ZGC ve Shenandoah GC:<\/strong> Modern GC algoritmalar\u0131d\u0131r, \u00e7ok d\u00fc\u015f\u00fck duraklama s\u00fcreleri (genellikle milisaniyelerin alt\u0131nda) sunarak b\u00fcy\u00fck \u00f6l\u00e7ekli ve y\u00fcksek performansl\u0131 uygulamalar i\u00e7in tasarlanm\u0131\u015flard\u0131r.<\/li>\n<\/ul>\n\n\/\/ Basit bir nesne olu\u015fturma ve referans\u0131n kaybolmas\u0131\npublic class GcOrnegi {\n public static void main(String[] args) {\n String s1 = new String(\"Merhaba\"); \/\/ Nesne 1 olu\u015fturuldu\n String s2 = new String(\"D\u00fcnya\"); \/\/ Nesne 2 olu\u015fturuldu\n\n s1 = s2; \/\/ Nesne 1'e olan referans kayboldu. \"Merhaba\" nesnesi art\u0131k uygun (eligible) hale geldi.\n \/\/ Ancak GC'nin ne zaman \u00e7al\u0131\u015faca\u011f\u0131 JVM'e ba\u011fl\u0131d\u0131r.\n\n s2 = null; \/\/ Nesne 2'ye olan referans da kayboldu. \"D\u00fcnya\" nesnesi de uygun hale geldi.\n\n \/\/ System.gc(); \/\/ JVM'e GC \u00e7al\u0131\u015ft\u0131rmas\u0131 i\u00e7in bir \u00f6neri g\u00f6ndeririz, garanti etmez.\n }\n}\n <\/pre>\n<\/code><\/p>\n
Performans\u0131 optimize etmek i\u00e7in do\u011fru GC algoritmas\u0131n\u0131 se\u00e7mek ve JVM arg\u00fcmanlar\u0131yla ayarlar yapmak (\u00f6rne\u011fin -Xms<\/code>, -Xmx<\/code> ile Heap boyutunu ayarlamak) kritik \u00f6neme sahiptir.<\/p>\n\u00c7oklu \u0130\u015f Par\u00e7ac\u0131\u011f\u0131 (Multithreading) ve E\u015fzamanl\u0131l\u0131k: Performans\u0131 Art\u0131rma Yollar\u0131<\/h2>\n
Modern uygulamalar genellikle e\u015fzamanl\u0131 ve paralel i\u015flemlere ihtiya\u00e7 duyar. Java, bu ihtiyac\u0131 kar\u015f\u0131lamak i\u00e7in g\u00fc\u00e7l\u00fc \u00e7oklu i\u015f par\u00e7ac\u0131\u011f\u0131 (multithreading) ve e\u015fzamanl\u0131l\u0131k (concurrency) mekanizmalar\u0131 sunar. Birden fazla i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131 kullanarak, bir uygulaman\u0131n ayn\u0131 anda farkl\u0131 g\u00f6revleri y\u00fcr\u00fctmesini sa\u011flayabiliriz. Bu, \u00f6zellikle UI tepkiselli\u011fini art\u0131rmak, arka plan g\u00f6revlerini i\u015flemek veya yo\u011fun hesaplama gerektiren i\u015flemleri h\u0131zland\u0131rmak i\u00e7in hayati \u00f6neme sahiptir. Ancak e\u015fzamanl\u0131 programlama, beraberinde senkronizasyon ve veri tutarl\u0131l\u0131\u011f\u0131 gibi yeni zorluklar da getirir.<\/p>\n
Thread Olu\u015fturma ve Y\u00f6netimi<\/h3>\n
Java'da bir i\u015f par\u00e7ac\u0131\u011f\u0131 (thread) olu\u015fturman\u0131n ve \u00e7al\u0131\u015ft\u0131rman\u0131n iki temel yolu vard\u0131r:<\/p>\n
\nThread<\/code> S\u0131n\u0131f\u0131n\u0131 Extend Etmek:<\/strong> Kendi s\u0131n\u0131f\u0131n\u0131z\u0131 java.lang.Thread<\/code> s\u0131n\u0131f\u0131ndan t\u00fcreterek ve run()<\/code> metodunu ge\u00e7ersiz k\u0131larak (override) bir i\u015f par\u00e7ac\u0131\u011f\u0131 olu\u015fturabilirsiniz.<\/li>\nRunnable<\/code> Aray\u00fcz\u00fcn\u00fc Implement Etmek:<\/strong> Kendi s\u0131n\u0131f\u0131n\u0131z\u0131 java.lang.Runnable<\/code> aray\u00fcz\u00fcn\u00fc uygulayarak ve run()<\/code> metodunu implement ederek bir i\u015f par\u00e7ac\u0131\u011f\u0131 olu\u015fturabilirsiniz. Bu daha esnek bir yakla\u015f\u0131md\u0131r \u00e7\u00fcnk\u00fc Java'da sadece tek kal\u0131t\u0131m desteklendi\u011fi i\u00e7in s\u0131n\u0131f\u0131n\u0131z ba\u015fka bir s\u0131n\u0131ftan da kal\u0131t\u0131m alabilir.