym": {"type": "dynamic_synonym","synonyms_path": "http://192.168.78.37:10001/synonym.txt","interval": 10},"edgeNgramFilter": {"type": "edge_ngram","min_gram": 1,"max_gram": 50}},"analyzer": {"ngramIndex": {"type": "custom","tokenizer": "keyword","filter": ["edgeNgramFilter","lowercase"]},"ngramLowercase": {"filter": ["lowercase"],"type": "custom","tokenizer": "ngramTokenizer"},"ikIndex": {"filter": ["remote_synonym"],"type": "custom","tokenizer": "ik_max_word"},"ikSearch": {"type": "custom","tokenizer": "ik_smart"},"keywordLowercase": {"type": "custom","filter": ["lowercase"],"tokenizer": "keyword"}},"tokenizer": {"ngramTokenizer": {"type": "nGram","min_gram": "1","max_gram": "3"}}}}},"mappings": {"type1": {"properties": {"title": {"type": "text","fields": {"ik": {"analyzer": "ikIndex","search_analyzer": "ikSearch","type": "text"},"keywordLowercase": {"analyzer": "keywordLowercase","type": "text"},"ngram": {"analyzer": "ngramLowercase","type": "text"}}}}}}
}

{ "mappings": { "employee": { "properties": { "name":{ "type": "text", "analyzer": "ik-index", //指定索引时候用的解析器 "search_analyzer": "ik-smart" //指定搜索时候用的解析器 } } } } , "settings": { "analysis": { "filter": { "remote_synonym": { "type" : "dynamic_synonym", "synonyms_path" : "http://localhost/synonym/list", "interval": 60 // 没60s调取一次接口 }, "local_synonym" : { "type" : "dynamic_synonym", "synonyms_path" : "analysis/synonym.txt" } }, "analyzer": { "ik-index": { "type": "custom", "tokenizer": "ik_max_word", "filter": [ "remote_synonym", //对远程同义词进行了过滤 "local_synonym" //对本地同义词进行了过滤 ] }, "ik-smart": { "type": "custom", "tokenizer": "ik_smart", "filter": [ "local_synonym" ] } } } }