![\"elasticsearch-nest\"](\"\/wp-content\/uploads\/2016\/07\/elasticsearch-nest.png\")
<\/a><\/p>\nBir \u00f6nceki seriden hat\u0131rlarsak olu\u015fturmu\u015f oldu\u011fumuz index i\u00e7erisine, hem tek olarak hem de bulk olarak product’lar eklemi\u015ftik. Bu noktaya kadar art\u0131k her \u015feyimiz mevcut. Bir adet “product_search<\/em>” alias’\u0131na sahip indeximiz ve i\u00e7erisinde de bir ka\u00e7 \u00fcr\u00fcn var. Geriye art\u0131k yava\u015f yava\u015f ElasticSearch’\u00fcn as\u0131l amac\u0131\u00a0olan search<\/strong> i\u015flemlerine girmenin zaman\u0131 geldi.<\/p>\n Bu makale i\u00e7erisinde\u00a0ElasticSearch \u00fczerinde temel olarak\u00a0nas\u0131l search i\u015flemlerini ger\u00e7ekle\u015ftirebiliriz, filtre<\/strong> nedir ve nas\u0131l kullan\u0131l\u0131r gibi maddeleri ele al\u0131yor olaca\u011f\u0131z.<\/p>\n \u00d6ncelikle bir \u00f6nceki seride implemente etmi\u015f oldu\u011fumuz “BulkIndex” method’unu kullanarak bir ka\u00e7 adet daha product\u00a0ekleyelim. Eklemi\u015f oldu\u011fumuz product’lar \u00fczerinden search<\/strong> ve filter<\/strong> i\u015flemlerini ger\u00e7ekle\u015ftirece\u011fiz. “ElasticSearch.DataTransfer” projesi\u00a0i\u00e7erisinde \u00e7a\u011f\u0131rm\u0131\u015f oldu\u011fumuz\u00a0“BulkIndex” method’unu a\u015fa\u011f\u0131daki gibi de\u011fi\u015ftirip, \u00e7al\u0131\u015ft\u0131ral\u0131m.<\/p>\n Bu i\u015flemin ard\u0131ndan Sense \u00fczerinden a\u015fa\u011f\u0131daki query’i execute edelim\u00a0ve response’a bir bakal\u0131m.<\/p>\n Query sonucundaki response a\u015fa\u011f\u0131daki gibi olacakt\u0131r.<\/p>\n “BulkIndex” method’u ile eklemi\u015f oldu\u011fumuz product’lar\u0131n,\u00a0geldi\u011fini g\u00f6rebilmekteyiz. Yeterli say\u0131da product\u00a0elde etti\u011fimize\u00a0g\u00f6re art\u0131k temel olarak search i\u015flemlerine\u00a0ba\u015flayabiliriz.<\/p>\n Query yap\u0131s\u0131n\u0131 anlat\u0131m k\u0131sm\u0131nda ilk \u00f6nce Sense \u00fczerinden i\u015flemlerimizi pratik bir \u015fekilde ger\u00e7ekle\u015ftirece\u011fiz ve ard\u0131ndan C# taraf\u0131nda bunu nas\u0131l\u00a0implemente edebilece\u011fimize bir bakaca\u011f\u0131z. Bu i\u015flemlerin \u00f6ncesinde biraz Query DSL<\/em> yap\u0131s\u0131na bir g\u00f6z atal\u0131m.<\/p>\n ElasticSearch query’leri tan\u0131mlayabilmek i\u00e7in,\u00a0JSON<\/em>\u00a0temelli\u00a0bir Query DSL<\/em>\u00a0(Domain Specific Language) yap\u0131s\u0131 kullan\u0131r. Bu yap\u0131 ise iki maddeye dayan\u0131r:<\/p>\n Query structure’\u0131n\u0131 daha iyi kavrayabilmek\u00a0i\u00e7in dilerseniz basic\u00a0bir search<\/strong> query’si yazal\u0131m.<\/p>\n \u0130lk olarak URI\u00a0terminolojisine bakt\u0131\u011f\u0131m\u0131zda\u00a0HTTP verb’lerinden “GET” verb’\u00fcn\u00fc yazd\u0131ktan sonra<\/p>\n \u015feklinde bir URI\u00a0path’i\u00a0olu\u015fturuyoruz ve devam\u0131ndaki JSON objesine bakt\u0131\u011f\u0131m\u0131zda ise, i\u00e7erisinde bir adet “query” context’i bulunmakta. Bu query context’i i\u00e7erisinde kullan\u0131lacak olan herhangi bir Leaf Query veya Compound Query, ElasticSearch’\u00fcn kendi referans sitesinde de bahsetti\u011fi gibi \u015fu soruya cevap verir “How well does this document match this query clause?<\/em>” ve ayr\u0131ca e\u015fle\u015fenler aras\u0131nda bir skorlama<\/strong>(scoring) i\u015flemi ger\u00e7ekle\u015ftirir. Yukar\u0131daki query context i\u00e7erisinde kullanm\u0131\u015f oldu\u011fumuz “term” query,\u00a0leaf query’e bir \u00f6rnektir. Burada “product” type’\u0131n\u0131n “name” property’si \u00fczerinde bir contain i\u015flemi yapmaktad\u0131r. Farkl\u0131 ihtiya\u00e7lara y\u00f6nelik di\u011fer query \u00f6rneklerine ise buraya<\/a><\/em>\u00a0t\u0131klayarak, sa\u011f b\u00f6l\u00fcmde\u00a0bulunan\u00a0“Query DSL” tree’si alt\u0131ndan eri\u015febilirsiniz.