<\/li>\n<\/ol>\n\n\/\/ Runnable Aray\u00fcz\u00fcn\u00fc Kullanan \u00d6rnek\npublic class MyRunnable implements Runnable {\n private String threadName;\n\n public MyRunnable(String name) {\n this.threadName = name;\n System.out.println(threadName + \" olu\u015fturuluyor...\");\n }\n\n @Override\n public void run() {\n System.out.println(threadName + \" \u00e7al\u0131\u015ft\u0131r\u0131l\u0131yor.\");\n try {\n for (int i = 5; i > 0; i--) {\n System.out.println(\"Thread: \" + threadName + \", \" + i);\n Thread.sleep(500); \/\/ 500 milisaniye bekle\n }\n } catch (InterruptedException e) {\n System.out.println(\"Thread \" + threadName + \" kesintiye u\u011frad\u0131.\");\n }\n System.out.println(\"Thread \" + threadName + \" sonlan\u0131yor.\");\n }\n\n public static void main(String[] args) {\n MyRunnable runnable1 = new MyRunnable(\"Thread-1\");\n Thread t1 = new Thread(runnable1);\n t1.start(); \/\/ Thread'i ba\u015flat\n\n MyRunnable runnable2 = new MyRunnable(\"Thread-2\");\n Thread t2 = new Thread(runnable2);\n t2.start(); \/\/ Thread'i ba\u015flat\n }\n}\n <\/pre>\n<\/code><\/p>\n
start()<\/code> metodu, JVM'in yeni bir i\u015f par\u00e7ac\u0131\u011f\u0131 olu\u015fturmas\u0131n\u0131 ve bu i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n run()<\/code> metodunu \u00e7a\u011f\u0131rmas\u0131n\u0131 sa\u011flar. Do\u011frudan run()<\/code> metodunu \u00e7a\u011f\u0131rmak, kodu ayn\u0131 i\u015f par\u00e7ac\u0131\u011f\u0131 \u00fczerinde \u00e7al\u0131\u015ft\u0131r\u0131r ve yeni bir i\u015f par\u00e7ac\u0131\u011f\u0131 olu\u015fturmaz.<\/p>\nE\u015fzamanl\u0131l\u0131k Sorunlar\u0131 ve \u00c7\u00f6z\u00fcmleri<\/h3>\n
Birden fazla i\u015f par\u00e7ac\u0131\u011f\u0131 ayn\u0131 payla\u015f\u0131lan kayna\u011fa (de\u011fi\u015fken, dosya, veritaban\u0131 ba\u011flant\u0131s\u0131 vb.) eri\u015fmeye \u00e7al\u0131\u015ft\u0131\u011f\u0131nda \u00e7e\u015fitli e\u015fzamanl\u0131l\u0131k sorunlar\u0131 ortaya \u00e7\u0131kabilir. En yayg\u0131n sorunlar \u015funlard\u0131r:<\/p>\n
\n- Race Condition (Yar\u0131\u015f Durumu):<\/strong> Birden fazla i\u015f par\u00e7ac\u0131\u011f\u0131 payla\u015f\u0131lan bir kayna\u011f\u0131 e\u015fzamanl\u0131 olarak de\u011fi\u015ftirmeye \u00e7al\u0131\u015ft\u0131\u011f\u0131nda, sonucun i\u015f par\u00e7ac\u0131klar\u0131n\u0131n \u00e7al\u0131\u015fma s\u0131ras\u0131na ba\u011fl\u0131 olmas\u0131 durumudur. Bu, \u00f6ng\u00f6r\u00fclemeyen ve hatal\u0131 sonu\u00e7lara yol a\u00e7abilir.<\/li>\n
- Deadlock (Kilitlenme):<\/strong> \u0130ki veya daha fazla i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n, birbirlerinin sal\u0131vermesini bekledi\u011fi ancak hi\u00e7birinin asla sal\u0131vermeyece\u011fi bir durumu ifade eder. Her i\u015f par\u00e7ac\u0131\u011f\u0131 bir kayna\u011f\u0131 tutar ve di\u011ferinin tuttu\u011fu kayna\u011f\u0131 bekler.<\/li>\n
- Livelock:<\/strong> \u0130\u015f par\u00e7ac\u0131klar\u0131n\u0131n kilitlenmedi\u011fi ancak ilerleme kaydedemedi\u011fi bir durumdur. S\u00fcrekli olarak durumlar\u0131n\u0131 de\u011fi\u015ftirirler ancak hi\u00e7bir i\u015f yapamazlar.<\/li>\n
- Starvation (A\u00e7l\u0131k):<\/strong> Bir veya daha fazla i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n, kaynaklara veya i\u015flem s\u00fcresine s\u00fcrekli olarak eri\u015fememesi durumudur, \u00e7\u00fcnk\u00fc di\u011fer i\u015f par\u00e7ac\u0131klar\u0131 s\u00fcrekli olarak bu kaynaklar\u0131 kullan\u0131r.<\/li>\n<\/ul>\n
