逐行读取STDIN的最快方法? [英] Fastest way for line-by-line reading STDIN?

问题描述

我正在寻找最省时的方式逐行读取STDIN.

I'm looking for the most time-efficient way to read STDIN line-by-line.

第一行是要测试的条件数. 以下所有行都是条件(字符串)，最多100000个字符.

The first line is the number of conditions to test. All the following lines are conditions (strings) with a maximum of 100 000 characters.

我已经尝试了以下方法(加上4次90 000个字符的结果:

I have already tried the following (plus result for 4 times 90 000 characters:

• 扫描仪带有while循环(7255毫秒)

• Scanner with a while-loop (7255 ms)

Scanner sc = new Scanner(System.in);
int numberOfLines = Integer.parseInt(sc.nextLine());
long start = 0;
int i = 1;
while (i<=numberOfLines){
    start = System.currentTimeMillis();
    sc.nextLine();
    Debug.println((System.currentTimeMillis()-start) + "ms for scanner while");
    i++;
}

• 结果:
  • Results :
    1. 3228毫秒的扫描器时间
    2. 扫描仪时为2264毫秒
    3. 扫描仪时为1​​309毫秒
    4. 扫描仪同时运行454毫秒

  具有for循环功能的扫描仪(7078毫秒)

  Scanner with a for-loop (7078 ms)

  Scanner sc = new Scanner(System.in);
  int numberOfLines = Integer.parseInt(sc.nextLine());
  long start = 0;
  for (int i = 1; i<= numberOfLines;i++){
      start = System.currentTimeMillis();
      sc.nextLine();
      Debug.println((System.currentTimeMillis()-start) + "ms for scanner for");
      //i++;     
  }
  

  • 结果:
    • Results :
      1. 3168毫秒用于扫描器
      2. 2207毫秒用于扫描器
      3. 扫描仪的1236毫秒用于
      4. 467ms用于扫描器

    带for循环(7403 ms)的BufferedReader

    BufferedReader with a for-loop (7403 ms)

    try {
    BufferedReader br = new BufferedReader(new InputStreamReader(System.in));
    
    int numberOfLines = Integer.parseInt(br.readLine());
    long start = 0;
    for (int i = 0; i< numberOfLines;i++){
        start = System.currentTimeMillis();
        br.readLine();
        Debug.println((System.currentTimeMillis()-start) + "ms for bufferreader for");
        //i++;
    }
     } catch (Exception e) {
    System.err.println("Error:" + e.getMessage());
    

    }

    • 结果:
    1. 用于缓冲读取器的时间为3273ms
    2. bufferreader的时间为2330ms
    3. 用于缓冲区读取器的1293毫秒用于
    4. 缓冲读卡器的时间为507ms

    带while循环(7461 ms)的BufferedReader

    BufferedReader with a while-loop (7461 ms)

    try {
    BufferedReader br = new BufferedReader(new InputStreamReader(System.in));
    
    int numberOfLines = Integer.parseInt(br.readLine());
    int i=0;
    long start = 0;
    while(i< numberOfLines){
        start = System.currentTimeMillis();
        br.readLine();
        Debug.println((System.currentTimeMillis()-start) + "ms for bufferreader while");
        i++;
    }
     } catch (Exception e) {
    System.err.println("Error:" + e.getMessage());
    

    }

    • 结果:
    1. 3296毫秒用于bufferreader，
    2. 2358毫秒用于bufferreader，
    3. 缓冲读取器时为1307毫秒
    4. 500毫秒缓冲读取器时间

    在调试所花费的时间时，我注意到每次读取后所花费的时间都会减少. 是否可以限制初始化的字节(fe:如果您最多有100.000个字符，请将扫描器/缓冲读取器限制为仅初始化100 000个字符.读取后，将需要用接下来的100 000个字符重新填充自身)

    While debugging the time taken, i noticed that the time-taken decreases after each read. Is it possible to restrict the bytes that are initialized (f.e. : If you have a maximum of 100.000 chars, limit the scanner/bufferedreader to only initialize 100 000 chars. After a read it will need to refill itself with the next 100 000 chars)

    任何有关此事的想法都值得欢迎.

    Any ideas on this matter are more than welcome.

    添加了每种方案的代码以及每行读取所花费的时间.还将100.000更改为100000，以便于阅读.

    EDIT : Added the code for each scenario along with the time-taken per line read. Also changed 100.000 to 100 000 to read more easily.

    推荐答案

    在BufferedReader#readLine源内部查找.我看到了几个问题:

    Looked inside BufferedReader#readLine source. There're several problems I see:

    1. 它使用StringBuffer代替StringBuilder，这会产生同步开销.
    2. 似乎还有数据复制的开销-不能完全确定，最好检查一下.
    3. BufferedReader中的专用监视对象以及更多的同步开销.

    您可能会碰两件事:

    1. 编写自己的缓冲，这可以节省一些数据双重复制的时间.
    2. 编写自己的nextLine方法，该方法将使用StringBuilder并以简单的周期遍历源数据.

    这篇关于逐行读取STDIN的最快方法?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！