<\/p>\n JSON objesi \u00fczerine query contextden farkl\u0131 olarak kullanabilece\u011fimiz \u00f6nemli bir ka\u00e7 alana daha\u00a0bakmak gerekirse:<\/p>\n Terminolojiye bakt\u0131ktan sonra\u00a0\u015fimdi yazm\u0131\u015f oldu\u011fumuz basic search query’sini, Sense \u00fczerinden execute edelim ve response’a bir bakal\u0131m.<\/p>\n Result’a bakt\u0131\u011f\u0131m\u0131zda term query t\u00fcm “name” field’lar\u0131 i\u00e7erisinde “iphone” kelimesi ge\u00e7en product’lar\u0131 getirmi\u015f\u00a0ve bir scoring i\u015flemi\u00a0ger\u00e7ekle\u015ftirerek, en uygun olan\u0131na g\u00f6re s\u0131rlama i\u015flemini yapm\u0131\u015ft\u0131r.<\/p>\n Dilerseniz\u00a0uygulam\u0131\u015f oldu\u011fumuz bu basic search i\u015flemini, di\u011fer makale serilerinde\u00a0de oldu\u011fu gibi ElasticSearch projesi i\u00e7erisine implemente edelim. Bunun i\u00e7in ilk olarak “ElasticSearch.Data.Contracts” projesine giderek, i\u00e7erisinde a\u015fa\u011f\u0131daki gibi yeni bir DTO tan\u0131mlayal\u0131m.<\/p>\n SearchResponseDTO<\/strong>, “IndexResponseDTO” class’\u0131ndan t\u00fcreyerek i\u00e7erisinde bir adet IEnumerable tipinde “Documents” bar\u0131nd\u0131r\u0131yor. Bu\u00a0dok\u00fcmanlar\u00a0search i\u015flemi sonucunda, geriye d\u00f6necek olan dok\u00fcmanlard\u0131r. \u015eimdi search method’unu “IElasticContext” interface’i \u00fczerinde tan\u0131mlayabiliriz.<\/p>\n Eklemi\u015f oldu\u011fumuz “Search” method’u parametre olarak NEST’e ait olan ISearchRequest interface’ini almaktad\u0131r ve geriye biraz \u00f6nce eklemi\u015f oldu\u011fumuz “SearchResponseDTO” yu d\u00f6nmektedir. “ElasticContext” \u00fczerine yeni method’u implemente edebiliriz.<\/p>\n Search method’unda tek yapt\u0131\u011f\u0131m\u0131z, d\u0131\u015far\u0131dan ald\u0131\u011f\u0131m\u0131z “searchRequest” parametresini “_elasticClient” \u00fczerindeki “Search” method’una ge\u00e7mektir. Bu i\u015flemin sonucunda ise NEST, verilen kriterler do\u011frultusunda search i\u015flemini ger\u00e7ekle\u015ftirerek resoponse’u d\u00f6nmektedir. “SearchResponseDTO” yu ise “IndexResponseDTO” dan t\u00fcretti\u011fimiz i\u00e7in “IsValid”, “StatusMessage” ve “Exception” parametrelerini tekrar yazman\u0131n \u00f6n\u00fcne ge\u00e7mi\u015f olduk\u00a0ve ilgili property’leri response’a g\u00f6re doldurarak, geriye d\u00f6n\u00fcyoruz.<\/p>\n \u015eimdi geldik as\u0131l search logic’in i\u015fleyece\u011fi k\u0131sma. Yani “Search” method’una parametre olarak ge\u00e7ti\u011fimiz “searchRequest” objesinin\u00a0haz\u0131rlanmas\u0131na. Burada \u00f6yle bir design yapal\u0131m\u00a0ki istedi\u011fimiz query’leri, filter’lar\u0131 kolayca eklemeye ve geli\u015ftirilmeye\u00a0a\u00e7\u0131k bir yap\u0131 olsun. Uzun uzaya giden method overload’lar\u0131 gibi anti-pattern<\/strong>‘lere gitmeden, esnek bir yap\u0131 kuraca\u011f\u0131z. Logic i\u015flemlerine ba\u015flamak i\u00e7in \u00f6ncelikle “ElasticSearch” solution’\u0131na “ElasticSearch.Business.Contracts” isminde yeni bir class library ekleyelim ve i\u00e7erisinde “IElasticSearchEngine” isminde bir interface tan\u0131mlayal\u0131m.