Elasticsearch Java Rest Client 上手指南(下)

High Level Rest Clent到现在还不是完成版。我试了一下,5.6版本的RestHighLevelClient就这么些API


包含了基本的增删改查和批量操作

我翻了一下官方文档,凉凉。确实像官方文档说的那样,需要完善。虽然是High Level的Client,但是东西少的可怜。
增(index)删(delete)改(update)查(get)操作都是和Index,type,id严格绑定的。
不能跨Index操作

目前几乎所有的High Level Rest Clent的中文介绍全部是照搬ES的文档啊。我懒得抄,而且我司用的Elasticsearch 5.6
API少的可怜

明显特性比版本6少了很多。所以,我倒是想填这个坑,但是太大了。还是拉倒吧。强烈建议直接去翻官方文档,这个API版本不同版本的差别很大,一定去看自己使用的版本!现有的中文博客参考价值有限。包括本篇。

0x1 基本增删改查

  • 第一步创建高级Client

    1
    2
    3
    4
    5
    RestClient restClient = RestClient
    .builder(new HttpHost("localhost", 9200, "http"))
    .build();
    RestHighLevelClient highLevelClient = new RestHighLevelClient(restClient);
  • 一次演示增删改查

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    //增, source 里对象创建方式可以是JSON字符串,或者Map,或者XContentBuilder 对象
    IndexRequest indexRequest = new IndexRequest("指定index", "指定type", "指定ID") .source(builder);
    highLevelClient.index(indexRequest);
    //删
    DeleteRequest deleteRequest = new DeleteRequest("指定index", "指定type", "指定ID");
    highLevelClient.delete(deleteRequest);
    //改, source 里对象创建方式可以是JSON字符串,或者Map,或者XContentBuilder 对象
    UpdateRequest updateRequest = new UpdateRequest("指定index", "指定type", "指定ID").doc(builder);
    highLevelClient.update(updateRequest);
    //查
    GetRequest getRequest = new GetRequest("指定index", "指定type", "指定ID");
    highLevelClient.get(getRequest);
  • 以上四个方法都有一个*Async的方法是异步回调的,只需添加ActionListener对象即可
  • Get查询不是唯一的查询方法,还有SearchRequest等, 但是这个GetRequest只支持单Index操作
  • Get操作支持限定查询的字段,传入fetchSourceContext对象即可
  • Update 操作演示的并不是全量替换,而是和现有文档作合并,除了doc操作还有使用Groovy script操作。
  • upsert类似update操作,不过如果文档不存在会作为新的doc存入ES

0x2 Bulk批量操作

其实就是把一大堆IndexRequest, UpdateRequest, DeleteRequest操作放在一起。
所以缺点就是必须指定Index,否则操作没戏。
简单示例

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
BulkRequest request = new BulkRequest();
request.add(new IndexRequest("指定index", "指定type", "指定ID_1").source(XContentType.JSON,"field", "foo"));
request.add(new DeleteRequest("指定index", "指定type", "指定ID_2"));
request.add(new UpdateRequest("指定index", "指定type", "指定ID_3") .doc(XContentType.JSON,"other", "test"));
BulkResponse bulkResponse = client.bulk(request);
for (BulkItemResponse bulkItemResponse : bulkResponse) {
if (bulkItemResponse.isFailed()) {
BulkItemResponse.Failure failure = bulkItemResponse.getFailure();
continue;
}
DocWriteResponse itemResponse = bulkItemResponse.getResponse();
if (bulkItemResponse.getOpType() == DocWriteRequest.OpType.INDEX
|| bulkItemResponse.getOpType() == DocWriteRequest.OpType.CREATE) {
IndexResponse indexResponse = (IndexResponse) itemResponse;
} else if (bulkItemResponse.getOpType() == DocWriteRequest.OpType.UPDATE) {
UpdateResponse updateResponse = (UpdateResponse) itemResponse;
} else if (bulkItemResponse.getOpType() == DocWriteRequest.OpType.DELETE) {
DeleteResponse deleteResponse = (DeleteResponse) itemResponse;
}
}

0x3 SearchRequest高级查询

支持多文档查询、聚合操作。可以完全取代GetRequest。

1
2
3
4
5
// 创建
SearchRequest searchRequest = new SearchRequest();
SearchSourceBuilder builder = new SearchSourceBuilder();
searchSourceBuilder.query(xxxQuery);
searchRequest.source(builder);

可以在创建的时候指定index,SearchRequest searchRequest = new SearchRequest("some_index*");,支持带*号的模糊匹配

