1. elasticsearch 命令的基本格式
RESTful接口URL的格式:
http://localhost:9200/<index>/<type>/[<id>]
其中index、type是必須提供的。id是可選的,不提供es會自動生成。index、type將信息進行分層,利於管理。index可以理解為數據庫;type理解為數據表;id相當於數據庫表中記錄的主鍵,是唯一的。
注:在url網址后面加"?pretty",會讓返回結果以工整的方式展示出來,適用所有操作數據類的url。"?"表示引出條件,"pretty"是條件內容。
2. elasticsearch基本的增刪改
2.1 elasticSearch增加
向store索引中添加一些書籍
curl -H "Content-Type: application/json" -XPUT 'http://192.168.187.201:9200/store/books/1?pretty' -d '{ "title": "Elasticsearch: The Definitive Guide", "name" : { "first" : "Zachary", "last" : "Tong" }, "publish_date":"2015-02-06", "price":"49.99" }'
注:curl是linux下的http請求,-H "Content-Type: application/json"需要添加,否則會報錯{"error":"Content-Type header [application/x-www-form-urlencoded] is not supported","status":406}
加"pretty"
不加"pretty"
加"pretty"與不加"pretty"的區別就是返回結果工整與不工整的差別,其他操作類似。為了使返回結果工整,以下操作都在url后添加"pretty"
2.2 elasticSearch刪除
刪除一個文檔
curl -XDELETE 'http://hadoop1:9200/store/books/1?pretty'
2.3 elasticSearch更新(改)
①可以通過覆蓋的方式更新
curl -H "Content-Type:application/json" -XPUT 'http://hadoop1:9200/store/books/1?pretty' -d '{ "title": "Elasticsearch: The Definitive Guide", "name" : { "first" : "Zachary", "last" : "Tong" }, "publish_date":"2016-02-06", "price":"99.99" }'
② 通過_update API的方式單獨更新你想要更新的
curl -H "Content-Type: application/json" -XPOST 'http://hadoop1:9200/store/books/1/_update?pretty' -d '{ "doc": { "price" : 88.88 } }'
3. elasticSearch查詢
elasticSearch查詢分三種,一是瀏覽器查詢,二是curl查詢,三是請求體查詢GET或POS。
注:采用_search的模糊查詢(包括bool過濾查詢、 嵌套查詢、range范圍過濾查詢等等),url可以不必指定type,只用指定index查詢就行,具體例子看"2.1.4 elasticSearch查詢 ③query基本匹配查詢"節點的具體查詢實例
3.1 瀏覽器查詢
通過瀏覽器IP+網址查詢
http://hadoop1:9200/store/books/1?pretty
3.2 在linux通過curl的方式查詢
3.2.1 通過ID獲得文檔信息
curl -XGET 'http://hadoop1:9200/store/books/1?pretty'
3.2.2 通過_source獲取指定的字段
curl -XGET 'http://hadoop1:9200/store/books/1?_source=title&pretty' curl -XGET 'http://hadoop1:9200/store/books/1?_source=title,price&pretty' curl -XGET 'http://hadoop1:9200/store/books/1?_source&pretty'
3.2.3 query基本匹配查詢
查詢數據前,可以批量導入1000條數據集到elasticsearch里,具體參考"4 elasticSearch批處理命令 4.1 導入數據集"節點,以便數據查詢方便。
① "q=*"表示匹配索引中所有的數據,一般默認只返回前10條數據。
curl 'hadoop1:9200/bank/_search?q=*&pretty' #等價於: curl -H "Content-Type:applicatin/json" -XPOST 'localhost:9200/bank/_search?pretty' -d ' { "query": { "match_all": {} } }'
② 匹配所有數據,但只返回1個
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match_all": {}}, "size": 1 }'
注:如果size不指定,則默認返回10條數據。
③ 返回從11到20的數據(索引下標從0開始)
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {“match_all”: {}}, "from": 10, "size": 10 }
④ 匹配所有的索引中的數據,按照balance字段降序排序,並且返回前10條(如果不指定size,默認最多返回10條)
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match_all": {}}, "sort": {"balance":{"order": "desc"}} }'
⑤ 返回特定的字段(account_number balance) ,與②通過_source獲取指定的字段類似
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match_all": {}}, "_source": ["account_number", "balance"] }'
⑥ 返回account_humber為20的數據
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match": {"account_number":20}} }'
⑦ 返回address中包含mill的所有數據
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match":{"address": "mill"}} }'
⑧ 返回地址中包含mill或者lane的所有數據
curl -H "Content_Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ “query": {"match": {"address": "mill lane"}} }'
⑨ 與第8不同,多匹配(match_phrase是短語匹配),返回地址中包含短語"mill lane"的所有數據
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": {"match_phrase": {"address": "mill lane"}} }'
3.2.4 bool過濾查詢,可以做組合過濾查詢、嵌套查詢等
SELECT * FROM books WHERE (price = 35.99 OR price = 99.99) AND (publish_date != "2016-02-06")
類似的,Elasticsearch也有 and, or, not這樣的組合條件的查詢方式,格式如下:
{ ”bool“ : { "filter": [], "must" : [], "should": [], "must_not": [] } }
說明:
filter:過濾
must:條件必須滿足,相當於and
should:條件可以滿足也可以不滿足,相當於or
must_not:條件不需要滿足,相當於not
3.2.4.1 filter查詢
①filter指定單個值
# SELECT * FROM books WHERE price = 35.99 # filtered 查詢價格是35.99的 curl -H "Content-Type:application/json" -XGET 'http://hadoop1:9200/store/books/_search?pretty' -d '{ "query" : { "bool" : { "must" : { "match_all" : {} }, "filter" : { "term" : { "price" : 35.99 } } } } }'
注:帶有key-value鍵值對的都需要加 -H “Content-Type: application/json”