<\/p>\n “IElasticSearchEngine” interface’inin i\u00e7erisine tan\u0131mlam\u0131\u015f oldu\u011fumuz “Execute” method’u i\u00e7erisinde search i\u015flemlerimizi ger\u00e7ekle\u015ftirece\u011fiz. Engine\u00a0<\/strong>katman\u0131 eklememizin temel\u00a0sebebi ise, bu tarz logic i\u00e7eren i\u015flemleri\u00a0burada toplamakt\u0131r. “ElasticSearch” solution’\u0131na “ElasticSearch.Business” isminde bir class library daha ekleyelim. Burada ise interface’leri implemente edece\u011fimiz as\u0131l concrete engine’ler yer alacakt\u0131r. Library’nin eklenmesinden sonra “ElasticSearch.Business.Contracts”, “ElasticSearch.Data.Contracts” ve “ElasticSearch.Data” library’lerini referans olarak ekleyelim ve Nuget Manager \u00fczerinden “NEST” k\u00fct\u00fcphanesini de kural\u0131m.<\/p>\n “ElasticSearch.Business”\u00a0library’si i\u00e7erisine “Business Engines” isminde bir folder ekleyip,\u00a0i\u00e7erisine “ElasticSearchEngine” isminde bir class tan\u0131ml\u0131yorum. Eklemi\u015f oldu\u011fumuz class’a “IElasticSearchEngine” interface’ini \u015fimdilik bo\u015f olacak \u015fekilde implemente edelim.<\/p>\n \u0130\u00e7erisinde private olarak tan\u0131mlam\u0131\u015f oldu\u011fumuz field’lar\u0131 constructor arac\u0131l\u0131\u011f\u0131 ile daha sonra inject edece\u011fiz. Yukar\u0131daki b\u00f6l\u00fcmde hat\u0131rlarsak, uzun uzaya giden method overload’lar\u0131 gibi anti-pattern’lerden ka\u00e7\u0131naca\u011f\u0131m\u0131z\u0131 ve kolayca geli\u015ftirmeye a\u00e7\u0131k bir design yapaca\u011f\u0131m\u0131zdan bahsetmi\u015ftik. “ElasticSearchEngine” i bir Builder<\/strong> ile Fluently<\/strong> bir \u015fekilde olu\u015fturmaya ne dersiniz? Yani d\u00fc\u015f\u00fcnelim ki \u015fu \u015fekilde bir kullan\u0131m\u0131 olsa:<\/p>\n Ne kadar da ak\u0131c\u0131 duruyor de\u011fil mi? E haydi ozaman implementasyon i\u015flemlerine ba\u015flayal\u0131m.<\/p>\n “ElasticSearch.Business” projesi i\u00e7erisine “ElasticSearchBuilder” isminde bir class ekleyelim. Bu class i\u00e7erisinde Builder pattern’ini, search logic’ine y\u00f6nelik olarak\u00a0implemente edece\u011fiz. Builder pattern’i \u00f6zetle kompleks class’lar\u0131n nesne \u00fcretim s\u00fcre\u00e7lerini, client’dan gizleyen bir pattern’dir ve GOF desenleri aras\u0131ndan Creational<\/strong> grubunda yer almaktad\u0131r.<\/p>\n “ElasticSearchBuilder” i\u00e7erisinde “ElasticSearchEngine” i\u00e7erisinde de tan\u0131mlam\u0131\u015f oldu\u011fumuz field’lar\u0131, internal<\/strong> olarak tan\u0131mlad\u0131k. Ayn\u0131 assembly i\u00e7erisinden eri\u015filmesi bizim i\u00e7in yeterli olacakt\u0131r. \u00c7\u00fcnk\u00fc bu parametrelere daha sonra “ElasticSearchEngine” i\u00e7erisinden eri\u015fece\u011fiz. Constructor’da ise, “indexName” ve “elasticContext” parametrelerini\u00a0zorunlu oldu\u011fu i\u00e7in costructor arac\u0131l\u0131\u011f\u0131 ile inject ediyoruz ve ard\u0131ndan “QueryContainer” \u0131 initialize ediyoruz. Bunun sebebi ise\u00a0NEST’in elastic client’\u0131 query container arac\u0131l\u0131\u011f\u0131 ile query’leri almaktad\u0131r ve bizde Builder i\u00e7erisinde Fluently bir yap\u0131 kuraca\u011f\u0131m\u0131z i\u00e7in, ilk ba\u015fta “QueryContainer” \u0131\u00a0initialize ediyoruz.<\/p>\n Di\u011fer parametreler elastic\u00a0i\u00e7in must olmad\u0131klar\u0131ndan dolay\u0131 “SetSize”, “SetFrom” ve “AddTermQuery” olarak farkl\u0131 method’lara b\u00f6ld\u00fck ve hepsi ilgili i\u015flemlerini yap\u0131p geriye yine kendi context’ini\u00a0d\u00f6nmektedir. Son olarak “Build” method’u ise art\u0131k Fluently olarak eklememiz bittikten sonra \u00e7a\u011f\u0131raca\u011f\u0131m\u0131z son method olacakt\u0131r. Burada yeni bir “ElasticSearchEngine” initialize edip, constructor arac\u0131l\u0131\u011f\u0131 ile kendisini yani “ElasticSearchBuilder” \u0131 parametreleri ile beraber inject etmektedir.