logo标签列表

在Java中解析日志文件

javaparsingloggingtokenize
3

大家好，

我有一个包含以下内容的日志文件。

Request from Centercord.
2010-12-14 12:42:13.724 [ 6796] ****************************
2010-12-14 12:42:13.724 [ 6796] 1111111111111111
2010-12-14 12:42:13.724 [ 6796]22222222222

Response from Centercord.
2010-12-14 12:42:21.802 [ 5960] 11111111111111
2010-12-14 12:42:21.802 [ 5960]  ffffffffffffffffffffffffffff
2010-12-14 12:42:21.802 [ 5960]  tttttttttttttttttttttttttttt

Request from Centercord.
2010-12-14 12:42:13.724 [ 6796] ****************************

我需要创建两个日志文件，一个用于存储所有请求细节，另一个用于存储所有响应细节。我该如何解析并准备这两个日志文件？
Log 1:

Request from Centercord.
2010-12-14 12:42:13.724 [ 6796] ****************************
2010-12-14 12:42:13.724 [ 6796] 1111111111111111
2010-12-14 12:42:13.724 [ 6796]22222222222

2010-12-14 12:42:13.724 [ 6796] ****************************

Log 2:

Response from Centercord.
2010-12-14 12:42:21.802 [ 5960] 11111111111111
2010-12-14 12:42:21.802 [ 5960]  ffffffffffffffffffffffffffff
2010-12-14 12:42:21.802 [ 5960]  tttttttttttttttttttttttttttt

敬礼， Kanagaraj

4个回答
10

这是我会做的方式：

import java.io.*;
import java.util.Scanner;

public class Test {

    public static void main(String[] args) {

        try {

            PrintWriter requests = new PrintWriter("requests.txt");
            PrintWriter responses = new PrintWriter("responses.txt");
            PrintWriter currentLog = null;

            Scanner s = new Scanner(new File("log.txt"));
            while (s.hasNextLine()) {
                String line = s.nextLine();
                if (line.startsWith("Request from"))
                    currentLog = requests;
                else if (line.startsWith("Response from"))
                    currentLog = responses;
                else if (currentLog != null)
                    currentLog.println(line);
            }

            requests.close();
            responses.close();
            s.close();
        } catch (IOException ioex) {
            // handle exception...
        }
    }
}

给定 log.txt

Request from Centercord.
2010-12-14 12:42:13.724 [ 6796] ****************************
2010-12-14 12:42:13.724 [ 6796] 1111111111111111
2010-12-14 12:42:13.724 [ 6796]22222222222

Response from Centercord.
2010-12-14 12:42:21.802 [ 5960] 11111111111111
2010-12-14 12:42:21.802 [ 5960]  ffffffffffffffffffffffffffff
2010-12-14 12:42:21.802 [ 5960]  tttttttttttttttttttttttttttt

Request from Centercord.
2010-12-14 12:42:13.724 [ 6796] ****************************

它生成了requests.txt

2010-12-14 12:42:13.724 [ 6796] ****************************
2010-12-14 12:42:13.724 [ 6796] 1111111111111111
2010-12-14 12:42:13.724 [ 6796]22222222222

2010-12-14 12:42:13.724 [ 6796] ****************************

...和responses.txt

2010-12-14 12:42:21.802 [ 5960] 11111111111111
2010-12-14 12:42:21.802 [ 5960]  ffffffffffffffffffffffffffff
2010-12-14 12:42:21.802 [ 5960]  tttttttttttttttttttttttttttt
0
StringBuilder requestBuilder = new StringBuilder("Request from Centercord.\n");
StringBuilder responseBuilder = new StringBuilder("Response from Centercord.\n");

boolean isResponse = false;
for (String line:getLinesFromLogFile()) {
  if (line.startsWith("Response")) {
      if(isResponse)
        responseBuilder.append("\n");
      isResponse = true;
  } else if (line.startsWith("Request")) {
      if(!isResponse)
        requestBuilder.append("\n");
      isResponse = false;
  } else {
      if (isResponse) {
         responseBuilder.append(line).append("\n");
      } else {
         requestBuilder.append(line).append("\n");
      }
  }
}

现在将两个 StringBuilder 的内容转储到目标文件中。
0
为了减少样板代码，请下载Google Guava并使用以下代码：
File inputFile = new File("input.log");
File outputLog1 = new File("log1.log");
File outputLog2 = new File("log2.log");

String match1 = "[ 6796]";
String match2 = "[ 5960]";

Files.append("Request from Centercord.\n", outputLog1, Charsets.UTF_8);
Files.append("Response from Centercord.\n", outputLog2, Charsets.UTF_8);

List<String> lines = Files.readLines(inputFile, Charsets.UTF_8);

for (String line : lines) {
    if(line.contains(match1)) Files.append(line + "\n", outputLog1, Charsets.UTF_8);
    if(line.contains(match2)) Files.append(line + "\n", outputLog2, Charsets.UTF_8);
}
3
不要这样做！所有行都会进入内存。对于大多数日志文件来说，这太多了！另外，你的静态类设计模式是一个不好的实践。 - Karussell
0
你可以使用正则表达式或 RandomAccessFiles 以你想要的方式获取数据。
网页内容由stack overflow 提供, 点击上面的
可以查看英文原文，
原文链接