{"id":9097,"date":"2018-08-15T14:27:59","date_gmt":"2018-08-15T14:27:59","guid":{"rendered":"http:\/\/www.gencayyildiz.com\/blog\/?p=9097"},"modified":"2018-08-15T14:27:59","modified_gmt":"2018-08-15T14:27:59","slug":"nosql-mongodb-islevsel-fonksiyonlar-ve-aggregate-operatorleri","status":"publish","type":"post","link":"https:\/\/www.gencayyildiz.com\/blog\/nosql-mongodb-islevsel-fonksiyonlar-ve-aggregate-operatorleri\/","title":{"rendered":"NoSQL – MongoDB \u0130\u015flevsel Fonksiyonlar ve Aggregate Operat\u00f6rleri"},"content":{"rendered":"
<\/div>\n

Merhaba,<\/p>\n

Klasik olarak veritaban\u0131ndan elde edilen veriler \u00fczerinde belli ba\u015fl\u0131 i\u015flemleri NoSQL sistemlerinde de uygulayabilmekteyiz. Bu i\u00e7eri\u011fimizde MongoDB veritaban\u0131nda bar\u0131nd\u0131r\u0131lan veriler \u00fczerinde s\u0131ralama, limit belirleme vs. gibi s\u0131radan i\u015flevlerle beraber $group, $lookup vs. gibi aggregate operat\u00f6rleri ile elde edilen verilerin daha farkl\u0131 modifiyeler sa\u011flanarak nas\u0131l anlamland\u0131r\u0131labilece\u011fini inceliyor olaca\u011f\u0131z.<\/p>\n

\u0130\u00e7eri\u011fimizde \u00f6rneklendirme a\u015famas\u0131nda model olarak;<\/p>\n