<\/p>\n \u015eimdi “ElasticSearchEngine” class’\u0131na geri d\u00f6nelim ve art\u0131k constructor arac\u0131l\u0131\u011f\u0131 ile bir “ElasticSearchBuilder” alabilecek bir hale getirelim, ard\u0131ndan ilgili field’lar\u0131 builder’a g\u00f6re\u00a0set’leyelim.<\/p>\n Constructor arac\u0131l\u0131\u011f\u0131 ile bir adet “ElasticSearchBuilder” ald\u0131k ve yukar\u0131daki field’lar\u0131 setledik. Bu i\u015flemin ard\u0131ndan ise “Execute” method’unda “_elasticContext” \u00fczerindeki “Search” method’unu \u00e7a\u011f\u0131rd\u0131k ve ilgili parametreleri verdik. E\u011fer response sonucu “IsValid” ise, response \u00fczerinden gelen dok\u00fcmanlar\u0131 List of T tipinde\u00a0geriye d\u00f6n\u00fcyoruz.<\/p>\n Implementasyonumuz \u015fuan bitmi\u015f durumdad\u0131r. Hemen “ElasticSearch.DataTransfer” projesi alt\u0131ndaki “Program.cs” class’\u0131n\u0131 a\u015fa\u011f\u0131daki gibi\u00a0d\u00fczenleyelim ve bir test yapal\u0131m.<\/p>\n “SearchProduct” method’unda parametre olarak “indexName” ve “elasticContext” i g\u00f6nderiyoruz ve “ElasticSearchBuilder” arac\u0131l\u0131\u011f\u0131 ile, doldurmu\u015f oldu\u011fumuz parametrelere g\u00f6re bir “ElasticSearchEngine” initialize ediyoruz. Bu i\u015flemin sonucunda “Execute” method’una istemi\u015f oldu\u011fumuz tipi verip, search i\u015flemini ba\u015flat\u0131yoruz. Console ekran\u0131ndaki sonu\u00e7 ise a\u015fa\u011f\u0131daki gibi olacakt\u0131r.<\/p>\n Her ne kadar query’ler ile istedi\u011fimiz d\u00fczeyde i\u015flem\u00a0yapabiliyor olsak da, baz\u0131 durumlarda query’leri filter kullanarak daha fazla \u00f6zelle\u015ftirmeye ihtiya\u00e7 duyabiliriz. Filter’lar\u0131 t\u0131pk\u0131 SQL’de oldu\u011fu gibi “WHERE”\u00a0statement’\u0131na benzetebiliriz. DSL \u00fczerinde iki farkl\u0131 filter bulunmaktad\u0131r. Bunlar:<\/p>\n Filtered query Filter’a g\u00f6re daha performansl\u0131 \u00e7al\u0131\u015fmaktad\u0131r. Filter executing i\u015fleminden sonra filtreleme\u00a0i\u015flemi ger\u00e7ekle\u015ftirdi\u011fi i\u00e7in, CPU’ya olan maliyeti daha fazlad\u0131r. Tabi bu filtre’yi kullanmak, CPU’ya olan maliyetine g\u00f6re de\u011fil istenilen sonucun do\u011frultusunda olmal\u0131d\u0131r. \u00c7\u00fcnk\u00fc atm\u0131\u015f oldu\u011funuz query’deki\u00a0Aggregation<\/strong>‘lar\u0131n ve Hits<\/strong>‘lerin filtrelemeden etkilenmemesini isteyebilirsiniz. \u0130\u015fte bu gibi durumlarda Filter yani postfiltering i\u015flemi uygulanmaktad\u0131r.<\/p>\n \u00d6rne\u011fimiz \u00fczerinden devam etmek gerekirse, a\u015fa\u011f\u0131daki gibi basit bir filtre yazal\u0131m.<\/p>\n Yukar\u0131daki sorguda query context i\u00e7erisine “filtered” ve “filter” olmak \u00fczere iki filtreyi de ekledik. Filtered kapsam\u0131nda prefiltering olarak bir adet term filter<\/strong> ekledik ve “name” field’lar\u0131nda “iphone” kelimesi ge\u00e7enleri getirmesini s\u00f6yledik. Filter kapsam\u0131nda ise postfiltering olarak bir adet range filter<\/strong> takt\u0131k. Bu filtre ile “3900” de\u011ferinden b\u00fcy\u00fck ve e\u015fitleri, “5000” de\u011ferinden ise k\u00fc\u00e7\u00fck ve e\u015fitleri filtrelemesini s\u00f6yledik. Bu i\u015flemin sonucunda ise response, Sense \u00fczerinde a\u015fa\u011f\u0131daki gibi olacakt\u0131r:<\/p>\n Sonuca bakt\u0131\u011f\u0131m\u0131zda “3950” ve “5000” price de\u011ferlerine sahip iki adet product’\u0131n geldi\u011fini g\u00f6r\u00fcyoruz. Dilerseniz \u015fimdi\u00a0filter ekleme i\u015flemlerini, geli\u015ftirmi\u015f oldu\u011fumuz builder yap\u0131s\u0131na implemente edelim.