Bu sorunlar\u0131 \u00e7\u00f6zmek i\u00e7in Java'da \u00e7e\u015fitli senkronizasyon mekanizmalar\u0131 mevcuttur:<\/p>\n
\nsynchronized<\/code> Anahtar Kelimesi:<\/strong> Metotlar\u0131 veya kod bloklar\u0131n\u0131 senkronize etmek i\u00e7in kullan\u0131l\u0131r. Bir i\u015f par\u00e7ac\u0131\u011f\u0131 senkronize bir metot veya blo\u011fa girdi\u011finde, ilgili nesnenin (veya statik metotlar i\u00e7in s\u0131n\u0131f\u0131n) kilidini al\u0131r. Ba\u015fka bir i\u015f par\u00e7ac\u0131\u011f\u0131 ayn\u0131 kilidi almaya \u00e7al\u0131\u015ft\u0131\u011f\u0131nda, kilit serbest b\u0131rak\u0131lana kadar bekler. Bu, Race Condition'lar\u0131 \u00f6nler.<\/li>\nvolatile<\/code> Anahtar Kelimesi:<\/strong> Bir de\u011fi\u015fkene ait de\u011ferin her zaman ana bellekten okunup, ana belle\u011fe yaz\u0131lmas\u0131n\u0131 sa\u011flar. Bu, birden fazla i\u015f par\u00e7ac\u0131\u011f\u0131n\u0131n ayn\u0131 de\u011fi\u015fkene eri\u015fti\u011fi durumlarda g\u00f6r\u00fcn\u00fcrl\u00fck sorunlar\u0131n\u0131 \u00e7\u00f6zer. Ancak atomik operasyonlar\u0131 garanti etmez.<\/li>\njava.util.concurrent<\/code> Paketi:<\/strong> Java 5 ile birlikte gelen bu paket, e\u015fzamanl\u0131 programlamay\u0131 basitle\u015ftiren ve daha verimli hale getiren geli\u015fmi\u015f ara\u00e7lar sunar. \u00d6rnekler:\n\nExecutorService<\/code> ve ThreadPoolExecutor<\/code>:<\/strong> \u0130\u015f par\u00e7ac\u0131klar\u0131n\u0131 manuel olarak y\u00f6netmek yerine, bir havuzdan al\u0131p kullanmay\u0131 sa\u011flar.<\/li>\nSemaphore<\/code>:<\/strong> Belirli bir say\u0131da kayna\u011fa e\u015fzamanl\u0131 eri\u015fimi kontrol eder.<\/li>\nCountDownLatch<\/code> ve CyclicBarrier<\/code>:<\/strong> \u0130\u015f par\u00e7ac\u0131klar\u0131n\u0131n birbirlerini beklemesini sa\u011flayan senkronizasyon yard\u0131mc\u0131lar\u0131d\u0131r.<\/li>\nConcurrentHashMap<\/code>:<\/strong> Thread-safe ve y\u00fcksek performansl\u0131 bir Map<\/code> implementasyonudur.<\/li>\nAtomic<\/code> S\u0131n\u0131flar (AtomicInteger<\/code>, AtomicLong<\/code>):<\/strong> K\u00fc\u00e7\u00fck \u00f6l\u00e7ekli atomik operasyonlar i\u00e7in kullan\u0131l\u0131r, synchronized<\/code> bloklar\u0131ndan daha hafif bir alternatiftir.<\/li>\n<\/ul>\n<\/li>\nLock<\/code> Aray\u00fcz\u00fc (ReentrantLock<\/code>):<\/strong> synchronized<\/code> anahtar kelimesine g\u00f6re daha esnek kilit mekanizmalar\u0131 sunar. Ko\u015fullu beklemeler (Condition<\/code>), kilit zaman a\u015f\u0131m\u0131 gibi \u00f6zellikler sa\u011flar.<\/li>\n<\/ol>\n\n\/\/ Synchronized Metot \u00d6rne\u011fi\npublic class Sayac {\n private int count = 0;\n\n public synchronized void increment() { \/\/ Metot senkronize edildi\n count++;\n }\n\n public int getCount() {\n return count;\n }\n\n public static void main(String[] args) throws InterruptedException {\n Sayac sayac = new Sayac();\n\n Runnable task = () -> {\n for (int i = 0; i < 1000; i++) {\n sayac.increment();\n }\n };\n\n Thread t1 = new Thread(task);\n Thread t2 = new Thread(task);\n\n t1.start();\n t2.start();\n\n t1.join(); \/\/ t1 bitene kadar bekle\n t2.join(); \/\/ t2 bitene kadar bekle\n\n System.out.println(\"Son Say\u0131: \" + sayac.getCount()); \/\/ Her zaman 2000 olmal\u0131\n }\n}\n <\/pre>\n<\/code><\/p>\n
