Elasticsearch Date类型使用技巧
elasticsearch原生支持date类型。这里简单记录下使用的方法。
使用date类型可以用如下两种方式：
-
使用毫秒的时间戳，直接将毫秒值传入即可。
-
传入格式化的字符串，默认是ISO 8601标准，例如
2015-02-27T00:07Z(零时区)、2015-02-27T08:07+08:00(东八区),这两个时间实际是同一个，只是时区不同。另外还可以自定义时间格式，参见es的文档。但个人不建议使用自定义格式，设置不当容易遇到时区问题。在php中获取ISO 8601标准的时间很简单，date('c',time()）即可。
elasticsearch默认会自动识别date类型，如果想关闭该功能，修改mapping的设置'date_detection' => false即可 。
贴下我的代码：
package com.osp.log.service.impl;
import java.text.DecimalFormat;
import java.util.HashMap;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import javax.ws.rs.core.Response;
import org.elasticsearch.action.search.SearchResponse;
import org.elasticsearch.action.search.SearchType;
import org.elasticsearch.client.transport.TransportClient;
import org.elasticsearch.common.text.Text;
import org.elasticsearch.common.xcontent.XContentType;
import org.elasticsearch.index.query.BoolQueryBuilder;
import org.elasticsearch.index.query.QueryBuilders;
import org.elasticsearch.search.SearchHit;
import org.elasticsearch.search.SearchHits;
import org.elasticsearch.search.aggregations.AggregationBuilders;
import org.elasticsearch.search.aggregations.bucket.terms.Terms;
import org.elasticsearch.search.aggregations.bucket.terms.Terms.Bucket;
import org.elasticsearch.search.aggregations.bucket.terms.TermsAggregationBuilder;
import org.elasticsearch.search.aggregations.metrics.avg.AvgAggregationBuilder;
import org.elasticsearch.search.aggregations.metrics.avg.InternalAvg;
import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder;
import org.elasticsearch.search.sort.SortOrder;
import org.springframework.stereotype.Service;
import com.osp.common.json.JsonUtil;
import com.osp.log.config.ESConfig;
import com.osp.log.model.SearchModel;
import com.osp.log.model.SearchResultBean;
import com.osp.log.model.TomcatModel;
import com.osp.log.service.SearchService;
import com.osp.log.util.ESUtil;
import com.osp.log.util.TimeUtils;
/**
* 关键字搜索
*
* @author zhangmingcheng 2017-09-26
*/
@Service
public class SearchServiceImpl implements SearchService {
public static final String INDEX_NAME = "logstash-apacheaccesslog*"; // 索引名称
public static final String INDEX_TYPE = "logs"; // 索引类型
/**
* 关键字搜索
*
* @param q
* 关键词
* @param page
* 页码
* @param pageSize
* 页数大小
* @param ip
* 请求客户端ip
* @param city
* 请求客户端所在城市
* @return
*/
@Override
public SearchResultBean getSearchesult(String keyword, String ip, String city, Integer page, Integer pagesize) {
TransportClient client = ESUtil.getClient();
BoolQueryBuilder boolQueryBuilder = new BoolQueryBuilder();
if (keyword.isEmpty()) {
boolQueryBuilder.must(QueryBuilders.matchAllQuery());
} else {
boolQueryBuilder.must(QueryBuilders.multiMatchQuery(keyword, "request", "verb", "clientip", "message",
"path", "response", "host", "timestamp"));
}
/**
* 高亮设置
*/
HighlightBuilder hiBuilder = new HighlightBuilder();
hiBuilder.preTags("<span class=\'pointKey\'>");
hiBuilder.postTags("</span>");
hiBuilder.field("request", 50);
hiBuilder.field("message", 30);
/**
* 开始搜索
*/
SearchResponse response = client.prepareSearch(SearchServiceImpl.INDEX_NAME)
.setTypes(SearchServiceImpl.INDEX_TYPE).setSearchType(SearchType.DEFAULT).setQuery(boolQueryBuilder)
.setFrom(pagesize * (page - 1)).setSize(pagesize).highlighter(hiBuilder).setExplain(true) // 设置是否按查询匹配度排序
.get();
SearchHits myhits = response.getHits();
LinkedList<TomcatModel> newsList = new LinkedList<>();
for (SearchHit hit : myhits) {
Map<String, Object> map = hit.getSource();
TomcatModel tomcatModel = new TomcatModel();
String request = this.getHighlightFieldString(hit, "request");
String message = this.getHighlightFieldString(hit, "message");
if (request.isEmpty()) {
tomcatModel.setRequest((String) map.get("request"));
} else {
tomcatModel.setRequest(request);
}
if (message.isEmpty()) {
tomcatModel.setMessage((String) map.get("message"));
} else {
tomcatModel.setMessage(message);
}
tomcatModel.setPath((String) map.get("path"));
tomcatModel.setClientip((String) map.get("clientip"));
tomcatModel.setResponse((String) map.get("response"));