<\/p>\n “ElasticSearchBuilder” class’\u0131n\u0131 art\u0131k filtered ve filter query’ler \u00fczerinden build\u00a0yapabilecek \u015fekilde\u00a0g\u00fcncelleyelim.<\/p>\n \u00d6ncelikle constructor’da “QueryContainer” \u0131n “Filtered” property’sini initialize ediyoruz. “AddTermQuery” method’unda ise, direkt olarak “QueryContainer” \u0131n “Query” property’sine set etmek yerine “Filtered” i\u00e7erisindeki “Query” e set ediyoruz. Sorgumuzda bir adette range filter vard\u0131 hat\u0131rlarsak. Bunun i\u00e7in ise “AddRangeFilter” isminde bir method tan\u0131ml\u0131yoruz ve i\u00e7erisinde “QueryContainer.Filtered” i\u00e7erisindeki “Filter” property’sine “NumericRangeQuery” initialize edip, alm\u0131\u015f oldu\u011fumuz parametrelere g\u00f6re\u00a0property’lerini set ediyoruz.<\/p>\n \u00d6rne\u011fimiz gere\u011fi kurmu\u015f oldu\u011fumuz bu yap\u0131da bir adet “Filtered Query” ve bir adet “Filter” ekleyebilmekteyiz. Sizler ise bu builder yap\u0131s\u0131n\u0131 kullanarak NEST’in interface’leri arac\u0131l\u0131\u011f\u0131 ile istedi\u011finiz Query’leri array olarak toplay\u0131p, birden fazla filter’lar\u0131 “QueryContainer” a ekleyebilme gibi logic’leri\u00a0kazand\u0131rabilirsiniz.<\/p>\n \u015eimdi “ElasticSearch.DataTransfer” projesi i\u00e7erisindeki “Program.cs” class’\u0131n\u0131 a\u015fa\u011f\u0131daki gibi g\u00fcncelleyelim.<\/p>\n Tek fark olarak Fluently bir \u015fekilde sadece “AddRangeFilter” method’unu \u00e7a\u011f\u0131rd\u0131k. Console uygulamas\u0131n\u0131 \u00e7al\u0131\u015ft\u0131ral\u0131m ve sonucu birde ekran \u00fczerinden g\u00f6relim.<\/p>\nprivate static void BulkIndex(ElasticContext elasticContext, string indexName)\r\n{\r\n var response = elasticContext.BulkIndex(indexName, new List<Product>()\r\n {\r\n new Product()\r\n {\r\n Id = 1,\r\n Name = \"Iphone 6s Plus\",\r\n Description = \"64gb Iphone 6s Plus, Renk: Space Gray\",\r\n Price = 5000\r\n },\r\n new Product()\r\n {\r\n Id = 2,\r\n Name = \"Sony Xperia Z Ultra\",\r\n Description = \"Full HD ekran, Renk: Siyah\",\r\n Price = 1200\r\n },\r\n new Product()\r\n {\r\n Id = 3,\r\n Name = \"Samsung Galaxy S7 Edge\",\r\n Description = \"Amelod ekran, Renk: Beyaz\",\r\n Price = 2800\r\n },\r\n new Product()\r\n {\r\n Id = 4,\r\n Name = \"Samsung Galaxy S7\",\r\n Description = \"Amelod ekran, Renk: Beyaz\",\r\n Price = 2500\r\n },\r\n new Product()\r\n {\r\n Id = 5,\r\n Name = \"Iphone 6s\",\r\n Description = \"64gb Iphone 6s, Renk: Space Gray\",\r\n Price = 3800\r\n },\r\n new Product()\r\n {\r\n Id = 6,\r\n Name = \"Iphone 6s\",\r\n Description = \"128gb Iphone 6s, Renk: Space Gray\",\r\n Price = 3950\r\n },\r\n });\r\n\r\n Console.WriteLine(response.StatusMessage);\r\n}<\/pre>\nGET product_search\/product\/_search<\/pre>\n
{\r\n \"took\": 25,\r\n \"timed_out\": false,\r\n \"_shards\": {\r\n \"total\": 5,\r\n \"successful\": 5,\r\n \"failed\": 0\r\n },\r\n \"hits\": {\r\n \"total\": 6,\r\n \"max_score\": 1,\r\n \"hits\": [\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"5\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 5,\r\n \"name\": \"Iphone 6s\",\r\n \"description\": \"64gb Iphone 6s, Renk: Space Gray\",\r\n \"price\": 3800\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"2\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 2,\r\n \"name\": \"Sony Xperia Z Ultra\",\r\n \"description\": \"Full HD ekran, Renk: Siyah\",\r\n \"price\": 