E\u015fzamanl\u0131 programlama karma\u015f\u0131kt\u0131r, ancak do\u011fru ara\u00e7lar ve yakla\u015f\u0131mlarla g\u00fcvenli, verimli ve \u00f6l\u00e7eklenebilir uygulamalar geli\u015ftirmek m\u00fcmk\u00fcnd\u00fcr. \u00d6zellikle kritik b\u00f6l\u00fcmleri do\u011fru bir \u015fekilde senkronize etmek, hatalar\u0131 \u00f6nlemek ve uygulaman\u0131n tutarl\u0131l\u0131\u011f\u0131n\u0131 sa\u011flamak i\u00e7in temel bir beceridir.<\/p>\n
Java Koleksiyon \u00c7at\u0131s\u0131 ve Stream API: Veri \u0130\u015flemenin G\u00fcc\u00fc<\/h2>\n
Java Koleksiyon \u00c7at\u0131s\u0131 (Java Collections Framework - JCF), veri yap\u0131lar\u0131n\u0131 y\u00f6netmek i\u00e7in bir dizi aray\u00fcz ve s\u0131n\u0131f sunar. Bu \u00e7er\u00e7eve, geli\u015ftiricilere veri depolama, eri\u015fim, s\u0131ralama ve manip\u00fclasyon i\u00e7in g\u00fc\u00e7l\u00fc ve esnek ara\u00e7lar sa\u011flar. JCF'in yan\u0131 s\u0131ra, Java 8 ile tan\u0131t\u0131lan Stream API, veri k\u00fcmeleri \u00fczerinde fonksiyonel programlama tarz\u0131nda i\u015flemler yapmay\u0131 kolayla\u015ft\u0131r\u0131r ve modern Java geli\u015ftirmenin vazge\u00e7ilmez bir par\u00e7as\u0131 haline gelmi\u015ftir. Bu iki konu, m\u00fclakatlarda s\u0131kl\u0131kla ele al\u0131nan ve Java geli\u015ftiricilerin g\u00fcnl\u00fck i\u015flerinde yo\u011fun olarak kulland\u0131\u011f\u0131 alanlard\u0131r.<\/p>\n
Java Koleksiyonlar\u0131 Nelerdir?<\/h3>\n
JCF, temel olarak \u00fc\u00e7 ana aray\u00fcz etraf\u0131nda d\u00f6ner: List<\/code>, Set<\/code> ve Map<\/code>. Her birinin kendine \u00f6zg\u00fc kullan\u0131m senaryolar\u0131 ve davran\u0131\u015flar\u0131 vard\u0131r:<\/p>\n\nList<\/code> Aray\u00fcz\u00fc:<\/strong>\n\n- S\u0131ral\u0131 elemanlar koleksiyonudur. Elemanlar eklenme s\u0131ras\u0131na g\u00f6re depolan\u0131r ve indeksleri arac\u0131l\u0131\u011f\u0131yla eri\u015filebilir.<\/li>\n
- Tekrarlayan elemanlara izin verir.<\/li>\n
- Ba\u015fl\u0131ca implementasyonlar:\n
\nArrayList<\/code>: Dinamik bir dizi olarak implement edilmi\u015ftir. Rastgele eri\u015fim h\u0131zl\u0131d\u0131r (O(1)), ancak ortas\u0131na eleman ekleme veya \u00e7\u0131karma maliyetli olabilir (O(n)).<\/li>\nLinkedList<\/code>: \u00c7ift y\u00f6nl\u00fc ba\u011fl\u0131 liste olarak implement edilmi\u015ftir. Ba\u015f\u0131na veya sonuna eleman ekleme\/\u00e7\u0131karma h\u0131zl\u0131d\u0131r (O(1)), ancak rastgele eri\u015fim yava\u015ft\u0131r (O(n)).<\/li>\nVector<\/code>: ArrayList<\/code>'e benzer ancak t\u00fcm metotlar\u0131 synchronized<\/code>'dir, bu da onu thread-safe yapar ancak performans maliyetine neden olur. Genellikle daha modern e\u015fzamanl\u0131 koleksiyonlar tercih edilir.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\nSet<\/code> Aray\u00fcz\u00fc:<\/strong>\n\n- Tekrarlayan elemanlara izin vermeyen bir koleksiyondur. Her eleman benzersiz olmal\u0131d\u0131r.<\/li>\n
- Elemanlar\u0131n belirli bir s\u0131ras\u0131 garanti edilmez (
LinkedHashSet<\/code> hari\u00e7).<\/li>\n- Ba\u015fl\u0131ca implementasyonlar:\n