\r\nconst mongoose = require("mongoose");\r\n \r\nmodule.exports = mongoose.model("Employees", new mongoose.Schem a({\r\n    _id: Number,\r\n    userName: String,\r\n    name: String,\r\n    surName: String,\r\n    age: Number\r\n}));\r\n<\/pre>\n
router yap\u0131s\u0131 olarak;<\/p>\n
\r\nconst express = require("express");\r\nconst router = express.Router();\r\nrouter.get("\/", (request, response, next) => {\r\n    response.send("Merhaba d\u00fcnya...");\r\n});\r\nmodule.exports = router;\r\n<\/pre>\n
ve ana mod\u00fcl olarak;<\/p>\n
\r\nconst express = require("express");\r\nconst app = express();\r\nconst mongoose = require("mongoose");\r\nmongoose.connect("mongodb:\/\/localhost\/companydb")\r\n    .then(() => console.log("Veritaban\u0131 ba\u011flant\u0131s\u0131 ba\u015far\u0131yla sa\u011flanm\u0131\u015ft\u0131r..."))\r\n    .catch(error => console.log("Veritaban\u0131na ba\u011flant\u0131 olu\u015fturulurken beklenmeyen bir hatayla kar\u015f\u0131la\u015f\u0131ld\u0131...", error.message));\r\nconst employeeRouter = require(".\/routers\/employeeRouter");\r\napp.use("\/employee", employeeRouter);\r\napp.listen(5000);\r\n<\/pre>\n
yap\u0131lar\u0131n\u0131 kullanacak ve bunlardan route yap\u0131s\u0131nda \u00e7al\u0131\u015faca\u011f\u0131m\u0131z\u0131 ba\u015ftan bildirmekte fayda g\u00f6r\u00fcyorum.<\/p>\n
S\u0131ralama(sort)<\/h3>\n
Belirtilen kolona g\u00f6re verileri s\u0131ralamaktad\u0131r. Klasik SQL’de ki “Order By” komutana kar\u015f\u0131l\u0131k gelmektedir. Say\u0131sal tipte kolonlarda b\u00fcy\u00fckl\u00fck-k\u00fc\u00e7\u00fckl\u00fck k\u0131yaslamas\u0131 yaparken metinsel kolonlarda alfabetik s\u0131ralama ger\u00e7ekle\u015ftirmektedir.<\/p>\n
\r\nconst express = require("express");\r\nconst router = express.Router();\r\nconst employeeModel = require("..\/models\/employee");\r\nrouter.get("\/", (request, response, next) => {\r\n    employeeModel.find({}, (error, data) => {\r\n        if (!error)\r\n            response.json(data);\r\n        else\r\n            response.send("Beklenmeyen bir hata ile kar\u015f\u0131la\u015f\u0131ld\u0131.", error.message);\r\n    }).sort({ "_id": -1 });\r\n});\r\nmodule.exports = router;\r\n<\/pre>\n
Dikkat ederseniz e\u011fer belirtilen “_id” kolonuna “-1” de\u011feri veirlmi\u015ftir. Bu de\u011fer say\u0131sal olarak b\u00fcy\u00fckten k\u00fc\u00e7\u00fc\u011fe, metinsel olarakta z’den a’ya anlam\u0131n\u0131 ifade etmektedir. E\u011fer ki tam tersi bir s\u0131ralama isteniyorsa “1” de\u011feri verilmesi yeterlidir.<\/p>\n
\r\nconst express = require("express");\r\nconst router = express.Router();\r\nconst employeeModel = require("..\/models\/employee");\r\nrouter.get("\/", (request, response, next) => {\r\n    employeeModel.find({}, (error, data) => {\r\n        if (!error)\r\n            response.json(data);\r\n        else\r\n            response.send("Beklenmeyen bir hata ile kar\u015f\u0131la\u015f\u0131ld\u0131.", error.message);\r\n    }).sort({ "name": 1, "_id": -1 });\r\n});\r\n\r\nmodule.exports = router;\r\n<\/pre>\n
Birden fazla kolonda s\u0131ralama yapmak istenirse yukar\u0131daki gibi \u00e7al\u0131\u015f\u0131labilmektedir. Burada \u00f6nce “name” kolonuna g\u00f6re a’dan z’ye ard\u0131ndan “_id” kolonuna g\u00f6rede b\u00fcy\u00fckten k\u00fc\u00e7\u00fc\u011fe bir s\u0131ralama ger\u00e7ekle\u015ftirilmi\u015ftir.<\/p>\n
Limit Belirleme(limit)<\/h3>\n
Eldeki veri listesinin en ba\u015ftaki verisinden belirtilen adet kadar veriyi getirecektir. Klasik SQL’de ki “Top n” komutuna kar\u015f\u0131 gelmektedir.<\/p>\n
\r\nconst express = require("express");\r\nconst router = express.Router();\r\nconst employeeModel = require("..\/models\/employee");\r\nrouter.get("\/", (request, response, next) => {\r\n    employeeModel.find({}, (error, data) => {\r\n        if (!error)\r\n            response.json(data);\r\n        else\r\n            response.send("Beklenmeyen bir hata ile kar\u015f\u0131la\u015f\u0131ld\u0131.", error.message);\r\n    }).limit(2);\r\n});\r\nmodule.exports = router;\r\n<\/pre>\n
Atlama(skip)<\/h3>\n
Veri listesinin ba\u015f\u0131ndan belirtilen de\u011fer kadar veriyi g\u00f6rmemezlikten gelecek gerisini getirecektir.<\/p>\n
\r\nconst express = require("express");\r\nconst router = express.Router();\r\nconst employeeModel = require("..\/models\/employee");\r\nrouter.get("\/", (request, response, next) => {\r\n    employeeModel.find({}, (error, data) => {\r\n        if (!error)\r\n            response.json(data);\r\n        else\r\n            response.send("Beklenmeyen bir hata ile kar\u015f\u0131la\u015f\u0131ld\u0131.", error.message);\r\n    }).skip(2);\r\n});\r\nmodule.exports = router;\r\n<\/pre>\n
Aggregate Operat\u00f6rleri<\/h3>\n
MongoDB’de aggregate operat\u00f6rleri klasik SQL’de ki “aggregate” fonksiyonlar\u0131yla ayn\u0131 \u00f6zellikleri bar\u0131nd\u0131rmakta ve orada hangi mant\u0131kta hangi durumlarda kullan\u0131l\u0131yorsa burada da ayn\u0131 mant\u0131k ve durumlar i\u00e7in kullan\u0131lan, aradaki tek fark\u0131n oradakinin fonksiyon buradakinin ise operat\u00f6r olmas\u0131 olan i\u015flevlerdir.<\/p>\n
Aggregate parametrelerini kullanabilmek i\u00e7in model \u00fczerinden aggregate fonksiyonunu \u00e7a\u011f\u0131rmal\u0131s\u0131n\u0131z. Birazdan g\u00f6rece\u011fimiz t\u00fcm operat\u00f6rler i\u00e7in a\u015fa\u011f\u0131daki yap\u0131sal durumu genelleyerek zihninizde oturtman\u0131z sizler i\u00e7in yap\u0131lar\u0131n kullan\u0131m\u0131na dair olduk\u00e7a kolayl\u0131k sa\u011flayacakt\u0131r.<\/p>\n
\r\n    model.aggregate([\r\n        {\r\n            [$parametre]: {\r\n                [...i\u015flemler...]\r\n            }\r\n        }\r\n    ], (error, data) => {\r\n\r\n    });\r\n<\/pre>\n
Yukar\u0131daki prototipi incelerseniz “aggregate” fonksiyonu i\u00e7erisinde bir dizi olu\u015fturulmakta ve kullan\u0131lacak operat\u00f6r bu dizi i\u00e7erisinde yarat\u0131lm\u0131\u015f bir object i\u00e7erisinde \u00e7a\u011fr\u0131lmaktad\u0131r. Ayn\u0131 \u015fekilde operat\u00f6rde i\u015flevselli\u011fi i\u00e7in bir ba\u015fka object’e ihtiya\u00e7 duymakta ve t\u00fcm i\u015flemler en alttaki objede ger\u00e7ekle\u015ftirilmektedir. T\u00fcm parametreler bu iskelette kullan\u0131ld\u0131\u011f\u0131ndan dolay\u0131 bu yap\u0131y\u0131 genelleyebilirsiniz.<\/p>\n