摘要：緩沖輸入流從被稱為緩沖區(qū)的存儲(chǔ)器區(qū)域讀出數(shù)據(jù)僅當(dāng)緩沖區(qū)是空時(shí)，本地輸入才被調(diào)用。同樣，緩沖輸出流，將數(shù)據(jù)寫入到緩存區(qū)，只有當(dāng)緩沖區(qū)已滿才調(diào)用本機(jī)輸出。

：https://segmentfault.com/blog...

前陣子休息天日常在尋找項(xiàng)目里不好的代碼,看到了這樣的一段代碼:

    private Result sshSameExec(Session session, String cmd) {
        if (log.isDebugEnabled()) {
            log.debug("shell command: {}", cmd);
        }
        UserInfo ui = getUserInfo();
        session.setUserInfo(ui);
        int exitStatus = 0;
        StringBuilder builder = new StringBuilder();
        ChannelExec channel;
        InputStream in;
        InputStream err;
        try {
            session.connect(connectTimeout);
            channel = (ChannelExec) session.openChannel("exec");
            channel.setCommand(cmd);
            in = channel.getInputStream();
            err = channel.getErrStream();
            channel.connect();
        } catch (Exception e) {
            throw new CloudRuntimeException(e);
        }

        try {
            long lastRead = Long.MAX_VALUE;
            byte[] tmp = new byte[1024];
            while (true) {
                while (in.available() > 0 || err.available() > 0) {
                    int i = 0;
                    if (in.available() > 0) {
                        i = in.read(tmp, 0, 1024);
                    } else if (err.available() > 0) {
                        i = err.read(tmp, 0, 1024);
                    }
                    if (i < 0) {
                        break;
                    }
                    lastRead = System.currentTimeMillis();
                    builder.append(new String(tmp, 0, i));
                }
                if (channel.isClosed()) {
                    if (in.available() > 0) {
                        continue;
                    }
                    exitStatus = channel.getExitStatus();
                    break;
                }
                if (System.currentTimeMillis() - lastRead > exeTimeout) {
                    break;
                }
            }
        } catch (IOException e) {
            throw new CloudRuntimeException(e);
        } finally {
            channel.disconnect();
            session.disconnect();
        }

        if (0 != exitStatus) {
            return Result.createByError(ErrorData.builder()
                    .errorCode(ResultCode.EXECUTE_SSH_FAIL.getCode())
                    .detail(builder.toString())
                    .title(ResultCode.EXECUTE_SSH_FAIL.toString())
                    .build());
        } else {
            return Result.createBySuccess(builder.toString());
        }
    }

簡單解釋一下這段代碼——即通過ssh到一臺(tái)機(jī)器上,然后執(zhí)行一些命令.對命令輸出的東西,開了一個(gè)循環(huán),每一次讀一定的位置,然后以字節(jié)流的形式讀回來.

這段代碼有點(diǎn)丑,于是我聞到了學(xué)習(xí)的味道.

首先是對兩個(gè)Stream的消費(fèi),很顯然,在多核環(huán)境下,我們同時(shí)也只能夠消費(fèi)其中一個(gè)Stream.其次,這代碼太挫了,自己定義一個(gè)tmp,然后1024、1024這樣的去取出來.

在改良之前,我們先來回顧一下JavaIO的接口定義.

這里有同學(xué)可能問了,為啥叫它低抽象接口呢?因?yàn)樗x底層太近了,計(jì)算機(jī)本來就是處理二進(jìn)制的,而這兩個(gè)接口正是用來處理二進(jìn)制數(shù)據(jù)流的.