\nHashSet<\/code>: Elemanlar\u0131n s\u0131ralamas\u0131n\u0131 garanti etmez. Performans a\u00e7\u0131s\u0131ndan genellikle en iyisidir, eleman ekleme, silme ve arama O(1) ortalama karma\u015f\u0131kl\u0131\u011fa sahiptir.<\/li>\nLinkedHashSet<\/code>: Elemanlar\u0131n eklenme s\u0131ras\u0131n\u0131 korur. Performans\u0131 HashSet<\/code>'e yak\u0131nd\u0131r.<\/li>\nTreeSet<\/code>: Elemanlar\u0131 do\u011fal s\u0131ralamalar\u0131na g\u00f6re veya belirtilen bir Comparator<\/code>'a g\u00f6re s\u0131ralar. Operasyonlar\u0131 O(log n) karma\u015f\u0131kl\u0131\u011fa sahiptir.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\nMap<\/code> Aray\u00fcz\u00fc:<\/strong>\n\n- Anahtar-de\u011fer \u00e7iftleri depolayan bir yap\u0131d\u0131r. Her anahtar benzersiz olmal\u0131d\u0131r, ancak de\u011ferler tekrarlanabilir.<\/li>\n
- Anahtarlar arac\u0131l\u0131\u011f\u0131yla de\u011ferlere eri\u015filir.<\/li>\n
- Ba\u015fl\u0131ca implementasyonlar:\n
\nHashMap<\/code>: Anahtarlar\u0131n s\u0131ras\u0131n\u0131 garanti etmez. Performans\u0131 HashSet<\/code>'e benzer, O(1) ortalama karma\u015f\u0131kl\u0131\u011fa sahiptir.<\/li>\nLinkedHashMap<\/code>: Anahtarlar\u0131n eklenme s\u0131ras\u0131n\u0131 korur.<\/li>\nTreeMap<\/code>: Anahtarlar\u0131 do\u011fal s\u0131ralamalar\u0131na g\u00f6re veya belirtilen bir Comparator<\/code>'a g\u00f6re s\u0131ralar. Operasyonlar\u0131 O(log n) karma\u015f\u0131kl\u0131\u011fa sahiptir.<\/li>\nHashTable<\/code>: HashMap<\/code>'e benzer ancak t\u00fcm metotlar\u0131 synchronized<\/code>'dir (thread-safe). Eski bir implementasyondur ve genellikle ConcurrentHashMap<\/code> tercih edilir.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\/\/ Koleksiyon \u00d6rnekleri\nimport java.util.*;\n\npublic class KoleksiyonOrnekleri {\n public static void main(String[] args) {\n \/\/ List (S\u0131ral\u0131, Tekrarlayan Elemanlara \u0130zin Verir)\n List isimler = new ArrayList<>();\n isimler.add(\"Ay\u015fe\");\n isimler.add(\"Mehmet\");\n isimler.add(\"Ay\u015fe\");\n System.out.println(\"List: \" + isimler); \/\/ [Ay\u015fe, Mehmet, Ay\u015fe]\n\n \/\/ Set (S\u0131ras\u0131z, Benzersiz Elemanlar)\n Set benzersizIsimler = new HashSet<>();\n benzersizIsimler.add(\"Ay\u015fe\");\n benzersizIsimler.add(\"Mehmet\");\n benzersizIsimler.add(\"Ay\u015fe\"); \/\/ Bu eklenmez\n System.out.println(\"Set: \" + benzersizIsimler); \/\/ [Ay\u015fe, Mehmet] (s\u0131ra garanti de\u011fil)\n\n \/\/ Map (Anahtar-De\u011fer \u00c7iftleri)\n Map yaslar = new HashMap<>();\n yaslar.put(\"Ay\u015fe\", 30);\n yaslar.put(\"Mehmet\", 25);\n yaslar.put(\"Fatma\", 30);\n System.out.println(\"Map: \" + yaslar); \/\/ {Ay\u015fe=30, Mehmet=25, Fatma=30} (s\u0131ra garanti de\u011fil)\n System.out.println(\"Mehmet'in Ya\u015f\u0131: \" + yaslar.get(\"Mehmet\"));\n }\n}\n <\/pre>\n<\/code><\/p>\n
Stream API ile Veri D\u00f6n\u00fc\u015f\u00fcm\u00fc<\/h3>\n
Java 8 ile birlikte tan\u0131t\u0131lan Stream API, koleksiyonlar \u00fczerinde fonksiyonel tarzda i\u015flem yapmay\u0131 sa\u011flayan g\u00fc\u00e7l\u00fc bir ara\u00e7t\u0131r. Veri i\u015fleme operasyonlar\u0131n\u0131 daha okunakl\u0131, \u00f6zl\u00fc ve \u00e7o\u011fu durumda daha performansl\u0131 hale getirir. Bir Stream, bir veri kayna\u011f\u0131 (koleksiyon, dizi, I\/O kanallar\u0131 vb.) \u00fczerinde ard\u0131\u015f\u0131k veya paralel olarak i\u015flem yapmaya olanak tan\u0131r.<\/p>\n
Stream operasyonlar\u0131 iki ana kategoriye ayr\u0131l\u0131r:<\/p>\n
\n- Arac\u0131 Operasyonlar (Intermediate Operations):<\/strong> Bir Stream d\u00f6nd\u00fcr\u00fcrler ve genellikle ba\u015fka Stream operasyonlar\u0131yla zincirlenebilirler (
