Problem Description
I wrote a small search project, I want to use es to realize the highlight search, but suddenly I found that my search function can only find English such as java, dior, vue, etc., but it cannot search for medicine, advanced mathematics, etc. Out!
Solutions
At first, I thought it was my request sent in the past that caused Chinese garbled characters
new Vue({
el:'#app',
data:{
keyword:'' ,//搜索的关键字
results:[] //搜索的结果
},
methods:{
searchKey(){
var keyword = this.keyword;
console.log(keyword);
//对接后端接口
axios.get('search/'+keyword+"/1/20").then(response=>{
console.log(response);
//绑定数据
this.results = response.data;
})
}
}
})
I thought that the keyword search keyword received by my backend was garbled, but I printed it in the background and found that there was no garbled code! !
Look for solutions again
I looked up the information later and it seemed that it was a problem with the tokenizer! ! ! Hey, it makes sense.
I tried to search for a word "medicine" or "learning" on the trial sheet, and found that it could be found . Then there is a bit of eyebrows, but should this tokenizer be changed? At first I thought it was naive that the thesaurus had no words like medicine! ! People's thesaurus is too large, it is impossible to not have medical words, and then asked the boss, read the blog, and later found a similar
original link to the blogger : https://blog.csdn.net/qq_44961149 /article/details/107300665 If you
are interested, you can take a look. The
reason is:
//实现搜索功能
public List<Map<String,Object>> searchPage(String keyword,int pageNo,int pageSize) throws IOException {
if (pageNo<1){
pageNo = 1;
}
//条件搜索
SearchRequest searchRequest = new SearchRequest("jd_goods");
//构建搜索条件
SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
//实现分页功能
searchSourceBuilder.from(pageNo);
searchSourceBuilder.size(pageSize);
//精确匹配
// TermQueryBuilder termQueryBuilder = QueryBuilders.termQuery("title",keyword);
MatchPhraseQueryBuilder matchPhraseQueryBuilder = QueryBuilders.matchPhraseQuery("title", keyword);
searchSourceBuilder.query(matchPhraseQueryBuilder);
//60s加载时间
searchSourceBuilder.timeout(new TimeValue(60, TimeUnit.SECONDS));
//高亮
HighlightBuilder highlightBuilder = new HighlightBuilder();
highlightBuilder.field("title");
//这里可以配置多个字段，信息等等
//关闭多个高亮
highlightBuilder.requireFieldMatch(false);
highlightBuilder.preTags("<span style='color:red'>");
highlightBuilder.postTags("</span>");
searchSourceBuilder.highlighter(highlightBuilder);
//执行搜索
searchRequest.source(searchSourceBuilder);
SearchResponse searchResponse = restHighLevelClient.search(searchRequest, RequestOptions.DEFAULT);
//解析结果
ArrayList<Map<String,Object>> list = new ArrayList<>();
for (SearchHit documentFields : searchResponse.getHits().getHits()) {
//解析我们的高亮字段
Map<String, HighlightField> highlightFields = documentFields.getHighlightFields();
HighlightField title = highlightFields.get("title");
//原来的结果
Map<String, Object> sourceAsMap = documentFields.getSourceAsMap();
if (documentFields != null)
{
Text[] fragments = title.fragments();
String n_title = "";
//将高亮替换原来的字段
for (Text text : fragments) {
n_title += text;
}
sourceAsMap.put("title",n_title);
}
list.add(sourceAsMap);
}
return list;
}
//精确匹配
// TermQueryBuilder termQueryBuilder = QueryBuilders.termQuery("title",keyword);
MatchPhraseQueryBuilder matchPhraseQueryBuilder = QueryBuilders.matchPhraseQuery("title", keyword); At
this point, when I first watched the video, I had some impressions, saying that the term does not segment words. The keyword field is not segmented, but I have loaded the ik tokenizer. The ik tokenizer will segment the word by default, so you can no longer use TermQueryBuilder termQueryBuilder = QueryBuilders.termQuery("title",keyword); this method should be changed Into MatchPhraseQueryBuilder matchPhraseQueryBuilder = QueryBuilders.matchPhraseQuery("title", keyword); The pattern of exact match! ! !
Learned, have a deeper understanding of the exchange! ! ! !