先簡單看一眼這兩個(gè)接口:

InputStream

**
 * This abstract class is the superclass of all classes representing
 * an input stream of bytes.
 *
 * 
 Applications that need to define a subclass of InputStream
 * must always provide a method that returns the next byte of input.
 *
 * @author  Arthur van Hoff
 * @see     java.io.BufferedInputStream
 * @see     java.io.ByteArrayInputStream
 * @see     java.io.DataInputStream
 * @see     java.io.FilterInputStream
 * @see     java.io.InputStream#read()
 * @see     java.io.OutputStream
 * @see     java.io.PushbackInputStream
 * @since   JDK1.0
 */
public abstract class InputStream implements Closeable {.....}

OutputStream

/**
 * This abstract class is the superclass of all classes representing
 * an output stream of bytes. An output stream accepts output bytes
 * and sends them to some sink.
 * 

 * Applications that need to define a subclass of
 * OutputStream must always provide at least a method
 * that writes one byte of output.
 *
 * @author  Arthur van Hoff
 * @see     java.io.BufferedOutputStream
 * @see     java.io.ByteArrayOutputStream
 * @see     java.io.DataOutputStream
 * @see     java.io.FilterOutputStream
 * @see     java.io.InputStream
 * @see     java.io.OutputStream#write(int)
 * @since   JDK1.0
 */
public abstract class OutputStream implements Closeable, Flushable {...}

我們可以發(fā)現(xiàn),它們都實(shí)現(xiàn)了Closeable的接口.因此大家在使用這些原生類時(shí),要注意在結(jié)束時(shí)調(diào)用Close方法哦.

這兩個(gè)接口的常用實(shí)現(xiàn)類有:
-?FileInputStream和FileOutputStream

DataInputStream和DataOutputStream

?ObjectInputStream和ObjectOutputStream

為啥說它是高級(jí)抽象接口呢?我們先來看看它們的注釋:

Writer

/**
 * Abstract class for writing to character streams.  The only methods that a
 * subclass must implement are write(char[], int, int), flush(), and close().
 * Most subclasses, however, will override some of the methods defined here in
 * order to provide higher efficiency, additional functionality, or both.
 *
 * @see Writer
 * @see   BufferedWriter
 * @see   CharArrayWriter
 * @see   FilterWriter
 * @see   OutputStreamWriter
 * @see     FileWriter
 * @see   PipedWriter
 * @see   PrintWriter
 * @see   StringWriter
 * @see Reader
 *
 * @author      Mark Reinhold
 * @since       JDK1.1
 */

public abstract class Writer implements Appendable, Closeable, Flushable {

Reader

/**
 * Abstract class for reading character streams.  The only methods that a
 * subclass must implement are read(char[], int, int) and close().  Most
 * subclasses, however, will override some of the methods defined here in order
 * to provide higher efficiency, additional functionality, or both.
 *
 *
 * @see BufferedReader
 * @see   LineNumberReader
 * @see CharArrayReader
 * @see InputStreamReader
 * @see   FileReader
 * @see FilterReader
 * @see   PushbackReader
 * @see PipedReader
 * @see StringReader
 * @see Writer
 *
 * @author      Mark Reinhold
 * @since       JDK1.1
 */

public abstract class Reader implements Readable, Closeable {

我們可以看到,這個(gè)抽象類是用來面向character的,也就是字符.字符的抽象等級(jí)必然比字節(jié)高,因?yàn)樽址拷蠈?即人類.

如果我們直接使用上述實(shí)現(xiàn)類去打開一個(gè)文件(如FileWriter 、FileReader 、FileInputStream 、FileOutputStream ),對其對象調(diào)用read、write、readLine等,每個(gè)請求都是由基礎(chǔ)OS直接處理的,這會(huì)使一個(gè)程序效率低得多——因?yàn)樗鼈兌紩?huì)引發(fā)磁盤訪問or網(wǎng)絡(luò)請求等.

為了減少這種開銷，Java 平臺(tái)實(shí)現(xiàn)緩沖 I/O 流。緩沖輸入流從被稱為緩沖區(qū)（buffer）的存儲(chǔ)器區(qū)域讀出數(shù)據(jù);僅當(dāng)緩沖區(qū)是空時(shí)，本地輸入 API 才被調(diào)用。同樣，緩沖輸出流，將數(shù)據(jù)寫入到緩存區(qū)，只有當(dāng)緩沖區(qū)已滿才調(diào)用本機(jī)輸出 API。

用于包裝非緩存流的緩沖流類有4個(gè)：BufferedInputStream和BufferedOutputStream·用于創(chuàng)建字節(jié)緩沖字節(jié)流,?BufferedReader和BufferedWriter`用于創(chuàng)建字符緩沖字節(jié)流.

