use of org.graylog.shaded.elasticsearch7.org.elasticsearch.search.aggregations.Aggregations in project elasticsearch by elastic.
the class SignificantTermsSignificanceScoreIT method testBackgroundVsSeparateSet.
// compute significance score by
// 1. terms agg on class and significant terms
// 2. filter buckets and set the background to the other class and set is_background false
// both should yield exact same result
public void testBackgroundVsSeparateSet(SignificanceHeuristic significanceHeuristicExpectingSuperset, SignificanceHeuristic significanceHeuristicExpectingSeparateSets) throws Exception {
SearchResponse response1 = client().prepareSearch(INDEX_NAME).setTypes(DOC_TYPE).addAggregation(terms("class").field(CLASS_FIELD).subAggregation(significantTerms("sig_terms").field(TEXT_FIELD).minDocCount(1).significanceHeuristic(significanceHeuristicExpectingSuperset))).execute().actionGet();
assertSearchResponse(response1);
SearchResponse response2 = client().prepareSearch(INDEX_NAME).setTypes(DOC_TYPE).addAggregation(filter("0", QueryBuilders.termQuery(CLASS_FIELD, "0")).subAggregation(significantTerms("sig_terms").field(TEXT_FIELD).minDocCount(1).backgroundFilter(QueryBuilders.termQuery(CLASS_FIELD, "1")).significanceHeuristic(significanceHeuristicExpectingSeparateSets))).addAggregation(filter("1", QueryBuilders.termQuery(CLASS_FIELD, "1")).subAggregation(significantTerms("sig_terms").field(TEXT_FIELD).minDocCount(1).backgroundFilter(QueryBuilders.termQuery(CLASS_FIELD, "0")).significanceHeuristic(significanceHeuristicExpectingSeparateSets))).execute().actionGet();
StringTerms classes = response1.getAggregations().get("class");
SignificantTerms sigTerms0 = ((SignificantTerms) (classes.getBucketByKey("0").getAggregations().asMap().get("sig_terms")));
assertThat(sigTerms0.getBuckets().size(), equalTo(2));
double score00Background = sigTerms0.getBucketByKey("0").getSignificanceScore();
double score01Background = sigTerms0.getBucketByKey("1").getSignificanceScore();
SignificantTerms sigTerms1 = ((SignificantTerms) (classes.getBucketByKey("1").getAggregations().asMap().get("sig_terms")));
double score10Background = sigTerms1.getBucketByKey("0").getSignificanceScore();
double score11Background = sigTerms1.getBucketByKey("1").getSignificanceScore();
Aggregations aggs = response2.getAggregations();
sigTerms0 = (SignificantTerms) ((InternalFilter) aggs.get("0")).getAggregations().getAsMap().get("sig_terms");
double score00SeparateSets = sigTerms0.getBucketByKey("0").getSignificanceScore();
double score01SeparateSets = sigTerms0.getBucketByKey("1").getSignificanceScore();
sigTerms1 = (SignificantTerms) ((InternalFilter) aggs.get("1")).getAggregations().getAsMap().get("sig_terms");
double score10SeparateSets = sigTerms1.getBucketByKey("0").getSignificanceScore();
double score11SeparateSets = sigTerms1.getBucketByKey("1").getSignificanceScore();
assertThat(score00Background, equalTo(score00SeparateSets));
assertThat(score01Background, equalTo(score01SeparateSets));
assertThat(score10Background, equalTo(score10SeparateSets));
assertThat(score11Background, equalTo(score11SeparateSets));
}
use of org.graylog.shaded.elasticsearch7.org.elasticsearch.search.aggregations.Aggregations in project elasticsearch by elastic.
the class SignificantTermsSignificanceScoreIT method testScoresEqualForPositiveAndNegative.