1200\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"4\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 4,\r\n \"name\": \"Samsung Galaxy S7\",\r\n \"description\": \"Amelod ekran, Renk: Beyaz\",\r\n \"price\": 2500\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"6\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 6,\r\n \"name\": \"Iphone 6s\",\r\n \"description\": \"128gb Iphone 6s, Renk: Space Gray\",\r\n \"price\": 3950\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"1\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 1,\r\n \"name\": \"Iphone 6s Plus\",\r\n \"description\": \"64gb Iphone 6s Plus, Renk: Space Gray\",\r\n \"price\": 5000\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"3\",\r\n \"_score\": 1,\r\n \"_source\": {\r\n \"id\": 3,\r\n \"name\": \"Samsung Galaxy S7 Edge\",\r\n \"description\": \"Amelod ekran, Renk: Beyaz\",\r\n \"price\": 2800\r\n }\r\n }\r\n ]\r\n }\r\n}<\/pre>\n1) Query DSL & Basic Search<\/h3>\n
\n
GET product_search\/product\/_search\r\n{\r\n \"query\": {\r\n \"term\": {\r\n \"name\": {\r\n \"value\": \"iphone\"\r\n }\r\n }\r\n }\r\n}<\/pre>\n{index ad\u0131 veya alias}\/{type}\/_search<\/pre>\n\n
{\r\n \"took\": 6,\r\n \"timed_out\": false,\r\n \"_shards\": {\r\n \"total\": 5,\r\n \"successful\": 5,\r\n \"failed\": 0\r\n },\r\n \"hits\": {\r\n \"total\": 3,\r\n \"max_score\": 0.8784157,\r\n \"hits\": [\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"6\",\r\n \"_score\": 0.8784157,\r\n \"_source\": {\r\n \"id\": 6,\r\n \"name\": \"Iphone 6s\",\r\n \"description\": \"128gb Iphone 6s, Renk: Space Gray\",\r\n \"price\": 3950\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"5\",\r\n \"_score\": 0.19178301,\r\n \"_source\": {\r\n \"id\": 5,\r\n \"name\": \"Iphone 6s\",\r\n \"description\": \"64gb Iphone 6s, Renk: Space Gray\",\r\n \"price\": 3800\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"1\",\r\n \"_score\": 0.15342641,\r\n \"_source\": {\r\n \"id\": 1,\r\n \"name\": \"Iphone 6s Plus\",\r\n \"description\": \"64gb Iphone 6s Plus, Renk: Space Gray\",\r\n \"price\": 5000\r\n }\r\n }\r\n ]\r\n }\r\n}<\/pre>\nusing System.Collections.Generic;\r\n\r\nnamespace ElasticSearch.Data.Contracts\r\n{\r\n public class SearchResponseDTO<T> : IndexResponseDTO\r\n {\r\n public IEnumerable<T> Documents { get; set; }\r\n }\r\n}<\/pre>\nusing Nest;\r\nusing System.Collections.Generic;\r\n\r\nnamespace ElasticSearch.Data.Contracts\r\n{\r\n public interface IElasticContext\r\n {\r\n IndexResponseDTO CreateIndex<T>(string indexName, string aliasName) where T : class;\r\n IndexResponseDTO Index<T>(string indexName, T document) where T : class;\r\n IndexResponseDTO BulkIndex<T>(string indexName, List<T> document) where T : class;\r\n SearchResponseDTO<T> Search<T>(ISearchRequest searchRequest) where T : class;\r\n }\r\n}<\/pre>\npublic SearchResponseDTO<T> Search<T>(ISearchRequest searchRequest) where T : class\r\n{\r\n var response = _elasticClient.Search<T>(searchRequest);\r\n\r\n return new SearchResponseDTO<T>()\r\n {\r\n IsValid = response.IsValid,\r\n StatusMessage = response.DebugInformation,\r\n Exception = response.OriginalException,\r\n Documents = response.Documents\r\n };\r\n}<\/pre>\nusing System.Collections.Generic;\r\n\r\nnamespace ElasticSearch.Business.Contracts\r\n{\r\n public interface IElasticSearchEngine\r\n {\r\n List<T> Execute<T>() where T : class;\r\n }\r\n}<\/pre>\nusing System.Collections.Generic;\r\nusing System.Linq;\r\nusing Nest;\r\nusing ElasticSearch.Business.Contracts;\r\nusing ElasticSearch.Data.Contracts;\r\n\r\nnamespace ElasticSearch.Business\r\n{\r\n