②filter指定多個值
curl -XGET 'http://hadoop1:9200/store/books/_search?pretty' -d '{ "query" : { "bool" : { "filter" : { "terms" : { "price" : [35.99, 99.99] } } } } }'
3.2.4.2 must、should、must_not查詢
①must、should、must_not與term結合使用:
curl -H "Content-Type:application/json" -XGET 'http://hadoop1:9200/store/books/_search?pretty' -d '{ "query" : { "bool" : { "should" : [ { "term" : {"price" : 35.99}}, { "term" : {"price" : 99.99}} ], "must_not" : { "term" : {"publish_date" : "2016-06-06"} } } } }'
②must、should、must_not與match結合使用
bool表示查詢列表中只要有任何一個為真則認為匹配:
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": { "bool": { "must_not": [ {"match": {"address": "mill"}}, {"match": {"address": "lane"}} ] } } }'
返回age年齡大於40歲、state不是ID的所有數據:
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": { "bool": { "must": [ {"match": {"age": "40"}} ], "must_not": [ {"match": {"state": "ID"}} ] } } }'
3.2.4.3 bool嵌套查詢
# 嵌套查詢 # SELECT * FROM books WHERE price = 35.99 OR ( publish_date = "2016-02-06" AND price = 99.99 ) curl -H "Content-Type:application/json" -XGET 'http://hadoop1:9200/store/books/_search?pretty' -d '{ "query" : { "bool" : { "should" : [ { "term" : {"price" : 35.99 }}, { "bool" : { "must" : [ { "term" : {"publish_date" : "2016-06-06"}}, { "term" : {"price" : 99.99}} ] } } ] } } }'
3.2.4.4 filter的range范圍過濾查詢
第一個示例,查找price價錢大於20的數據:
# SELECT * FROM books WHERE price >= 20 AND price < 100 # gt : > 大於 # lt : < 小於 # gte : >= 大於等於 # lte : <= 小於等於 curl -H "Content-Type:application/json" -XGET 'http://hadoop1:9200/store/books/_search?pretty' -d '{ "query" : { "bool" : { "filter" : { "range" : { "price" : { "gt" : 20.0, "boost" : 4.0 } } } } } }'
注:boost:設置boost查詢的值,默認1.0
第二個示例,使用布爾查詢返回balance在20000到30000之間的所有數據:
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "query": { "bool": { "must": {"match_all": {}}, "filter": { "range": { "balance": { "gte": 20000, "lte": 30000 } } } } } }'
3.2.4 elasticSearch聚合查詢
第一個示例,將所有的數據按照state分組(group),然后按照分組記錄數從大到小排序(默認降序),返回前十條(默認)
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "size": 0, "aggs": { "group_by_state": { "terms": { "field": "state" } } } }'
可能遇到的問題:elasticsearch 進行排序的時候,我們一般都會排序數字、日期,而文本排序則會報錯:Fielddata is disabled on text fields by default. Set fielddata=true on [state] in order to load fielddata in memory by uninverting the inverted index. Note that this can however use significant memory. Alternatively use a keyword field instead.如:
解決方案:5.x后對排序,聚合這些操作,用單獨的數據結構(fielddata)緩存到內存里了,需要單獨開啟,官方解釋在此fielddata。聚合前執行如下操作,用以開啟fielddata:
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_mapping/account?pretty' -d '{ "properties": { "state": { "type": "text", "fielddata": true } } }'
說明:bank為index,_mapping為映射,account為type,這三個要素為必須,”state“為聚合"group_by_state"操作的對象字段
聚合查詢成功示例:
第二個示例,將所有的數據按照state分組(group),降序排序,計算每組balance的平均值並返回(默認)
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/_search?pretty' -d '{ "size": 0, "aggs": { "group_by_state": { "terms": { "field": "state" }, "aggs": { "average_balance": { "avg": { "field":"balance" } } } } } }'
4. elasticSearch批處理命令
4.1 導入數據集
你可以點擊這里下載示例數據集:accounts.json
導入示例數據集:
curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/account/_bulk?pretty' --data-binary "@accounts.json" curl -H "Content-Type:application/json" -XPOST 'hadoop1:9200/bank/account/_bulk?pretty' --data-binary "@/home/hadoop/accounts.json"
注:_bulk表示批量處理,"@accounts.json"或者"@/home/hadoop/accounts.json"可以用相對路徑或者絕對路徑表示。
查看accounts.json導入情況,使用 curl 'hadoop1:9200/_cat/indices?v'
可以看到已經成功導入1000條數據記錄
4.2 批量創建索引
5. elasticSearch其他常用命令
注:url后面的"v"表示 verbose 的意思,這樣可以更可讀(有表頭,有對齊),如果不加v,不會顯示表頭
5.1. 查看所有index索引,輸入命令 curl 'hadoop1:9200/_cat/indices?v'
說明:index:索引為store,pri:5個私有的分片,rep:1個副本,docs.count:store索引里面有2個文檔(即里面有2條數據記錄),docs.deleted:刪除了0條記錄,store.size:數據存儲總大小(包括副本),pri.store.size:分片數據存儲的大小。
不加v,不會顯示表頭,可讀性差
5.2. 檢測集群是否健康,確保9200端口號可用 curl 'hadoop1:9200/_cat/health?v'
5.3. 獲取集群的節點列表 curl 'hadoop1:9200/_cat/nodes?v'
參考資料:
ElasticSearch權威指南(中文版):https://es.xiaoleilu.com/010_Intro/25_Tutorial_Indexing.html
https://www.cnblogs.com/pilihaotian/p/5830754.html