当然,这并不是最厉害的地方,最NB的地方是,支持QueryBuilder,兼容之前TransportClient的代码

  • 我自己写的跨Index模糊查询

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    SearchRequest searchRequest = new SearchRequest("gdp_tops*");
    SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
    sourceBuilder.query(QueryBuilders.termQuery("city", "北京市"));
    sourceBuilder.timeout(new TimeValue(60, TimeUnit.SECONDS));
    searchRequest.source(sourceBuilder);
    try {
    SearchResponse response = highLevelClient.search(searchRequest);
    Arrays.stream(response.getHits().getHits())
    .forEach(i -> {
    System.out.println(i.getIndex());
    System.out.println(i.getSource());
    System.out.println(i.getType());
    });
    System.out.println(response.getHits().totalHits);
    } catch (IOException e) {
    e.printStackTrace();
    }
  • 官方给出的聚合查询

    1
    2
    3
    4
    5
    6
    SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
    TermsAggregationBuilder aggregation = AggregationBuilders.terms("by_company")
    .field("company.keyword");
    aggregation.subAggregation(AggregationBuilders.avg("average_age")
    .field("age"));
    searchSourceBuilder.aggregation(aggregation);
  • 当然还支持异步查询
    官方示例

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    client.searchAsync(searchRequest, new ActionListener<SearchResponse>() {
    @Override
    public void onResponse(SearchResponse searchResponse) {
    }
    @Override
    public void onFailure(Exception e) {
    }
    });
  • 查询结果处理
    查询结束后会得到一个SearchResponse对象,可以拿到查询状态,消耗时间,查询到的总条目数等参数,具体结果操作

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    SearchHit[] searchHits = hits.getHits();
    for (SearchHit hit : searchHits) {
    // 结果的Index
    String index = hit.getIndex();
    // 结果的type
    String type = hit.getType();
    // 结果的ID
    String id = hit.getId();
    // 结果的评分
    float score = hit.getScore();
    // 查询的结果 JSON字符串形式
    String sourceAsString = hit.getSourceAsString();
    // 查询的结果 Map的形式
    Map<String, Object> sourceAsMap = hit.getSourceAsMap();
    // Document的title
    String documentTitle = (String) sourceAsMap.get("title");
    // 结果中的某个List
    List<Object> users = (List<Object>) sourceAsMap.get("user");
    // 结果中的某个Map
    Map<String, Object> innerObject = (Map<String, Object>) sourceAsMap.get("innerObject");
    }
  • 聚合查询
    前面演示的是正常查询,聚合查询官方文档也有展示

    1
    2
    3
    4
    5
    6
    SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
    TermsAggregationBuilder aggregation = AggregationBuilders.terms("by_company")
    .field("company.keyword");
    aggregation.subAggregation(AggregationBuilders.avg("average_age")
    .field("age"));
    searchSourceBuilder.aggregation(aggregation);

和query查询一样,searchSourceBuilder使用aggregation()方法即可
查询到的结果处理也跟普通查询类似,处理一下Bucket就可以展示到接口了

1
2
3
4
5
Aggregations aggregations = searchResponse.getAggregations();
Terms byCompanyAggregation = aggregations.get("by_company");
Bucket elasticBucket = byCompanyAggregation.getBucketByKey("Elastic");
Avg averageAge = elasticBucket.getAggregations().get("average_age");
double avg = averageAge.getValue();

0x4 分页和滚动搜索

有时候结果需要分页查询,推荐使用searchSourceBuilder

1
2
sourceBuilder.from(0);
sourceBuilder.size(5);

有时候需要查询的数据太多,可以考虑使用SearchRequest.scroll()方法拿到scrollId;之后再使用SearchScrollRequest
其用法如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
SearchRequest searchRequest = new SearchRequest("posts");
SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder();
searchSourceBuilder.query(matchQuery("title", "Elasticsearch"));
searchSourceBuilder.size(size);
searchRequest.source(searchSourceBuilder);
searchRequest.scroll(TimeValue.timeValueMinutes(1L));
SearchResponse searchResponse = client.search(searchRequest);
String scrollId = searchResponse.getScrollId();
SearchScrollRequest scrollRequest = new SearchScrollRequest(scrollId);
scrollRequest.scroll(TimeValue.timeValueSeconds(30));
SearchResponse searchScrollResponse = client.searchScroll(scrollRequest);
scrollId = searchScrollResponse.getScrollId();
hits = searchScrollResponse.getHits();
assertEquals(3, hits.getTotalHits());
assertEquals(1, hits.getHits().length);
assertNotNull(scrollId);

Scroll查询的使用场景是密集且前后有关联的查询。如果只是一般的分页,可以使用size from来处理

官方还列出了查询构造的方式。可以根据自己的需要,详细翻阅。

需要了解基础的,请查看:Elasticsearch Java Rest Client 上手指南(上)

转载请注明出处:https://micorochio.github.io/2018/07/22/elasticsearch_rest_high_level_client/
如有错误,请不吝指正。谢谢