public class ElasticSearchEngine : IElasticSearchEngine\r\n {\r\n private readonly string _indexName;\r\n private readonly int _size;\r\n private readonly int _from;\r\n private readonly IQueryContainer _queryContainer;\r\n private readonly IElasticContext _elasticContext;\r\n\r\n public ElasticSearchEngine()\r\n {\r\n }\r\n\r\n #region IElasticSearchEngine Members\r\n public List<T> Execute<T>() where T : class\r\n {\r\n return null;\r\n }\r\n #endregion\r\n }\r\n}<\/pre>\nvar elasticSearchEngine = new ElasticSearchBuilder(indexName)\r\n .SetElasticContext(elasticContext)\r\n .SetSize(5)\r\n .SetFrom(0)\r\n .AddTermQuery(\"iphone\", \"name\")\r\n .AddBlaBlaQuery()\r\n .AddBlaBlaFilter()\r\n .Build()\r\n .Execute<Product>();<\/pre>\n
using ElasticSearch.Data.Contracts;\r\nusing Nest;\r\n\r\nnamespace ElasticSearch.Business\r\n{\r\n public class ElasticSearchBuilder\r\n {\r\n internal string IndexName;\r\n internal int Size;\r\n internal int From;\r\n internal IQueryContainer QueryContainer;\r\n internal IElasticContext ElasticContext;\r\n\r\n public ElasticSearchBuilder(string indexName, IElasticContext elasticContext)\r\n {\r\n IndexName = indexName;\r\n ElasticContext = elasticContext;\r\n\r\n QueryContainer = new QueryContainer();\r\n }\r\n\r\n public ElasticSearchBuilder SetSize(int size)\r\n {\r\n Size = size;\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchBuilder SetFrom(int from)\r\n {\r\n From = from;\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchBuilder AddTermQuery(string term, string field)\r\n {\r\n QueryContainer.Term = new TermQuery()\r\n {\r\n Field = new Field()\r\n {\r\n Name = field\r\n\r\n },\r\n Value = term\r\n };\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchEngine Build()\r\n {\r\n return new ElasticSearchEngine(this);\r\n }\r\n }\r\n}<\/pre>\nusing System.Collections.Generic;\r\nusing System.Linq;\r\nusing Nest;\r\nusing ElasticSearch.Business.Contracts;\r\nusing ElasticSearch.Data.Contracts;\r\n\r\nnamespace ElasticSearch.Business\r\n{\r\n public class ElasticSearchEngine : IElasticSearchEngine\r\n {\r\n private readonly string _indexName;\r\n private readonly int _size;\r\n private readonly int _from;\r\n private readonly IQueryContainer _queryContainer;\r\n private readonly IElasticContext _elasticContext;\r\n\r\n public ElasticSearchEngine(ElasticSearchBuilder elasticSearchBuilder)\r\n {\r\n _indexName = elasticSearchBuilder.IndexName;\r\n _size = elasticSearchBuilder.Size;\r\n _from = elasticSearchBuilder.From\r\n _queryContainer = elasticSearchBuilder.QueryContainer;\r\n _elasticContext = elasticSearchBuilder.ElasticContext;\r\n }\r\n\r\n #region IElasticSearchEngine Members\r\n public List<T> Execute<T>() where T : class\r\n {\r\n var response = _elasticContext.Search<T>(new SearchRequest(_indexName, typeof(T))\r\n {\r\n Size = _size,\r\n From = _from,\r\n Query = (QueryContainer)_queryContainer\r\n });\r\n\r\n if (response.IsValid)\r\n {\r\n return response.Documents.ToList();\r\n }\r\n\r\n return null;\r\n }\r\n #endregion\r\n }\r\n}<\/pre>\nstatic void Main(string[] args)\r\n{\r\n string indexName = ConfigurationManager.AppSettings[\"ElasticSearchIndexName\"];\r\n\r\n var elasticClient = new ElasticClient(ElasticHelper.Instance.GetConnectionSettings());\r\n var elasticContext = new ElasticContext(elasticClient);\r\n\r\n \/\/CreateIndex(elasticContext, indexName);\r\n\r\n \/\/Index(elasticContext, indexName);\r\n\r\n \/\/BulkIndex(elasticContext, indexName);\r\n\r\n SearchProduct(indexName, elasticContext);\r\n\r\n Console.ReadKey();\r\n}\r\n\r\nprivate static void SearchProduct(string indexName, ElasticContext elasticContext)\r\n{\r\n var elasticSearchEngine = new ElasticSearchBuilder(indexName, elasticContext)\r\n .SetSize(5)\r\n .SetFrom(0)\r\n .AddTermQuery(\"iphone\", \"name\")\r\n .Build()\r\n .Execute<Product>();\r\n\r\n elasticSearchEngine.ForEach(x =>\r\n Console.WriteLine(\"Id: {0} Name: {1} Description: {2} Price: {3}\", x.Id, x.Name, x.Description, x.Price));\r\n}<\/pre>\n<\/a><\/p>\n2) Filter Kullan\u0131m\u0131<\/h3>\n