filter()<\/code>, map()<\/code>, sorted()<\/code>, distinct()<\/code>, limit()<\/code>). Bu operasyonlar tembel (lazy)dir, yani bir sonland\u0131rma operasyonu \u00e7a\u011fr\u0131lana kadar y\u00fcr\u00fct\u00fclmezler.<\/li>\n- Sonland\u0131rma Operasyonlar\u0131 (Terminal Operations):<\/strong> Bir sonu\u00e7 (
count()<\/code>, sum()<\/code>), bir yan etki (forEach()<\/code>) veya bir koleksiyon (collect()<\/code>) d\u00f6nd\u00fcr\u00fcrler. Bu operasyonlar Stream zincirini kapat\u0131r ve t\u00fcm arac\u0131 operasyonlar\u0131n y\u00fcr\u00fct\u00fclmesini tetikler.<\/li>\n<\/ol>\n\n\/\/ Stream API \u00d6rne\u011fi\nimport java.util.Arrays;\nimport java.util.List;\nimport java.util.stream.Collectors;\n\npublic class StreamApiOrnegi {\n public static void main(String[] args) {\n List kelimeler = Arrays.asList(\"elma\", \"armut\", \"kiraz\", \"\u00fcz\u00fcm\", \"nar\", \"kivi\", \"erik\");\n\n \/\/ Kelimelerden 'k' ile ba\u015flayanlar\u0131 filtrele, b\u00fcy\u00fck harfe \u00e7evir ve listeye topla\n List kIleBaslayanlarBuyukHarf = kelimeler.stream()\n .filter(s -> s.startsWith(\"k\")) \/\/ 'k' ile ba\u015flayanlar\u0131 filtrele (intermediate)\n .map(String::toUpperCase) \/\/ B\u00fcy\u00fck harfe \u00e7evir (intermediate)\n .collect(Collectors.toList()); \/\/ Sonu\u00e7lar\u0131 yeni bir Listeye topla (terminal)\n\n System.out.println(\"K ile ba\u015flayanlar (b\u00fcy\u00fck harf): \" + kIleBaslayanlarBuyukHarf); \/\/ [K\u0130RAZ, K\u0130V\u0130]\n\n \/\/ Uzunlu\u011fu 4'ten b\u00fcy\u00fck olan kelimeleri say\n long uzunKelimelerSayisi = kelimeler.stream()\n .filter(s -> s.length() > 4) \/\/ Uzunlu\u011fu 4'ten b\u00fcy\u00fck olanlar\u0131 filtrele\n .count(); \/\/ Say (terminal)\n\n System.out.println(\"Uzunlu\u011fu 4'ten b\u00fcy\u00fck kelime say\u0131s\u0131: \" + uzunKelimelerSayisi); \/\/ 3 (armut, kiraz, \u00fcz\u00fcm)\n\n \/\/ Benzersiz elemanlar\u0131 al ve alfabetik s\u0131rala\n List benzersizVeSirali = Arrays.asList(\"elma\", \"armut\", \"elma\", \"kiraz\", \"armut\").stream()\n .distinct() \/\/ Benzersiz elemanlar\u0131 al\n .sorted() \/\/ Alfabetik s\u0131rala\n .collect(Collectors.toList());\n System.out.println(\"Benzersiz ve S\u0131ral\u0131: \" + benzersizVeSirali); \/\/ [armut, elma, kiraz]\n }\n}\n <\/pre>\n<\/code><\/p>\n
\nUzman \u0130pucu: Stream API'nin paralel (parallelStream()<\/code>) versiyonunu kullan\u0131rken dikkatli olun. K\u00fc\u00e7\u00fck veri k\u00fcmelerinde veya G\/\u00c7 yo\u011fun i\u015flemlerde performans art\u0131\u015f\u0131 sa\u011flamayabilir, hatta senkronizasyon maliyetleri nedeniyle performans\u0131 d\u00fc\u015f\u00fcrebilir. Yaln\u0131zca CPU yo\u011fun ve b\u00fcy\u00fck veri k\u00fcmeleri i\u00e7in paralel Stream'leri de\u011ferlendirin.<\/p>\n<\/blockquote>\nKoleksiyon \u00c7at\u0131s\u0131 ve Stream API, Java uygulamalar\u0131nda veri i\u015fleme ve manip\u00fclasyonun temelini olu\u015fturur. Bu konular\u0131 iyi anlamak, daha temiz, daha etkili ve daha modern Java kodu yazman\u0131z i\u00e7in size gerekli becerileri kazand\u0131racakt\u0131r.<\/p>\n
\u0130leri D\u00fczey Konular: API'ler, Frameworkler ve Mimari Yakla\u015f\u0131mlar<\/h2>\n