tomcatModel.setType((String) map.get("verb"));
tomcatModel.setTimestamp((String) map.get("timestamp"));
newsList.add(tomcatModel);
}
/**
* 记录本次客户端查询：翻页不存储
*/
String usetime = response.getTook().toString();
if (page == 1) {
String json = JsonUtil.beanToJson(new SearchModel(keyword, TimeUtils.getCurrentTime(), ip, city,
Integer.parseInt(usetime.substring(0, usetime.length() - 2)), System.currentTimeMillis()));
client.prepareIndex(ESConfig.SEARCHINDEX, ESConfig.SEARCHTYPE).setSource(json, XContentType.JSON).get();
}
/**
* 开始存储结果
*/
SearchResultBean searchResult = new SearchResultBean();
searchResult.setPage(page);
searchResult.setPagesize(pagesize);
searchResult.setTotal(myhits.getTotalHits());
searchResult.setUsetime(usetime);
searchResult.setNewsList(newsList);
return searchResult;
}
/**
* 获取搜索统计数据
*/
@Override
public Response getcount() {
HashMap<String, Object> RealReult = new HashMap<>();
TransportClient client = getClient();
TermsAggregationBuilder agg_clientip = AggregationBuilders.terms("cilentip_count").field("clientip");
AvgAggregationBuilder avg_usetime = AggregationBuilders.avg("avg_usetime").field("usetime");
TermsAggregationBuilder agg_message = AggregationBuilders.terms("agg_message").field("message");
SearchResponse response = client.prepareSearch(ESConfig.SEARCHINDEX).setTypes(ESConfig.SEARCHTYPE)
.addAggregation(agg_clientip).addAggregation(avg_usetime).addAggregation(agg_message)
.setQuery(QueryBuilders.matchAllQuery()).execute().actionGet();
/**
* 搜索次数
*/
RealReult.put("total", response.getHits().getTotalHits());
/**
* 统计访问人数
*/
Terms terms = response.getAggregations().get("cilentip_count");
List<Bucket> buckets = terms.getBuckets();
RealReult.put("people", buckets.size());
/**
* 统计平均搜索用时
*/
InternalAvg avg = response.getAggregations().get("avg_usetime");
DecimalFormat df = new DecimalFormat("#.00");
RealReult.put("avgUsetime", df.format(avg.getValue()).trim() + "ms");
/**
* 统计最多搜索的10个词(聚合默认返回10个值)
*/
Terms messageTerms = response.getAggregations().get("agg_message");
List<Bucket> messageBuckets = messageTerms.getBuckets();
LinkedList<TopWord> result = new LinkedList<TopWord>();
for (Bucket bucket : messageBuckets) {
TopWord topWord = new TopWord();
topWord.set_id((String) bucket.getKey());
topWord.setCount((int) bucket.getDocCount());
result.add(topWord);
}
RealReult.put("Top", result);
return Response.status(200).entity(RealReult).build();
}
/**
* 获取历史搜索记录
*/
@Override
public String getHistoryList(Integer page, Integer pagesize) {
LinkedList<Map<String, Object>> result = new LinkedList<>();
TransportClient client = getClient();
SearchResponse response = client.prepareSearch(ESConfig.SEARCHINDEX).setTypes(ESConfig.SEARCHTYPE)
.setQuery(QueryBuilders.matchAllQuery()).addSort("createDate", SortOrder.DESC)
.setFrom(pagesize * (page - 1)).setSize(pagesize).get();
SearchHits myhits = response.getHits();
for (SearchHit hit : myhits) {
Map<String, Object> hitmap = hit.getSource();
HashMap<String, Object> map = new HashMap<>();
map.put("q", hitmap.get("message"));
map.put("usetime", hitmap.get("usetime"));
map.put("city", hitmap.get("city"));
map.put("ip", hitmap.get("clientip"));
map.put("total", myhits.getTotalHits());
map.put("time", hitmap.get("timestamp"));
result.add(map);
}
HashMap<String, Object> RealReult = new HashMap<>();
RealReult.put("rows", result);
RealReult.put("total", myhits.getTotalHits());
return JsonUtil.beanToJson(RealReult);
}
public String getHighlightFieldString(SearchHit hit, String field) {
String content = "";
if (hit.getHighlightFields().containsKey(field)) {
Text[] text = hit.getHighlightFields().get(field).getFragments();
for (Text str : text) {
content = content + str;
}
}
return content;
}
public TransportClient getClient() {
return ESUtil.getClient();
}
class TopWord {
private String _id;
private int count;
public String get_id() {
return _id;
}
public void set_id(String _id) {
this._id = _id;
}
public int getCount() {
return count;
}
public void setCount(int count) {
this.count = count;
}
}
}