public void testScoresEqualForPositiveAndNegative(SignificanceHeuristic heuristic) throws Exception {
//check that results for both classes are the same with exclude negatives = false and classes are routing ids
SearchResponse response = client().prepareSearch("test").addAggregation(terms("class").field("class").subAggregation(significantTerms("mySignificantTerms").field("text").executionHint(randomExecutionHint()).significanceHeuristic(heuristic).minDocCount(1).shardSize(1000).size(1000))).execute().actionGet();
assertSearchResponse(response);
StringTerms classes = response.getAggregations().get("class");
assertThat(classes.getBuckets().size(), equalTo(2));
Iterator<Terms.Bucket> classBuckets = classes.getBuckets().iterator();
Aggregations aggregations = classBuckets.next().getAggregations();
SignificantTerms sigTerms = aggregations.get("mySignificantTerms");
Collection<SignificantTerms.Bucket> classA = sigTerms.getBuckets();
Iterator<SignificantTerms.Bucket> classBBucketIterator = sigTerms.getBuckets().iterator();
assertThat(classA.size(), greaterThan(0));
for (SignificantTerms.Bucket classABucket : classA) {
SignificantTerms.Bucket classBBucket = classBBucketIterator.next();
assertThat(classABucket.getKey(), equalTo(classBBucket.getKey()));
assertThat(classABucket.getSignificanceScore(), closeTo(classBBucket.getSignificanceScore(), 1.e-5));
}
}
use of org.graylog.shaded.elasticsearch7.org.elasticsearch.search.aggregations.Aggregations in project elasticsearch by elastic.
the class ScriptedMetricIT method testInitMapCombineReduceWithParamsAsSubAgg.
public void testInitMapCombineReduceWithParamsAsSubAgg() {
Map<String, Object> varsMap = new HashMap<>();
varsMap.put("multiplier", 1);
Map<String, Object> params = new HashMap<>();
params.put("_agg", new ArrayList<>());
params.put("vars", varsMap);
Script initScript = new Script(ScriptType.INLINE, CustomScriptPlugin.NAME, "vars.multiplier = 3", Collections.emptyMap());
Script mapScript = new Script(ScriptType.INLINE, CustomScriptPlugin.NAME, "_agg.add(vars.multiplier)", Collections.emptyMap());
Script combineScript = new Script(ScriptType.INLINE, CustomScriptPlugin.NAME, "sum agg values as a new aggregation", Collections.emptyMap());
Script reduceScript = new Script(ScriptType.INLINE, CustomScriptPlugin.NAME, "sum aggs of agg values as a new aggregation", Collections.emptyMap());
SearchResponse response = client().prepareSearch("idx").setQuery(matchAllQuery()).setSize(1000).addAggregation(histogram("histo").field("l_value").interval(1).subAggregation(scriptedMetric("scripted").params(params).initScript(initScript).mapScript(mapScript).combineScript(combineScript).reduceScript(reduceScript))).get();
assertSearchResponse(response);
assertThat(response.getHits().getTotalHits(), equalTo(numDocs));
Aggregation aggregation = response.getAggregations().get("histo");
assertThat(aggregation, notNullValue());
assertThat(aggregation, instanceOf(Histogram.class));
Histogram histoAgg = (Histogram) aggregation;
assertThat(histoAgg.getName(), equalTo("histo"));
List<? extends Bucket> buckets = histoAgg.getBuckets();
assertThat(buckets, notNullValue());
for (Bucket b : buckets) {
assertThat(b, notNullValue());
assertThat(b.getDocCount(), equalTo(1L));
Aggregations subAggs = b.getAggregations();
assertThat(subAggs, notNullValue());
assertThat(subAggs.asList().size(), equalTo(1));
Aggregation subAgg = subAggs.get("scripted");
assertThat(subAgg, notNullValue());
assertThat(subAgg, instanceOf(ScriptedMetric.class));
ScriptedMetric scriptedMetricAggregation = (ScriptedMetric) subAgg;
assertThat(scriptedMetricAggregation.getName(), equalTo("scripted"));
assertThat(scriptedMetricAggregation.aggregation(), notNullValue());
assertThat(scriptedMetricAggregation.aggregation(), instanceOf(ArrayList.class));
List<?> aggregationList = (List<?>) scriptedMetricAggregation.aggregation();
assertThat(aggregationList.size(), equalTo(1));
Object object = aggregationList.get(0);
assertThat(object, notNullValue());
assertThat(object, instanceOf(Number.class));
assertThat(((Number) object).longValue(), equalTo(3L));
}
}
use of org.graylog.shaded.elasticsearch7.org.elasticsearch.search.aggregations.Aggregations in project pancm_project by xuwujing.
the class EsAggregationSearchTest method havingSearch.