Java ekosistemi, temel dil \u00f6zelliklerinin \u00e7ok \u00f6tesine ge\u00e7en zengin bir k\u00fct\u00fcphaneler, API'ler ve frameworkler d\u00fcnyas\u0131na sahiptir. Bir Java geli\u015ftiricinin kariyerinde ilerlemesi i\u00e7in, bu ileri d\u00fczey konulara hakim olmas\u0131 ve ger\u00e7ek d\u00fcnya projelerinde bunlar\u0131 nas\u0131l uygulayaca\u011f\u0131n\u0131 bilmesi vazge\u00e7ilmezdir. M\u00fclakatlarda, \u00f6zellikle orta ve k\u0131demli seviye pozisyonlar i\u00e7in bu konulara dair derinlemesine bilgi beklenir. Veritaban\u0131 ba\u011flant\u0131lar\u0131ndan modern web servislerine, b\u00fcy\u00fck \u00f6l\u00e7ekli uygulamalar\u0131n mimari yakla\u015f\u0131mlar\u0131na kadar geni\u015f bir yelpaze bu b\u00f6l\u00fcmde ele al\u0131n\u0131r.<\/p>\n
JDBC ile Veritaban\u0131 Ba\u011flant\u0131s\u0131 Nas\u0131l Kurulur?<\/h3>\n
JDBC (Java Database Connectivity), Java uygulamalar\u0131n\u0131n veritabanlar\u0131yla etkile\u015fim kurmas\u0131n\u0131 sa\u011flayan bir API'dir. Farkl\u0131 veritaban\u0131 t\u00fcrleriyle (MySQL, PostgreSQL, Oracle, SQL Server vb.) standart bir aray\u00fcz \u00fczerinden ileti\u015fim kurmay\u0131 m\u00fcmk\u00fcn k\u0131lar. JDBC kullanarak bir veritaban\u0131na ba\u011flanmak i\u00e7in genel ad\u0131mlar \u015funlard\u0131r:<\/p>\n
\n- S\u00fcr\u00fcc\u00fcy\u00fc Y\u00fckle:<\/strong> \u0130lgili veritaban\u0131 i\u00e7in JDBC s\u00fcr\u00fcc\u00fcs\u00fcn\u00fc (JAR dosyas\u0131) projenizin classpath'ine ekleyin.<\/li>\n
- Ba\u011flant\u0131 Kur:<\/strong>
DriverManager.getConnection()<\/code> metodunu kullanarak veritaban\u0131 URL'si, kullan\u0131c\u0131 ad\u0131 ve \u015fifre ile bir Connection<\/code> nesnesi olu\u015fturun.<\/li>\n- Statement Olu\u015ftur:<\/strong> SQL sorgular\u0131n\u0131 \u00e7al\u0131\u015ft\u0131rmak i\u00e7in
Connection<\/code> nesnesinden bir Statement<\/code> veya PreparedStatement<\/code> nesnesi elde edin. PreparedStatement<\/code>, SQL enjeksiyon sald\u0131r\u0131lar\u0131n\u0131 \u00f6nlemek ve performans\u0131 art\u0131rmak i\u00e7in parametrele\u015ftirilmi\u015f sorgulara olanak tan\u0131r.<\/li>\n- Sorguyu \u00c7al\u0131\u015ft\u0131r:<\/strong> SQL sorgular\u0131n\u0131 (
executeQuery()<\/code> i\u00e7in SELECT, executeUpdate()<\/code> i\u00e7in INSERT, UPDATE, DELETE) \u00e7al\u0131\u015ft\u0131r\u0131n.<\/li>\n- Sonu\u00e7lar\u0131 \u0130\u015fle:<\/strong> SELECT sorgular\u0131 i\u00e7in d\u00f6nen
ResultSet<\/code> nesnesini kullanarak verileri i\u015fleyin.<\/li>\n- Kaynaklar\u0131 Kapat:<\/strong> En \u00f6nemlisi, kullan\u0131lan t\u00fcm kaynaklar\u0131 (
ResultSet<\/code>, Statement<\/code>, Connection<\/code>) ters s\u0131rada kapat\u0131n. Genellikle try-with-resources<\/code> yap\u0131s\u0131 bu i\u015flemi otomatik olarak halleder ve kaynak s\u0131z\u0131nt\u0131lar\u0131n\u0131 \u00f6nler.<\/li>\n<\/ol>\n\n\/\/ JDBC Ba\u011flant\u0131 \u00d6rne\u011fi (try-with-resources ile)\nimport java.sql.*;\n\npublic class JdbcOrnek {\n private static final String DB_URL = \"jdbc:mysql:\/\/localhost:3306\/testdb\";\n private static final String USER = \"root\";\n private static final String PASS = \"password\";\n\n public static void main(String[] args) {\n \/\/ try-with-resources statement: Kaynaklar otomatik kapan\u0131r\n try (Connection conn = DriverManager.getConnection(DB_URL, USER, PASS);\n Statement stmt = conn.createStatement();\n ResultSet rs = stmt.executeQuery(\"SELECT id, name, email FROM users\")) {\n\n while (rs.next()) {\n \/\/ ResultSet'ten verileri al\n int id = rs.getInt(\"id\");\n String name = rs.getString(\"name\");\n String email = rs.getString(\"email\");\n\n System.out.println(\"ID: \" + id + \", \u0130sim: \" + name + \", Email: \" + email);\n }\n } catch (SQLException e) {\n e.printStackTrace();\n }\n }\n}\n <\/pre>\n<\/code><\/p>\n