之前,我們提到了這段代碼寫得搓的地方:

首先是對兩個(gè)Stream的消費(fèi),很顯然,在多核環(huán)境下,我們同時(shí)也只能夠消費(fèi)其中一個(gè)Stream.

其次,這代碼太挫了,自己定義一個(gè)tmp,然后1024、1024這樣的去取出來.

故此,我們可以考慮對每個(gè)Stream都進(jìn)行包裝,支持用線程去消費(fèi),其次我們可以用高級(jí)抽象分接口去適配Byte,然后去裝飾成Buffer.

接下來,我們來看一段ZStack里的工具類ShellUtils,為了節(jié)省篇幅,我們僅僅截出它在IDE里的
概覽:

run方法的核心:

        public ShellResult run() {
            StopWatch watch = new StopWatch();
            watch.start();
            try {
                if (withSudo) {
                    command = String.format("sudo %s", command);
                }

                ProcessBuilder pb = new ProcessBuilder(Arrays.asList("/bin/bash", "-c", command));
                if (baseDir == null) {
                    baseDir = System.getProperty("user.home");
                }
                pb.directory(new File(baseDir));

                process = pb.start();
                if (!suppressTraceLog && logger.isTraceEnabled()) {
                    logger.debug(String.format("exec shell command[%s]", command));
                }

                Writer stdout;
                int stdoutLog = stdoutLogStrategy();
                if (stdoutLog == LOG_TO_FILE) {
                    stdout = new BufferedWriter(new FileWriter(stdoutFile));
                } else if (stdoutLog == LOG_TO_SCREEN) {
                    stdout = new BufferedWriter(new OutputStreamWriter(System.out));
                } else {
                    stdout = new StringWriter();
                }

                Writer stderr;
                int stderrLog = stderrLogStrategy();
                if (stderrLog == LOG_TO_FILE) {
                    stderr = new BufferedWriter(new FileWriter(stderrFile));
                } else if (stderrLog == LOG_TO_SCREEN) {
                    stderr = new BufferedWriter(new OutputStreamWriter(System.err));
                } else {
                    stderr = new StringWriter();
                }

                StreamConsumer stdoutConsumer = new StreamConsumer(process.getInputStream(), new PrintWriter(stdout, true), stdoutLog != LOG_TO_FILE);
                StreamConsumer stderrConsumer = new StreamConsumer(process.getErrorStream(), new PrintWriter(stderr, true), stderrLog != LOG_TO_FILE);

                stderrConsumer.start();
                stdoutConsumer.start();
                process.waitFor();
                stderrConsumer.join(TimeUnit.SECONDS.toMillis(30));
                stdoutConsumer.join(TimeUnit.SECONDS.toMillis(30));

                ShellResult ret = new ShellResult();
                ret.setCommand(command);
                ret.setRetCode(process.exitValue());
                if (stderrLog == LOG_TO_STRING) {
                    ret.setStderr(stderr.toString());
                } else if (stderrLog == LOG_TO_FILE) {
                    stderr.close();
                }
                if (stdoutLog == LOG_TO_STRING) {
                    ret.setStdout(stdout.toString());
                } else if (stdoutLog == LOG_TO_FILE) {
                    stdout.close();
                }

                return ret;
            } catch (Exception e) {
                StringBuilder sb = new StringBuilder();
                sb.append("Shell command failed:
");
                sb.append(command);
                throw new ShellException(sb.toString(), e);
            } finally {
                if (process != null) {
                    process.destroy();
                }
                watch.stop();
                if (!suppressTraceLog && logger.isTraceEnabled()) {
                    logger.trace(String.format("shell command[%s] costs %sms to finish", command, watch.getTime()));
                }
            }
        }
    }

我們可以看到StreamConsumer這個(gè)類,我們來看一下它的代碼:

    private static class StreamConsumer extends Thread {
        final InputStream in;
        final PrintWriter out;
        final boolean flush;

