Java的跨行比對－Jemmy Walker

　　在家加班會有點不甘願，但研究和工作有關的技能倒還可以。Java在Regex的跨行比對有點彆扭，用來解析非XML、具Block性質的配置檔會比較清晰。需求如下：

*
*
Definition for table TESTUTF8.T1
Record length: 112
Syskey: 0
Columns: 2
C1 64 50 0 0 0 1 0 50 50 50 0 0 0 0 1 0 1 2
C2 64 50 56 0 0 1 0 50 50 0 0 0 0 0 1 0 0 0
End of definition

* TEST T2
Definition for table TESTUTF8.T2
Record length: 112
Syskey: 0
Columns: 2
C1 64 50 0 0 0 1 0 50 50 50 0 0 0 0 1 0 1 2
C2 64 50 56 0 0 1 0 50 50 0 0 0 0 0 1 0 0 0
End of definition

　　如何在Java去group上述粗黑字的多行字串？逆向思考，跨行比對其實就是把它們當作一個單行，在Perl是用m(mulitline)參數，在Java使用Pattern物件要帶Patter.DOTALL參數，dot是一點，表示連換行字元都列入比對範圍，是故要group上述的需求的Sample code如下：

import java.io.FileInputStream;
import java.io.StringReader;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class TestRegexMultiLines {

    public static void main(String[] args) throws Exception {
        FileInputStream fis = new FileInputStream("D:/javaue.def");
        byte[] ba = new byte[fis.available()];
        fis.read(ba);
        String javadef = new String(ba);
        Pattern pat = Pattern.compile("Definition for table .*?End of definition",
            Pattern.DOTALL);
        Matcher mat = pat.matcher(javadef);
        while (mat.find()) {
            System.out.println("========================");
            System.out.println(mat.group());
        }
    }
}