Modern uygulamalarda genellikle JDBC'nin do\u011frudan kullan\u0131m\u0131 yerine, Spring Data JPA veya Hibernate gibi ORM (Object-Relational Mapping) frameworkleri tercih edilir. Ancak JDBC'nin temellerini anlamak, bu frameworklerin nas\u0131l \u00e7al\u0131\u015ft\u0131\u011f\u0131n\u0131 kavramak i\u00e7in \u00f6nemlidir.<\/p>\n
Spring ve Hibernate Neden \u00d6nemlidir?<\/h3>\n
Spring Framework:<\/strong> Java geli\u015ftirmenin de facto standard\u0131 haline gelmi\u015f, kapsaml\u0131 bir uygulama geli\u015ftirme framework'\u00fcd\u00fcr. Mod\u00fcler yap\u0131s\u0131 sayesinde, web uygulamalar\u0131ndan mikroservislere, veri eri\u015fiminden g\u00fcvenli\u011fe kadar bir\u00e7ok alanda kullan\u0131labilir. Spring'i \u00f6nemli k\u0131lan temel \u00f6zellikler:<\/p>\n\n- Ba\u011f\u0131ml\u0131l\u0131k Enjeksiyonu (Dependency Injection - DI) \/ Inversion of Control (IoC):<\/strong> Uygulama bile\u015fenlerinin ba\u011f\u0131ml\u0131l\u0131klar\u0131n\u0131 y\u00f6netmeyi kolayla\u015ft\u0131r\u0131r. Kodun daha mod\u00fcler, test edilebilir ve yeniden kullan\u0131labilir olmas\u0131n\u0131 sa\u011flar.<\/li>\n
- Aspect-Oriented Programming (AOP):<\/strong> Logging, g\u00fcvenlik, transaction y\u00f6netimi gibi cross-cutting (\u00e7apraz kesen) endi\u015feleri ana i\u015f mant\u0131\u011f\u0131ndan ay\u0131rmay\u0131 sa\u011flar.<\/li>\n
- Spring Boot:<\/strong> Spring tabanl\u0131 uygulamalar\u0131 h\u0131zl\u0131 bir \u015fekilde kurmak ve \u00e7al\u0131\u015ft\u0131rmak i\u00e7in bir ara\u00e7t\u0131r. Otomatik yap\u0131land\u0131rma, g\u00f6m\u00fcl\u00fc sunucular ve \"starter\" ba\u011f\u0131ml\u0131l\u0131klar\u0131 ile geli\u015ftirme s\u00fcrecini b\u00fcy\u00fck \u00f6l\u00e7\u00fcde basitle\u015ftirir.<\/li>\n
- Spring MVC \/ Spring WebFlux:<\/strong> Web uygulamalar\u0131 ve RESTful API'ler geli\u015ftirmek i\u00e7in g\u00fc\u00e7l\u00fc mod\u00fcller sunar.<\/li>\n
- Spring Data JPA:<\/strong> Veritaban\u0131 i\u015flemlerini ORM (Hibernate gibi) kullanarak \u00e7ok daha az kodla yapmay\u0131 sa\u011flayan bir abstraksiyon katman\u0131d\u0131r.<\/li>\n<\/ul>\n
Hibernate:<\/strong> Java i\u00e7in \u00f6nde gelen bir ORM (Object-Relational Mapping) arac\u0131d\u0131r. Nesneye dayal\u0131 bir Java modelini ili\u015fkisel bir veritaban\u0131 \u015femas\u0131yla e\u015fle\u015ftirerek, geli\u015ftiricilerin SQL sorgular\u0131 yazmak yerine Java nesneleriyle \u00e7al\u0131\u015fmas\u0131na olanak tan\u0131r. Hibernate'in faydalar\u0131:<\/p>\n\n- Veritaban\u0131 Abstraksiyonu:<\/strong> Veritaban\u0131 t\u00fcr\u00fcnden ba\u011f\u0131ms\u0131z geli\u015ftirme imkan\u0131 sunar. Veritaban\u0131 de\u011fi\u015fti\u011finde kodda b\u00fcy\u00fck de\u011fi\u015fiklikler yapma gereksinimini azalt\u0131r.<\/li>\n
- CRUD \u0130\u015flemleri:<\/strong> Nesneleri veritaban\u0131na kaydetmek, g\u00fcncellemek, silmek ve okumak i\u00e7in kolay bir API sa\u011flar.<\/li>\n
- Performans Optimizasyonlar\u0131:<\/strong> Otomatik \u00f6nbellekleme (caching) ve lazy loading gibi \u00f6zelliklerle performans\u0131 art\u0131rabilir.<\/li>\n