        StreamConsumer(InputStream in, PrintWriter out, boolean flushEveryWrite) {
            this.in = in;
            this.out = out;
            flush = flushEveryWrite;
        }

        @Override
        public void run() {
            BufferedReader br = null;
            try {
                br = new BufferedReader(new InputStreamReader(in));
                String line;
                while ( (line = br.readLine()) != null) {
                    out.println(line);
                    if (flush) {
                        out.flush();
                    }
                }
            } catch (Exception e) {
                logger.warn(e.getMessage(), e);
            } finally {
                try {
                    if (br != null) {
                        br.close();
                    }
                } catch (IOException e) {
                    logger.warn(e.getMessage(), e);
                }
            }
        }
    }

這段代碼已經(jīng)達(dá)到了我們的理想狀態(tài):線程消費(fèi),高級(jí)抽象.

閑話不多說,先貼代碼為敬:

import java.io.InputStream
import java.io.InputStreamReader

class StreamGobbler(private val inputStream: InputStream, private var result: StringBuilder) : Runnable {

    override fun run() {
        InputStreamReader(inputStream).buffered().use {
            it.lines().forEach { r -> result.append(r) }
        }
    }
}

還是一樣熟悉的配方,我們逐行來解讀:

定義一個(gè)類,并且要求構(gòu)造函數(shù)必須傳入InputStream和一個(gè)StringBuilder.且實(shí)現(xiàn)了Runnable接口,這意味著它可以被線程消費(fèi).

覆寫run方法.我們可以看到InputStream被適配成了InputStreamReader,這意味著它可以輸出字符流了,然后我們使用了Kotlin的接口將其裝飾成了Buffer.

讀每一行buffer,并appned到result這個(gè)StringBuilder里去.

讀完就可以告辭了,close.(use會(huì)將其關(guān)閉)

先看一下上面的圖,我們都知道內(nèi)核態(tài)線程是由OS調(diào)度的,但當(dāng)一個(gè)線程拿到時(shí)間片時(shí),卻調(diào)到了阻塞IO,那么只能等在那邊,浪費(fèi)時(shí)間.

而協(xié)程則可以解決這個(gè)問題,當(dāng)一個(gè)Jobhang住的時(shí)候,可以去做別的事情,繞開阻塞.更好的利用時(shí)間片.

最后,我們來看一下成品代碼:

    override fun sshExecWithCoroutine(session: Session, cmd: String): SimpleResult {
        val ui = InnerUserInfo()
        session.userInfo = ui
        val exitStatus: Int
        var channel = ChannelExec()
        val inputBuilder = StringBuilder()
        val errorBuilder = StringBuilder()
        try {
            session.connect(connectTimeout)
            channel = session.openChannel("exec") as ChannelExec
            channel.setCommand(cmd)
            channel.connect()
            val inputStream = StreamGobbler(channel.inputStream, inputBuilder)
            val errStream = StreamGobbler(channel.errStream, errorBuilder)

            val customJob = GlobalScope.launch {
                customStream(inputStream, errStream)
            }

            while (!customJob.isCompleted) {
                // wait job be done
            }

            exitStatus = channel.exitStatus
        } catch (e: IOException) {
            throw java.lang.RuntimeException(e)
        } finally {
            if (channel.isConnected) {
                channel.disconnect()
            }
            if (session.isConnected) {
                session.disconnect()
            }
        }

        return if (0 != exitStatus) {
            return SimpleResult.createByError(ErrorData.Builder()
                    .errorCode(ResultCode.EXECUTE_SSH_FAIL.value)
                    .detail(errorBuilder.toString())
                    .title(ResultCode.EXECUTE_SSH_FAIL.toString())
                    .build())

        } else {
            SimpleResult.createBySuccess(inputBuilder.toString())
        }
    }


    private suspend fun customStream(inputStream: StreamGobbler, errorStream: StreamGobbler) {
        val inputDeferred = GlobalScope.async {
            inputStream.run()
        }
        val errorDeferred = GlobalScope.async {
            errorStream.run()
        }

        inputDeferred.join()
        errorDeferred.join()
    }