/**
* @Author pancm
* @Description having
* @Date 2020/8/21
* @Param []
* @return void
*/
private static void havingSearch() throws IOException {
String index = "";
SearchRequest searchRequest = new SearchRequest(index);
searchRequest.indices(index);
SearchSourceBuilder sourceBuilder = new SearchSourceBuilder();
BoolQueryBuilder boolQueryBuilder = new BoolQueryBuilder();
searchRequest.indicesOptions(IndicesOptions.lenientExpandOpen());
String alias_name = "nas_ip_address_group";
String group_name = "nas_ip_address";
String query_name = "acct_start_time";
String query_type = "gte,lte";
String query_name_value = "2020-08-05 13:25:55,2020-08-20 13:26:55";
String[] query_types = query_type.split(",");
String[] query_name_values = query_name_value.split(",");
for (int i = 0; i < query_types.length; i++) {
if ("gte".equals(query_types[i])) {
boolQueryBuilder.must(QueryBuilders.rangeQuery(query_name).gte(query_name_values[i]));
}
if ("lte".equals(query_types[i])) {
boolQueryBuilder.must(QueryBuilders.rangeQuery(query_name).lte(query_name_values[i]));
}
}
AggregationBuilder aggregationBuilder = AggregationBuilders.terms(alias_name).field(group_name).size(Integer.MAX_VALUE);
// 声明BucketPath,用于后面的bucket筛选
Map<String, String> bucketsPathsMap = new HashMap<>(8);
bucketsPathsMap.put("groupCount", "_count");
// 设置脚本
Script script = new Script("params.groupCount >= 1000");
// 构建bucket选择器
BucketSelectorPipelineAggregationBuilder bs = PipelineAggregatorBuilders.bucketSelector("having", bucketsPathsMap, script);
aggregationBuilder.subAggregation(bs);
sourceBuilder.aggregation(aggregationBuilder);
// 不需要解释
sourceBuilder.explain(false);
// 不需要原始数据
sourceBuilder.fetchSource(false);
// 不需要版本号
sourceBuilder.version(false);
sourceBuilder.query(boolQueryBuilder);
searchRequest.source(sourceBuilder);
System.out.println(sourceBuilder);
// 同步查询
SearchResponse searchResponse = client.search(searchRequest, RequestOptions.DEFAULT);
// 查询条数
long count = searchResponse.getHits().getHits().length;
Aggregations aggregations = searchResponse.getAggregations();
// agg(aggregations);
Map<String, Object> map = new HashMap<>();
List<Map<String, Object>> list = new ArrayList<>();
agg(list, aggregations);
// System.out.println(map);
System.out.println(list);
}
use of org.graylog.shaded.elasticsearch7.org.elasticsearch.search.aggregations.Aggregations in project pancm_project by xuwujing.
the class EsAggregationSearchTest2 method test.
private static void test() throws IOException {
// TermsAggregationBuilder termsBuilder = AggregationBuilders.terms("ecid").field("ecid").size(99999);
// List<FieldSortBuilder> fieldSorts=new ArrayList<>();
// fieldSorts.add(new FieldSortBuilder("sum_field").order(SortOrder.DESC));
// termsBuilder.subAggregation(new BucketSortPipelineAggregationBuilder("bucket_field", fieldSorts).from(6000).size(10));
//
AggregationBuilder aggregation = AggregationBuilders.terms("ecid").field("ecid");
List<FieldSortBuilder> fieldSorts = new ArrayList<>();
fieldSorts.add(new FieldSortBuilder("id_").order(SortOrder.DESC));
aggregation.subAggregation(new BucketSortPipelineAggregationBuilder("id", fieldSorts).from(0).size(10));
SearchResponse searchResponse = search(aggregation);
// 获取聚合结果
Aggregations aggregations = searchResponse.getAggregations();
Map<String, Object> map = new HashMap<>();
List<Map<String, Object>> list = new ArrayList<>();
agg(map, list, aggregations);
logger.info("聚合查询结果:" + list);
logger.info("------------------------------------");
}
Aggregations