\n
GET product_search\/product\/_search\r\n{\r\n \"from\": 0,\r\n \"size\": 5,\r\n \"query\": {\r\n \"filtered\": {\r\n \"query\": {\r\n \"term\": {\r\n \"name\": {\r\n \"value\": \"iphone\"\r\n }\r\n }\r\n },\r\n \"filter\": {\r\n \"range\": {\r\n \"price\": {\r\n \"gte\": 3900,\r\n \"lte\": 5000\r\n }\r\n }\r\n }\r\n }\r\n }\r\n}<\/pre>\n{\r\n \"took\": 62,\r\n \"timed_out\": false,\r\n \"_shards\": {\r\n \"total\": 5,\r\n \"successful\": 5,\r\n \"failed\": 0\r\n },\r\n \"hits\": {\r\n \"total\": 2,\r\n \"max_score\": 0.8784157,\r\n \"hits\": [\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"6\",\r\n \"_score\": 0.8784157,\r\n \"_source\": {\r\n \"id\": 6,\r\n \"name\": \"Iphone 6s\",\r\n \"description\": \"128gb Iphone 6s, Renk: Space Gray\",\r\n \"price\": 3950\r\n }\r\n },\r\n {\r\n \"_index\": \"product_search_201607242334\",\r\n \"_type\": \"product\",\r\n \"_id\": \"1\",\r\n \"_score\": 0.15342641,\r\n \"_source\": {\r\n \"id\": 1,\r\n \"name\": \"Iphone 6s Plus\",\r\n \"description\": \"64gb Iphone 6s Plus, Renk: Space Gray\",\r\n \"price\": 5000\r\n }\r\n }\r\n ]\r\n }\r\n}<\/pre>\nusing ElasticSearch.Data.Contracts;\r\nusing Nest;\r\n\r\nnamespace ElasticSearch.Business\r\n{\r\n public class ElasticSearchBuilder\r\n {\r\n internal string IndexName;\r\n internal int Size;\r\n internal int From;\r\n internal IQueryContainer QueryContainer;\r\n internal IElasticContext ElasticContext;\r\n\r\n public ElasticSearchBuilder(string indexName, IElasticContext elasticContext)\r\n {\r\n IndexName = indexName;\r\n ElasticContext = elasticContext;\r\n\r\n QueryContainer = new QueryContainer();\r\n QueryContainer.Filtered = new FilteredQuery();\r\n }\r\n\r\n public ElasticSearchBuilder SetSize(int size)\r\n {\r\n Size = size;\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchBuilder SetFrom(int from)\r\n {\r\n From = from;\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchBuilder AddTermQuery(string term, string field)\r\n {\r\n QueryContainer.Filtered.Query = new TermQuery()\r\n {\r\n Field = new Field()\r\n {\r\n Name = field\r\n\r\n },\r\n Value = term\r\n };\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchBuilder AddRangeFilter(double gte, double lte, string field)\r\n {\r\n QueryContainer.Filtered.Filter = new NumericRangeQuery()\r\n {\r\n Field = new Field()\r\n {\r\n Name = field\r\n },\r\n GreaterThanOrEqualTo = gte,\r\n LessThanOrEqualTo = lte\r\n };\r\n\r\n return this;\r\n }\r\n\r\n public ElasticSearchEngine Build()\r\n {\r\n return new ElasticSearchEngine(this);\r\n }\r\n }\r\n}<\/pre>\nprivate static void SearchProduct(string indexName, ElasticContext elasticContext)\r\n{\r\n var elasticSearchEngine = new ElasticSearchBuilder(indexName, elasticContext)\r\n .SetSize(5)\r\n .SetFrom(0)\r\n .AddTermQuery(\"iphone\", \"name\")\r\n .AddRangeFilter(3900, 5000, \"price\")\r\n .Build()\r\n .Execute<Product>();\r\n\r\n elasticSearchEngine.ForEach(x =>\r\n Console.WriteLine(\"Id: {0} Name: {1} Description: {2} Price: {3}\", x.Id, x.Name, x.Description, x.Price));\r\n}<\/pre>\n