ElasticSearch AggregationBuilders java api經常使用聚會查詢

以球員信息爲例，player索引的player type包含5個字段，姓名，年齡，薪水，球隊，場上位置。
index的mapping爲：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

"mappings" : {      "player" : {          "properties" : {              "name" : {                  "index" :  "not_analyzed" ,                  "type" :  "string"              },              "age" : {                  "type" :  "integer"              },              "salary" : {                  "type" :  "integer"              },              "team" : {                  "index" :  "not_analyzed" ,                  "type" :  "string"              },              "position" : {                  "index" :  "not_analyzed" ,                  "type" :  "string"              }          },          "_all" : {              "enabled" :  false          }      } }

　　

索引中的所有數據：

 
首先，初始化Builder：

1	SearchRequestBuilder sbuilder = client.prepareSearch( "player" ).setTypes( "player" );

　　

接下來舉例說明各類聚合操做的實現方法，由於在es的api中，多字段上的聚合操做須要用到子聚合(subAggregation)，初學者可能找不到方法（網上資料比較少，筆者在這個問題上折騰了兩天，最後度了源碼才完全搞清楚T_T），後邊會特地說明多字段聚合的實現方法。另外，聚合後的排序也會單獨說明。

• group by/count

例如要計算每一個球隊的球員數，若是使用SQL語句，應表達以下：

select team, count(*) as player_count from player group by team;

ES的java api：

1 2 3

TermsBuilder teamAgg= AggregationBuilders.terms( "player_count " ).field( "team" ); sbuilder.addAggregation(teamAgg); SearchResponse response = sbuilder.execute().actionGet();

　　

 

• group by多個field

例如要計算每一個球隊每一個位置的球員數，若是使用SQL語句，應表達以下：

select team, position, count(*) as pos_count from player group by team, position;

ES的java api：

1 2 3 4

TermsBuilder teamAgg= AggregationBuilders.terms( "player_count " ).field( "team" ); TermsBuilder posAgg= AggregationBuilders.terms( "pos_count" ).field( "position" ); sbuilder.addAggregation(teamAgg.subAggregation(posAgg)); SearchResponse response = sbuilder.execute().actionGet();

　　

 

• max/min/sum/avg

例如要計算每一個球隊年齡最大/最小/總/平均的球員年齡，若是使用SQL語句，應表達以下：

select team, max(age) as max_age from player group by team;

ES的java api：

1 2 3 4

TermsBuilder teamAgg= AggregationBuilders.terms( "player_count " ).field( "team" ); MaxBuilder ageAgg= AggregationBuilders.max( "max_age" ).field( "age" ); sbuilder.addAggregation(teamAgg.subAggregation(ageAgg)); SearchResponse response = sbuilder.execute().actionGet();

　　

 

• 對多個field求max/min/sum/avg

例如要計算每一個球隊球員的平均年齡，同時又要計算總年薪，若是使用SQL語句，應表達以下：

select team, avg(age)as avg_age, sum(salary) as total_salary from player group by team;

ES的java api：

1 2 3 4 5 6

TermsBuilder teamAgg= AggregationBuilders.terms( "team" ); AvgBuilder ageAgg= AggregationBuilders.avg( "avg_age" ).field( "age" ); SumBuilder salaryAgg= AggregationBuilders.avg( "total_salary " ).field( "salary" ); sbuilder.addAggregation(teamAgg.subAggregation(ageAgg).subAggregation(salaryAgg)); SearchResponse response = sbuilder.execute().actionGet();

　　

• 聚合後對Aggregation結果排序

例如要計算每一個球隊總年薪，並按照總年薪倒序排列，若是使用SQL語句，應表達以下：

select team, sum(salary) as total_salary from player group by team order by total_salary desc;

ES的java api：

1 2 3 4

TermsBuilder teamAgg= AggregationBuilders.terms( "team" ).order(Order.aggregation( "total_salary " ,  false ); SumBuilder salaryAgg= AggregationBuilders.avg( "total_salary " ).field( "salary" ); sbuilder.addAggregation(teamAgg.subAggregation(salaryAgg)); SearchResponse response = sbuilder.execute().actionGet();

　　

須要特別注意的是，排序是在TermAggregation處執行的，Order.aggregation函數的第一個參數是aggregation的名字，第二個參數是boolean型，true表示正序，false表示倒序。 

• Aggregation結果條數的問題

默認狀況下，search執行後，僅返回10條聚合結果，若是想反悔更多的結果，須要在構建TermsBuilder 時指定size：

TermsBuilder teamAgg= AggregationBuilders.terms("team").size(15);

 

• Aggregation結果的解析/輸出

獲得response後：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Map<String, Aggregation> aggMap = response.getAggregations().asMap(); StringTerms teamAgg= (StringTerms) aggMap.get( "keywordAgg" ); Iterator<Bucket> teamBucketIt = teamAgg.getBuckets().iterator(); while  (teamBucketIt .hasNext()) { Bucket buck = teamBucketIt .next(); //球隊名 String team = buck.getKey(); //記錄數 long  count = buck.getDocCount(); //獲得全部子聚合 Map subaggmap = buck.getAggregations().asMap(); //avg值獲取方法 double  avg_age= ((InternalAvg) subaggmap.get( "avg_age" )).getValue(); //sum值獲取方法 double  total_salary = ((InternalSum) subaggmap.get( "total_salary" )).getValue(); //... //max/min以此類推 }

　　

 

• 總結

綜上，聚合操做主要是調用了SearchRequestBuilder的addAggregation方法，一般是傳入一個TermsBuilder，子聚合調用TermsBuilder的subAggregation方法，能夠添加的子聚合有TermsBuilder、SumBuilder、AvgBuilder、MaxBuilder、MinBuilder等常見的聚合操做。 從實現上來說，SearchRequestBuilder在內部保持了一個私有的 SearchSourceBuilder實例， SearchSourceBuilder內部包含一個List<AbstractAggregationBuilder>，每次調用addAggregation時會調用 SearchSourceBuilder實例，添加一個AggregationBuilder。一樣的，TermsBuilder也在內部保持了一個List<AbstractAggregationBuilder>，調用addAggregation方法（來自父類addAggregation）時會添加一個AggregationBuilder。有興趣的讀者也能夠閱讀源碼的實現。