woyanyouxin 2016-04-07
调用Zookeeper的构造方法来实例化一个Zookeeper。我们以一个构造方法进行一步一步的分析。
下面是一个Zookeeper的构造方法:
public ZooKeeper(String connectString, int sessionTimeout, Watcher watcher, boolean canBeReadOnly, HostProvider aHostProvider) throws IOException { LOG.info("Initiating client connection, connectString=" + connectString + " sessionTimeout=" + sessionTimeout + " watcher=" + watcher); watchManager = defaultWatchManager(); watchManager.defaultWatcher = watcher; ConnectStringParser connectStringParser = new ConnectStringParser( connectString); hostProvider = aHostProvider; cnxn = new ClientCnxn(connectStringParser.getChrootPath(), hostProvider, sessionTimeout, this, watchManager, getClientCnxnSocket(), canBeReadOnly); cnxn.start(); }
在《Zoopkeeper-会话创建流程》文章中已经介绍了初始化阶段所做的事情。这里就从每行代码开始分析。
第一行代码:
LOG.info("Initiating client connection, connectString=" + connectString + " sessionTimeout=" + sessionTimeout + " watcher=" + watcher);
这了一看就是在进行日志的记录。日志我们每个项目都在,而且定义LOG的方式也见过很多。这里看看zookeeper是如何创建的,并且思考why!跟踪的代码如下:
private static final Logger LOG; static { //Keep these two lines together to keep the initialization order explicit LOG = LoggerFactory.getLogger(ZooKeeper.class); Environment.logEnv("Client environment:", LOG); }
LOG的定义是static final 形式的。那么为什么会这样定义呢?那就需要知道这样定义的好处。网上找到了2片文章写的不错。地址:
总结一下就是这样定义可以提高性能。
这里除了初始化一个LOG还多了一行代码:Environment.logEnv("Client environment:", LOG);这行代码会对客户端一些环境参数进行日志记录。哪些是环境参数呢?譬如客户端ip地址、内存。那么java中怎么获取这些信息。平时很少会关注这样的方式。今天看源码就当扩展一下对java基础类的了解。
put(l, "host.name",InetAddress.getLocalHost().getCanonicalHostName()); 获取主机地址 // Get memory information. Runtime runtime = Runtime.getRuntime(); int mb = 1024 * 1024; put(l, "os.memory.free",Long.toString(runtime.freeMemory() / mb) + "MB"); jvm空闲的内存 put(l, "os.memory.max",Long.toString(runtime.maxMemory() / mb) + "MB");jvm能获取的最大内存 put(l, "os.memory.total",Long.toString(runtime.totalMemory() / mb) + "MB");jvm当前占用的内存
这行代码引发了我的一个思考:当我们开发的不是web程序时,是否需要对一些环境信息进行一些记录呢。
接下来的2行代码:
watchManager = defaultWatchManager(); watchManager.defaultWatcher = watcher;
创建了一个ZKWatchManager对象,然后把watcher赋值给了ZKWatchManager对象的defaultWatcher 属性。ZKWatchManager实现了ClientWatchManager接口。也就是创建了一个客户端Watcher管理器。
接下来的3行代码:
ConnectStringParser connectStringParser = new ConnectStringParser( connectString);
首先创建了一个ConnectStringParser类,从字面上可以理解这个类是一个连接字符串的解析类。跟踪代码去了解一下这类的真是面貌:
首先印入眼帘的就是这个类是final类型的。
/** * A parser for ZooKeeper Client connect strings. * * This class is not meant to be seen or used outside of ZooKeeper itself. * * The chrootPath member should be replaced by a Path object in issue * ZOOKEEPER-849. * * @see org.apache.zookeeper.ZooKeeper */ public final class ConnectStringParser
注解写的很清楚,这个类不能被ZooKeeper 以外的类看见或者使用。这个类具体解析连接字符串的逻辑是在构造函数中编写的。这里对字符串的处理并非使用了String自带的split方法。而是使用Zookeeper项目中StringUtils类。看来对字符串的处理是每个项目都必须要做的啊。我见过的项目几乎都会自己写一个名叫StringUtils的类。我们看看Zookeeper怎么实现split方法的。
/** * This method returns an immutable List<String>, but different from String's split() * it trims the results in the input String, and removes any empty string from * the resulting List. * */ public static List<String> split(String value, String separator) { String[] splits = value.split(separator); List<String> results = new ArrayList<String>(); for (int i = 0; i < splits.length; i++) { splits[i] = splits[i].trim(); if (splits[i].length() > 0) { results.add(splits[i]); } } return Collections.unmodifiableList(results); }
从注解上看只是对字符串进行了一些自己业务需求的处理。底层用到的还是String的split方法。但是有一点是值的关注的。那就是返回的是immutable 集合。记得在《重构》这本书有一种思想叫做:Encapsulate Collection (封装集群) 。也就是这个集合在返回之后是不允许再修改的。那么调用者怎么知道不能修改呢?所以这里使用了Collections.unmodifiableList方法返回一个不可变的集合。如果修改则会抛出java.lang.UnsupportedOperationException。使用这个方式之后代码变的更加优美,更加健壮。
我们再回到ConnectStringParser 类中,最终地址是放到了一个集合中:serverAddresses。而这个集合的初始化时这样的:
private final ArrayList<InetSocketAddress> serverAddresses = new ArrayList<InetSocketAddress>();
再次用到了final字段。目的应该就是上面文章中分析的。然而这里使用到了InetSocketAddress。对于IP地址和端口的存放,Zookeeper是存放在了jdk net包中的InetSocketAddress。存放的代码为:serverAddresses.add(InetSocketAddress.createUnresolved(host, port)); 如果是我的话,我可能就直接自己定义一个类表示serverAddress了。我理解使用这个类的好处有:
接下来的4行代码:
hostProvider = aHostProvider;
在ZooKeeper的构造函数中最后一个参数是传递一个HostProvider(接口)。然而一般不会自己去创建,ZooKeeper自己有默认的创建实现:
// default hostprovider private static HostProvider createDefaultHostProvider(String connectString) { return new StaticHostProvider( new ConnectStringParser(connectString).getServerAddresses()); }
最终返回的是StaticHostProvider(HostProvider的实现类)。这个构造方法实现了些什么呢,我们继续跟踪看看:
/** * Most simple HostProvider, resolves only on instantiation. * */ public final class StaticHostProvider implements HostProvider 首先这个类也是final 类型的。其次是我们要看的构造方法: public StaticHostProvider(Collection<InetSocketAddress> serverAddresses) { sourceOfRandomness = new Random(System.currentTimeMillis() ^ this.hashCode()); this.serverAddresses = resolveAndShuffle(serverAddresses); if (this.serverAddresses.isEmpty()) { throw new IllegalArgumentException( "A HostProvider may not be empty!"); } currentIndex = -1; lastIndex = -1; }
这里构造方法大致可以理解为:构造一个随机种子,然后对服务器地址进行解析和打乱操作:
private List<InetSocketAddress> resolveAndShuffle(Collection<InetSocketAddress> serverAddresses) { List<InetSocketAddress> tmpList = new ArrayList<InetSocketAddress>(serverAddresses.size()); for (InetSocketAddress address : serverAddresses) { try { InetAddress ia = address.getAddress(); String addr = (ia != null) ? ia.getHostAddress() : address.getHostString(); InetAddress resolvedAddresses[] = InetAddress.getAllByName(addr); for (InetAddress resolvedAddress : resolvedAddresses) { InetAddress taddr = InetAddress.getByAddress(address.getHostString(), resolvedAddress.getAddress()); tmpList.add(new InetSocketAddress(taddr, address.getPort())); } } catch (UnknownHostException ex) { LOG.warn("No IP address found for server: {}", address, ex); } } Collections.shuffle(tmpList, sourceOfRandomness); return tmpList; }
在StaticHostProvider的成员变量中,有一个地方让我在意了一下:
private final List<InetSocketAddress> oldServers = new ArrayList<InetSocketAddress>( 5);
代码非常的简单。就是创建一个集合。但是它设置了大小。那么问题来了,我们一般创建ArrayList不会去设置大小。那么设置大小和不设置大小有什么区别呢?我们看看集合的构造方法。
public ArrayList(int initialCapacity) { super(); if (initialCapacity < 0) throw new IllegalArgumentException("Illegal Capacity: "+ initialCapacity); this.elementData = new Object[initialCapacity]; } /** * Constructs an empty list with an initial capacity of ten. */ public ArrayList() { this(10); }
上面的代码解释了区别。默认是10.然后就这么点不同吗?我在代码中发现了一段如何扩充大小的代码:
/** * The maximum size of array to allocate. * Some VMs reserve some header words in an array. * Attempts to allocate larger arrays may result in * OutOfMemoryError: Requested array size exceeds VM limit */ private static final int MAX_ARRAY_SIZE = Integer.MAX_VALUE - 8; /** * Increases the capacity to ensure that it can hold at least the * number of elements specified by the minimum capacity argument. * * @param minCapacity the desired minimum capacity */ private void grow(int minCapacity) { // overflow-conscious code int oldCapacity = elementData.length; int newCapacity = oldCapacity + (oldCapacity >> 1); if (newCapacity - minCapacity < 0) newCapacity = minCapacity; if (newCapacity - MAX_ARRAY_SIZE > 0) newCapacity = hugeCapacity(minCapacity); // minCapacity is usually close to size, so this is a win: elementData = Arrays.copyOf(elementData, newCapacity); } private static int hugeCapacity(int minCapacity) { if (minCapacity < 0) // overflow throw new OutOfMemoryError(); return (minCapacity > MAX_ARRAY_SIZE) ? Integer.MAX_VALUE : MAX_ARRAY_SIZE; }
结合注释我们可以理解: