使用java进行纯文本解析

Question

I have an output from ssh like below. I want to parse the below output to an hashmap using java. any suggestions can be helpful...

Name        : mysql                        Relocations: (not relocatable)
Version     : 5.1.61                            Vendor: CentOS
Release     : 4.el6                         Build Date: Fri 22 Jun 2012 05:58:59 AM PDT
Install Date: Tue 13 Nov 2012 02:23:23 AM PST      Build Host: c6b10.bsys.dev.centos.org
URL         : http://www.mysql.com
Summary     : MySQL client programs and shared libraries

My output should be a hashmap like

Key Value

Name mysql

Relocations (not relocatable)

Version 5.1.61

Release 4.el6

Answer 1

A regular expression should do the trick here:

public static void main(String[] args) {
    StringBuilder sb = new StringBuilder();
    sb.append("Name        : mysql                        Relocations: (not relocatable)\n");
    sb.append("Version     : 5.1.61                            Vendor: CentOS\n");
    sb.append("Release     : 4.el6                         Build Date: Fri 22 Jun 2012 05:58:59 AM PDT\n");
    sb.append("Install Date: Tue 13 Nov 2012 02:23:23 AM PST      Build Host: c6b10.bsys.dev.centos.org\n");
    sb.append("URL         : http://www.mysql.com\n");
    sb.append("Summary     : MySQL client programs and shared libraries\n");

    Pattern p = Pattern.compile("([^\\r\\n:]+):\\s(.+?)(\\s{2,}|\\r\\n|\\r|\\n|$)");
    Matcher m = p.matcher(sb.toString());
    while(m.find()) {
        String key = m.group(1).trim();
        String value = m.group(2);

        System.out.println(key + " = \"" + value + "\"");
    }
}

which outputs:

Name = "mysql"
Relocations = "(not relocatable)"
Version = "5.1.61"
Vendor = "CentOS"
Release = "4.el6"
Build Date = "Fri 22 Jun 2012 05:58:59 AM PDT"
Install Date = "Tue 13 Nov 2012 02:23:23 AM PST"
Build Host = "c6b10.bsys.dev.centos.org"
URL = "http://www.mysql.com"
Summary = "MySQL client programs and shared libraries"

Answer 2

Try this regex as a starting point:

([a-zA-Z][a-zA-Z ]*): (.*?)(( {2,})|$)

First group should capture the key, second group the value. It assumes two things:

1) There are at least two spaces or the end of the line after a value. 2) There is never two spaces one after another within a value.

(It is important that these assumptions are really true. They are true in your example but you would need to verify that it is always true for your input.)

I tested it against your example above and it seems to work, try: http://regexpal.com/ (You need to enable checkbox "^$ match at line breaks" at the top to make it work)

If that regex is OK, use Pattern and Matcher from the Java API to build up your hashmap. Ah, and you should trim() your matched keys and values to get rid of the extra spaces at end.

Answer 3

try to create some regular expressions. Since many of the items are pretty defined well you can catch the starting and ending points using String.indexOf() . Then get the substring() , and develop the hashmap.